Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Handbook on Supply and Use

Tables and Input Output-Tables

with Extensions and Applications

ST/ESA/STAT/SER.F/

/Rev.1

Department of Economic and Social Affairs

Statistics Division

Studies in Methods

Handbook of National Accounting

Series F No.74, Rev.1

Handbook on Supply and Use Tables and

Input-Output Tables with Extensions and

Applications

Edited white cover version

United Nations

New York 2018

Department of Economic and Social Affairs

The Department of Economic and Social Affairs of the United Nations is a vital interface between

global policies in the economic, social and environmental spheres and national action. The

Department works in three main interlinked areas: (i) it compiles, generates and analyses a wide

range of economic, social and environmental data and information on which United Nations

Member States draw to review common problems and to take stock of policy options; (ii) it

facilitates the negotiations of Member States in many intergovernmental bodies on joint courses

of action to address ongoing or emerging global challenges; and (iii) it advises interested

Governments on the ways and means of translating policy frameworks developed in United

Nations conferences and summits into programmes at the country level and, through technical

assistance, helps build national capacities.

Note

The designations employed and the presentation of the material in the present publication

do not imply the expression of any opinion whatsoever on the part of the United Nations

concerning the legal status of any country or of its authorities, or the delimitations of its frontiers.

The term “country” as used in this report also refers, as appropriate, to territories or areas. The

designations of country groups are intended solely for statistical or analytical convenience and do

not necessarily express a judgement about the stage reached by a particular country, territory or

area in the development process. Mention of the names of firms and commercial products does not

imply endorsement by the United Nations. The symbols of United Nations documents are

composed of capital letters and numbers.

ST/ESA/STAT/SER.F/74/Rev.1

UNITED NATIONS PUBLICATION

Sales No.:

ISBN: 978-92-1-1

eISBN: 978-92-1-0

Preface and acknowledgements

The present Handbook on Supply and Use Tables and Input-Output Tables with Extensions

and Applications has been prepared as part of a series of handbooks on national accounting in

support of the implementation of the System of National Accounts 2008 (2008 SNA). The

objective of this Handbook is to provide step-by-step guidance for the compilation of supply and

use tables (SUTs) and input-output tables (IOTs) and an overview of the possible extensions of

SUTs and IOTs which increase their usefulness as analytical tools.

Preparation of the Handbook started as an update of the 1999 United Nations publication

entitled Handbook of National Accounting: Input-Output Table Compilation and Analysis,

with

the aim of incorporating changes in the underlying international economic accounting standards,

most notably the 2008 SNA, and classifications; extending the scope of the Handbook to provide

fuller coverage of SUTs; and providing practical compilation guidance for countries with advanced

and less advanced statistical systems. In this process, however, the Handbook has also evolved to

include an innovative approach to the compilation of SUTs and IOTs in the following three main

areas: first, the underlying use of an integrated approach to statistics; second, the use of a business

model for the compilation of SUTs and IOTs linking the various parts through the compilation

scheme known as the “H-Approach”; and, third, the mainstreaming of environmental

considerations.

The Handbook builds on the experience, practices and guidance available at national and

regional level, including the Eurostat Manual of Supply, Use and Input-Output Tables (Eurostat,

2008). It provides a consistent worked example of SUTs and IOTs, which runs throughout the

chapters (as far as practically possible) in order to facilitate understanding of the various

compilation steps. It also provides examples of best practices to illustrate certain aspects of the

compilation of SUTs, along with clear recommendations, principles and guidelines in order to

ensure best practice.

For the preparation and drafting of the Handbook, an editorial board was established in

May 2013, comprising 12 members and the United Nations Statistics Division. The editorial board

members were leading international experts, including members of the International Input-Output

Association, with decades of accumulated knowledge and experience from different regions and

from different institutions, such as national statistical offices, central banks, international

organizations and the academic community.

An editor (Sanjiv Mahajan, Office for National Statistics, United Kingdom) was appointed

to lead the work of the editorial board and coordinate the contributions of experts for the various

ST/ESA/STAT/SER.F/74, Sales No. E.99.XVII.9.

chapters. Initial drafts of the chapters were prepared by members of the editorial board, including

the editor. These were further refined and aligned by the editor in liaison with respective members

of the board and the United Nations Statistics Division into a coherent set of chapters. This was

achieved through many bilateral electronic communications between the editor and chapter

authors, a face-to-face meeting of all board members in New York in May 2014, and a final

editorial board review prior to a global consultation.

The Handbook is therefore the outcome of a collaborative team effort led by the editor in

liaison with the United Nations Statistics Division and the editorial board. This team comprises

the following:

• Sanjiv Mahajan, editor Office for National Statistics, United Kingdom

• Joerg Beutel Konstanz University of Applied Sciences, Germany

• Simon Guerrero Central Bank of Chile

• Satoshi Inomata Institute of Developing Economies, Japan External Trade

Organization

• Soren Larsen Statistics Denmark

• Brian Moyer Bureau of Economic Analysis, United States of America

• Isabelle Remond-Tiedrez European Commission, Eurostat

• José M. Rueda-Cantuche European Commission, Joint Research Centre

• Liv Hobbelstad Simpson Norway

• Bent Thage Denmark

• Catherine Van Rompaey Statistics Canada

• Piet Verbiest Statistics Netherlands

• Ilaria Di Matteo United Nations Statistics Division

The editorial board members contributed initial draft chapters and a detail review of all the

chapters in the various rounds of consultation. Substantive contributions on specific topics,

including initial draft chapters, were provided by the editorial board members as follows: Joerg

Beutel (transforming SUTs into IOTs, compiling physical SUTs (PSUTs) and environmentally

extended IOTs (EE-IOTs), extension of SUTs and IOTs and modelling applications of IOTs);

Simon Guerrero (examples of country practices); Satoshi Inomata (multi-country SUTs and IOTs);

Soren Larsen (compiling the use table); Brian Moyer (compiling the import use table and domestic

use table); José M. Rueda-Cantuche (transforming SUTs into IOTs and projecting SUTs and

IOTs); Liv Hobbelstad Simpson (guidance for countries with limited statistical resources and

examples of country practices); Bent Thage (classification of industries and products, compiling

the supply table, use table, valuation matrices, import use table and domestic use table, and

transforming SUTs into IOTs); Catherine Van Rompaey (regional SUTs); and Piet Verbiest

(compiling SUTs in volume terms and balancing). The editor also provided substantive

contributions to these topics, initial draft chapters and all other topics in the Handbook, and brought

iii

all the material together through numerous iterations with editorial board members reflecting many

changes and improvements.

The contributions by the editor and the members of the editorial board and their

commitment to the Handbook are very much acknowledged and appreciated. The following

specific contributions are also acknowledged: Joerg Beutel, in formatting and standardizing tables,

charts, boxes and figures throughout the Handbook; Ilaria Di Matteo, in reorganizing the chapters

and ensuring overall coherence and consistency of the Handbook; and Erwin Kolleritsch (Statistics

Austria), in kindly providing and checking much of the empirical data supporting the SUTs and

IOTs in parts two and three of the Handbook.

The Handbook also benefited from specific inputs provided by Issam Alsammak (Statistics

Canada), Gary Brown (Office for National Statistics, United Kingdom), Andrew Cadogan

(Australian Bureau of Statistics), Duncan Elliot (Office for National Statistics, United Kingdom),

Antonio F. Amores (European Commission Joint Research Centre), Ziad Ghanem (Statistics

Canada), Manfred Lenzen (University of Sydney, Australia), Bo Meng (Institute of Developing

Economies, Japan External Trade Organization), Louis de Mesnard (University of Bourgogne,

France), Carol Moylan and Tom Howells (Bureau of Economic Analysis, United States), Jan

Oosterhaven (University of Groningen, Netherlands), Ole Gravgard Pedersen (Statistics

Denmark), Xesús Pereira (University of Santiago de Compostela, Spain), Joao Rodrigues

(Technical University of Lisbon, Portugal), Jaroslav Sixta (Czech Statistical Office), Silke Stapel-

Weber (European Commission, Eurostat), Umed Temursho (European Commission Joint

Research Centre), Norihiko Yamano and Nadim Ahmad (OECD), and Herman Smith, Julian

Chow, Gulab Singh, Benson Sim and Alessandra Alfieri (United Nations Statistics Division).

Feedback was also received from participants at various meetings and conferences, most

notably the annual International Input-Output Association (2014, 2015 and 2016) and various

regional national accounts meetings. The Handbook has benefited greatly from the numerous

useful comments and suggestions made by national statistical offices, central banks, regional

commissions, academic associations and international organizations, and also by the

Intersecretariat Working Group on National Accounts during the global consultation in the period

August to October 2017.

The Handbook was prepared under the supervision of Herman Smith and the overall

responsibility of Ivo Havinga, both of the United Nations Statistics Division.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Contents

PREFACE AND ACKNOWLEDGEMENTS ........................................................................................................... I

CONTENTS ................................................................................................................................................................ V

ABBREVIATIONS ............................................................................................................................................... XVII

PART ONE ................................................................................................................................................................... 1

CHAPTER 1. INTRODUCTION ......................................................................................................................... 3

A. BACKGROUND ................................................................................................................................................... 3

B. USES OF SUTS AND IOTS .................................................................................................................................. 5

C. SYSTEM OF NATIONAL ACCOUNTS .................................................................................................................... 7

D. OBJECTIVES OF THIS HANDBOOK .................................................................................................................... 12

E. STRUCTURE OF THE HANDBOOK ...................................................................................................................... 15

PART TWO ............................................................................................................................................................... 21

CHAPTER 2. OVERVIEW OF THE SUPPLY AND USE TABLES AND INPUT-OUTPUT TABLES ... 23

A. INTRODUCTION ................................................................................................................................................ 23

B. OVERVIEW OF SUTS ....................................................................................................................................... 23

C. OVERVIEW OF IOTS ........................................................................................................................................ 35

D. STRUCTURE OF SUTS AND IOTS: BASIC ELEMENTS......................................................................................... 38

E. COMPILING SUTS AS AN INTEGRAL PART OF THE NATIONAL ACCOUNTS ......................................................... 57

CHAPTER 3. BUSINESS PROCESSES AND PRODUCTION STAGES ..................................................... 69

A. INTRODUCTION ................................................................................................................................................ 69

B. INSTITUTIONAL ARRANGEMENTS..................................................................................................................... 70

C. OVERVIEW OF THE GENERIC STATISTICAL BUSINESS PROCESS MODEL .......................................................... 71

D. OVERALL STRATEGY FOR THE COMPILATION OF SUTS AND IOTS ................................................................... 75

ANNEX A TO CHAPTER 3: EXAMPLES OF INSTITUTIONAL ARRANGEMENTS IN COUNTRIES .... 90

A. CENTRALIZED PRODUCTION OF ECONOMIC STATISTICS: CANADA ................................................................... 90

B. CENTRALIZED PRODUCTION OF ECONOMIC STATISTICS: NORWAY .................................................................. 91

C. CENTRALIZED PRODUCTION OF ECONOMIC STATISTICS: UNITED KINGDOM .................................................... 92

D. DECENTRALIZED PRODUCTION OF ECONOMIC STATISTICS: CHILE ................................................................... 93

E. DECENTRALIZED PRODUCTION OF ECONOMIC STATISTICS: UNITED STATES OF AMERICA ............................... 95

CHAPTER 4. SPECIFY NEEDS, DESIGN, BUILD AND COLLECT STAGE ........................................... 99

A. INTRODUCTION ................................................................................................................................................ 99

B. SPECIFY NEEDS, DESIGN AND BUILD PHASES ................................................................................................... 99

C. COLLECT PHASE ............................................................................................................................................ 120

PART THREE ......................................................................................................................................................... 129

CHAPTER 5. COMPILING THE SUPPLY TABLE .................................................................................... 131

A. INTRODUCTION .............................................................................................................................................. 131

B. STRUCTURE OF THE SUPPLY TABLE ............................................................................................................... 131

C. DOMESTIC OUTPUT ........................................................................................................................................ 135

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

D. IMPORTS OF GOODS AND SERVICES ................................................................................................................ 145

ANNEX A TO CHAPTER 5: SAMPLE QUESTIONNAIRE COLLECTING SALES OF GOODS AND

SERVICES, INVENTORIES OF GOODS AND TRADE-RELATED DATA .................................................. 153

CHAPTER 6. COMPILING THE USE TABLE ............................................................................................ 157

A. INTRODUCTION .............................................................................................................................................. 157

B. STRUCTURE OF THE USE TABLE ..................................................................................................................... 157

C. INTERMEDIATE CONSUMPTION PART OF THE USE TABLE ................................................................................ 162

D. GVA PART OF THE USE TABLE ....................................................................................................................... 170

E. FINAL CONSUMPTION EXPENDITURE PART OF THE USE TABLE ....................................................................... 173

F. GROSS CAPITAL FORMATION PART OF THE USE TABLE ................................................................................... 182

G. EXPORTS ....................................................................................................................................................... 196

ANNEX A TO CHAPTER 6: SAMPLE QUESTIONNAIRE COLLECTING PURCHASES OF GOODS AND

SERVICES FOR INTERMEDIATE CONSUMPTION ...................................................................................... 198

ANNEX B TO CHAPTER 6: IMPACT OF CAPITALIZING THE COSTS OF RESEARCH AND

DEVELOPMENT IN SUTS AND IOTS ............................................................................................................... 201

A. RESEARCH AND DEVELOPMENT AS FIXED CAPITAL FORMATION .................................................................... 201

B. IMPLICATIONS OF VALUATION OF OUTPUT AS SUM OF COSTS ......................................................................... 202

C. OWN-ACCOUNT RESEARCH AND DEVELOPMENT AS PRINCIPAL OR SECONDARY OUTPUT ............................... 203

D. BALANCING SUPPLY AND USE OF RESEARCH AND DEVELOPMENT SERVICES .................................................. 205

CHAPTER 7. COMPILING THE VALUATION MATRICES .................................................................... 207

A. INTRODUCTION .............................................................................................................................................. 207

B. VALUATION OF PRODUCT FLOWS ................................................................................................................... 207

C. TRADE MARGINS ........................................................................................................................................... 216

D. TRANSPORT MARGINS ................................................................................................................................... 229

E. TAXES ON PRODUCTS AND SUBSIDIES ON PRODUCTS ..................................................................................... 236

ANNEX A TO CHAPTER 7. EXAMPLE FOR DERIVING TRADE MARGINS IN SUTS BASED ON

SURVEY DATA ...................................................................................................................................................... 241

A. SUPPLY TABLE ............................................................................................................................................... 243

B. USE TABLE .................................................................................................................................................... 245

CHAPTER 8. COMPILING THE IMPORTS USE TABLE AND DOMESTIC USE TABLE ................. 249

A. INTRODUCTION .............................................................................................................................................. 249

B. STRUCTURE OF THE IMPORTS USE TABLE AND DOMESTIC USE TABLE ............................................................ 250

C. COMPILATION OF THE IMPORTS USE TABLE ................................................................................................... 255

D. SPECIFIC ISSUES IN THE COMPILATION OF THE IMPORTS USE TABLE .............................................................. 259

E. ENHANCEMENTS TO THE IMPORTS USE TABLE FOR ANALYTICAL USES .......................................................... 268

CHAPTER 9. COMPILING SUTS IN VOLUME TERMS .......................................................................... 271

A. INTRODUCTION .............................................................................................................................................. 271

B. RECOGNITION OF ALTERNATIVE APPROACHES............................................................................................... 272

C. OVERVIEW OF THE STEPS IN THE H-APPROACH WITH A FOCUS ON VOLUMES ................................................ 273

D. PRICE AND VOLUME INDICATORS IN THEORY................................................................................................. 284

E. PRICE AND VOLUME INDICATORS IN PRACTICE .............................................................................................. 285

F. INPUT-OUTPUT TABLES IN VOLUME TERMS .................................................................................................... 301

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

vii

CHAPTER 10. LINKING THE SUPPLY AND USE TABLES TO THE INSTITUTIONAL SECTOR

ACCOUNTS 303

A. INTRODUCTION .............................................................................................................................................. 303

B. INSTITUTIONAL SECTORS AND SUBSECTORS .................................................................................................. 304

C. TABLE LINKING SUTS AND INSTITUTIONAL SECTOR ACCOUNTS .................................................................... 307

D. COMPILATION METHODS ............................................................................................................................... 315

CHAPTER 11. BALANCING THE SUPPLY AND USE TABLES ............................................................... 319

A. INTRODUCTION .............................................................................................................................................. 319

B. OVERVIEW OF THE SYSTEM AND BASIC IDENTITIES ....................................................................................... 320

C. BALANCING ................................................................................................................................................... 324

D. STEP-BY-STEP PROCEDURE FOR SIMULTANEOUS BALANCING ........................................................................ 332

E. ALTERNATIVE BALANCING METHODS ............................................................................................................ 336

F. EXTENDING THE BALANCING OF SUTS TO INCLUDE INSTITUTIONAL SECTOR ACCOUNTS, IOTS, PSUTS AND EE-

IOT

S 338

G. PRACTICAL ASPECTS OF BALANCING ............................................................................................................. 342

ANNEX A TO CHAPTER 11. BALANCING SUPPLY AND USE TABLES ................................................... 350

CHAPTER 12. TRANSFORMING THE SUPPLY AND USE TABLES INTO INPUT-OUTPUT TABLES

369

A. INTRODUCTION .............................................................................................................................................. 369

B. OVERVIEW OF THE RELATIONSHIP BETWEEN IOTS AND SUTS ...................................................................... 369

C. CONVERSION OF SUTS TO IOTS .................................................................................................................... 374

D. INPUT-OUTPUT FRAMEWORK ......................................................................................................................... 379

E. EMPIRICAL APPLICATION OF THE TRANSFORMATION MODELS ....................................................................... 400

ANNEX A TO CHAPTER 12. MATHEMATICAL DERIVATION OF DIFFERENT IOTS ......................... 411

A. PRODUCT-BY-PRODUCT IOTS AND INDUSTRY-BY-INDUSTRY IOTS ............................................................... 411

B. PRODUCT-BY-PRODUCT IOTS ........................................................................................................................ 412

C. INDUSTRY-BY-INDUSTRY IOTS ..................................................................................................................... 414

D. USE OF A HYBRID TECHNOLOGY ASSUMPTION FOR PRODUCT-BY-PRODUCT IOTS ......................................... 416

ANNEX B TO CHAPTER 12. CLASSICAL CAUSES AND TREATMENT OF NEGATIVE CELL ENTRIES

IN THE PRODUCT TECHNOLOGY ................................................................................................................... 419

A. CLASSICAL CAUSES OF NEGATIVE ELEMENTS IN THE PRODUCT TECHNOLOGY ............................................... 419

B. OVERALL STRATEGY FOR REMOVING NEGATIVES .......................................................................................... 421

C. SPECIFIC APPROACHES TO DEALING WITH NEGATIVES ................................................................................... 421

ANNEX C TO CHAPTER 12. EXAMPLES OF REVIEWS OF APPROACHES TO THE TREATMENT OF

SECONDARY PRODUCTS ................................................................................................................................... 427

CHAPTER 13. COMPILING PHYSICAL SUPPLY AND USE TABLES AND ENVIRONMENTALLY

EXTENDED INPUT-OUTPUT TABLES ............................................................................................................. 429

A. INTRODUCTION .............................................................................................................................................. 429

B. OVERVIEW OF PSUTS ................................................................................................................................... 430

C. COMPILATION OF PSUTS .............................................................................................................................. 442

D. ENVIRONMENTAL EXTENDED IOTS ............................................................................................................... 449

E. COMPILATION OF EE-IOTS ........................................................................................................................... 453

F. COUNTRY EXAMPLES .................................................................................................................................... 454

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

viii

CHAPTER 14. SUPPLY AND USE TABLES AND QUARTERLY NATIONAL ACCOUNTS ................ 467

A. INTRODUCTION .............................................................................................................................................. 467

B. QUARTERLY NATIONAL ACCOUNTS ............................................................................................................... 468

C. SUTS AND QUARTERLY NATIONAL ACCOUNTS .............................................................................................. 474

CHAPTER 15. DISSEMINATING SUPPLY, USE AND INPUT-OUTPUT TABLES ................................ 487

A. INTRODUCTION .............................................................................................................................................. 487

B. USER IDENTIFICATION ................................................................................................................................... 487

C. DISSEMINATION STRATEGY ........................................................................................................................... 488

D. COMMUNICATIONS OF SUTS AND IOTS WITH USERS .................................................................................... 493

E. DISSEMINATION FORMAT FOR SUTS AND IOTS............................................................................................. 494

F. STATISTICAL DATA AND METADATA EXCHANGE INITIATIVE ........................................................................ 497

PART FOUR ............................................................................................................................................................ 499

CHAPTER 16. REGIONAL SUPPLY AND USE TABLES ........................................................................... 501

A. INTRODUCTION .............................................................................................................................................. 501

B. ISSUES ARISING IN AND METHODS FOR THE COMPILATION OF REGIONAL SUTS AND IOTS ............................ 501

C. EXAMPLE OF BOTTOM-UP METHODS FOR REGIONAL SUTS: CANADIAN EXPERIENCE .................................... 504

CHAPTER 17. MULTI-COUNTRY SUPPLY AND USE TABLES AND INPUT-OUTPUT TABLES ..... 521

A. INTRODUCTION .............................................................................................................................................. 521

B. OVERVIEW OF MULTI-COUNTRY SUTS AND IOTS AND MAIN COMPILATION ISSUES ...................................... 522

C. COMPILATION PROCEDURE ............................................................................................................................ 530

D. MULTI-COUNTRY INPUT-OUTPUT DATABASE INITIATIVES ............................................................................. 541

E. WAY AHEAD .................................................................................................................................................. 543

CHAPTER 18. PROJECTING SUPPLY, USE AND INPUT-OUTPUT TABLES ....................................... 549

A. INTRODUCTION .............................................................................................................................................. 549

B. SITUATIONS NEEDING PROJECTION METHODS ................................................................................................ 549

C. GENERAL APPROACHES TO PROJECTION FROM A HISTORICAL PERSPECTIVE .................................................. 551

D. NUMERICAL EXAMPLES ................................................................................................................................. 567

E. CRITERIA TO CONSIDER WHEN CHOOSING A METHOD .................................................................................... 580

CHAPTER 19. EXTENSIONS OF SUTS AND IOTS AS PART OF SATELLITE SYSTEMS .................. 583

A. INTRODUCTION .............................................................................................................................................. 583

B. OVERVIEW OF POSSIBLE EXTENSIONS ............................................................................................................ 584

C. SOCIAL ACCOUNTING MATRIX ....................................................................................................................... 591

D. EXTENDED INPUT-OUTPUT TABLES ................................................................................................................ 597

E. OTHER EXAMPLES OF SATELLITE SYSTEMS .................................................................................................... 601

CHAPTER 20. MODELLING APPLICATIONS OF IOTS ........................................................................... 603

A. INTRODUCTION .............................................................................................................................................. 603

B. NUMERICAL EXAMPLE OF IOTS AS A STARTING POINT .................................................................................. 604

C. DISTINCTION BETWEEN PRICE, VOLUME, QUANTITY, QUALITY AND PHYSICAL UNITS .................................... 605

D. INPUT COEFFICIENTS ..................................................................................................................................... 609

E. OUTPUT COEFFICIENTS .................................................................................................................................. 611

F. QUANTITY MODEL OF INPUT-OUTPUT ANALYSIS ........................................................................................... 612

G. PRICE MODEL OF INPUT-OUTPUT ANALYSIS ................................................................................................... 616

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

H. INPUT-OUTPUT MODELS WITH INPUT AND OUTPUT COEFFICIENTS ................................................................. 622

I. CENTRAL MODEL OF INPUT-OUTPUT ANALYSIS ............................................................................................. 623

J. INDICATORS .................................................................................................................................................. 626

K. MULTIPLIERS................................................................................................................................................. 629

L. INTER-INDUSTRIAL LINKAGE ANALYSIS ........................................................................................................ 636

CHAPTER 21. EXAMPLES OF COMPILATION PRACTICES .................................................................. 641

A. INTRODUCTION .............................................................................................................................................. 641

B. BASIC CONSIDERATIONS FOR THE COMPILATION OF NATIONAL ACCOUNTS AND SUTS ................................. 642

C. EFFECT ON GDP OF INTEGRATING SUTS IN THE NATIONAL ACCOUNTS FOR MALAWI................................... 655

D. DEVELOPMENT OF THE APPLICATION OF THE INPUT-OUTPUT FRAMEWORK IN THE CZECH REPUBLIC ............ 659

E. CONTINUAL CHANGE, DEVELOPMENT AND IMPROVEMENT IN CHILE ............................................................. 663

REFERENCES ........................................................................................................................................................ 675

ADDITIONAL READING ..................................................................................................................................... 695

Box 1.1 Evolution of the SUTs and IOTs within the national accounts ....................................... 13

Box 2.1 Numerical example of the SUTs system ......................................................................... 28

Box 2.2 Numerical example showing a use table split between consumption of domestic

production and imports ................................................................................................................. 29

Box 2.3 SUTs and product-by-product IOTs ................................................................................ 37

Box 2.4 SUTs and industry-by-industry IOTs .............................................................................. 38

Box 2.5 Three approaches to measuring GDP .............................................................................. 41

Box 2.6 Other classifications of products ..................................................................................... 46

Box 2.7 SNA recommendations on partitioning of vertically and horizontally integrated enterprises

....................................................................................................................................................... 53

Box 2.8 Overview of the valuation in SUTs and IOTs ................................................................. 57

Box 2.9 Calculation of output for market and non-market producers .......................................... 60

Box 2.10 Example of derivation of GDP from balanced SUTs .................................................... 64

Box 3.1 Examples of the main recommendations, principles and guidelines provided in this

Handbook ...................................................................................................................................... 85

Box 4.1 Example of in-house custom-built software: Statistics Netherlands ............................. 100

Box 4.2 ERETES ........................................................................................................................ 101

Box 4.3 Data sources generally used .......................................................................................... 121

Box 5.1 Redefinitions ................................................................................................................. 144

Box 5.2 Consistency issues with the CIF/FOB adjustment ........................................................ 150

Box 6.1 Example of a calculation of the values of an input column .......................................... 168

Box 6.2 Classification of individual consumption according to purpose ................................... 174

Box 6.3: Non-durable, semi-durable and durable goods ............................................................ 176

Box 6.4 Classification of the purposes of NPISHs ..................................................................... 179

Box 6.5 Classification of functions of government .................................................................... 181

Box 6.6 Gross fixed capital formation by type of asset .............................................................. 183

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 7.1 Compilation process for trade margins ......................................................................... 221

Box 7.2 Examples of transport costs which do not form transport margins ............................... 231

Box 7.3 Options to consider where no data exists on transport margins .................................... 234

Box 8.1 Standard services components of BPM 6 ...................................................................... 257

Box 9.1 Treatment of newly introduced and disappearing taxes and subsidies ......................... 297

Box 10.1 Compilation methods used to link SUTs to the institutional sector accounts ............. 315

Box 11.1 Example of discrepancies balanced in current prices and in volume terms ................ 331

Box 11.2 Example of simultaneous balancing comparing volume indices ................................ 331

Box 11.3 Methods used for automated balancing SUTs ............................................................. 345

Box 12.1 Clarification of IOTs terminology ............................................................................... 373

Box 12.2 Input-output framework for domestic output and imports .......................................... 381

Box 12.3 Basic transformations of SUTs to IOTs ...................................................................... 383

Box 13.1 Selected reference material ......................................................................................... 444

Box 15.1 Fundamental Principles of Official Statistics .............................................................. 488

Box 15.2 Reference metadata in the SDMX metadata structure for SUTs and IOTs ................ 493

Box 17.1 Background papers of each database initiative ........................................................... 541

Box 17.2 Overview of the main features of the various databases ............................................. 543

Box 18.1 Methods for projection of SUTs and IOTs.................................................................. 554

Box 18.2 SUTs and IOTs for Austria, 2005 and 2006................................................................ 568

Box 18.3 Results using the GRAS Method ................................................................................ 570

Box 18.4 Flow diagram of the GRAS method ............................................................................ 571

Box 18.5 Results using the SUT-RAS method ........................................................................... 574

Box 18.6 Flow diagram of the SUT-RAS method ...................................................................... 575

Box 18.7 Results using the SUT-Euro method ........................................................................... 578

Box 18.8 Flow diagram of the SUT-EURO method................................................................... 580

Box 19.1 Measurement performance and social progress: overview of Stiglitz-Sen-Fitoussi

Commission 2009 report ............................................................................................................. 588

Box 20.1 Quantities, prices, values and volumes in IOTs .......................................................... 607

Box 20.2 Quantity input-output model ....................................................................................... 620

Box 20.3 Price input-output model ............................................................................................. 621

Box 20.4 Multipliers in the input-output model ......................................................................... 631

Box 21.1 Material product system and Phare projects ............................................................... 660

Figure 1.1 Overview of the links between SUTs and the SNA framework .................................... 9

Figure 2.1: Graphical overview of supply and use tables ............................................................. 27

Figure 2.2 Schematic overview of the compilation of SUTs and IOTs: H-Approach .................. 30

Figure 2.3 System of national accounts in matrix form ................................................................ 41

Figure 2.4 Overview of SUTs and IOTs as part of the SNA compilation .................................... 59

Figure 3.1 Phases of the GSBPM ................................................................................................. 72

Figure 3.2 Simplified business processing model for compiling SUTs, IOTs, and PSUTs ......... 74

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 3.3 Structure of the SUTs and the links covered in this Handbook .................................. 77

Figure 3.4 Compilation of SUTs and IOTs in current prices and in volume terms ...................... 79

Figure 3.5 Evolution of compiling SUTs and IOTs in the first three years .................................. 80

Figure 3.6 First year of compilation ............................................................................................. 82

Figure 3.7 Second year of compilation ......................................................................................... 84

Figure 4.1 Overview of SUTs and IOTs as part of the SNA compilation .................................. 123

Figure 5.1 Link between valuation matrices in the supply table and the use table..................... 135

Figure 6.1 Three-dimensional view of SUTs .............................................................................. 162

Figure 7.1 Schematic representation of the valuation matrices in the SUTs .............................. 210

Figure 7.2 Alternative distribution channels of goods ................................................................ 227

Figure 9.1 Overview of the compilation schematic layout linking SUTs in current prices and in

volume terms ............................................................................................................................... 274

Figure 9.2 Link between SUTs in current prices and in volume terms ...................................... 278

Figure 10.1 Links between the industry accounts and the institutional sector accounts ............ 304

Figure 10.2 Link between the SUTs and institutional sector accounts ....................................... 309

Figure 11.1 Simplified SUTs system .......................................................................................... 321

Figure 11.2 Six-pack ................................................................................................................... 325

Figure 11.3 Overview of the SUTs balancing framework for simultaneous balancing.............. 333

Figure 11.4 Sources of feedback loops emanating from the balancing process ......................... 342

Figure 12.1 Transformation of SUTs into IOTs ......................................................................... 372

Figure 12.2 Basic transformation models ................................................................................... 378

Figure 13.1 Physical flows of natural inputs, products and residuals ......................................... 431

Figure 13.2 Overview of the compilation system for PSUTs ..................................................... 446

Figure 13.3 Key feedback loops in producing and balancing the PSUTs and environmental

extended IOTs ............................................................................................................................. 449

Figure 13.4 Danish SUTs framework extended with physical flows ......................................... 456

Figure 13.5 From source data to PSUTs ..................................................................................... 458

Figure 14.1 Quarterly GDP production (output) aggregate: data availability and estimation in the

United Kingdom.......................................................................................................................... 470

Figure 14.2 Quarterly GDP expenditure components: data availability and estimation in the United

Kingdom ..................................................................................................................................... 471

Figure 15.1 Release calendar covering SUTs, IOTs and national accounts: Statistics Denmark

..................................................................................................................................................... 490

Figure 15.2 Measuring United Kingdom GDP and SUTs: revision policy ................................ 491

Figure 17.1 Schematic representation of multi-country SUTs (three-country case) .................. 523

Figure 17.2 Schematic representation of multi-country IOTs (three country case) ................... 524

Figure 17.3 System of multi-country SUTs and its conceptual correspondence to a national SUTs

framework ................................................................................................................................... 532

Figure 17.4 Splitting the import matrix by country of origin ..................................................... 535

Figure 17.5 Converting valuation scheme .................................................................................. 536

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

xii

Figure 17.6 Formation of the export vector to rest of the world................................................. 538

Figure 17.7 Transformation to multi-country IOTs .................................................................... 540

Figure 21.1: Illustration of a database for the product-flow method used in smaller countries . 649

Figure 21.2 Supply and use table ................................................................................................ 670

Table 2.1: Simplified structure of the supply table ....................................................................... 24

Table 2.2: Simplified structure of the use table ............................................................................ 24

Table 2.3: Supply and use tables framework ................................................................................ 25

Table 2.4 Schematic view of the physical supply table ................................................................ 34

Table 2.5 Schematic view of the physical use table ..................................................................... 34

Table 2.6 Simplified IOT (product by product) ............................................................................ 36

Table 2.7 Links between the use table and functional classifications .......................................... 51

Table 2.8 Simplified table linking the SUTs to the institutional sector accounts ......................... 65

Table 4.1 Examples of the size of published and internal working level SUTs and IOTs ......... 107

Table 5.1 Numerical example of a supply table at basic prices .................................................. 132

Table 5.2 Supply table at basic prices, including a transformation into purchasers’ prices ....... 133

Table 5.3 Data adjustment for external trade of goods and services .......................................... 148

Table 5.4 CIF and FOB adjustment row ..................................................................................... 149

Table 6.1 Use table at purchasers’ prices.................................................................................... 158

Table 6.2 Intermediate consumption of selected inputs into “Manufacture of rubber and plastic

products” ..................................................................................................................................... 169

Table 6.3 Sample product balance for "Gelatine and gelatine derivatives"................................ 170

Table 6.4 Categories of final consumption expenditure ............................................................. 173

Table 6.5 Table linking final expenditures by purpose (COICOP) and product (CPC) ............. 175

Table 6.6 Final consumption expenditure of households (by COICOP headings) ..................... 176

Table 6.7 Table linking final consumption expenditures of NPISHs by purpose (COPNI) and by

product (CPC) ............................................................................................................................. 180

Table 6.8 Table linking final consumption expenditure of general government by COFOG and

CPC ............................................................................................................................................. 182

Table 6.9 Categories of gross capital formation ......................................................................... 183

Table 6.10 Table linking gross fixed capital formation by industries, assets and products ....... 185

Table 6.11 Gross fixed capital formation by investing industry ................................................. 187

Table 6.12 Table linking change in inventories industries, assets and products ........................ 192

Table 7.1 Supply table at basic prices, including a transformation into purchasers’ prices ....... 212

Table 7.2 Use table at purchasers’ prices.................................................................................... 213

Table 7.3 Use-side valuation matrices ........................................................................................ 214

Table 7.4 Use table at basic prices .............................................................................................. 216

Table 7.5 Trade turnover and trade margins for wholesale and retail trade margins ................. 224

Table 8.1 Structure of the imports use table ............................................................................... 251

Table 8.2 Numerical example of the imports use table .............................................................. 251

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

xiii

Table 8.3 Structure of the domestic use table ............................................................................. 252

Table 8.4 Numerical example of a domestic use table ............................................................... 252

Table 8.5 Example of an input table for imports at basic prices ................................................. 253

Table 8.6 Input-output table for domestic output at basic prices ................................................ 254

Table 8.7 Input-output table for domestic output at basic prices, net exports with adjustment items

..................................................................................................................................................... 255

Table 8.8 Processing within the country ..................................................................................... 261

Table 8.9 Goods sent abroad for processing ............................................................................... 262

Table 8.10 Industry 1 – Alternative input structures .................................................................. 264

Table 9.1 Supply table in current prices and in volume terms .................................................... 280

Table 9.2 Use table in current prices and in volume terms ......................................................... 281

Table 9.3 Gross domestic product in current prices and in volume terms .................................. 283

Table 10.1 Summary of institutional sectors and subsectors ...................................................... 306

Table 10.2 Main features of the SUT approach and the institutional sector approach ............... 307

Table 10.3 Numerical example showing the table linking the SUTs and institutional sector

accounts....................................................................................................................................... 310

Table 10.4 Goods and services for the whole economy ............................................................. 312

Table 10.5 Production account for the whole economy ............................................................. 313

Table 10.6 Link between GDP and industry GVA ..................................................................... 313

Table 10.7 Generation of income account for the whole economy ............................................ 314

Table 12.1 Product-by-product IOT at basic prices .................................................................... 370

Table 12.2 Numerical example of rectangular SUTs for the transformation ............................. 375

Table 12.3 Numerical example of square SUTs for the transformation ..................................... 376

Table 12.4 Transformation matrix for the product technology assumption ............................... 386

Table 12.5 Product-by-product IOTs based on product technology ........................................... 387

Table 12.6 Transformation matrix for industry technology assumption .................................... 388

Table 12.7 Product-by-product IOTs based on industry technology .......................................... 388

Table 12.8 Matrix for hybrid technology .................................................................................... 389

Table 12.9 Transformation matrix for hybrid technology assumption ....................................... 390

Table 12.10 IOTs based on the hybrid technology assumption .................................................. 390

Table 12.11 Transformation matrix for the fixed industry sales structure assumption ............. 391

Table 12.12 IOTs based on the fixed industry sales structure assumption ................................. 392

Table 12.13 Transformation matrix for the fixed product sales structure assumption for rectangular

SUTs ........................................................................................................................................... 394

Table 12.14 IOTs based on the fixed product sales structure assumption derived from rectangular

SUTs ........................................................................................................................................... 394

Table 12.15 Transformation matrix for the fixed product sales structure assumption for square

SUTs ........................................................................................................................................... 395

Table 12.16 IOTs based on the fixed product sales structure assumption for square SUTs....... 396

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

xiv

Table 12.17 Absolute deviation of IOTs based on rectangular SUTs less IOTs based on square

SUTs for Model D ...................................................................................................................... 397

Table 12.18 Alternative presentations of product-by-product IOTs ........................................... 399

Table 12.19 Empirical example of product-by-product IOTs .................................................... 402

Table 12.20 Empirical example of industry-by-industry IOTs ................................................... 404

Table 13.1 General PSUT ........................................................................................................... 432

Table 13.2 Classes of natural input ............................................................................................. 434

Table 13.3 Typical components for groups of residuals ............................................................. 436

Table 13.4 List of individual components of SEEA physical PSUTs ........................................ 443

Table 13.5 Common national data sources and links to SEEA component accounts ................. 443

Table 13.6 Single region IOT with environmental data ............................................................. 450

Table 13.7 Single region IOT in hybrid units ............................................................................. 452

Table 13.8 Industry-by -industry IOTs (upper block of Table 13.6) .......................................... 453

Table 13.9 Environmental data by industry (lower block of Table 13.6) ................................... 453

Table 13.10 PSUTs in Denmark ................................................................................................. 460

Table 13.11 SUTs for the Netherlands, 2010 ............................................................................. 462

Table 13.12 PSUTs for the Netherlands, 2010 ........................................................................... 463

Table 14.1 Balancing supply and use of products ...................................................................... 477

Table 16.1 SUTs framework for interregional SUTs.................................................................. 507

Table 16.2 Interregional and international trade flows by province and territory, 2010 ........... 510

Table 17.1 Adjustment targets for national tables of selected countries in the Asian international

input-output table for the year 2000............................................................................................ 530

Table 18.1 Categorization of methods ........................................................................................ 561

Table 19.1 Structure of a social accounting matrix .................................................................... 594

Table 19.2 Numerical example of a social accounting matrix .................................................... 595

Table 19.3 Extended IOT with satellite systems ........................................................................ 599

Table 20.1 IOT at basic prices .................................................................................................... 605

Table 20.2 Input coefficients of IOTs ......................................................................................... 610

Table 20.3 Output coefficients of IOTs ...................................................................................... 612

Table 20.4 Input coefficients for domestic intermediate consumption ....................................... 613

Table 20.5 Leontief matrix ......................................................................................................... 614

Table 20.6 Leontief inverse ........................................................................................................ 615

Table 20.7 Quantity input-output model based on monetary data .............................................. 616

Table 20.8 Price input-output model based on monetary data .................................................... 619

Table 20.9 Emission model......................................................................................................... 625

Table 20.10 Input indicators for production activities per unit of output ................................... 628

Table 20.11 Output multipliers (Leontief inverse) ..................................................................... 629

Table 20.12 Multipliers for products .......................................................................................... 632

Table 20.13 Input content of final use by category .................................................................... 635

Table 20.14 Backward linkages .................................................................................................. 637

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Table 20.15 Forward linkages..................................................................................................... 637

Table 20.16 Forward and backward linkages ............................................................................. 638

Table 20.17 Normalized forward and backward linkages .......................................................... 638

Table 21.1 Historical benchmark exercises ................................................................................ 664

Table 21.2 Annual business surveys ........................................................................................... 666

Table 21.3 Administrative records .............................................................................................. 667

Table 21.4 SUTs for tobacco products, year 2008, current prices .............................................. 672

Table 21.5 SUT for cleaning and toiletry products, year 2008, current prices ........................... 672

Table 21.6 IOTs for domestic output at basic prices, 2008 ........................................................ 673

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

xvi

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

xvii

Abbreviations

AIIOT

Asian International Input-Output Tables

ANZSIC

Australian and New Zealand Standard Industrial Classification

ARIMA

autoregressive integrated moving average

BEC

classification by broad economic categories

BPM

Balance of Payments and International Investment Position Manual

BTDIxE

Bilateral Trade Database by Industry and End-Use

methane

CIF

cost, insurance and freight

carbon dioxide

COFOG

Classification of the Functions of Government

COICOP

Classification of the Purposes of Non-profit Institutions Serving Households

COPNI

Classification of the Purposes of Non-profit Institutions Serving Households

COPP

Classification of the Outlays of Producers According to Purpose

CPA

Classification of Products by Activity

CPC

Central Product Classification

CPI

consumer price index

CRAS

cell-corrected RAS method

CREEA

compiling and refining of economic and environmental accounts

CSPI

corporate services price index

EBOPS

Extended Balance of Payments Services Classification

ECE

Economic Commission for Europe

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

xviii

EE-IOT

environmentally extended input-output tables

EPI

export price index

ERETES

equilibre ressources-emplois et tableau entrées-sorties

ESA

European system of accounts

European Union

FAO

Food and Agriculture Organization of the United Nations

FBS

finance and business services

fixed industry

FIGARO

full international and global accounts for research in input-output

analysis

FISIM

financial intermediation services indirectly measured

FOB

free on board

fixed product

GDP

gross domestic product

GENESIS

online databank of the Federal Statistical Office of Germany

GNI

gross national income

GRAS

generalized RAS

GSBPM

Generic Statistical Business Process Model

GTAP

global trade analysis project

GTAP-MRIO

multi-region input–output table based on the global trade analysis project

database

GVA

gross value added

Harmonized Commodity Description and Coding System

ICIO

inter-country input-output

ICPIs

intermediate consumption price indices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

xix

ICT

information and communications technology

ILO

International Labour Organization

IMF

International Monetary Fund

IMTS

International Merchandise Trade Statistics: Concepts and Definitions

INSEE

Institut national de la statistique et des études économiques

IOTs

input-output tables

IPCC

Intergovernmental Panel on Climate Change

IPIs

import price indices

ISIC

International Standard Industrial Classification of All Economic Activities

KLEMS

Integrated Industry-Level Production Account

KRAS

Konfliktfreies RAS

MPS

Material Product System

MRIO

multi-region input-output

nitrous oxide

NACE

Statistical Classification of Economic Activities in the European Community

NAICS

North American Industry Classification System

NMVOC

non-methane volatile organic compounds

NOx

mono-nitrogen oxides

NPISH

non-profit institution serving households

OECD

Organization for Economic Cooperation and Development

PIOT

physical input-output tables

PPP

purchasing power parities

PSUT

physical supply and use table

PYPs

previous years’ prices

R&D

research and development

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

RAS

ranking and scaling data reconciliation method

RPI

retail prices index

SAM

social accounting matrix

SDMX

Statistical Data and Metadata Exchange

SEEA

System of Environmental and Economic Accounts

SESAME

system of economic and social accounting matrices and extensions

SITC

Standard International Trade Classification

SNA

System of National Accounts

SO2

sulphur dioxide

SUT-Euro

Euro method for SUTs

SUTs

supply and use tables

TEC

trade by enterprise characteristics data

TiVA

trade in value added

TLS

taxes less subsidies on products

TRAS

three-stage RAS

TTM

trade and transport margins

UNSD

United Nations Statistics Division

UNWTO

World Tourism Organization

VAT

value added taxes

WCO

World Customs Organization

WIOD

World Input-Output Database

WPI

wholesale price index

WTO

World Trade Organization

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Part one

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Chapter 1. Introduction

A. Background

1.1. The supply and use tables (SUTs) are an integral part of the System of National Accounts

2008 (2008 SNA) forming the central framework for the compilation of a single and coherent

estimate of gross domestic product (GDP) integrating all the components of production, income

and expenditure approaches, and providing key links to other parts of the SNA framework.

1.2. In their simplest form, the SUTs describe how products (goods and services) are brought

into an economy (either as a result of domestic production or imports from other countries) in the

supply table, and how those same products (intermediate consumption; final consumption by

household, non-profit institutions serving households, and general government; gross capital

formation; and exports) are used in the use table.

1.3. The SUTs also provide the link between components of gross value added (GVA), industry

inputs and outputs. Although typically they show only the industry dimension, SUTs can also be

formulated to show the role of different institutional sectors (for example, non-financial

corporations, government, and others) providing an important linking mechanism to the different

accounts of the SNA framework (the goods and services account, production account, generation

of income account and the capital account).

1.4. Importantly, and by design, these interlinkages facilitate data confrontation and the

examination of the consistency of data on goods and services obtained from different statistical

sources, such as business surveys, household surveys and administrative data within a single

detailed framework. As such, they provide a powerful mechanism for feedback on the quality and

coherency of primary data sources.

1.5. The SUTs do not just provide a framework to ensure the best quality estimates of GDP and

its components: they are also an important analytical resource in their own right, showing the

interaction between producers and consumers. When measured in volume terms, the SUTs provide

the basis for a rich stream of analyses, notably in the field of structural analysis, and in particular

productivity, where in recent years SUTs have been widely accepted as an important tool for

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

KLEMS-type

productivity measures. Just as important is their growing use as the basis for

deriving the input-output tables (IOTs).

1.6. In many respects, the IOTs, which show the links between final uses and intermediate uses

of goods and services defined according to industry outputs (industry-by-industry tables) or

according to product outputs (product-by-product tables) predate the SUTs. The IOTs also show

separately the consumption of domestically produced and imported goods and services. The

widespread availability of SUTs has meant, however, that the SUTs form the starting point for

constructing IOTs and, in turn, an entire swathe of related analytical products and indicators, such

as the Leontief inverse and other type of analyses, including output multipliers, employment

multipliers, and others.

1.7. The SUTs and IOTs are compiled by many countries in the course of producing their core

national accounts, thereby improving the coherency and consistency of their national account

estimates. The ability to readily create IOTs from SUTs (as shown in chapter 12) has helped to

reinforce the momentum behind the evolution, role and use of SUTs.

1.8. SUTs and IOTs have received much attention in recent years. This is because their

analytical properties allow for a much wider set of analyses, not only of the national economy and

the regions within a nation but also of the interlinkages between economies at the global level and

also of environmental impacts.

1.9. Further momentum has been generated for the role of SUTs and IOTs in step with the

rapidly growing impact of globalization and the international fragmentation of production. For a

full understanding of international interdependencies and their impact on important policy areas,

such as trade, competitiveness and sustainable development, there is increasing need to view

production and consumption through a global value chain lens. In other words, multi-country and

regional SUTs and IOTs have become essential tools to inform policy and policymakers. Over the

past five years, a number of efforts have been made by the international statistics community to

meet these needs, such as the trade in value added database prepared by the Organization for

Economic Cooperation and Development (OECD) and the World Trade Organization (WTO), and

other comparable databases such as the World Input-Output Database (WIOD) and the Handbook

on Accounting for Global Value Chains prepared by the Expert Group on International Trade and

Economic Globalization Statistics.

1.10. Given these developments and, in particular, the heightened importance of SUTs and IOTs,

the timing of the present Handbook is important and highly relevant. The present chapter provides

a general introduction to the various issues considered in greater detail in the various chapters that

follow. Section B of this introductory chapter provides a general overview of the roles and uses of

KLEMS is an industry-level growth and productivity research project, based on the analysis of capital (K), labour

(L), energy (E), materials (M) and service (S) inputs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

SUTs and IOTs. Section C covers the SNA and its links to SUTs and IOTs. Section D covers the

objectives of the Handbook and its new features compared to previous manuals on the subject.

Lastly, section E briefly outlines the structure and content of the Handbook.

B. Uses of SUTs and IOTs

1.11. The uses of SUTs and IOTs are multiple and their statistical and analytical importance has

increased with time and in response to new and emerging issues, such as globalization and

sustainable development, with its three pillars of social, economic and environmental

development. Where possible, the analytical uses of SUTs and IOTs are presented below in

parallel. As SUTs form the basis for the compilation of IOTs, the uses of the two types of tables

are treated in the same way in this section.

1.12. As mentioned above, the SUTs combine in a single framework the three approaches to

measuring GDP, namely, the production approach, the income approach and the expenditure

approach. All three approaches are based on sets of data with various levels of detail and a range

of different sources. Combining the data in a single statistical framework compels compilers to

use harmonized and unique classifications of producers, users and income receivers, together with

harmonized and unique classifications and definitions of products and income categories. Under

these conditions, corresponding data can be related and compared in an organized manner.

Combining the three data sets provides an opportunity to analyse the causes of discrepancies, make

necessary adjustments and fill data gaps when necessary.

1.13. An important objective of national accounts is to estimate year-to-year and quarter-to-

quarter changes in a number of macroeconomic variables. When dealing with production, use and

the generation of value added, it is important to divide the current price changes into volume

changes (representing what is termed “real” growth) and price changes. When SUTs are compiled

simultaneously in current prices and in volume measures (as recommended in this Handbook,

using what is known as the “H-Approach”), there are considerable advantages in the overall quality

and consistency of the information provided. During the entire statistical process – from the

processing and analysis of the source data through to, and including, the balancing of the SUTs –

data in current prices and deflated data are obtained simultaneously and consistently with each

other.

1.14. In addition to annual national accounts, SUTs can be used in the compilation of quarterly

national accounts. This may range from the compilation and balancing of quarterly SUTs to the

mere use of the SUTs framework to highlight possible discrepancies between quarterly product

supply and use. The annual estimates of GVA can, for example, be used as weights in the quarterly

estimate of GDP in volume terms to reflect the most recent period. In addition, SUTs can provide

weighting schemes for price and volume indices.

1.15. The SUTs and IOTs serve also as the basis for compiling a range of accounts –regional,

environmental, labour, tourism, etc. The clear links of these satellite systems with both the SUTs

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

and the IOTs ensure the consistency of the satellite systems with the concepts and methods of the

core national accounts and allow for feedback loops with the SUTs during the compilation and

balancing process of the frameworks involved. For instance, the SUTs can support the compilation

of regional accounts by including clear links to variables like regional GVA. When physical

environmental flows are linked to the SUTs and IOTs in the environmental accounts, they provide

feedback loops to the compilation of SUTs by contrasting physical and monetary measures of the

supply and use of products. When SUTs are linked to labour and capital, they can be used for

productivity analyses that link economic growth to the use of intermediate inputs. Lastly, social

accounting matrices elaborate the linkages between SUTs and sector accounts. They capture

transactions and transfers between all economic agents in the accounting system and measures

effects of macroeconomic policies on distribution.

1.16. The SUTs and IOTs also provide the basis for different types of analytical uses at micro

and macro levels (see, for example, United Nations, 2002; Mahajan, 2004a; and Mahajan, 2006).

Various examples are included in the list of additional reading at the end of this Handbook.

Examples include the following:

• Economic analyses: export shares, import penetration, concentration ratios, links between

prices and costs, links between energy production, consumption and emissions, etc.

• Impact and policy analyses: sensitivity analyses, analyses of the impacts of taxation changes,

price changes, introduction of a minimum wage, specific economic crisis, earthquakes, etc.,

analyses of consumption and demand-based accounting and analyses of air emissions,

material flows, energy, water, etc.

• Industrial and sectoral analyses: changes over time to specific sectors, such as information

and communications technology (ICT), oil and gas, food, sport, creative arts, tourism, health,

etc., and, more recently, analyses covering the digital economy, sharing economy,

collaborative economy and also product-specific global value chains.

• Local government type investment planning: construction projects, shopping centres, new

motorways, rural planning, etc.

• Base structures for modelling: computable general equilibrium models, environmental

analyses, supply-side-based models, etc.

1.17. The role of SUTs and IOTs in understanding global value chains is of particular

importance, given the interconnected nature of today’s global economy. SUTs constitute the

centrepiece of the internationally compatible accounting framework for a systematic and detailed

description of the economy, its various components on the supply and use side and its relations to

other economies. The construction of international SUTs and IOTs makes it possible, in

combination with trade statistics, to follow the trade in value added and to understand who

ultimately benefits from the trade of finished goods in terms of value added, employment, and

other factors. The compilation of international or global SUTs and IOTs tables poses a number of

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

compilation challenges (including, for example, the recording of goods sent abroad for processing

and the recording of the production abroad and merchanting operations affecting SUTs and IOTs)

and relies on the availability of national SUTs and IOTs on a comparable basis.

1.18. In addition, the inclusion of the environmental dimension in the SUTs and IOTs further

enhances the usefulness of these tables by allowing the integration and consistency of the

economic and environmental information and an understanding of the interlinkages between the

economy and the environment. Incorporating environmental considerations as part of the regular

compilation of SUTs improves the quality, coherence and consistency of the related outputs and

the process provides powerful feedback loops for identifying improvements.

C. System of National Accounts

1.19. The SNA provides an internationally compatible framework for a systematic and detailed

description of a total economy (namely, that of a region, country or group of countries), its

components and its relations with other total economies. The 2008 SNA (United Nations,

European Commission, IMF, OECD and World Bank, 2009) is the latest version of the SNA,

which was adopted by the United Nations Statistical Commission in 2008.

1.20. The SNA describes the basic features of the accounting system in terms of concepts,

principles, statistical units and their groupings, etc. The SNA gives an overview of the sequence

of accounts, the balancing items associated with each account, a brief description of key aggregates

and the role of SUTs and the input-output framework. The key accounting sequence includes the

following stages: production of goods and services, transactions relating to products (goods and

services) and also to non-produced assets, transactions which distribute and redistribute income

and wealth, financial transactions and balance sheets.

1.21. The SNA framework also draws in other aspects, such as price and volume measurement,

population, labour market measures, regional accounts and various specific conceptual issues.

Figure 1.1 provides an overview of how SUTs and IOTs fit within the SNA framework. In

particular, it shows which accounts in the SNA sequence of accounts are more directly linked with

SUTs and IOTs, namely, production accounts, generation of income accounts, use of disposable

income accounts and capital accounts.

1.22. Producing annual SUTs simultaneously both in current prices and in volume terms, not

only ensures consistency for price volume measures, it also allows for the estimation of the volume

of GVA through what is termed “double deflation”, which is recommended in the 2008 SNA.

1.23. As noted above, the SUTs are an integral part of the SNA, determining a single estimate

of GDP both in current prices and in volume terms and linked to the institutional sector accounts.

For example, the goods and services account for the total economy can be directly compiled from

the SUTs through appropriate aggregation. In addition, by using the breakdown of GVA by

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

components in the use table, the production account and the generation of income account can

easily be compiled from the SUTs and linked to the institutional sectors.

1.24. Another important aspect linking the SUTs and the institutional sector accounts is the

statistical unit. The SNA uses two types of units and two corresponding ways of subdividing the

economy, which are quite different and serve separate analytical purposes. The units can be

classified to an industry for use in the SUTs and to an institutional sector for use in the institutional

sector accounts.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 1.1 Overview of the links between SUTs and the SNA framework

Compiled by Sanjiv Mahajan, May 2014

1.25. The first purpose of describing production, income, expenditure and financial flows, and

balance sheets, is served by grouping institutional units into institutional sectors on the basis of

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

their principal functions, behaviour and objectives. The SNA enables a complete set of flow

accounts and balance sheets to be compiled for each sector and subsector, and also for the total

economy. The five institutional sectors distinguished in the SNA are the following:

• Non-financial corporations

• Financial corporations

• General government

• Households

• Non-profit institutions serving households

1.26. The SNA also describes the transactions between these five institutional sectors and the

rest of the world. These institutional sectors can be further split into subsectors, for example,

general government can be split into central government and local government.

1.27. The second purpose of describing processes of production and for input-output analysis is

served by the grouping of local kind-of-activity units (or establishments) into industries, on the

basis of their type of activity. An activity is characterized by an input of products, a production

process and an output of products.

1.28. In order to ensure consistency between SUTs and the institutional sector accounts, a link

table is compiled as an integrated part of the system. In this link table, a cross-classification of

output, intermediate consumption, components of value added (and other possible variables of

industries) between the industries and the institutional sectors is shown. Thus, this table links the

main macroeconomic variables from the SUTs to the institutional sector accounts, providing a

picture of local kind-of-activity units and one based on institutional units. As both types of units

are classified differently, the link table also provides a picture of the relations of output,

intermediate consumption, value added, and other variables, originating in the different industries

and institutional sectors.

1.29. The SUTs – consistent with the national accounts – are normally produced in connection

with the final or benchmarked versions of the macroeconomic data some two or three years after

first preliminary results of the national accounts are published. The SUTs, however, should play a

more vital role at the heart of national accounts in the production of preliminary annual or even

quarterly accounts. Once the SUTs compilation system is in place on an annual basis, the statistical

benefits are significant.

1.30. SUTs can play various roles in the national accounts. One, for example, is to update SUTs

– often in a more aggregated version – from the previous year with information available for the

preliminary estimates in order to have a complete set of SUTs available that are consistent with

the national accounts. This procedure is a good method for revealing inconsistencies in the

aggregated preliminary national accounts. Another role of SUTs could arise from new information

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

in a situation in which new, and more, detailed information on total supply and exports is available

at an early stage, the structure and relationships in the SUTs of the previous year could be used to

project SUTs for domestic output and imports.

1.31. The compilation of SUTs was in the past associated solely with the construction of IOTs.

The SUTs were therefore seen as an intermediate step in the compilation of IOTs. This meant,

effectively, that the SUTs were only compiled out after the compilation of the national accounts

had been completed. This approach, in fact, has significant limitations because the independently

calculated national accounts aggregates had to be kept unchanged despite inconsistencies

identified through the SUTs system.

1.32. SUTs are now seen as more than just as a step in the construction of IOTs: it is the SUTs

that provide the ideal framework guaranteeing the coherency and consistency of supply and use of

products in the system in current prices, and in volume terms, thereby improving the quality of the

national accounts, and in turn the key economic aggregates.

1.33. The compilation of SUTs is thus recommended as part of the regular annual compilation

of national accounts. The annual compilation of SUTs is also one of the recommended data sets

used in assessing the scope of implementation of the 2008 SNA.

The compilation of SUTs on a

quarterly basis can also play a role in improving the quality and coherence of quarterly national

accounts (the role played by SUTs in quarterly national accounts is further elaborated in chapter

14).

1.34. The approach to the compilation of SUTs as an integral component of the production of

national accounts may be formulated in general terms as follows:

• SUTs are produced as a central element of the compilation of national accounts with a view

to providing a key link to various parts of the SNA framework.

• SUTs provide a statistical framework representing the most efficient means of incorporating

all basic data – aggregated or detailed – covering the components of the three approaches to

measuring GDP, and linking to the institutional sector accounts in a systematic way.

• SUTs effectively ensure the consistency and reconciliation of results at a detailed level and

thereby improve the overall quality of the national accounts.

• SUTs are compiled and balanced in both current prices and in volume terms.

• SUTs are produced annually or even, if possible, on a quarterly basis, the ideal option.

See table 2 of the 2011 report of the Intersecretariat Working Group on National Accounts to the United Nations

Statistical Commission at its forty-second session (E/CN.3/2011/6), available at:

http://unstats.un.org/unsd/statcom/doc11/2011-6-NationalAccounts-E.pdf.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• SUTs can provide a powerful feedback mechanism on the coherency and consistency of

source data, such as business surveys, and on the classification of units on the business

1.35. When balanced, SUTs provide a coherent, consistent and wholly integrated suite of

statistics for a single period (for example, a year), which include:

• A single estimate of GDP in current prices and in volume terms, which is underpinned with

components of the production, income (only in current prices) and expenditure approaches

to measuring GDP

• Detailed goods and services account in current prices and in volume terms (not by

institutional sector)

• Production accounts by industry and by institutional sector in current prices and in volume

terms.

• Generation of income accounts by industry and by institutional sector (both in current prices

only)

• Link to the use of disposable income account through the flows of final consumption

expenditures and capital account through gross capital formation (and its components)

balanced via SUTs.

1.36. These guidelines should form part of the strategic tools used to improve the quality of the

national accounts.

D. Objectives of this Handbook

1.37. The theoretical development of IOTs has a long history. Box 1.1 provides a description of

the evolution of both IOTs and SUTs within the context of national accounts. The United Nations

Statistics Division has followed the theoretical development and the practical work of national

statistics offices on IOTs and SUTs from the outset. Starting in 1996, it has prepared a number of

publications, under the guidance of the United Nations Statistical Commission, such as those listed

among the references under United Nations, 1966, 1973 and 1999, to share practices, update the

methodology in line with the updates of the SNA, and provide guidance on the compilation of

IOTs.

1.38. This Handbook continues those efforts in cooperation with other international

organizations and experts, providing practical and step-by-step guidance for the compilation of

SUTs and IOTs based on the latest international statistical standards set out in the 2008 SNA and

the sixth edition of the International Monetary Fund (IMF) Balance of Payments and International

Investment Position Manual (BPM 6) (IMF, 2009).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 1.1 Evolution of the SUTs and IOTs within the national accounts

The national accounting system is continuously evolving across the various domains to reflect developments in

and improvements to the quality of economic statistics and the evolution of economies, with a view to providing

a relevant measurement of the economy. Over the past four centuries many significant contributions have been

provided by people from various disciplines and countries, resulting in the system as it stands today and how it

relates to SUTs and IOTs. Below is a short description of this evolution.

Wassily Leontief (1905–1999), who is often referred to as the pioneer of input-output-based economics, made

the first of many key contributions with the publication of his article “Quantitative input and output relations in

the economic system of the United States”. This article discussed the construction of an economic transactions

table that Leontief had based on the tableau économique, proposed by François Quesnay in 1758.

The framework was developed and applied as an economic tool with the construction of the first IOTs for the

United States covering the years 1919 and 1929 published in 1936. Later, Leontief developed the first input-

output-based model, which was based on theories developed by Leon Walras published in 1874 and 1877.

Leontief‘s pioneering work was recognized by the award to him of the Nobel Prize in Economics in 1973. As

a result, input-output analysis has become a major tool in developing quantitative economics as a science.

The role of SUTs and IOTs has evolved within national accounts. The 1953 SNA (United Nations, 1953)

included no reference to SUTs or IOTs. The 1968 SNA (United Nations, 1968), however, presented the

integration of an input-output framework into the integrated economic accounts of the SNA. The conceptual

development of the integrated economic accounts of the SNA earned Richard Stone the Nobel Prize in

Economic Science in 1984, for having made fundamental contributions to the development of the SNA and

having thereby greatly improved the basis for empirical economic analysis.

Alongside Leontief and Stone, other Noble laureates include Ragnar Frisch and Jan Tinbergen in 1969, Paul

Samuelson in 1970, Simon Kuznets in 1971, John Hicks in 1972 and James Meade in 1977, who have all

contributed to the foundations of the measurements used in today’s SNA and the interlinkages between various

sectors and activities in an economy.

The latest evolution of SUTs was recognized in the 1993 SNA (United Nations, CEC, IMF, OECD and World

Bank, 1993), chapter XV of which covered both SUTs and IOTs. With the latest version of the SNA, the 2008

SNA, the role and applications of SUTs have been further enhanced, and this in turn will help to meet many

analytical needs, as reflected in chapters 14 and 28 of the 2008 SNA.

1.39. This Handbook may, therefore, be viewed as an update of the United Nations Handbook

of Input-Output Table Compilation and Analysis (United Nations, 1999). In response, however, to

the ever-increasing importance of SUTs in their own right, this Handbook extends the scope of the

previous publication by providing a more detailed description and compilation guidance for SUTs.

As stated in the 2008 SNA, “only supply and use tables provide a sufficiently rigorous framework

to eliminate discrepancies in the measured flows of goods and services throughout the economy

to ensure the alternative measures of GDP converge to the same value” (2008 SNA, para. 14.15).

1.40. The Handbook builds on the experience, practices and guidance available at national and

regional level, such as the Eurostat Manual of Supply, Use and Input-Output Tables (Eurostat,

2008). At the same time, however, it provides an innovative approach to the compilations of SUTs

and IOTs in the following three main areas:

• Underlying use of an integrated approach to statistics

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Use of a business model for the compilation of SUTs and IOTs linking the various parts

through an H-Approach compilation scheme

• Mainstreaming of environmental considerations, through the inclusion of the environmental

focus of chapter 13 at the core of part three of the Handbook

1.41. The compilation guidance provided in this Handbook relies on an integrated statistics

approach whereby the production of statistics in the various domains is not seen in isolation but as

part of an integrated process using common concepts, definitions, business registers and frames,

statistical units, estimation methods and data sources to improve the consistency of the statistics

compiled, to reduce the respondent burden, and potentially to reduce the statistical agency costs.

In particular, the consistency of the basic economic information that feeds into the national

accounts and the SUTs with the classifications, concepts and definitions of the 2008 SNA greatly

reduces the discrepancies across data from different sources, thus facilitating their reconciliation

as part of the integration process. The integrated statistics approach is described in the Guidelines

on Integrated Economic Statistics (United Nations, 2013).

1.42. This Handbook follows the Generic Statistical Business Process Model (GSBPM)

(UNECE, 2013) to describe the production of statistics in a general and process-oriented manner.

The underlying concepts and principles of the GSBPM have been followed in describing the

business process and stages of the statistical production processes underpinning the compilation

of SUTs and IOTs. Chapter 3 of this Handbook describes these links in more detail in the context

of SUTs and IOTs. In addition, the chapters in parts two and three of the Handbook are linked to

the different parts of these stages of the statistical production process.

1.43. With the adoption of the System of Environmental-Economic Accounting (SEEA) (United

Nations, European Commission, FAO, IMF, OECD and World Bank, 2014) by the United Nations

Statistical Commission, the extension of SUTs and IOTs to include environmental flows in

monetary and physical terms has become an internationally agreed standard. Including

environmental consideration from the outset in the compilation of SUTs brings a number of

advantages. It facilitates the integration and reconciliation of the information, it enhances the

quality of the information, and it significantly increases the uses to which the tabulations may be

put.

1.44. In line with the United Nations Statistical Commission,

this Handbook recommends the

annual compilation of SUTs. In addition, the Handbook promotes the compilation of these tables

as an integral part of the compilation of national accounts in order to ensure full consistency of the

basic data and also of the macroeconomic estimates that are derived from the accounts.

See the 2011 report of the Intersecretariat Working Group on National Accounts to the United Nations Statistical

Commission at its forty-second session (E/CN.3/2011/6), available online at:

http://unstats.un.org/unsd/statcom/doc11/2011-6-NationalAccounts-E.pdf

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

1.45. The Handbook provides a consistent numerical example of SUTs and IOTs that runs

throughout the chapters (as far as practically possible), in order to facilitate understanding of the

various compilation steps. It also provides examples of best practices to illustrate certain aspects

of the compilation of SUTs. It should be noted that, in the numerical examples provided in the

Handbook, the numbers may not add up exactly to the totals because of rounding.

1.46. The target audience for this Handbook mainly includes compilers of SUTs and IOTs with

a basic knowledge and understanding of the SNA. Since, however, the Handbook provides an

overview of the whole statistical production process, managers or staff in charge of the programme

of national accounts and of economic and environmental accounts would also benefit from the

Handbook in gaining an overall understanding of the requirements for the compilation of SUTs

and IOTs. Lastly, analytical users may also benefit from perusing the Handbook, as it would

provide them with a better understanding of the compilation steps, thus increasing the analytical

applications of SUTs and IOTs.

E. Structure of the Handbook

1.47. The Handbook consists of the following four main parts:

• Part one, providing an introduction to the Handbook, set out in chapter 1.

• Part two, describing the overview of SUTs and IOTs, the fundamental building blocks

required, cross-cutting issues and the main stages of the GSBPM – namely, the design,

building and collection phases. Part two comprises chapters 2–4.

• Part three, describing the compilation, balancing and dissemination phases of SUTs and

IOTs. This also includes the physical SUTs (PSUTs), environmentally extended IOTs (EE-

IOTs) and the SUTs links to the quarterly national accounts. Part three comprises chapters

5–15.

• Part four, providing examples of the extensions and applications of SUTs and IOTs. Part

four comprises chapters 16–21.

Part one

1.48. As indicated above, chapter 1 provides an introduction to the Handbook; it describes the

importance of SUTs and IOTs for statistical purposes (for example, compilation of annual and

quarterly national accounts, etc.), for policymaking and for analytical purposes (for example,

economic forecasting, assessing the impact of globalization). It also provides a general description

of the SNA and where the SUTs fit within the SNA framework. This chapter also describes the

overall approach of the Handbook (including in comparison to previous handbooks) to the

compilation of SUTs and IOTs and provides a general outline of its contents.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Part two

1.49. Chapter 2 provides a conceptual overview of SUTs and IOTs and describes the basic

elements determining their structure and compilation. These include the accounting principles of

the SNA, the classifications of economic activities and products, the choice of the statistical units

and how they affect SUTs and IOTs, and the valuation in SUTs and IOTs. The chapter identifies

the advantages of compiling SUTs as an integral part of the national accounts and how the SUTs

are used to obtain consistent estimates of GDP. It also describes in more detail the extended

perspective adopted in this Handbook to SUTs and IOTs, incorporating an environmental

dimension which makes possible an integrated overview of the framework from the very outset.

1.50. Chapter 3 provides an overview of the different phases that constitute the statistical

production process of SUTs and IOTs, based on the stages of the GSBPM and in line with the

United Nations Guidelines on Integrated Economic Statistics (United Nations, 2013). This chapter

also provides an overview of the different institutional set-ups in countries which may have an

impact on the compilation process. The compilation phases specific to SUTs and IOTs are

presented in this chapter together with a link to the relevant chapters of the Handbook.

1.51. Chapter 4 covers specific phases of the GSBPM, namely the specify needs, design, build

and collect phases. It provides a description of the elements that should be considered and carefully

evaluated at the beginning of the compilation process, such as the level of detail of the industry

and products in the tables, the compilation schedule, the revision policy, resources, typical data

sources, and others. These and other issues are covered in this chapter, thus providing a foundation

for the compilation of SUTs and IOTs.

Part three

1.52. Chapter 5 describes the conceptual and practical aspects of the compilation of the supply

table and how the so-called “unbalanced” supply table is put together from the typical data sources

for SUTs, such as business surveys, administrative data and others.

1.53. Chapter 6 describes the conceptual and practical aspects of the compilation of the use table.

As in chapter 5, this chapter shows how an unbalanced use table is constructed on the basis of

typical data sources.

1.54. Chapter 7 describes how to compile the valuation matrices necessary to bridge the different

valuation concepts of the product flows. This chapter covers the main concepts and methodologies

of compiling matrices for trade margins, transport margins, taxes on products and subsidies on

products.

1.55. Chapter 8 describes the structure of the imports use table and the domestic use table and

the steps necessary to disaggregate the use table into an imports use table and a domestic use table.

Historically, the compilation of these tables was largely viewed as an intermediate – though not

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

essential – step in the compilation of IOTs. That said, however, the imports use table and the

domestic use table are becoming increasingly important in their own right for analytical purposes.

1.56. Chapter 9 covers the compilation of SUTs in volume terms. It follows the recommendation

that SUTs should simultaneously be compiled in current prices and in volume terms. The

compilation of SUTs in volume terms can start after the SUTs have been compiled in current prices

(although the current price tables do not need to be balanced) but there is need for a simultaneous

presentation of volume and price indices.

1.57. Chapter 10 describes the importance of linking SUTs and the institutional sector accounts,

which involve data by industry that are to be subdivided according to the institutional sectors to

which the units within each industry are assigned. The chapter provides guidance on how to

compile the cross-tabulation between industries and institutional sectors and presents various

approaches that may be followed in establishing the link between the SUTs and the institutional

accounts. It also identified certain issues that may arise in the compilation of the linking table.

1.58. Chapter 11 describes the manual and automated balancing procedures of SUTs in both

current prices and in volume terms. This is important for full consistency of the detailed

information. The various checks related to product, industry and macro identities, benchmarking

with national accounts, and comparison with previous SUTs, if available, are explained. It is

recommended that SUTs should be produced and balanced simultaneously at basic and purchasers’

prices and also for domestic and imported products, all of which should be both in current prices

and in volume terms. A further dimension, and challenge to be surmounted, is the need to cover

both annual SUTs and, if possible, quarterly SUTs.

1.59. The sequence of chapters represents the preferred scenario for the compilation of SUTs.

Different approaches may be followed, however. An increasing number of countries have achieved

the preferred scenario. This scenario for the compilation of SUTs and IOTs may be seen as

ambitious but it can be realized through gradual improvements in source data, production

processes and the information technology environment.

1.60. Chapter 12 provides an overview of the IOTs (product by product and industry by industry)

and describes the methods and the underlying assumptions for transforming SUTs into IOTs. The

compilation of IOTs is quite different in nature from that of SUTs and relies on the availability of

SUTs. The compilation of IOTs is considered more as an analytical step than a compilation process

and, for this purpose, is viewed as a transition from statistics to modelling.

1.61. Chapter 13 describes the structure of the SUTs in physical units where additional rows and

columns are added to show flows from the environment to the economy and vice versa. This

chapter also describes typical data sources for the compilation of these tables and examples of

specific issues in which the SEEA and the SNA differ (for example, the treatment of international

flows and the treatment of goods for processing), and shows how standard economic IOTs in

monetary units may be extended to include information on the environment in physical units in

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

the environmentally extended IOTs. Physical IOTs are also an extension of the SUTs framework,

extended to take into account environmental considerations. They consist of a transformation of

the PSUTs. Given certain conceptual and practical issues in the compilation of physical IOTs,

however, the focus of the 2012 SEEA – and thus of this chapter – has shifted towards the

compilation of environmentally extended IOTs rather than physical IOTs. Examples of two

country practices are also presented in this chapter.

1.62. Chapter 14 provides an overview of how SUTs may be used to improve the quarterly

national accounts. Since there are various scenarios that can be used in practice, this chapter

focuses only on three main situations which illustrate the use of SUTs to various degrees in the

compilation of the quarterly national accounts.

1.63. Data dissemination is an important activity for any statistical production process, as it

provides users with a range of statistics produced to internationally agreed guidelines. Presenting

SUTs and IOTs to users in a clear, transparent and user-friendly manner is thus an important task

for the statisticians. Chapter 15 provides an overview of the elements that should be considered

when disseminating SUTs and IOTs, such as the identification of users’ needs in order to tailor

dissemination to the main types of users of SUTs and IOTs, the importance of having a

dissemination strategy and the elements that should be covered in the strategy. Reference to the

Statistical Data and Metadata Exchange (SDMX) for SUTs and IOTs is also provided in this

chapter.

Part four

1.64. Chapter 16 describes the methods for compiling regional (subnational) SUTs and the main

compilation issues, such as the disaggregation of the information at subnational level, among

others. Different issues and challenges are covered through a bottom-up and top-down compilation

approach.

1.65. Although the focus of this Handbook is mainly on the compilation of national SUTs and

national IOTs, there is a growing demand for these instruments to capture the structure and

mechanism of the cross-border fragmentation of production activities. Chapter 17 provides an

overview of multi-country SUTs and IOTs, the main compilation issues, and a simplified

compilation procedure. This chapter also reviews current international initiatives in this area.

1.66. Chapter 18 deals with the projections of SUTs and IOTs. Many users also require

comparable input-output products that are comparable in terms of frequencies and timeliness. For

example, some countries produce quarterly SUTs, some countries produce annual SUTs and some

countries produce SUTs on a less regular basis. Consequently, a variety of methods, techniques

and approaches exist for the projection of SUTs and IOTs and for dealing with data gaps. These

techniques also can help producers, for example, to deal with periods between benchmarked years.

This chapter examines various methods and techniques used, along with a range of literature

available on how to surmount the problem of incomplete data, thus allowing the estimation and

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

projection of IOTs. The chapter also presents a numerical example for three methods: the

generalized RAS, the SUT-RAS and the Euro methods.

1.67. Chapter 19 describes the main extensions of supply, use and input-output tables as part of

a satellite system which is routinely used for economic analysis. Several examples of the

disaggregation of the use table and various satellite accounts are reviewed, including such

extensions as social accounting matrices, extended IOTs and other satellite systems.

1.68. Chapter 20 describes the different types of input-output models and provides a broad

overview illustrating the benefits and the approaches used. The traditional quantity model and

price model of input-output analysis are presented for monetary IOTs and physical IOTs. Input

and output coefficients, the Leontief inverse, price and quantity models, indicators, multipliers and

inter-industrial linkages were developed for an empirical extended IOT, with extensions for gross

fixed capital formation, capital stock, employment, energy, air emissions, waste, sewage and

water.

1.69. Chapter 21 provides examples of compilation practices from various countries with

different statistical systems. In general, the compilation practices can vary greatly depending on

the resources available, the statistical infrastructure, registers, surveys, methodologies and other

factors. This chapter provides guidance for countries with limited statistical resources and

illustrates differences and challenges in the compilation of SUTs and IOTs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Part two

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Chapter 2. Overview of the supply and use tables and input-output tables

A. Introduction

2.1. Before providing step-by-step guidance on the compilation of SUTs and IOTs, it is

important to ensure a general understanding of the two sets of tables. The main objective of this

chapter is, therefore, to provide an overview of SUTs and IOTs, which may be found in sections

B and C, respectively. Section D introduces the fundamental elements of the SUTs and IOTs, such

as the underlying classifications, the statistical units and the valuation methods. Some of these

elements are discussed in more detail in subsequent chapters. Lastly, section E elaborates on the

importance of compiling SUTs as an integral part of the national accounts.

2.2. Although this chapter covers a wide range of challenges and issues that must be tackled

when planning and building a new system of SUTs and IOTs, all aspects may not be achievable in

countries with limited resources. It is worth recognizing that a system may be established with a

moderate level of ambition using available data, even if these are incomplete. Nonetheless, it is

preferable to have a SUTs-type environment for reconciliation of the various statistical sources,

rather than only an unbalanced series of national accounts aggregates.

B. Overview of SUTs

2.3. The SUTs describe the whole economy by industry (for example, the motor vehicle

industry) and by product (for example, sports goods). The tables show links between components

of GVA, industry inputs and outputs, and product supply and use. The SUTs link different

institutional sectors of the economy (for example, non-financial corporations) together with details

of imports and exports of goods and services, final consumption expenditure of government,

household and non-profit institutions serving households (referred to as NPISHs), and capital

formation.

2.4. As their name suggests, SUTs consist of two interlinked tables: the supply table and the

use table. The supply table shows the supply of goods and services by type of product and by type

of industry, distinguishing between supply by domestic industries and imports of goods and

services. In other words, the supply table provides information on the output (by product)

generated by economic activities and the imports (by product) from abroad. The totals in the last

column represent the total supply by products and the totals in the bottom row represent the total

output by economic activity and total imports. A simplified supply table is presented in Table 2.1

below.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Table 2.1: Simplified structure of the supply table

2.5. The second table is the use table, which provides information on the uses of the different

products. The use table shows the use of goods and services by type of product and by type of use,

in other words, as intermediate consumption by industry, final consumption, gross capital

formation or exports. Furthermore, the table shows the components of gross value added by

industry – namely, compensation of employees, other taxes less subsidies on production,

consumption of fixed capital and net operating surplus. While the totals by row represent the total

uses by product, the total by column represent the total output by economic activity, total final

consumption, total gross fixed capital formation and total exports. Table 2.2 below shows the

simplified structure of the use table.

Table 2.2: Simplified structure of the use table

2.6. The classification of products, in practice, is often more detailed than the classification of

industries, thus generating rectangular SUTs. For example, the output of the dairy industry is

separately shown in the SUTs for the products of processed milk, butter, yoghurt, cheese and so

forth, and not as only one aggregate product for all dairy products.

2.7. There are three basic identities that hold between the supply table and the use table. The

first identity corresponds to the fundamental identity in national accounts, whereby for each

economic activity the following holds:

Identity (1) Output = Intermediate consumption + GVA

Agriculture,

forestry, etc.

Mining and

quarrying

… Services

Agriculture, forestry, etc.

Ores and minerals, etc.

…

Services

Total

Total imports Total supply

Total output by industry

Industries

Products

Industries

Imports

Total

Output by product by industry

Imports by

product

Total supply

by product

Agriculture,

forestry, etc.

Mining and

quarrying

… Services

Final

consumption

Gross capital

formation

Exports

Agriculture, forestry, etc.

Ores and minerals, etc.

…

Services

Value added

Total

Empty cells by definition

Total

Intermediate consumption by product and by industry

Final uses by product and by category

Total use by

product

Value added by component and by industry

Total output by industry

Total final uses by category

Industries

Products

Industries

Final uses

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2.8. The second identity is that the total supply by product is equal to the total use by product.

This means that the amount of products available for use in an economy must have been supplied

either by domestic production or by imports, and the same amount of products entering an

economy in an accounting period must be used for intermediate consumption, final consumption,

capital formation or exports. This means that, for each product (or group of products):

Identity (2) Output + Imports = Intermediate consumption + Final consumption +

Capital formation + Exports

2.9. Another important identity which is also key when linking the production and income

approaches to calculating GDP and the industry and institutional sector dimension through the

SUTs is the following:

Identity (3) For each industry, the GVA using the production approach equals the

GVA estimate using the income approach.

2.10. These identities are fundamental in the balancing process that is carried out when

compiling SUTs both in current prices and in volume terms, all through a time series dimension.

2.11. Once balanced, the supply table and the use table can be integrated into a single matrix –

often referred to as the SUTs framework, which is shown in Table 2.3 below. This table clearly

shows the two basic identities linking the SUTs. The total supply by product (left part of the bottom

row of Table 2.3) equals the total use by product (top part of the last column of Table 2.3) and the

total outputs by industry are identical in both SUTs (the middle part of the bottom row equals the

middle part of the last column). The schematic view of SUTs in Table 2.3 also serves as the

underlying matrix for projection methods (see chapter 18).

Table 2.3: Supply and use tables framework

2.12. SUTs thus bring together the components of each of the three approaches to measuring

GDP– namely, the production, income and expenditure approaches:

(a) Production approach:

Agriculture,

forestry, etc.

Ores and

minerals, etc.

… Services

Agriculture,

forestry, etc.

Mining and

quarrying

… Services

Final

consumption

Gross capital

formation

Exports

Agriculture, forestry, etc.

Ores and minerals, etc.

…

Services

Agriculture, forestry, etc.

Mining and quarrying

…

Services

Value added

Imports

Total

imports

Total

Empty cells by definition

Total output

by industry

Value added by component and by industry

Total imports by product

Total

Products

Intermediate consumption by product and by

industry

Final uses by product and by category

Total use by

product

Products

Industries

Final uses

Total supply by product

Total output by industry

Total final uses by category

Industries

Output by product by industry

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

GDP = Output (at basic prices) - Intermediate consumption + Taxes less subsidies

on products

(b) Income approach:

GDP = Compensation of employees + Gross operating surplus + Other taxes less

subsidies on production + Taxes less subsidies on products

GDP = Final consumption + Gross capital formation + Exports - Imports

2.13. When balanced, SUTs show, by definition, a single estimate of GDP both in current prices

and in volume terms. This underlines the importance of the recommendation that SUTs be

compiled as part of the annual regular compilation of the national accounts, as they ensure the

consistency and coherence of the national accounts components, namely, goods and services

account, production account (by industry and by institutional sector) and generation of income

account (by industry and by institutional sector), and make it possible to derive a single estimate

of GDP. The institutional sector links are covered in more detail in chapter 10.

2.14. The SUTs also have links to other accounts, such as disposable income accounts (covering

variables like household final consumption expenditure) and accumulation accounts (covering

variables like gross fixed capital formation as part of the capital account).

2.15. Producing annual SUTs simultaneously both in current prices and in volume terms

(preferably, when two successive years of current price SUTs are available) ensures coherence and

consistency for both price and volume measures. In addition, this approach allows for the

estimation of the volume of GVA through what may be termed “double deflation”, where GVA is

derived by deducting intermediate consumption in volume terms from total output in volume

terms. This can be achieved on the basis of an individual unit, industry, institutional sector, and

for the whole economy.

2.16. SUTs can also be compiled on a quarterly basis to derive official estimates of quarterly

GDP. Developing quarterly SUTs may be highly demanding in terms of resources, time and data

availability but have the advantage of significantly improving the quality of the estimate of

quarterly GDP.

2.17. Figure 2.1 provides a graphical overview of the SUTs, explicitly identifying the main

identities that are ensured in balanced SUTs. Box 2.1 presents a numerical example of balanced

SUTs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 2.1: Graphical overview of supply and use tables

2.18. The use table records the intermediate consumption and final uses by type of product but

it does not distinguish between the consumption of domestically produced goods and services and

that of imported goods and services. Although such a split is not a necessary condition for the

creation of balanced SUTs in current prices, it is a key step linking SUTs and IOTs. The

disaggregation of the use table into two tables, the domestic use table and the imports use table, is

shown in Box 2.2, with a numerical example.

2.19. The compilation of the imports use table is necessary to ensure good quality volume

estimates (in particular, GVA by industry) and these tables are becoming increasingly important

owing to the growing impact of globalization and the need to measure global value chains and

trade in value added.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 2.1 Numerical example of the SUTs system

The supply table shows the supply of goods and services by product and by type of supplier, distinguishing supply

by domestic industries and imports of goods and services. The domestic output of industries is shown by products.

The vector of imports comprises the nation’s total imports of goods and services by product.

The use table shows the use of goods and services by product and by type of use, i.e. as intermediate consumption

by industry, final consumption expenditure, gross capital formation and exports of goods and services. The

intermediate uses and final uses reflect the consumption of domestically produced goods and services and also of

imported goods and services. Furthermore, although the table is shown in summary form, it should be noted that

there are components underlying the headings, for example, GVA can be split between compensation of employees,

other taxes less subsidies on production, consumption of fixed capital and net operating surplus.

Note that, for illustrative purposes, it is assumed that the SUTs presented here are compiled on a consistent valuation

basis.

2.20. Once the imports use table is constructed, the domestic use table can be estimated by

subtracting the imports use table from the use table. The imports use table and the domestic use

table form the basis for the construction of input imports tables and domestic IOTs, respectively.

More detail may be found in chapters 8 and 12.

Supply table

Agriculture

Manufacturing

and

construction

Services

Agriculture

270 30 50 20 370

Manufacturing

6 380 87 42 515

Construction

4 50 13 8 75

Trade, transport and communication

10 15 210 7 242

Finance and business services

6 17 240 11 274

Other services

4 8 100 12 124

Total

300 500 700 100 1 600

Use table

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross capital

formation

Exports

Agriculture

34 59 93 131 21 32 370

Manufacturing

97 107 57 122 73 59 515

Construction

9 12 4 17 30 3 75

Trade, transport and communication

42 24 11 140 20 5 242

Finance and business services

14 53 42 116 31 18 274

Other services

14 35 22 35 10 8 124

Taxes less subsidies on products

4 5 12 52 6 1 80

GVA

86 205 459 750

Total 300 500 700 613 191 126 2 430

Supply and use tables framework

Agriculture Manufacturing Construction

Trade, transport

and

communication

Finance and

business

services

Other

services

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agriculture

34 59 93 131 21 32 370

Manufacturing

97 107 57 122 73 59 515

Construction

9 12 4 17 30 3 75

Trade, transport and communication

42 24 11 140 20 5 242

Finance and business services

14 53 42 116 31 18 274

Other services

14 35 22 35 10 8 124

Agriculture

270 6 4 10 6 4 300

Manufacturing and construction

30 380 50 15 17 8 500

Services

50 87 13 210 240 100 700

Taxes less subsidies on products

4 5 12 52 6 1 80

GVA

86 205 459 750

Imports

20 42 8 7 11 12 100

Total 370 515 75 242 274 124 300 500 700 613 191 126 4 030

= Zero by definition

Final use

Total

Total use

Industries

Imports

Total supply

Products

Industries

Products

Industries

Final use

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 2.2 Numerical example showing a use table split between consumption of domestic

production and imports

The domestic use table is derived by subtracting the imports use table from the total use table shown in box 2.1.

The imports of goods and services are then shown separately as a new row denoted as “Imports” in the domestic

use table. The domestic use table shows the input requirements of industries in terms of domestic intermediates,

imported intermediates and primary inputs (GVA). It also shows the use of domestic output of products for

intermediate uses and final uses.

The imports use table includes information on the use of imported products for intermediate consumption and

final uses and the column totals, which match the estimates shown in the “Imports” row.

1. Supply and use tables in current prices and in volume terms: H-Approach

2.21. The SUTs framework not only constrains the current value estimates of supply and use of

products to balance exactly, it also provides a means of ensuring that the corresponding volume

estimates in previous years’ prices are balanced and that the series of prices implied by the

existence of one table in current prices and one in volume terms are strictly consistent. In general,

the best way to ensure mutual consistency is to prepare the SUTs in current values and in volume

terms at the same time (2008 SNA, para. 14.136).

2.22. The compilation and balancing of SUTs in current prices and in volume terms for a

sequence of years also helps to balance the changes in volumes, values and prices in the best

possible way (the key condition for the attainment of this outcome is that SUTs are available in

current prices both for the current year and for the previous year). This approach ensures a high

degree of quality in terms of coherence and consistency over time – and recommended as the best

Supply table

Agricul-

ture

Manufact.

and

construction

Services

Agriculture

270

350

20 370

Manufacturing

6 380 87 473 42 515

Construction

4 50 13 67 8 75

Trade, transport and communication

210

235

242

Finance and business services

6 17 240 263 11 274

Other services

4 8 100 112 12 124

300 500 700 1 500 100 1 600

Use table Domestic use table

Agricul-

ture

Manufact.

and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agricul-

ture

Manufact. and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agriculture

34 59

131 21 32 370

Agriculture

30 50 90 130 20 30 350

Manufacturing

97 107 57 122 73 59 515

Manufacturing

85 90 51 120 70 57 473

Construction

9 12 4 17 30 3 75

Construction

5 10 3 16 30 3 67

Trade, transport and communication

42 24

140

5 242

Trade, transport and communication

140

5 235

Finance and business services

14 53 42 116 31 18 274

Finance and business services

10 50 40 115 30 18 263

Other services

14 35 22 35 10 8 124

Other services

10 30 20 35 10 7 112

Taxes less

subsidies on products

5 12 52 6 1 80

Imports

30 40 15 5 5 5 100

GVA

86 205 459

750

Taxes

less subsidies on products

12 52 6 1 80

Total 300 500

700 613 191 126 2 430

GVA

86 205 459 750

Output

300 500 700 613 191 126 2 430

Imports use table

Agricul-

ture

Manufact. and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agriculture

1 2 20

Manufacturing

12 17 6 2 3 2 42

Construction

4 2 1 1 8

Trade, transport and communication

1 7

Finance and business services

4 3 2 1 1 11

Other services

4 5 2 1 12

Total 30 40 15 5 5 5 100

Empty cells by definition

Products

Total

Final use

Total use

Industries

Output

Imports

Total

supply

Products

Industries

Final use

Total

use

Industries

Products

Industries

Final use

Total

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

approach for the production of SUTs. Producing annual SUTs simultaneously both in current

prices and in volume terms also allows estimation of the volume of GVA through double deflation,

whereby GVA in previous years’ prices is derived by deducting intermediate consumption in

previous years’ prices from total output in previous years’ prices. Then, the change in volume of

GVA between each pair of consecutive years is given by the change of GVA in previous years’

prices compared to GVA of the previous year at current prices of that year.

2.23. The SUTs at purchasers’ prices and at basic prices in current prices and in volume terms

can be compiled and balanced sequentially or simultaneously. In both cases, powerful feedback

loops covering quality in terms of consistency and coherence are available. More details are

provided in chapters 9 and 11.

2.24. Figure 2.2 shows an overview of the H-Approach for an integrated compilation of SUTs

(and IOTs) in current prices and volume terms. The H-Approach is the recommended compilation

approach, which brings together the compilation of SUTs in current prices and volume terms, the

valuation at basic prices, producers’ prices and purchasers’ prices, and the links with the

compilation of IOTs. The matrices covering other taxes on production, other subsidies on

production, trade margins and transport margins are the valuation matrices which link between

basic prices, producers’ prices and purchasers’ prices.

Figure 2.2 Schematic overview of the compilation of SUTs and IOTs: H-Approach

Volume terms

SUTs

at purchasers’ prices

IOTs

Assumption applied

SUTs

at basic prices

Current prices

SUTs

at purchasers’ prices

IOTs

Assumption applied

SUTs

at basic prices

- Valuation matrices

- Split domestic use table

and import use table

Deflation

processes

- Valuation matrices

- Split domestic use table

and import use table

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2.25. The diagram in Figure 2.2 may be visualized as the letter “H” with the left vertical arm

representing SUTs and IOTs in current prices and the right vertical arm representing the SUTs and

IOTs in previous years’ prices. The horizontal transition represents the deflation process using, for

example, a combination of prices, volume indicators and rates of the previous year applied to the

volumes.

2.26. The SUTs in current prices are decomposed into the component parts (imports and the

valuation matrices on the left-hand side of the H-Approach), each of which is deflated separately

as appropriate (the join in the middle), and then added back to get to a purchasers’ prices valuation

in previous years’ prices (the right-hand side). This means that basic prices play the dominant role

in the process, and the initial compilation flow is from top left to middle left and that of deflation

to middle right, and then to top right.

2.27. When final use deflators are deemed to be better for final use components at purchasers’

prices, then the H-Approach allows for the use of higher quality deflators, which are perhaps more

appropriate. In these types of examples, it is possible to work with purchasers’ prices where the

data are believed to be more reliable, making appropriate adjustments, working from top right to

middle right, then onto middle left and to top left, and still ensuring a balance at each stage.

Similarly, if high quality volume indicators are available, then this can better inform, for example,

the step between the middle left and middle right with adjustments as appropriate.

2.28. For balancing purposes, areas such as value added tax (VAT) on products and changes in

inventories may be separated out as balanced matrices (whereby the impact on production, income

and expenditure is equal and the matrices are in balance) to avoid any balancing adjustments, but

this may overcomplicate the system.

2.29. It is important to note that the scheme presented in Figure 2.2 should not be taken as one

to be implemented as a whole. In practice, for example, if a country wants to focus exclusively on

the compilation of annual SUTs, the focus should be on the compilation steps of the SUTs within

the bold line box in Figure 2.2 which are recommended in order to achieve balanced SUTs in

current and in volume terms. If, however, a country wants to compile SUTs and IOTs, all the steps

in Figure 2.2 should be completed in order to guarantee important feedback loops and enhance the

quality of the tabulations.

2.30. When planning for the compilation of SUTs, it is useful to keep in mind the compilation

approach in Figure 2.2, since it is naturally linked to the production of time series of SUTs (and

IOTs) both in current prices and in previous years’ prices using chain-linked volumes. More detail

of this may be found in chapter 9. Although SUTs in volume terms for one period can be compiled

using SUTs in current prices for one period and deflators, the preferred approach contains a time-

series dimension and the following principles:

• To compile SUTs in volume terms, the following are needed:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

 Balanced set of SUTs at purchasers’ prices in current prices for the present year and

the previous year

 Deflators for each flow

• Previous year’s SUTs in current prices of that year are needed to derive volume estimates.

The SUTs in figure 2.2 are balanced for illustrative purposes. In reality, however, they may

be unbalanced and an iterative balancing process may be necessary. This allows for the first

SUTs in previous years’ prices to be compiled.

• Each transition stage is created in a balanced format which provides a much easier systematic

process and build. This means that each of the transition matrices covering taxes, subsidies,

trade and transport margins and import of goods and services will be balanced, in other

words, the supply-side row total will equal the use-side row total.

• For some of the variables, like household final consumption expenditure, there are already

present deflation approaches using consumer price indices to generate the corresponding

estimates in previous years’ prices. These estimates are likely to differ from those generated

with the H-Approach but would also feature in the reconciliation and balancing process of

the estimates, and form an example of working from right to left through balanced

adjustments.

2.31. The H-Approach provides a transparent, coherent and consistent approach for the

compilation and balancing of SUTs. For example, balancing adjustments to one part of the SUTs

can be assessed in terms of their impact on other areas of the SUTs, and also in terms of time

series.

2. PSUTs

2.32. The SUTs described in the previous sections are part of the 2008 SNA framework. As such,

they reflect the production boundary of the SNA and they are compiled in monetary units. The

tables can, however, be extended to include the environment as providers of natural inputs into the

economy and as absorbers of residuals from the economy. The extension of these tables and, more

generally, of the accounting framework of the SNA to include environmental considerations is

carried out in the 2012 SEEA. The SEEA enables the analyses of the interaction between the

environment and the economy, such as an assessment of the use of natural resources, the generation

of waste by the economy and waste flows into the environment.

2.33. The SEEA central framework comprises a sequence of accounts – namely, the SUTs in

monetary and physical units, the asset accounts in physical and monetary units, and environmental

activity accounts and related flows. The present Handbook covers the SUTs of the SEEA and, in

particular, since the monetary SUTs of the SEEA are the same as the SNA, the Handbook focuses

on the PSUTs of the SEEA. For additional information on the complete set of accounts of the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

SEEA, the reader is directed to the publication United Nations, European Commission, FAO, IMF,

OECD and World Bank (2014).

2.34. PSUTs are used to assess how an economy supplies and uses energy, water, materials, and

also their changes in production and consumption patterns over time, and therefore, in combination

with data from monetary SUTs, changes in productivity and intensity in the use of natural inputs

and the release of residuals can be examined. The structure of PSUTs is based on the monetary

SUTs with extensions to incorporate a column for the environment and rows for natural inputs and

residuals.

2.35. Table 2.4 and Table 2.5 provide the simplified structure of the physical supply table and

use table, respectively. In order to address specific environmental domains (for example,

accounting for water, energy, timber and other resources), these tables are compiled for a

disaggregation of products and industries which is relevant for the particular environmental

domain of interest. In the case of energy, for example, the products of interest that can be explicitly

shown in the table include coal, peat and peat products, natural gas and others. The industries of

interest include the main suppliers of energy products (for example, electricity generation,

manufacture of gas and other products) and the main users of energy products (for example,

manufacturing, transportation and others). These tables are compiled in monetary units – within

the SNA context – and in physical units as shown in Table 2.4 and Table 2.5.

2.36. In the PSUTs, the SUTs of the SNA are augmented to include a block of rows for “natural

inputs” and a block of rows for “residuals”. The block for natural inputs is used to describe the

flows from the environment to the economy; in other words, this block describes the extraction of

natural inputs (for example, water, energy resources and others) from their location in the

environment as a part of economic production processes or that are directly used in production.

Natural inputs may be, first, natural resource inputs, such as mineral and energy resources or timber

resources; second, inputs from renewable energy sources, such as solar energy captured by

economic units; or, third, other natural inputs such as inputs from soil (for example, soil nutrients)

and inputs from air (for example, oxygen absorbed in combustion processes) (2012 SEEA, para.

2.89). When an industry, for example, extracts water as part of the economic production process,

this is recorded in the block of natural inputs in the use table in Table 2.5. It is assumed that the

environment provides (that is, supplies) all the natural inputs that are used into the economic

production process.

2.37. The blocks for “residuals” represent the flows of solid, liquid and gaseous materials, and

energy, that are discarded, discharged or emitted to the environment (for example, emission to air)

by establishments and households through processes of production, consumption or accumulation

but that may also flow within the economy, as is the case when, for example, solid waste is

collected as part of a waste collection scheme (2012 SEEA, para. 2.92).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2.38. The block for residuals in the supply table (Table 2.4) represents the flows of waste from

the economy to the environment and thus it includes the generation and disposal of waste during

economic production activities (generation of waste by industries) and generated during final

consumption (mainly by households). While the block of residuals in the use table (Table 2.5)

shows, for example, the collection and treatment of waste and other residuals by economic

activities, the accumulation of waste in controlled landfills and the residuals flows direct to the

environment.

Table 2.4 Schematic view of the physical supply table

Table 2.5 Schematic view of the physical use table

2.39. The supply and use identity applies to both physical and monetary flows. For each product

measured in physical terms (for example, cubic metres of timber), the quantity of output and

imports (total supply of products) must equal the quantity of intermediate consumption, household

final consumption, gross capital formation and exports (total use of products). The equality

between supply and use also applies to the total supply and use of natural inputs and the total

supply and use of residuals. In addition to the supply and use identity, the PSUTs incorporate the

Agriculture,

forestry, etc.

Mining and

quarrying

… Services

Mineral and energy resource

Water

…

Agriculture, forestry, etc.

Ores and minerals; etc.

…

Services

Solid waste

Wastewater

…

Empty cells by definition

C ells may contain relevant flows

Total supply

by residuals

Residuals

Residuals generated by industry

Residuals

generated

by final

consumtpion

Residuals from

scrapping and

demolition of

produced assets

Products

Output by product by industry

Imports by

product

Total supply

by product

Total

Natural

inputs

Flows from

the

environment

Total supply

by natural

inputs

Industries

Imports

Final

consumtpion

Gross capital

formation/

Accumulation

Environment

Agriculture,

forestry, etc.

Mining and

quarrying

… Services

Mineral and energy resource

Water

…

Agriculture, forestry, etc.

Ores and minerals; etc.

…

Services

Solid waste

Wastewater

…

Empty cells by definition

C ells may contain relevant flows

Residuals

C ollection and treatment of waste and other

residuals

Accumulation of

waste in

controlled

landfilled

Intermediate consumption by product and by

industry

Final uses by product and by category

Total use by

product

Total use by

residuals

Residual

flows direct

to the

enviornment

Total

Natural

inputs

Extraction of Natural inputs

Total use by

natural

inputs

Industries

Exports

Final

consumtpion

Gross capital

formation/Accu

mulation

Environment

Products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

input-output identity, implying that the total flows into the economy either are returned to the

environment or accumulate in the economy.

C. Overview of IOTs

2.40. For many analytical purposes, a transformation from a pair of SUTs into a single IOT where

total input (row totals) and total output (column totals) are equal brings considerable advantages.

IOTs have algebraic properties that make them particularly suitable for analyses that enable

estimates of the effects of changing relative prices, labour and capital requirements in the face of

changing output levels, the consequences of changing patterns of demand and so on. They may

also be used as the basis for an expanded version that may be used to estimate the demands made

by the economy on the environment (2008 SNA, para. 28.35).

2.41. An IOT is essentially derived from the use table, where either the columns representing

industries are replaced by products or where the rows representing the products are replaced by

industries through a transformation process which involves a range of assumptions. The resulting

intermediate consumption matrix is then square, showing products in both rows and columns or

industries in both. In both cases, the row totals for the complete matrix match the column totals

for the complete matrix, product-by-product matrix or industry-by-industry matrix as the case may

be (2008 SNA, para. 28.32). Of course, the classifications in the IOTs coincide with those in the

SUTs, as the former is a transformation of the latter.

2.42. It is recommended that the IOTs be derived from SUTs. The IOTs derived from the SUTs

further describe the interrelationships between industries and products, along with the sale and

purchase relationships between producers and consumers within an economy. They can be

produced to illustrate flows between the sales and purchases (final and intermediate) of industry

outputs (referred to as industry-by-industry tables) or to illustrate the sales and purchases (final

and intermediate) of product outputs (referred to as product-by-product tables).

2.43. The derivation of IOTs from the system of SUTs may also reveal inconsistencies and

weaknesses in the SUTs. This is made possible by the powerful quality-related feedback loop from

the IOTs to the SUTs and vice versa.

2.44. Table 2.6 provides a simplified IOT where the columns of the original use table referring

to industry-based structures are transformed into product-based structures. The relations between

output and input are now relations between products and primary inputs necessary to produce

outputs in similar production units. Primary inputs are inputs that are not outputs of other

industries. They include the imports of goods and services and the components of GVA, such as

compensation of employees, and others. They are necessary to the production process but are not

produced anywhere in the domestic economy.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Table 2.6 Simplified IOT (product by product)

2.45. For the transformation of SUTs into IOTs, various assumptions have to be made or

adjustments are required based on industry-by-industry or product-by-product assumptions:

• Product-by-product IOTs: these may be compiled using either the product technology

assumption (whereby each product is produced in its own specific way, irrespective of the

industry where it is produced) or the industry technology assumption (whereby each industry

has its own specific means of production, irrespective of its product mix).

• Industry-by-industry IOTs: these may be compiled using either the fixed industry sales

structure assumption (whereby each industry has its own specific sales structure, irrespective

of its product mix) or the fixed product sales structure assumption (whereby each product

has its own specific sales structure, irrespective of the industry where it is produced).

A mixture of both assumptions can also be applied by implementing a hybrid technology

assumption. The correct use and understanding of the terminology, transformation process and

assumptions applied are covered in more detail in chapter 12 of this Handbook.

2.46. The selection of the appropriate type of IOT – product-by-product or industry-by-industry

– depends on a number of statistical and practical considerations. For example, industry-by-

industry IOTs are closer to statistical sources and actual market transactions. Product-by-product

IOTs are believed to be more similar in terms of cost structure and production activities. This view

has been somewhat weakened, however, by changes introduced in the 2008 SNA, such as the strict

implementation of changes in economic ownership.

2.47. In the IOTs, two identities of the SUTs system are reduced to a single type of identity. It is

typical for IOTs that, for each product or industry, the input equals output and total input equal

total output.

2.48. The figures of total output and total input by product are the same as total supply and total

use by product of the SUTs – this holds for product-by-product IOTs. The industry-based

structures are transformed into product-based structures. In this transformation, the final use data

Agriculture,

forestry, etc.

Ores and

minerals;

etc.

… Services

Final

consumption

Gross capital

formation

Exports

Agriculture, forestry, etc.

Ores and minerals; etc.

…

Services

Imports

Value added

Value

added

Total

Empty cells by definition

Total

Intermediate consumption by product

Final uses by product and by category

Total use

product

Value added by component

Intermediate consumption of imported products

Products

Final uses

Total supply

Total final uses by category

Final use of imported products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

are left unchanged. The transformation only rearranges the data on the basis of the production

matrix of the intermediate use table by applying certain analytical assumptions to the relations

between primary and secondary outputs.

2.49. In general, and for analytical purposes, it is recommended to separate the use table into the

use table for domestic output and the imports use table. More detail on the compilation of the

domestic use table and the imports use table may be found in chapter 8.

2.50. Box 2.3 and Box 2.4 show a simplified numerical example of a sequence of tables – based

on the SUTs shown in Box 2.1 and Box 2.2 – necessary for compiling product-by-product IOTs

and industry-by-industry IOTs, respectively.

Box 2.3 SUTs and product-by-product IOTs

In product-by-product IOTs, all inputs are allocated to similar production units. They are derived from the SUTs

system on the basis of analytical assumptions (see chapter 12 for further detail on the derivation of IOTs).

Product-by-product IOTs are further away from statistical sources than industry-by-industry IOTs.

Supply table

Agriculture

Manufacturing

and construction

Services

Agriculture

270 30 50 350 20 370

Manufacturing and

construction

430 100 540 50 590

Services

20 40 550 610 30 640

Total 300 500 700 1500 100 1600

Use table Use table for domestic output

Agriculture

Manufacturing

and construction

Services

Final

consumption

expenditure

Gross capital

formation

Exports

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agriculture

34 59 93 131 21 32 370

Agriculture

30 50 90 130 20 30 350

Manufacturing and

construction

106 119 61 139 103 62 590

Manufacturing and

construction

90 100 54 136 100 60 540

Services

70 112 75 291 61 31 640

Services

60 100 70 290 60 30 610

Taxes less subsidies

on products

5 12 52 6 1 80

Imports

30 40 15 5 5 5 100

GVA

86 205 459 750

Taxes less subsidies on

products

5 12 52 6 1 80

Total 300 500 700 613 191 126

GVA

86 205 459 750

Total 300 500 700 613 191 126

Empty cells by definition

Imports use table

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agriculture

4 9 3 1 1 2 20

Manufacturing and

construction

16 19 7 3 3 2 50

Services

10 12 5 1 1 1 30

Total 30 40 15 5 5 5 100

Input-output table for domestic output (product by product)

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agriculture

34.08 52.23 83.69 130.00 20.00 30.00 350.00

Manufacturing and

construction

113.17 111.84 18.99 136.00 100.00 60.00 540.00

Services

73.23 114.19 42.58 290.00 60.00 30.00 610.00

Imports

37.73 46.07 1.20 5.00 5.00 5.00 100.00

Taxes less subsidies on

products

4.58 4.83 11.59 52.00 6.00 1.00 80.00

GVA

87.21 210.84 451.95 750.00

Total 350.00 540.00 610.00 613.00 191.00 126.00

Input table for imports (product by product)

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agriculture

4.91 10.75 0.34 1.00 1.00 2.00 20.00

Manufacturing and

construction

20.22 21.68 0.11 3.00 3.00 2.00 50.00

Services

12.60 13.65 0.74 1.00 1.00 1.00 30.00

Total

37.73 46.07 1.20 5.00 5.00 5.00 100.00

Products

Final use

Total

Products

Industries

Final use

Total

Products

Output

Products

Final use

Total use

Industries

Final use

Total

Products

Industries

Final use

Industries

Domestic

supply

Imports

Total

supply

Products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 2.4 SUTs and industry-by-industry IOTs

In industry-by-industry IOTs, inputs are allocated to industries. They are derived from the SUTs system on the

basis of pragmatic assumptions. The intermediate input of industries consists of output of industries rather than

products (of industry adjusted products) (see chapter 12 for details on the derivation of IOTs).

Industry-by-industry IOTs are closer to statistical sources and actual observations than product-by-product

IOTs.

D. Structure of SUTs and IOTs: basic elements

2.51. Defining the structure of SUTs and IOTs is a principal first step and depends on a number

of basic elements which form the backbone of these tabulations. These elements include:

• Principles of the accounting system underlying the SNA applied to SUTs and IOTs

• Classification of economic activities and its level of detail

• Classification of products and its level of detail

Supply table

Agriculture

Manufacturing

and

construction

Services

Agriculture

270

350 20

370

Manufacturing and

construction

430 100

540

50 590

Services

20 40 550

610 30 640

Total 300

500 700 1500

100

1600

Use table

Use table for domestic output

Agriculture

Manufacturing

and

construction

Services

Final

consumptio

expenditure

Gross

capital

formation

Exports

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross capital

formation

Exports

Agriculture

131 21

370

Agriculture

30 50 90 130 20 30 350

Manufacturing and

construction

106 119

139 103

62 590

Manufacturing and

construction

90 100 54 136 100 60 540

Services

70 112

75 291

31 640

Services

60 100 70 290 60 30 610

Taxes

less subsidies

on products

52 6

Imports

30 40 15 5 5 5 100

GVA

86 205

459 750

Taxes

less subsidies on

products

4 5 12 52 6 1 80

Total 300 500 700 613 191 126

GVA

86 205 459 750

Total 300 500 700 613 191 126

= Zero by definition

Imports use table

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross capital

formation

Exports

Agriculture

4 9 3 1 1 2 20

Manufacturing and

construction

16 19 7

3 3 2

Services

10 12 5 1 1 1 30

Total 30 40

15 5 5

5 100

Input-output table for domestic output (industry by industry)

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross capital

formation

Exports

Agriculture

26.78 43.70 72.72 112.31

19.25 25.24 300.00

Manufacturing and

construction

78.17 90.47 55.30 138.46 85.28 52.32

500.00

Services

75.05 115.83 85.97 305.23

75.47 42.45 700.00

Imports

30.00 40.00 15.00 5.00 5.00 5.00 100.00

Taxes less subsidies on

products

4.00 5.00 12.00

52.00 6.00

1.00 80.00

GVA

86.00 205.00 459.00

750.00

Total 300.00 500.00 700.00 613.00 191.00 126.00

Input table for imports (industry by industry)

Agriculture

Manufacturing

and

construction

Services

Final

consumption

expenditure

Gross capital

formation

Exports

Agriculture

3.71 7.69 2.61 0.86 0.86 1.61 17.34

Manufacturing and

construction

13.74 16.69 6.16 2.54

2.54 1.83

43.50

Services

12.55 15.62 6.23 1.60

1.60 1.56

39.17

Total 30.00

40.00 15.00 5.00

5.00 5.00

100.00

Industries

Final use

Total

Industries

Final use

Total

Products

Total

Industries

Final use

Total use

Industries

Final use

Total

Products

Industries

Final use

Industries

Domestic

supply

Imports

Total

supply

Products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Choice of the statistical units

• Valuation

2.52. Considerations on these elements reflect the specific context in which the tables are

compiled, which include the analytical objectives, the data availability, the economic structure of

the country, and others. Each of these elements is described below.

1. Principles of the accounting system underlying the SNA applied to SUTs and IOTs

2.53. The accounting system underlying the SNA is derived from broad bookkeeping principles

and is applied to the structure and links in the SUTs and IOTs. There are three bookkeeping

principles underlying the SNA accounting system:

• Vertical double-entry bookkeeping, also known as double-entry bookkeeping

• Horizontal double-entry bookkeeping

• Quadruple-entry bookkeeping

2.54. The main characteristic of vertical double-entry bookkeeping is that each transaction leads

to at least two entries, traditionally referred to as a credit entry and a debit entry. This principle

ensures that the total of all credit entries and all debit entries for all transactions are equal, thus

permitting a check on consistency of accounts for a single unit. Each transaction requires two

entries.

2.55. The concept of horizontal double-entry bookkeeping is useful for compiling accounts that

reflect the mutual economic relationships between different institutional units in a consistent

manner. It implies that, if unit A provides something to unit B, the accounts of both A and B show

the transaction for the same amount: as a payment in A’s account and as a receipt in B’s account.

Horizontal double-entry bookkeeping ensures the consistency of recording for each transaction

category by counterparties. For example, dividends payable throughout the economy should be

equal to dividends receivable throughout the economy once transactions with the rest of the world

are taken into account.

2.56. The simultaneous application of vertical and horizontal double-entry bookkeeping results

in quadruple-entry bookkeeping – which forms the accounting system underlying the SNA. It deals

in a coherent manner with multiple transactors or groups of transactors, each of which satisfies

vertical double-entry bookkeeping requirements. A single transaction between two counterparties

thus gives rise to four entries. In contrast to business bookkeeping, national accounts deal with

interactions among a multitude of units in parallel, and thus require special care from a consistency

point of view.

2.57. An account records and displays all the flows and stocks for a given aspect of economic

life. In each account, the sum of resources is equal to the sum of uses with a balancing item to

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

ensure this equality. Normally the balancing item will be an economic measure which is itself of

interest.

2.58. The accounts can be built up for different areas of the economy by employing a system of

economic accounts which highlight, for example, production, income and financial transactions.

In many cases, these accounts can be elaborated and set out for different institutional units and

groups of units (or institutional sectors). Usually a balancing item has to be introduced between

the total resources and total uses of these units or sectors and, when summed across the whole

economy, these balancing items constitute significant aggregates.

2.59. The accounting structure is uniform throughout the system and applies to all units in the

economy, whether they are institutional units, subsectors, sectors or the whole economy, although

some accounts (or transactions) may not be relevant for some institutional sectors.

2.60. The national accounting system uses two types of units and two corresponding ways of

subdividing the economy, which are quite different and serve separate analytical purposes:

• The first purpose, namely that of describing production, income, expenditure and financial

flows, and balance sheets, is served by grouping institutional units into institutional sectors

on the basis of their principal functions, behaviour and objectives. The national accounting

system enables a complete set of flow accounts and balance sheets to be compiled for each

sector, and subsector, and also for the total economy.

• The second purpose, namely that of describing processes of production and for input-output

analysis, is served by the system grouping establishments into industries on the basis of their

type of activity. An activity is characterized by an input of products, a production process

and an output of products.

2.61. Figure 2.3 shows in matrix form an overview of the structure of the SNA. The degree of

subdivisions of the columns and rows using the relevant classifications determines the degree of

detail of the accounts. The shaded rows and columns for goods and services and production by

industry indicate those parts of the system relevant for the compilation of SUTs and IOTs, and

clearly indicate that SUTs are at the core of the national accounts system.

2.62. The three approaches to measuring GDP (production, income and expenditure) are shown

in Box 2.5 and can be derived from the data in Figure 2.3 generating a single estimate of GDP.

2.63. All the aggregate components and detailed components are included in the SUTs and IOTs-

related part of the system.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 2.3 System of national accounts in matrix form

millions of euros

Table based on 2011 figures from Austria

Box 2.5 Three approaches to measuring GDP

Table based on 2011 figures from Austria

Receipts

Liabilities 1 2 3 4 5 6 7 8

303 492 226 258 74 612 165 648 770 009

Intermediate

consumption

Final

consumption

Gross capital

formation

Exports of goods

and services

578 360 578 360

Output at basic

prices

33 778 274 868 38 023 346 670

Taxes less

subsidies on

products

Value added at

basic prices

Primary incomes

and current

transfers from

rest of the w orld

79 669 - 5 057 74 612

Liabilities of

domestic sectors

Gross savings

Deficit on the

balance of

payments

Liabilities of

domestic sectors

157 871 40 743 198 614

Financial

liabilities of the

rest of the w orld

Imports of goods

and services

Primary incomes

and current

transfers to the

rest of the w orld

Financial liabilities

of the rest of the

w orld

Total excl.

balance

8 770 009 578 360 346 670 74 612 198 614

Closing

balance

Total excl.

balance

Assets

Opening balance

Assets of

domestic sectors

(real and

financial)

Financial assets

of the rest of the

w orld

Expenditures

Opening

balance

Goods and

services

Production by

industry

Income and

consumption

Accummu-

lation

Rest of the

world

Goods and

services

Production by

industry

Closing balance

Assets of

domestic sectors

(real and

financial )

Financial assets

of the rest of the

w orld

Income and

consumption

Accummulation

Rest of the world

millions of euros millions of euros millions of euros

Variable Value Variable Value Variable Value

Output at basic prices 578 360 Compensation of employees 144 343 Final consumptipon 226 258

+ Other taxes less subsidies on production 4 858

- Intermediate consumption - 303 492 + Consumption of fixed capital 53 469 + Gross capital formation 74 612

+ Net operating surplus 72 198

= Gross value added at basic prices 274 868 = Gross value added at basic prices 274 868 + Exports of goods and services 165 648

+ Taxes less subsidies on products 33 778 + Taxes less subsidies on products 33 778 - Imports of goods and services - 157 871

= Gross domestic product 308 647 = Gross domestic product 308 647 = Gross domestic product 308 647

Production approach

Income approach

Expenditure approach

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2. Classification of economic activities

2.64. The International Standard Industrial Classification of All Economic Activities (ISIC) is

the international reference classification of economic activities (also referred to as “industries”).

The fourth revision, ISIC Rev. 4, was issued by the United Nations in 2008 (United Nations, 2008).

Its main purpose is to provide a set of activity categories that can be used for collecting and

presenting internationally comparable statistics by economic activity.

2.65. In general, the scope of ISIC covers productive activity, that is, all economic activities

within the production boundary as described in the SNA (with one exception for activities in Class

9820– “Undifferentiated service-producing activities of private households for own use”). The

classification is used to classify statistical units such as establishments or enterprises, according to

the economic activity in which they mainly engage. All categories at each level of the classification

are mutually exclusive. ISIC Rev. 4 is the reference classification of production activities of the

2008 SNA.

2.66. The structure of ISIC consists of 21 sections, 88 divisions, 238 groups and 419 classes. The

principles and criteria used to define and delineate the categories are based on the inputs of goods,

services and factors of production, the process and technology of production, the characteristics of

outputs, and the use to which the outputs are put. At the class level of the classification, preference

has been given to the process and technology of production in defining individual ISIC classes, in

particular in the classes related to services. The list of products that defines a class is called the

principal products of that class. At the division and group levels, characteristics of outputs and the

use to which outputs are put become more important for the creation of analytically useful

aggregation categories.

2.67. At national and regional levels, there may be need for recourse to a level of detail that

reflects specific national and regional circumstances. It is important, however, that these

classifications are compatible with ISIC Rev. 4 at an aggregated level of detail. At its thirty-seventh

session, the United Nations Statistical Commission recommended that countries adapt their

national classifications in a way that allows them to report data at least at the two-digit level of

ISIC Rev. 4 without loss of information.

Examples of regional classifications include the

industrial classification used in the European Union, the second revision of the Statistical

Classification of Economic Activities in the European Community (NACE Rev. 2), which is

identical with ISIC Rev. 4 up to the two-digit level (divisions) of the classification. At lower levels,

NACE has provided more detail suitable for the European users of the classification. The

additional detail can always be aggregated to ISIC categories at the three-digit and four-digit

levels, within the same structure. The North American Industry Classification System (NAICS),

although it has a substantially different structure from ISIC, has been designed in a way that

See Official Records of the Economic and Social Council, 2002, Supplement No.4 (E/2006/24), chapter I, para. 3,

item 37/105 (b).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

statistical data collected according to NAICS can be re-aggregated into the two-digit divisions of

ISIC Rev. 4. The Australian and New Zealand Standard Industrial Classification (ANZSIC) was

revised in 2006. Its structure broadly follows the ISIC structure, so that categories at the division

and more detailed levels can be aggregated into the two-digit categories of ISIC.

2.68. An economic unit may engage in a variety of production activities. The classification of

the economic unit is made in accordance with the importance of the production activities. In this

regard, the activities of an economic unit are subdivided into principal activity, secondary activity

and ancillary activities. The principal activity of an economic entity is the activity that contributes

most to the value added of the entity, as determined by what is known as the “top-down method”.

This method follows a hierarchical order, starting with the identification of the relevant category

at the highest level (section) and progressing down through the levels of the classification to the

lowest level (classes). The effect of this top-down method is such that the principal activity need

not account for 50 per cent or more of the total value added of an entity or even that its generated

value added exceed that of all other activities carried out by the unit, although, in the majority of

cases, it will do so (United Nations, 2008, para. 57).

2.69. In practice, it is often impossible to obtain information on the GVA of the different

activities performed and the activity classification has to be determined by using substitute criteria,

such as employment and turnover.

2.70. Products resulting from a principal activity are either principal products or by-products.

By-products are products that are necessarily produced together with principal products, for

example, hides produced when producing meat by slaughtering animals. Since normal patterns of

horizontal integration have been taken into account when defining the ISIC classification, such

commonly integrated activities are usually included in the same ISIC class, even though the

outputs of the activities have quite different characteristics. Thus ISIC class 1010 “processing and

preserving of meat” also includes hides, skins, wool and feathers originating from slaughtered

animals (United Nations, 2008, paras. 57 and 120).

2.71. A secondary activity is a separate activity the products of which are ultimately intended for

third parties and that is not the principal activity of the entity in question. The outputs of secondary

activities are called secondary products, including any by-products associated with these outputs.

Most economic entities produce at least some secondary products.

2.72. Traditionally, the existence of by-products has been seen as creating problems in input-

output analysis as they would disturb supply-and-use relationships. Additional demand for the

principal product would therefore also result in more output of the by-product, without there

automatically being any additional demand for that increased output. In the case of more complex

production processes than meat and hides, for example in the chemical and electronic industries,

it will, however, be very difficult or even, in the absence of special technical insight, impossible

for the compilers of SUTs to identify by-products separately.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2.73. A distinction is made between principal and secondary activities on the one hand and

ancillary activities on the other. All economic units require some basic, routine services to support

their production activities. When they are provided in-house, they are called ancillary activities. In

general, an ancillary activity is a supporting activity undertaken within an enterprise in order to

create the conditions within which the principal and secondary activities can be carried out (2008

SNA, para. 5.36). Ancillary activities typically produce services that are commonly found as inputs

into almost any kind of economic activity. These outputs are always services and intended for

intermediate consumption within the same entity. They include, for example, the maintenance of

records, files or accounts in written form or on computers; the purchase of materials and

equipment; the provision of electronic and traditional written communication facilities; the hiring,

training, managing and paying of employees; the storing of materials or equipment; warehousing;

the provision of security and surveillance, and others.

2.74. Some of these activities are found in every economic entity. The output of an ancillary

activity is not explicitly recognized and recorded separately in the SNA. It follows that the use of

this output is also not recorded. All the inputs consumed by an ancillary activity – materials, labour,

consumption of fixed capital, and so forth – are treated as inputs into the principal or secondary

activity that it supports.

2.75. The following activities are not to be considered ancillary: producing goods or services as

part of gross fixed capital formation and research and development activities, which are considered

to be part of gross fixed capital formation in the 2008 SNA. These items will therefore appear as

either principal or secondary output. Goods that become embodied in the output of the principal

or secondary activities are not outputs of ancillary activities.

2.76. More details on principal products, secondary products and ancillary products specific to

the construction of the supply table may be found in chapter 5.

(a) Considerations for the compilation of SUTs and IOTs

2.77. In SUTs and IOTs, industries should be classified according to ISIC Rev. 4. The major

advantage of using established international industrial classifications is that comparability with

other types of economic statistics and the national accounts is not compromised. The choice is

therefore not which industrial classifications should be used in the SUTs and IOTs but rather the

level of detail.

2.78. At the working level, it is recommended to use the most detailed level of classification of

industry, taking into consideration user needs, the availability of data and the level of detail used

in the national accounts. Furthermore, certain compilation aspects also influence the choice of

working level, such as the distinction between industries which are allowed to deduct VAT and

those that are not, the distinction between market and non-market producers, and the explicit

identification of certain industry subdivisions that are relevant for the compilation of the trade and

transport margin matrices. In addition, the link between SUTs and the institutional sector accounts

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

should be reflected. These considerations are further elaborated in chapters 4–7. In general, the

level of detail in the published and disseminated SUTs differs from that at the working level: SUTs

tend to be published at a more aggregated level of detail which takes into account users’ needs and

confidentiality.

3. Classification of products

2.79. The international reference classification of products is the Central Product Classification

(CPC). The latest revision, CPC Version 2.1, was issued by the United Nations in 2015 (United

Nations, 2015). The primary purpose of CPC is to classify all goods and services that are the result

of production in the economy. CPC presents categories for all products that can be the object of

domestic or international transactions or that can be entered into stocks. It includes products that

are the output of economic activity, including transportable goods, non-transportable goods and

services. CPC in general follows the definition of products within the SNA.

2.80. The importance of the industrial origin of goods and services was underscored by the

attempt to group into one CPC subclass mainly the products that are the output of a single ISIC

class. Through their linkage to the criterion of industrial origin, the input structure, technology and

organization of production characteristics of products are also reflected in the structure of CPC.

The criterion of industrial origin of products is one of the classification principles applied by ISIC.

2.81. CPC was developed primarily to enhance harmonization among various fields of economic

and related statistics and to strengthen the role of national accounts as an instrument for the

coordination of economic statistics. It provides a basis for transforming basic statistics from their

original classifications into a standard classification for analytical use. As a general purpose

classification, CPC provides less detail than other specific classification systems in areas or

applications for which such systems are available, for example the Harmonized Commodity

Description and Coding System (Harmonized System). The Harmonized System

codes provide

building blocks for the part dealing with transportable goods and take into account the basic

categories of economic supply and use of products as specified in the SNA such as intermediate

consumption, final consumption, capital formation, and imports and exports.

2.82. CPC is a system of categories that are both exhaustive and mutually exclusive. This means

that if a product does not fit into one CPC category, it must automatically fit into another. In CPC

Version 2.1, in total there are 10 sections, 71 divisions, 329 groups, 1,299 classes and 2,887

subclasses. Each subclass in sections 0–4 of CPC is defined as the equivalent of one heading or

subheading or the aggregation of several headings or subheadings of the Harmonized System,

since the Harmonized System is a detailed classification of transportable goods that is widely

accepted for use in international trade statistics by virtually all countries. Other classifications of

The Harmonized Commodity Description and Coding System is the classification used for international trade

statistics.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

products may be used at country and regional level, however, these classifications are in general

broadly consistent with CPC Version 2.1. The Classification of Products by Activity (CPA) is

based on NACE – and therefore follows a different aggregation structure than CPC – and detailed

categories that are mostly aligned with CPC. Exceptions exist for areas where CPC deviates from

the Harmonized System, since CPA maintains a closer link with the Combined Nomenclature,

which is the European version of that classification.

2.83. CPC and ISIC are both general purpose classifications, with ISIC representing the activity

side. Each subclass of CPC consists of goods or services that are predominantly produced in a

specific class of ISIC. The relationship between industries and their products is complex, however,

and changes over time, and it should be noted that there has been no intention of establishing a

one-to-one correspondence between CPC and ISIC. Such an effort is considered neither practical

nor desirable as it might lead to an inadequate description of CPC categories, in particular at the

higher levels.

2.84. The classification of a product in the service part of CPC does not automatically imply that

the product cannot be a principal output of a goods producing industry. Thus the two CPC

divisions: (87) Maintenance, repair and installation (except construction) services; and (88)

Manufacturing services on physical inputs owned by others, both appear in the business and

production services section of CPC but the units carrying out these activities on a fee or contract

basis are classified in the same ISIC category as units producing the same goods or services for

their own account. The correspondence table between CPC Version 2.1 and ISIC Rev. 4 (see the

United Nations Statistics Division classification website at:

http://unstats.un.org/unsd/class/default.asp) shows that 125 subclasses of CPC division 88 are

defined to correspond to 125 manufacturing industry classes of ISIC. This implies that these

manufacturing services are the principal output (and not as might have been expected, the

secondary output) of the corresponding manufacturing activities. In other words, there are no

service industries producing these services. This example shows the need to ensure that services

of these kinds are correctly entered into the domestic output matrix, requiring a considerable

number of products.

2.85. Box 2.6 shows other classification of products, such as the Harmonized System, the

Standard International Trade Classification (SITC), the Classification by Broad Economic

Categories (BEC) and the Extended Balance of Payments Services Classification (EBOPS) and

how they relate to CPC. The basis for grouping products in the SUTs (and IOTs) is thus most

commonly an aggregation of CPC sections, divisions or groups (2008 SNA, para. 14.22).

Box 2.6 Other classifications of products

The Harmonized Commodity Description and Coding System, generally referred to as the Harmonized

System or simply HS, is a multipurpose international product nomenclature developed by the World Customs

Organization (WCO). The system is used by more than 200 countries and economies as a basis for their

collection of external trade statistics for goods. It is also extensively used by governments, international

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

organizations and the private sector for many other purposes, such as internal taxes, trade policies, monitoring

of controlled goods, rules of origin,

freight tariffs, transport statistics, price monitoring, quota controls,

compilation of national accounts, and economic research and analysis.

As a result of the intensive world-wide use and degree of detail that the Harmonized System provides, it is a

fundamental classification system and provides a key link for the definitions of all other classifications of

goods (including the goods part of CPC), as well as for the definition of the classes of ISIC. The latest version

now available is HS 2012.

The Harmonized System explanatory notes are a part of a commodity database giving the Harmonized System

classification of more than 200,000 products actually traded internationally. This high number of background

products also makes it evident that, at the level of external trade statistics (usually 5,000–10,000 products, as

most countries apply further subdivisions of the 5,000 Harmonized System

codes), there will be no

homogeneous products, and of course, even less so at the much higher level of aggregation applied in a SUTs

system.

SITC and BEC are both classifications of goods defined in terms of the Harmonized System

, and also

primarily used in relation to external trade data. SITC distinguishes around 3,000 products at its most detailed

level. It is primarily used as an alternative to the Harmonized System publication level of external trade

statistics, and there will usually be no advantage in applying it in SUTs, rather than using the Harmonized

System directly. Following the breakdown of products according to BEC (food, materials, fuel, capital goods,

transport equipment, consumer goods), these groupings may be used as a reference when deciding on uses of

some products but this breakdown is not applicable as the main product classification in the SUTs system.

Furthermore, BEC is not an international standard classification in the same sense as the Harmonized System

or SITC.

The relationship between CPC and SITC is similar to that between CPC and the Harmonized System, since

SITC also uses the subheadings of the Harmonized System as building blocks to create product groupings that

are more suitable for the economic analysis of trade. BEC is related to CPC through its close correlation with

SITC and is designed to serve as a means for converting external trade data compiled by use of the SITC into

end-use categories that are meaningful within the SNA framework. It is generally possible to rearrange entire

CPC subclasses into BEC categories through the correspondences between CPC and SITC, and between SITC

and BEC.

EBOPS 2010 (United Nations, European Commission, IMF, OECD and WTO, 2011) is a classification of

trade in services that was developed to provide further breakdowns of the BPM 6 classification so as to meet

a number of user requirements, including the provision of information required under the General Agreement

on Trade in Services. It builds upon the BPM 6 classification of services. In BPM 6, 12 main service categories

are identified and broken down into a list of standard and supplementary components. EBOPS 2010 consists

of a further breakdown of these components into more detailed sub-items. EBOPS 2010 also includes several

supplementary items for the recording of useful additional information regarding services transactions in

various sectors such as, travel and tourism or insurance services. Like the BPM 6 services classification,

EBOPS 2010 is primarily a product-based classification. Items of these classifications may be described in

terms of CPC. Correspondences cannot, however, be established in the areas of travel, construction, and

government goods and services, n.i.e., which focus on the mode of consumption of goods and services or the

status of the transactor, rather than on the type of product consumed. A detailed correspondence between

EBOPS 2010 and CPC Version

2, may be found online at

http://unstats.un.org/unsd/tradeserv/TFSITS/msits2010/annexes.htm.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

(a) Considerations for the compilation of products in the SUTs and IOTs

2.86. Based on the description of the product classifications and their level of detail, it is obvious

that products in the SUTs, even when a high level of detail is applied, as is the case with 2,000

products, will nonetheless represent aggregated groups of products when compared with the detail

applied in basic statistics, and even more so when compared with the real-world variety of

products. Accordingly, analogies to the notion of homogeneous products, which are often assumed

in standard economic theory, will in general be inappropriate, as there can be no homogeneous

products or production processes at the SUTs or IOTs level of aggregation. Many economies

usually consist of hundreds of thousands of producing units, of which virtually no two are

completely identical, and there are millions of different products and even more production

processes. It is therefore important to realize that national accounts and SUTs record economic

transactions, and not physically identifiable products or related technical production processes,

which will in general be outside the sphere of official statistics.

2.87. Even very detailed basic statistics already represent highly aggregated data when compared

to the number of real-world products. As previously mentioned, the HS contains descriptions of

200,000 products. Statistics on products are collected at a maximum detail of, say, 10,000

products, and only in selected areas such as external trade statistics and output from manufacturing

industries. Furthermore, products that are identical in a physical sense may be different in an

economic sense when they are sold at different prices to different purchasers. This may, for

example, happen because of the way transportation costs are invoiced. The concept of basic prices

is defined specifically to include this possibility. Statistics on the breakdown of products for

intermediate consumption will often give less detail than production statistics and may sometimes

be collected from enterprise units rather than establishment units, and in most cases the statistical

coverage of purchases is irregular or limited to certain industries, for example, mainly

manufacturing industries, but even in this case the compilers of SUTs may be confronted with the

task of further aggregation.

2.88. To gain a better understanding of the level of aggregation, it is useful to consider the

product definitions required when selecting items and collecting prices for compiling price indices

such as consumer price indices and producer price indices. Each item must be defined more

precisely than by just referring to even the most detailed product classification. The same applies

when collecting prices for use in the International Comparison Programme. Official statistics have

in these cases to make selections from a product universe at a much more detailed level than 10,000

product groups, in order to compile a sound price index. This places the notion of “homogeneity”,

as often applied in connection with SUTs and IOTs, in perspective (see 2008 SNA, para. 14.144).

2.89. As a result, the term “homogenous” in the context of the SUTs system usually means

“mutually exclusive”. As outlined above, international activity and product classifications aim at

mutually exclusive classification criteria. Yet within any group of products fulfilling this criterion,

there may be considerable “non-homogeneity” depending on the level of aggregation. The

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

classification of products in this (mutually exclusive) sense is statistically possible at any level of

aggregation but a product in the SUTs will usually represent a basket of products, and the contents

of the basket will, furthermore, vary from one cell to another along the rows of both the supply

table and the use table. For the classification of producer units into industries the same “mutually

exclusive” conditions basically hold, although the situation is somewhat different as the

statistically observed input structures will usually represent a mixture of intermediate consumption

structures for many individual products, some of which will also be produced in other industries.

As a consequence, industries producing mutually exclusive products can only be derived on certain

assumptions which do not in general form part of the compilation of the SUTs. Redefinitions (see

chapter 5) may be seen as an exception.

4. Other classifications relevant for SUTs and IOTs

2.90. The SUTs system distinguishes a large number of products and industries. Final uses,

however, often distinguish only final consumption, gross capital formation and exports at a very

aggregate level. The functional classifications help to support the compilation of SUTs and allow

for a wide range of other analyses. It is mainly the disaggregated SUTs which allow us to identify

the different purposes of expenditure on a product basis.

2.91. The SNA uses special classifications to analyse consumption and other outlays according

to the purpose for which the expenditure is undertaken. Such functional classifications and

associated detail – the Classifications of Expenditure According to Purpose (United Nations,

2000a) – can be found in chapter 29 of 2008 SNA, on satellite accounts and other extensions.

These classifications include, in particular, the classification of the functions of government

(COFOG); the classification of individual consumption according to purpose (COICOP), the

classification of the purposes of non-profit institutions serving households (COPNI) and the

classification of the outlays of producers according to purpose (COPP). The main purpose of these

classifications is to provide more detailed statistics for a wide variety of analytical uses. In 2018,

the United Nations issued the first revision of COICOP, COICOP 2018 (United Nations, 2018), to

reflect users’ need for more detail and several other issues that required a revision of the

classification.

2.92. For the SUTs, it is recommended that the lower-level detail be produced in the form of

disaggregated matrices as subsystems feeding into the central compilation of SUTs in current

prices and in volume terms. As a result, the correspondence between categories of these functional

classifications and CPC makes it possible to bring the basic data into the use tables.

2.93. The correspondence table between categories of CPC and COICOP has been established

and available on the United Nations Statistics Division classification website at

http://unstats.un.org/unsd/class/default.asp. When making decisions on the details of the product

classification to be applied in SUTs, the possibility of establishing transformation tables to

COICOP at group levels or class levels and to make use of the reverse transformation, from the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

products surveyed in the household budget surveys to the products of the SUTs, should be taken

into consideration. These transformation matrices are keys to the use of the consumer price index

(CPI) in the volume estimates, as sub-indices of the CPI will usually be based on the COICOP

classification. In addition, for the purposes of household budget surveys and purchasing power

parities (PPPs), COICOP is applied at a more detailed level, including as many as 300 or more

subclasses.

2.94. Table 2.7 shows the types of links and extensions. Some of the key areas are covered in

this section but more detail in terms of compilation is provided in chapter 6.

2.95. COPP provides detailed information on outlays of producers for current production,

infrastructure research and development, environmental protection, marketing and human resource

development. It should be noted that COPP is included here more for completeness of presentation

of the functional classifications. COPP is not widely used and does not fit well in the SUTs

framework, as its outlays include wages and other types of costs in addition to intermediate

consumption. In principle, COPP applies to all producers, whether market or non-market or for

own final use.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Table 2.7 Links between the use table and functional classifications

Use table

Households NPISH

General

government

Agriculture

Other services

Value added

Total

Final consumption expenditure by households (COICOP 2018)

Food and non-alcoholic

beverages

Alcoholic beverages, tobacco

and narcotics

Clothing and footwear

…

Personal care, social

protection and miscellaneous

goods and services

Total

Agriculture

Other services

Total

Final consumption expenditure by NPISH (COPNI)

Housing

Health

Recreation

and culture

…

Services

n.e.c.

Total

Agriculture

Other services

Total

Final consumption expenditure by government (COFOG)

General public

services

Defence

Public order

and safety

…

Social

protection

Total

Agriculture

Other services

Total

Products

Total

FINAL USE

COICOP

COPNI

Products

COFOG

Products

INDUSTRIES

Agricul-

ture

Other

services

Total

Final consumption expenditure

Gros s

capital

formation

Exports

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

5. Statistical units

2.96. In general, the same statistical unit is the basis for compiling the use table and for compiling

the supply table. Different choices of statistical units are available for the compiler and it is

important to have a clear understanding of the impact of the choice of different statistical units has

on the SUTs and on the IOTs.

2.97. Different types of statistical units may be defined (for example, enterprise group, local unit,

kind-of-activity unit, and others). For SUTs, however, the focus is on two specific statistical units:

enterprises and establishments (local kind-of-activity units).

2.98. An enterprise is defined as the view of an institutional unit as a producer of goods and

services (where institutional units are economic entities that have autonomy of decision making

and have clear links with the legal units). An establishment is an enterprise or part of an enterprise

that is situated in a single location and in which only a single productive activity is carried out or

in which the principal productive activity accounts for most of the value added.

2.99. The impact of globalization and the way in which multi-national businesses control and

operate their activities pose a number of challenges, including the basis of the statistical unit for

measurement of national activity versus global activity. Following the recommendations of the

2008 SNA, however, the establishment is the unit that is more suitable for the analysis of

production in which the technology of production plays an important role. The establishment is

therefore the recommended unit for the compilation of the production part of the national accounts

and therefore the compilation of SUTs. This means, as a rule, that multi-product enterprises must

be partitioned into smaller and more uniform units with regard to their kind of production, if

possible.

2.100. Trying to collect data on sub-establishment production processes as part of the input-output

compilation is an approach that has no natural limitation and that will, apart from the costs, almost

invariably become skewed by the specific knowledge and insight that the compilers happen to

possess and lead to non-transparent and uneven compilation processes.

2.101. In practice, the extent of partitioning enterprises into establishments varies across

countries, depending on whether the creation of establishments is based on a relatively modest

breakdown of institutional units or whether, alternatively, the starting point is a register of all local

producer units. The latter case follows the formal definitions set out in the 2008 SNA and would

lead to a purer activity classification than the former. Recommendations for partitioning vertical

and horizontal integrated enterprises are briefly outlined in Box 2.7.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 2.7 SNA recommendations on partitioning of vertically and horizontally integrated

enterprises

A horizontally integrated enterprise is one in which several different kinds of activities that produce different

kinds of goods or services for sale on the market are carried out simultaneously, using the same factors of

production (2008 SNA, para. 5.21).

Horizontal integration occurs when an activity results in end products with different characteristics. This could

theoretically be interpreted as activities carried out simultaneously using the same factors of production. In this

case, it will not be possible to separate them statistically into different processes, assign them to different units or

generally provide separate data for these activities. Another example would be the production of electricity

through a waste-incineration process. The activity of waste disposal and the activity of electricity production

cannot be separated in this case.

Within the SNA, a separate establishment should be identified for each different kind of activity wherever

possible (2008 SNA, para. 5.22).

A vertically integrated enterprise is one in which different stages of production, which are usually carried out

by different enterprises, are carried out in succession by different parts of the same enterprise (2008 SNA, para.

5.23). Vertical integration of activities occurs wherever the different stages of production are carried out in

succession by the same unit and the output of one process serves as the input to the next process. Examples of

common vertical integration include tree-felling and subsequent on-site sawmilling, mining of metal ores and

manufacture of basic iron and steel, operation of a clay pit combined with a brickworks or production of synthetic

fibres in a textile mill.

While the procedure for the treatment of vertically integrated activities could be applied to any unit, it should be

noted that the SNA recommends that, when a vertically integrated enterprise spans two or more sections of ISIC,

at least one establishment must be distinguished within each section. With such a treatment, activities of units

engaged in vertically integrated activities will not cross section boundaries of ISIC (2008 SNA, para. 5.26). If

this approach has not already been followed in basic statistics, the compilers of SUTs will exceptionally have to

deal with individual producer units.

6. Valuation in the SUTs

2.102. More than one set of prices may be used to value outputs and inputs depending on how

taxes and subsidies on products, and also transport charges and trade margins, are recorded. The

2008 SNA distinguishes three main valuation concepts of the flows of goods and services: basic

prices, producers’ prices and purchasers’ prices.

2.103. The valuation of the data for the use table (for example, intermediate and final

consumption) is different from the valuation of the data for the production side of the supply table.

In fact, the valuation of use table is based on the actual price paid by the users for the goods and

services (i.e., purchasers’ price) while the valuation of the production data in the supply table is

based on output at basic prices – this in line with the 2008 SNA.

2.104. In order to balance the SUTs, the same valuation should be used. For this purpose, specific

matrices have to be compiled for trade and transport margins and taxes and subsidies on products.

The compilation of these valuation matrices is an important component of the compilation of SUTs

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

and IOTs. Chapter 7 provides a detailed description of the compilation steps for the valuation

matrices and the compilation issues.

2.105. An overview of the three different valuations – basic prices, producers’ prices and

purchasers’ prices – is provided below. They differ as a result of the treatment of taxes on products

less subsidies on products, and trade and transport margins.

(a) Basic prices

2.106. Basic prices are the preferred method in the 2008 SNA for valuing output in the accounts.

This price basis reflects the amount receivable by the producer from the purchaser for a unit of

goods or services, minus any taxes payable, and plus any subsidy receivable on that unit as a

consequence of production or sale (for instance, the cost of production).

2.107. The value of output at basic prices reflects the sum of intermediate consumption of goods

and services at purchasers’ prices, compensation of employees, return to capital for market

producers’ own capital formation, and other taxes less subsidies on production. Other taxes on

production include items such as property taxes and business rates, business licences, motor

vehicle licenses, mission permits issued by governments under cap-and-trade schemes, and others.

Basic prices exclude any transport charges invoiced separately by the producer. When a valuation

at basic prices is definitely not feasible, then a proxy as close as possible to basic prices should be

used.

2.108. The basic price valuation is the preferred valuation for the construction of IOTs which in

turn are used in constructing structural models of the economy or modelling particular features of

economic behaviour. When compiling the IOTs, it is therefore necessary also to value the

purchases by products at basic prices, a process which is further explained in chapter 7.

(b) Producers’ prices

2.109. Producers’ prices may be thought of as the prices of goods and services “at the factory

gate”, so to speak. This valuation includes all taxes on production and taxes on products, for

example excise duties. Producers’ prices relate to basic prices as follow:

Producers’ prices equals basic prices

plus taxes on products (excluding invoiced VAT)

less subsidies on products.

2.110. Although the producers’ price valuation is valid and noted in the 2008 SNA, it not

recommended for use in the 2008 SNA. At the same time, this valuation still forms the basis for

some business survey data. Accordingly, if relevant, specific steps are needed to change data based

on business survey to basic prices, as appropriate, for use in national accounts and SUTs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2.111. Purchasers’ prices are those prices payable by the purchaser and include transport costs,

trade margins and taxes (unless the taxes are deductible by the purchaser). Purchasers’ prices are

defined as follows:

Purchasers’ prices equals producers’ prices

plus any non-deductible VAT or similar tax payable by the

purchaser

plus transport costs paid separately by the purchaser and not

included in the producers’ price.

plus trade margins.

2.112. Where taxes and subsidies on products and other taxes and subsidies on production are

concerned, some short definitions are provided below:

• Taxes on products include, in particular, value added taxes, taxes and duties on imports, and

taxes on products such as stamp taxes on the sale of petrol, diesel, alcoholic beverages and

tobacco.

• Subsidies on products include import subsidies and other subsidies on products.

• Other taxes on production consist of all taxes that enterprises incur as a result of engaging in

production, independently of the quantity or value of the goods and services produced or

sold. These may be payable on the land, fixed assets, business and property rates or labour

employed in the production process or on certain activities or transactions.

• Other subsidies on production consist of subsidies which resident producer units may receive

as a consequence of engaging in production, including in particular subsidies on payroll or

work force, subsidies to reduce pollution and grants for interest relief.

2.113. In the use table, transactions are recorded at purchasers’ prices. In the supply table,

domestic production is recorded at basic prices and imports by type of product at cost, insurance

and freight (CIF) prices. In the SNA and the balance of payments, total imports of goods are valued

at free on board (FOB) prices. Further details on these connections and the adjustments required

may be found in chapter 5, section D. Accordingly, additional columns are included in the supply

table in order to complete the valuation gap between total use and total supply of products. These

include information on trade and transport margins, taxes on products and subsidies on products.

(d) Value added tax

2.114. VAT is a wide-ranging tax usually designed to cover most or all goods and services. In

some countries, VAT may replace most other forms of taxes on products but it may also be levied

in addition to certain other taxes on products, such as excise duties on tobacco, alcoholic beverages

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

or fuel oils. VAT is a tax on products collected in stages by enterprises. Producers are required to

charge certain percentage rates of VAT on the goods or services that they sell. VAT is shown

separately on the sellers’ invoices so that purchasers know the amounts that they have paid.

Producers are generally not required, however, to remit to the government the full amounts of the

VAT invoiced to their customers because they are permitted to deduct the VAT that they

themselves have paid on goods and services purchased for their own intermediate consumption,

resale or gross fixed capital formation.

2.115. Deductible VAT is the VAT payable on purchases of goods or services intended for

intermediate consumption, gross fixed capital formation or for resale that producers are permitted

to deduct from their own VAT liability to the government in respect of VAT invoiced to their

customers. Non-deductible VAT is VAT payable by purchasers that is not deductible from their

own VAT liability, if any.

2.116. The SNA requires that the net system of recording VAT should be followed. In the net

system, outputs of goods and services are valued excluding invoiced VAT; imports are similarly

valued excluding invoiced VAT; and purchases of goods and services are recorded including non-

deductible VAT.

(e) Valuation in SUTs and IOTs

2.117. Box 2.8 presents an overview of the valuation in the compilation of SUTs and IOTs in a

simplified numerical example. this overview underlines the different valuation layers: the supply

table at basic prices including the transformation into purchasers’ prices is considered with the use

table at purchasers’ prices (total supply equals total use). In a second step, valuation matrices are

compiled – one for the trade and transport margins and the other for the taxes less subsidies on

products – in order to transform the use table from purchasers’ prices to basic prices. In this way,

the supply table at basic price can be considered in relation to the use table at basic prices (total

supply at basic prices equals total use at basic prices). The use table at basic prices is further split

between the domestic use table and imports use table at basic prices. The SUTs at basic prices are

the starting point for the compilation of IOTs, which are compiled at basic prices.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 2.8 Overview of the valuation in SUTs and IOTs

E. Compiling SUTs as an integral part of the national accounts

2.118. As mentioned before, the compilation of SUTs should be seen as an integral part of the

compilation of the national accounts. Figure 2.4 provides a general overview of how the

compilation of SUTs and IOTs fits within the compilation of national accounts conforming to the

same statistical standards (for example, 2008 SNA, BPM 6, 2012 SEEA, IMF Government Finance

Statistics, and others), and using the same basic sources generally used for the compilation of

national accounts.

Supply table at basic prices including a transformation into purchasers' prices Use table at purchasers' prices

Agric ul-

ture

Manuf. and

constr.

Services

Agric ul-

ture

Manuf. and

constr.

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agric ulture 270

30 50 350 20 370 28 20 418 Agriculture 38 65 103 155 24 33 418

Manufacturing 6 380 87 473 42 515 80 21 616 Manufacturing 115 123 64 167 85 62 616

Construction

50 13 67 8 75 22 4 101 Construction 12 16 6 24 39 4 101

Trade, transport and

communication

15 210 235 7 242 - 130 13 125

Trade, transport and

communication

21 2 2 98 1 1 125

Finance and business

services 6 17 240 263 11 274 15 289

Finance and business

services

14 54 43 128 32 18 289

Other servic es 4

8 100 112 12 124 7 131 Other services 14 35 23 41 10 8 131

Total 300

500 700 1 500 100 1 600 80 1 680 GVA 86 205 459 750

300 500 700 613 191 126

Trade and transport margins

Agric ul-

ture

Manuf. and

constr.

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agric ulture 3 5 4 13 2 1 28

Manufacturing 16 14 5 33 10 2 80

Construction 3 4 1 5 8 1 22

Trade, transport and

communication

- 22 - 23 - 10 - 51 - 20 - 4 - 130

Finance and business

services

Other servic es

Total

Taxes less subsidies on products

Agricul-

ture

Manuf. and

cons tr.

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agric ulture 1 1 6 11 1 20

Manufacturing 2 2 2 12 2 1 21

Construction 1 2 1 4

Trade, transport and

communication

1 1

1 9 1 13

Finance and business

services

1 1 12 1 15

Other servic es 1 6 7

Total 4 5 12 52 6 1 80

Supply table at basic prices Use table at basic prices

Agric ul-

ture

Manuf. and

constr.

Services

Agric ul-

ture

Manuf. and

constr.

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agric ulture 270 30 50 350 20 370 Agriculture 34 59 93 131 21 32 370

Manufacturing

6 380 87 473 42 515 Manufacturing 97 107 57 122 73 59 515

Construction 4 50 13 67 8 75 Construction 9 12 4 17 30 3 75

Trade, transport and

communication

10 15

210 235 7 242

Trade, transport and

communication

42 24

11 140 20 5 242

Finance and business

services 6 17 240 263 11 274

Finance and business

services

14 53 42 116 31 18 274

Other servic es

4 8 100 112 12 124 Other services 14 35 22 35 10 8 124

Total 300 500 700 1 500 100 1 600

Taxes less subsidies on

products

4 5 12 52 6 1 80

GVA 86 205 459 750

Total 300 500 700 613 191 126

Domestic use table for domestic output at basic prices

Agric ul-

ture

Manuf. and

constr.

Services

Final

cons.

exp.

Gross

capital

form.

Exports

Agric ul-

ture

Manuf. and

constr.

Services

Final

consumption

expenditure

Gross

capital

formation

Exports

Agric ulture

4 9 3 1 1 2 20 Agriculture 30 50 90 130 20 30 350

Manufacturing 12 17 6 2 3 2 42 Manufacturing 85 90 51 120 70 57 473

Construction 4 2 1 1 8 Construction 5 10 3 16 30 3 67

Trade, transport and

communication

2 4 1 7

Trade, transport and

communication

40 20 10 140 20 5 235

Finance and business

services 4 3 2 1 1 11

Finance and business

services

10 50 40 115 30 18 263

Other servic es 4 5 2 1 12 Other services 10 30 20 35 10 7 112

Total 30 40 15 5 5 5 100 Imports 30 40 15 5 5 5 100

Taxes less subsidies on

products

4 5 12 52 6 1 80

Empty cells by definition GVA 86 205 459 750

Total 300 500 700 613 191 126

BALANCED SUPPLY AND USE SYSTEM

Industries

Output

at basic

prices

Imports

Supply at

basic

prices

Trade and

transport

margins

Taxes

less

subsidies

Total supply

purchasers’

prices

Industries

Final use

TRANSFORMATION OF SUPPLY AND USE TABLES TO BASIC PRICES

Total use at

purchasers’

prices

Products

Output at basic prices

Industries

Final use

Total

Products

Industries

Final use

Total

Products

Industries

Final use

Total use at

basic prices

Industries

Output

at basic

prices

Imports

Supply at

basic

prices

Industries

Final use

COMPILATION OF VALUATION TABLES

Imports use table at basic prices

Final use

Total

Products

Total

Products

COMPILATION OF SEPARATE USE TABLES FOR DOMESTIC OUTPUT AND IMPORTS

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2.119. One important feature of the approach outlined in Figure 2.4 is the level at which the

traditional annual and quarterly balancing process of the national accounts and balance of

payments system takes place. Balanced macroeconomic data can be derived at a more aggregated

level by applying the production, income and expenditure approaches. A recommended, better

quality option is that the system be balanced at the same time for the institutional sector accounts

and the SUTs at a lower-level disaggregation of products and industries. In many countries, the

annual and quarterly estimates of GDP are obtained from the production, income and expenditure

approaches and reconciled using SUTs. Some countries have a long tradition and much experience

in using detailed production data based on establishments (local kind-of-activity units) as the

statistical unit for compiling GDP estimates following the production approach.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 2.4 Overview of SUTs and IOTs as part of the SNA compilation

1. Different approaches to measuring GDP

2.120. The three approaches to measuring GDP form the basis of estimating GDP both quarterly

and annually. The use of three different methods which, as far as possible, use independent sources

of information avoids sole reliance on one source and is conducive to greater confidence in the

overall estimation process. This in turn also underpins not only the quality of the key aggregates

but also of the underlying details. The SUTs combine the three approaches in a consistent manner.

Accumu-

lation

accounts

Supply and use tables at

basic prices

(balanced)

Institutional sector accounts

Total

economy

Rest of

the world

1. non-financial corporations

2. financial corporations

3. general government

4. households

5. NPISH

Product-by-

product

IOTs

Industry-by-

industry

IOTs

Goods

and

services

account

Production

account

Distribution

and use of

income

accounts

Balancing

Supply and

use tables at

purchasers'

prices

(balanced)

Valuation

matrices

(balanced)

Single estimate of GDP

(balanced)

System of National Accounts

Supply and use tables

Accounts

Supply and

use tables at

purchasers'

prices

(unbalanced)

Valuation

matrices

(unbalanced)

Production

approach

GDP

(unbalanced)

Income

approach

GDP

(unbalanced)

Expenditure

approach GDP

(unbalanced)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

(a) Production approach

2.121. The production approach looks at the contribution of each economic unit by estimating the

value of their output less the value of goods and services used up in the production process to

produce their output, this is also known as GVA. Using the production approach:

GVA at basic prices equals output at basic prices

less intermediate consumption at purchasers’ prices

and then,

GDP equals GVA at basic prices

plus taxes on products

less subsidies on products

GDP is also the balancing item of the production account for the whole economy.

2.122. The distinction between market and non-market producers (see 2008 SNA, para. 6.133, for

the definitions) is important for the determinants of both total output and gross value added which

is covered in this section. While the output of market producers is determined from the revenue

side, the output of non-market producers is calculated as the costs of all inputs including labour

cost and consumption of fixed capital. Box 2.9 provides an overview of the calculation of output

for market and non-market producers.

2.123. The estimate of output for producing units in the non-market sector is derived by summing

their costs, for example, intermediate consumption, compensation of employees, other taxes (less

subsidies) on production and consumption of fixed capital. GVA is the sum of compensation of

employees, other taxes (less subsidies) on production and consumption of fixed capital.

2.124. The production approach to measuring GDP, and the estimates of GVA, can be

implemented by using an industry dimension or by an institutional sector dimension. GVA is the

variable used when producing labour productivity estimates and also output per worker uses GVA

as the output measure.

Box 2.9 Calculation of output for market and non-market producers

Market producers and producers for own final use

Total output equals total sales of goods and services (as invoiced, excluding VAT)

(at basic prices) plus changes in inventories of work-in-progress and finished goods

plus output produced for own use, for example research and development (R&D), computer software

and construction (also known as own account capital formation) and household production of

agriculture products for own use

less purchases of goods or services for resale without further processing (thereby only including the

gross margin within output)

plus income earned in kind

less any taxes on products

plus any subsidies on products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Total intermediate equals total purchases of goods and services for use as inputs to the production process (excluding

consumption (at employment costs and fixed capital formation)

purchasers’ prices) less changes in inventories of materials and fuels

less any purchased or bought-in R&D, computer software (treated as capital expenditure, assuming

this is included in the purchases in the first place)

plus Financial intermediation services indirectly measured (FISIM)

plus any imputed insurance premium supplements

less any payments to employees such as income earned-in-kind

Gross value added equals total output (at basic prices)

(at basic prices) less total intermediate consumption (at purchasers’ prices)

Non-market producers

Total output equals total intermediate consumption (at purchasers’ prices)

(at basic prices) plus compensation of employees (labour costs)

plus imputed charge for consumption of fixed capital (sometimes called depreciation)

plus other taxes on production and imports

less other subsidies on production

Gross value added equals compensation of employees (labour costs)

(at basic prices) plus imputed charge for consumption of fixed capital (depreciation)

plus other taxes on production and imports

less other subsidies on production

Final consumption expenditure equals total intermediate consumption at purchasers’ prices

(at purchasers’ prices) plus gross value added at basic prices

equals total output at basic prices

less market output

less payment for non-market output

less output produced for own final use

equals non-market output

(b) Income approach

2.125. Using the income approach, GDP is obtained by adding together the income components

that make up value added. GDP by income approach covers only the income generated within the

domestic economy:

GDP equals compensation of employees

plus gross operating surplus and gross mixed income

plus other taxes less subsidies on production

plus taxes on products and imports.

less subsidies on products

The above income approach provides estimates of GDP market prices.

2.126. As its name suggests, the income approach adds up all income earned by resident

individuals or corporations in the production of goods and services and is therefore the sum of uses

in the generation of income account for the total economy (or alternatively the sum of primary

incomes distributed by resident producer units).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2.127. The income approach to measuring GDP can be analysed either by industry, by institutional

sector or by type of factor income. The type of factor income approach is often linked to the source

data and allows for the incorporation of various administrative data sources. These include, for

example, generating direct estimates of mixed income (using labour force data and administrative

data) and gross trading profit and loss (using company accounts data) as complementary estimates

and not as residuals.

2.128. Based on factor incomes, to estimate gross operating surplus the following categories are

added:

• Self-employment income (mixed income and quasi-corporations)

• Gross trading profits of private financial corporations

• Gross trading profits of private non-financial corporations

• Gross trading surplus of public corporations (financial and non-financial)

• Rental income

• Non-market consumption of fixed capital

and the following categories are deducted:

• Holding gains and losses on inventories

• Intermediate consumption of financial intermediation services indirectly measured (referred

to as FISIM)

2.129. Producing all three dimensions in a single, integrated SUTs framework provides a natural

link between the production account and generation of income account, both by industry and by

institutional sector. This approach also ensures a high degree of consistency and coherency across

the accounts.

2.130. It should be noted that the income approach to measuring GDP cannot be used to calculate

chained linked volume measures directly because it is not possible to separate income components

into prices and quantities in the same way as for goods and services. However, a chained linked

volume measure of the income based total can be obtained indirectly. The expenditure based GDP

deflator at market prices (also known as the index of total home costs) can be used to deflate the

current market price income based total estimate to provide a chained linked volume measure of

the total income component of GDP for balancing purposes.

2.131. In the expenditure approach, GDP is obtained by adding the final expenditures or uses by

consumers and producers of goods and services produced within the domestic economy. The total

is obtained from the sum of final consumption expenditure on goods and services by households,

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

NPISHs and government, gross capital formation (gross fixed capital formation on tangible and

intangible fixed assets, changes in inventories and acquisitions less disposals of valuables) and net

exports of goods and services.

2.132. Using the expenditure approach:

GDP equals Final consumption expenditure (households, NPISHs and government)

plus gross fixed capital formation

plus change in inventories

plus acquisitions less disposals of valuables

plus exports

less imports

2.133. The data for these categories are estimated from a wide variety of sources, including

business surveys, expenditure surveys, the government’s internal accounting system, surveys of

traders and the administrative documents used in importing and exporting goods.

2.134. To avoid double counting in this approach, it is important to classify consumption

expenditures as either final or intermediate. Final consumption expenditure involves the

consumption of goods purchased by or for the ultimate consumer or user. These expenditures are

final because the goods are no longer part of the economic flow or being traded in the market place.

Intermediate consumption, on the other hand, is consumption of goods and services that are used

or consumed in the production process. Gross capital formation is treated separately from

intermediate consumption as the goods (or services) involved are not used up within the production

process in an accounting period, except for depreciating over time.

2.135. Exports include all sales to non-residents, and exports of both goods and services have to

be regarded as final consumption expenditure, since they are final as far as the domestic economy

is concerned. Imports of goods and services are deducted because they are not part of the

production of the domestic economy but produced in another economy.

2.136. The expenditure approach to measuring GDP is also used to estimate chain-linked volume

measures of GDP. The chained-linked volume measure shows the change in GDP after the effects

of inflation have been removed.

2.137. Box 2.10 shows a numerical example of how a single estimate of GDP can be derived from

a balanced SUTs system by extracting the components of the production, income and expenditure

approaches to measuring GDP from the supply table and use table.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Box 2.10 Example of derivation of GDP from balanced SUTs

Table based on 2011 figures from Austria

The box below shows how a single estimate of GDP at market prices can be derived from the above balanced SUTs

system by extracting the components of the production, income and expenditure approaches to measuring GDP

from either the supply table or the use table.

Table based on 2011 figures from Austria

2. Linking SUTs to the institutional sector accounts

2.138. It is important to link the SUTs to the institutional sector accounts in order to have a

complete, consistent and integrated set of accounts, as highlighted in Figure 1.1. The SNA uses

two types of units and two ways to subdivide the economy. Both are quite different and serve

different analytical purposes. In order to describe production, income, expenditure and financial

Supply table at basic prices, including a transformation into purchasers' prices

Millions of euros

Agriculture Manufacturing Construction

Trade,

transport and

communication

Finance and

business

services

Other

services

Total

Wholesale

trade margins

Retail trade

margins

Transport

margins

Taxes on

products

Subsidies

on products

Total

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (15)

Agriculture (1)

8 782 0 0 0 0 0 8 782 3 271 12 052 1 052 873 274 386 - 107 2 479 14 532

Manufacturing (2)

796 182 982 643 1 808 133 44 186 405 124 590 310 995 29 777 19 061 2 540 21 041 - 49 72 370 383 364

Construc tion (3)

83 961 43 060 734 255 179 45 272 563 45 835 0 0 0 1 542 0 1 542 47 377

Trade (4)

1 4 773 311 54 204 640 257 60 187 600 60 787 - 31 301 - 21 040 0 586 0 - 51 755 9 032

Transport (5)

13 465 66 25 538 128 125 26 335 8 150 34 485 0 0 - 2 800 628 - 448 - 2 620 31 865

Communication (6)

160 1 781 139 43 912 1 253 982 48 228 6 234 54 463 472 1 021 9 3 592 - 34 5 059 59 522

Finance and business services (7)

29 8 902 698 7 588 106 909 3 381 127 508 7 061 134 569 0 0 - 22 4 865 0 4 842 139 411

Other services (8)

3 85 13 1 053 143 74 346 75 643 824 76 467 0 85 0 1 777 0 1 861 78 329

Total (9)

9 867 199 950 44 931 134 837 109 461 79 314 578 360 151 293 729 653 0 0 0 34 416 - 638 33 778 763 431

CIF/FOB adjustments on imports (10)

0 0 0 0 0 0 0 - 97 - 97 0 0 0 0 0 0 - 97

Direct purchases abroad by

residents

(11)

0 0 0 0 0 0 0 6 675 6 675 0 0 0 0 0 0 6 675

Total (12)

9 867 199 950 44 931 134 837 109 461 79 314 578 360 157 871 736 230 0 0 0 34 416 - 638 33 778 770 009

Total of w hich:

Market output (13)

9 763 195 916 41 462 127 401 88 330 18 116 480 989 0 480 989 0 0 0 0 0 0 480 989

Output for ow n final use (14)

104 4 029 3 468 2 134 19 890 2 670 32 295 0 32 295 0 0 0 0 0 0 32 295

Non-market output (15)

0 4 0 5 302 1 241 58 528 65 075 0 65 075 0 0 0 0 0 0 65 075

Use table at purchasers' prices

Millions of euros

Households NPISH

General

government

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture (1)

2 583 6 570 16 371 34 49 9 623 3 595 0 0 180 0 - 27 1 161 4 909 14 532

Manufacturing (2)

2 205 107 190 12 441 16 874 6 015 8 797 153 522 71 438 0 3 180 26 756 2 183 3 034 123 252 229 842 383 364

Construc tion (3)

105 2 440 9 528 2 446 3 907 1 604 20 029 1 667 0 0 25 155 0 - 38 563 27 348 47 377

Trade (4)

33 1 883 119 2 240 259 308 4 842 3 325 0 0 67 45 0 753 4 189 9 032

Transport (5)

14 4 386 267 8 399 822 321 14 208 5 833 0 3 370 0 0 0 8 453 17 656 31 865

Communication (6)

34 2 563 299 9 359 5 919 1 833 20 008 26 444 0 121 5 976 0 67 6 905 39 514 59 522

Finance and business services (7)

457 13 578 4 736 20 359 29 166 9 134 77 430 38 838 0 1 006 11 170 0 - 178 11 145 61 981 139 411

Other services (8)

8 382 59 1 171 415 1 794 3 829 14 923 5 416 53 373 113 107 1 567 74 500 78 329

Total at purchasers’ prices (9)

5 440 138 991 27 466 61 219 46 538 23 839 303 492 166 063 5 416 61 050 69 418 2 335 2 859 152 800 459 939 763 431

CIF/FOB adjustments on exports (10)

0 0 0 0 0 0 0 0 0 0 0 0 0 - 97 - 97 - 97

Direct purchases abroad by

residents

(11)

0 0 0 0 0 0 0 6 675 0 0 0 0 0 0 6 675 6 675

Purchases on the domestic

territory by non-residents

(12)

0 0 0 0 0 0 0 - 12 945 0 0 0 0 0 12 945 0 0

Total at purchasers’ prices (13)

5 440 138 991 27 466 61 219 46 538 23 839 303 492 159 792 5 416 61 050 69 418 2 335 2 859 165 648 466 517 770 009

Compensation of employees (14)

551 30 679 10 239 37 906 22 997 41 971 144 343 0 0 0 0 0 0 0 0 0

Other taxes less subsidies on

production

(15)

- 1 627 1 077 546 1 755 2 004 1 103 4 858 0 0 0 0 0 0 0 0 0

Consumption of f ixed capital (16)

1 845 12 750 1 542 10 917 18 934 7 480 53 469 0 0 0 0 0 0 0 0 0

Net operating surplus (17)

3 658 16 453 5 138 23 040 18 989 4 921 72 198 0 0 0 0 0 0 0 0 0

Gross operating surplus (18)

5 503 29 203 6 680 33 957 37 923 12 401 125 667 0 0 0 0 0 0 0 0

GVA (19)

4 427 60 959 17 465 73 618 62 923 55 475 274 868 0 0 0 0 0 0 0 0 0

Total input at basic prices (20)

9 867 199 950 44 931 134 837 109 461 79 314 578 360 0 0 0 0 0 0 0 0 0

Total

supply at

basic

prices

VALUATION

Total

supply at

purchasers’

prices

Adjustments

Products

INDUSTRIES

Imports

INDUSTRIES

FINAL USE

Total use at

purchasers’

prices

Agriculture

Manufacturing

Construc tion

Trade,

transport and

communication

Finance and

business

services

Exports

Total

Products

Adjustments

VALUE ADDED

Changes

inventorie

Other

services

Total

Final consumption expenditure

Gross fixed

capital

formation

Changes

valuables

Millions of euros

Total output at basic prices 578 360

Compensation of employees

144 343 Final consumption expenditure by Househods 159 792

- Intermedate consumption at purchasers’ prices - 303 492

+ Other taxes less subsidies on productio

4 858 + Final consumption expenditure by NPISH 5 416

+ Consumption of f ixed capital 53 469 + Final consumption expenditure by General government

61 050

+ Net operating surplus 72 198 + Gross fixed capital formation 69 418

= GV A 274 868 = GVA 274 868 + Changes in valuables 2 335

+ Changes in inventories 2 859

+ Taxes less subsidies on products 33 778 + Taxes less subsidies on products 33 778 + Exports 165 648

- Imports - 157 871

= GDP 308 647 = GDP 308 647 = GDP 308 647

Production approach

Income approach

Expenditure approach

Calculation of gross domestic product

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

flows, and balance sheets, the SNA uses institutional units which, on the basis of their principal

functions, behaviour and objectives, are grouped into institutional sectors like non-financial

corporations and financial corporations. For the institutional units, the full set of accounts is

covered in the system.

2.139. A simplified version of a table covering the main institutional sectors is shown in Table

2.8. Further details may be found in chapter 10, on linking the institutional sector accounts to the

SUTs.

Table 2.8 Simplified table linking the SUTs to the institutional sector accounts

2.140. When describing the processes of production (and input-output analyses), the system uses

the establishment as the statistical unit and groups it into industries on the basis of its principal

activity. For the establishment, only a limited set of accounts is feasible, namely those accounts of

the SUTs framework.

2.141. In order to show the relationships between the accounts of the production processes and

the accounts of the institutional units, a link table can be compiled as an integrated part of the

system. In this link table, a cross-classification of output, intermediate consumption, components

INSTITUTIONAL SECTORS

1 2 … n

1. Non-financial corporations

Total output

Market output

Output for own final use

Non-market output

Intermediate consumption

GVA at basic prices

Compensation of employees

Other net taxes on production and imports

Consumption of fixed capital

Operating surplus, net

Gross fixed capital formation

2. Financial corporations

Total output

Gross fixed capital formation

3. General government

Total output

Gross fixed capital formation

4. Households

Total output

Gross fixed capital formation

5. Non-profit institutions serving households

Total output

Gross fixed capital formation

6. Total

Total output

Gross fixed capital formation

INDUSTRIES

Total

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

of GVA (and possible other variables of industries) between the industries and the institutional

sectors is shown. This link table should help to ensure consistency of data compiled on the basis

of establishment and on the basis of institutional units. As both units are classified differently, the

link table also provides a picture of the relationships between output, intermediate consumption,

GVA, and other variables, originating in the different industries and institutional sectors.

3. Benefits of compiling SUTs as an integral part of the national accounts

2.142. There are a number of advantages of producing SUTs as an integral part of the national

accounts and this approach is therefore recommended in this Handbook.

2.143. From a methodological point of view, the following may be identified:

• SUTs provide the ideal framework for integrating the components of the three approaches

to measuring GDP both in current prices and in volume terms.

• When statistical discrepancies exist amongst the macroeconomic aggregates, it is less clear

where adjustments could be applied. Through their detailed examination of the supply and

use of products, however, the SUTs provide a powerful approach to identifying which areas

could be adjusted.

• SUTs allow for the data confrontation of different primary sources by bringing them together

into a single framework, and facilitate efforts to prioritize how resources could be allocated

to seek quality improvements.

• Where statistical information is incomplete or contradictory, as may happen with gross fixed

capital formation or household final consumption expenditure, alternative estimates can be

made in a transparent way using the SUTs framework, ensuring consistency and coherence.

• SUTs provide a full framework for establishing the connection between the various valuation

concepts in national accounts, from basic prices through to purchasers’ prices.

• SUTs form the ideal framework for estimating GVA through double deflation and GDP in

volume terms, while also ensuring coherence of deflation across the different areas.

2.144. In terms of practical benefits:

• SUTs offer new options to incorporate all existing information, including from primary

sources, on a consistent basis. This is also true for information that is only periodically

available, as well as a framework for making reliable estimates, including plausible

restrictions and identities.

• When SUTs are produced as an integral part of the national accounts, it is relatively easy to

compile IOTs. These IOTs derived from SUTs will be fully compatible and consistent with

all figures from the national accounts, adding to the credibility and analytical usefulness of

both products.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• SUTs that are consistent with the national accounts are normally produced in connection

with benchmarked macroeconomic data some two or three years after the initial preliminary

results of the national accounts are published. SUTs can also play a vital role in the

production of preliminary annual or even quarterly accounts.

2.145. Once the SUTs system is in place on an annual basis, the benefits are significant and can

take various forms:

• SUTs from the previous year can be updated with information available for the preliminary

year in order to produce a complete set of SUTs (albeit at a more aggregated level) that are

consistent with the preliminary figures. This procedure is a good method for revealing

inconsistencies in the aggregated preliminary figures at an early stage.

• SUTs can be used to incorporate new information; for example, when new detailed

information on total supply and exports is available earlier, then the structure of SUTs of the

previous year could be used to project SUTs for domestic output and imports.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Chapter 3. Business processes and production stages

A. Introduction

3.1. The compilation of monetary and physical SUTs and thus IOTs is viewed as part of a

statistical production process which starts from the identification of the objectives and users’ needs

to the dissemination of the tabulations and the evaluation of the production process. The various

stages of compilation of SUTs (and IOTs) presented in this Handbook follow those of the GSBPM

(United Nations Economic Commission for Europe, 2013). The GSBPM explicitly identifies and

organizes the compilation steps and the interdependencies between them in a generic statistical

business process. Thus it provides a useful flexible framework by which to describe the

compilation process for SUTs and IOTs.

3.2. It should be mentioned that country practices in the compilation of SUTs and IOTs vary

considerably, since they are specific to the particular context in which they take place. For

example, they depend on the specific institutional arrangements of the statistical system, the

statistical legal framework, the legal, political, regional and taxation arrangements, the statistical

units, the business registers, the range of processes, publication schedules, revision policies,

resources, data availability, confidentiality, and also the final outputs. Despite the great variability

in country practices, there are common steps in the compilation of SUTs and IOTs. In the

description of the compilation stages of the GSBPM in this Handbook, these common steps that

are flexible and applicable to all countries are identified.

3.3. There is an overarching framework within which the statistical production process takes

place and this should be taken into consideration in the design of the compilation process and also

in the actual compilation of SUTs and IOTs. This includes the statistical institutional arrangement

in the country and the data and metadata quality framework.

3.4. The objective of this chapter is to provide an overview of the compilation steps for SUTs

and IOTs. Section B presents an overview of the different institutional set-ups in various countries.

Section C outlines the GSBPM compilation stages that relate to SUTs and IOTs and, lastly, section

D provides a schematic summary of the compilation steps and their links with the relevant chapters

of the Handbook and a summary of the main recommendations, principles and guidelines for the

compilation of SUTs, IOTs, PSUTs and EE-IOTs that are covered by this Handbook. Annex A to

chapter 3 provides examples of institutional arrangements for the compilation of economic

statistics in countries.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

B. Institutional arrangements

3.5. The institutional arrangements are generally understood as a set of agreements between the

agencies involved regarding the division of responsibilities in the collection, processing,

compilation and dissemination of data. They are fundamental to an effective statistical system and

essential for the management of an integrated economic statistics programme. The functions and

responsibilities of the lead statistical agency in the country can be carried out more efficiently if it

is supported in its role by institutional arrangements such as advisory committees, relationship

meetings, memorandums of understanding, service-level agreements, technical cooperation and a

legal framework that protects the confidentiality and integrity of the data while allowing for the

sharing of data between partner statistical agencies (United Nations, 2013, para. 3.23).

3.6. Apart from the legal framework and other factors, institutional arrangements depend on the

kind of national statistical system that exists in a given country, namely, whether it is centralized

or decentralized. A national statistical service is considered centralized if the management and

operations of the statistical programmes are predominantly the responsibility of a single

autonomous government agency, and decentralized if the statistical programmes are managed and

operated under the authority of several government departments. Under this arrangement, a

particular agency is usually entrusted with the responsibility of coordinating the statistical

activities of the various departments.

3.7. In economic statistics, countries have different institutional arrangements under which

such bodies as, for example, the national statistical office and the central bank have different roles

and responsibilities. Countries often follow a decentralized approach, under which the collection

of economic statistics is split across different institutions within the country, so that, for example,

the national accounts (non-financial accounts) are compiled by the national statistics office, the

balance of payments and financial accounts are compiled by the central bank, and the government

finance statistics covering the public sector are compiled by the finance ministry.

3.8. When countries are considering either building or redesigning their systems or changing

the roles and responsibilities of the various institutions involved, the undertaking should be

approached with the aim of producing integrated economic accounts throughout the entire

statistical production process. The motivation for integrated economic statistics comes from the

benefits that such data sets provide for coordinated national and global policy initiatives in an

increasingly interconnected world. The integration of economic statistics involves the use of

common concepts, definitions, estimation methods and data sources for statistical reconciliation,

helping to improve the coherence and consistency of a wide range of economic statistics and to

reduce the respondent burden and overall costs. Integration therefore is not specific either to the

type of statistical system (centralized versus decentralized) or to the level of development of the

statistical system. This approach has the following prerequisites:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Adoption of the conceptual framework of the SNA and the SEEA as the umbrella framework

for organizing economic statistics

• Alignment of the interdependencies of the components of the statistical production process

(statistical units, classifications and others)

• Establishment of enabling institutional arrangements for statistical integration (United

Nations, 2013, para. 2.5)

3.9. Examples of institutional arrangements in different countries may be found in the annex to

this chapter. In general, it could be said that, beyond being conducive to the coherence and

consistency of official economic statistics, a centralized arrangement may provide more

comparability and harmonization both within the statistical system and with the statistical system

of other countries. Although the transition to an integrated system may incur large investment

costs, it would generate great benefits in terms of improved quality and reduced costs in both the

short term and the long term.

3.10. The roles and responsibilities of the various institutions in countries evolve over time and

aspects of the historical evolution of these arrangements are reflected in the country examples

covered in this chapter. One such example is that of Finland, where, in 2014, the compilation of

the balance of payments was transferred from the country’s central bank to its statistics office.

Finland now follows the practice of other countries such as Denmark, Ireland, Luxembourg, Malta,

Norway and the United Kingdom, where the balance of payments is compiled alongside the

national accounts within the statistical office and not in the central bank.

C. Overview of the Generic Statistical Business Process Model

3.11. The GSBPM describes, and defines, the set of business processes needed to produce

official statistics. It provides a standard framework and harmonized terminology to help statistical

organizations to modernize their statistical production processes, and to share methods and

components. The GSBPM can also be used for integrating data and metadata standards, as a

template for process documentation, for harmonizing statistical computing infrastructures, and to

provide a framework for process quality assessment and improvement. The GSBPM is a reference

model that can be used in a flexible manner to describe, document, organize and communicate the

statistical production process in question.

3.12. The GSBPM consists of a sequence of eight phases: (1) Specify needs; (2) Design; (3)

Build; (4) Collect; (5) Process; (6) Analyse; (7) Disseminate; and (8) Evaluate. An overview of

the phases, together with the sub-elements of each phase, may be seen in Figure 3.1.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 3.1 Phases of the GSBPM

3.13. The GSBPM is not a rigid framework in which all steps must be followed in a strict order;

rather, it helps to identify the possible steps in the statistical business process, and the

interdependencies between them. Although presentation of the GSBPM follows the logical

sequence of steps in most statistical business processes (for example, business surveys), in

different circumstances the elements of the model may occur in a different order. In addition, in

compiling SUTs and IOTs, some sub processes will be revisited a number of times, forming

iterative loops, in particular within the “Process” and “Analyse” phases.

3.14. This section focuses on the business processes in national accounts, in particular, the

compilation of SUTs, PSUTs and IOTs. The business process and stages of production covered in

this chapter therefore reflect the application of the underlying GSBPM. As a result, Figure 3.2

provides an overview of a simplified business processing model specific for the compilation of

SUTs, PSUTs and IOTs.

3.15. In the compilation of SUTs and IOTs, the sequential stages in the compilation of the

GSBPM may be summarized as follows, and as presented in Figure 3.2:

• Phases 1–3: Specify needs, design and build. This stage includes tasks related to the phases:

1 – “Specify needs”, 2 – “Design”, and 3 – “Build” of the GSBPM set out in Figure 3.1. It

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

covers all the pre-collection activities of setting up the system, including identifying users’

needs, designing the system, determining the size of the SUTs and IOTs, and other tasks.

• Phase 4: Collect. This relates to the activities of data gathering from various sources. In

general, the compilers of SUTs (and IOTs) rely on data already collected for the purposes of

national accounts which have often already been adjusted to fit into the national accounting

framework.

• Phase 5: Process. This stage corresponds to a number of activities related to the data

cleaning, adjustments and transformation that are needed in order to start putting the data

into an unbalanced SUT. This stage is very important in the compilation of SUTs and IOTs

and is therefore separated into two steps in Figure 3.2. The first corresponds to all the

activities necessary to put the data in the initial unbalanced SUTs. This involves data

cleaning, pre-processing, aggregation and disaggregation of the basic data and any other

adjustment to the basic data to fit into the national accounts concepts of the SUTs. The

second step in this phase corresponds to all the activities of setting up an initial (unbalanced)

set of SUTs at purchasers’ and basic prices and in current prices and volume terms.

• Phase 6: Analyse. This stage corresponds mainly to the activities of balancing (manual and

automated) SUTs and IOTs and the feedback loop to the source data to resolve

inconsistencies. As a result, there is a continuous loop between this and the previous phase,

making it possible to achieve balanced SUTs and IOTs. In this stage, the final output of the

compilation process is prepared, validated and finalized.

• Phases 7 and 8: Disseminate and evaluate. This stage refers to the activities related to the

dissemination of the output tables, which include the preparation of printed publications,

press releases and websites, the promotion of dissemination products and other tasks, as well

as the activities related to the evaluation of the production process and also of the output in

the light of internal or external feedback.

3.16. The grey boxes for each stage in Figure 3.2 include examples of the types of functions

undertaken. They are listed in no particular order of importance and are linked with one another.

3.17. The broad approach is to move and process data from left to right, with minimal backward

loops, even though effective feedback loops are critical at each phase, and the incorporation of

new, or improved, data deliveries are unavoidable. Good data version control at each stage is

needed, enabling the generation of a wide-range of outputs, articles and analyses such as revision

analysis.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 3.2 Simplified business processing model for compiling SUTs, IOTs, and PSUTs

3.18. Each phase should be viewed as cumulative, even when allowing for the iterative nature of

the balancing process. The incorporation of balancing adjustments should be viewed as a

cumulative step and not as creating a loop.

3.19. It is important to prepare proper documentation throughout the various compilation stages

and in particular during the stage of balancing and adjustment. The steps and links between the

source data through to the balanced data should be recorded and documented separately and

reviewed in subsequent balancing exercises to investigate source data incoherence, bias and other

factors. For example, moving from the original source data (such as business survey data,

administrative-based data, company accounts-based data, and other types of data) to the validated

2008 SNA data, a number of adjustments may need to be made in such areas as the following:

• Coverage (including exhaustiveness) adjustments

• Conceptual adjustments

• Quality adjustments

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Balancing and coherence adjustments

3.20. In general, for all the stages of the compilation process, it is important also to have in place

the following:

• Data version control – for example, data storage, conventions allowing easy access and

revision analyses, and also pre and post automated balancing analyses

• Clear controls and disciplines – for example, read and write access for each stage, setting

out operational standards, change controls and testing, and other measures

• Appropriate staffing – for example, ensuring that all staff are trained and skilled to undertake

the different functions and ensuring that sufficient staff are in place for each phase

• Clear organizational structure of the staff involved – for example, clear roles and

responsibilities of staff, as staff members can have more than one role in more than one phase

D. Overall strategy for the compilation of SUTs and IOTs

3.21. Within the stages of the overall production process presented in the previous section, the

“Process” and “Analyse” phases (5 and 6) have a particular structure in the compilation of SUTs

and IOTs. This section provides an overview of the steps that are generally undertaken to construct

SUTs and IOTs after the data have been gathered. In addition, since the compilation of SUTs and

IOTs is not seen as a one-time exercise but as part of a continuous programme, this section also

provides the strategy for compiling SUTs and IOTs in current prices and in previous years’ prices

for the first year of compilation and the subsequent years.

3.22. The first step in compiling SUTs and IOTs is to populate the various separate parts of the

supply table and use table (as shown in Figure 3.3) with the available data. This leads to the

construction of unbalanced SUTs which are then subjected to a balancing process to reconcile all

the entries.

3.23. The steps that are generally used by countries to construct an unbalanced version of the

SUTs are presented below:

• Step 1 – construction of the supply table: This consists of filling the available data into an

initial unbalanced supply table, which covers domestic output by product (part 1 in Figure

3.3) and the imports of goods and services and the valuation matrices comprising

information on taxes less subsidies on products, trade margins and transport margins (part

2 in Figure 3.3). These valuation matrices allow the transformation of total supply of

products at basic prices (formed by summing the domestic output and the imports) to total

supply of products at purchasers’ prices. The construction of this initial unbalanced supply

table is presented in chapter 5 of this Handbook.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Step 2 – construction of the use table: In a manner similar to step 1, this step consists of

filling the use table with the available data, which cover: the intermediate consumption at

purchasers’ prices (showing the input requirements of goods and services for the

production of the domestic output of each industry shown in the supply table – part 3 in

Figure 3.3); final uses at purchasers’ prices and for each category, such as final

consumption and gross fixed capital formation, for which separate compilation steps will

be needed (part 4 in Figure 3.3); and production-based GVA at basic prices shown by

industry (part 5 in Figure 3.3). This compilation step is covered in chapter 6 of this

Handbook.

• Step 3 – compilation of the valuation matrices: These matrices are essential to the

preparation of SUTs at basic prices. They expand the valuation columns in part 2 of the

supply table in Figure 3.3 into corresponding matrices for intermediate consumption and

final consumption of the use table. This compilation step is described in chapter 7 of this

Handbook.

• Step 4 – compilation of the imports use table and domestic use table at basic prices: This

step is essential to increasing the analytical uses of SUTs by distinguishing the use of

imported and domestic products. This compilation step is presented in chapter 8 of this

Handbook.

• Step 5 – compilation of the SUTs in volume terms (previous years’ prices): When balanced

both in current prices and in volume terms, the SUTs ensure coherent and consistent

deflation of the components of the production and expenditure approaches to measuring

GDP as well as coherent and consistent estimates of price and volume indices. This requires

that SUTs are compiled in volume term at this stage of the compilation process. The

compilation of SUTs in volume terms is described in chapter 9 of this Handbook.

• Step 6 – Linking SUTs with the institutional sector accounts: Linking SUTs and the

institutional sector accounts is an important step in the compilation of SUTs, ensuring the

full integration and consistency of SUTs with the national accounts. This link is provided

by compiling a linking table between the sectors and industries (part 6 in Figure 3.3). The

compilation of the linking tables is presented in chapter 10 of this Handbook.

3.24. These six steps above are generally followed in that order; however, there is a significant

level of interdependency in the compilation process. For example, trade and transport margins and

taxes less subsidies on products are necessary for the transformation of the use table from

purchasers’ prices to basic prices and also for conversion of the supply of products at basic prices

to purchasers’ prices in the supply table, to enable the balancing of products at purchasers’ prices.

This information may partly be derived from estimates based on the use tables and linked to

estimates from the supply table at basic prices.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

3.25. On the other hand, estimates of certain final uses may require basic supply side information

if, for example, the product flow

method is being applied. Nevertheless, allowing for

interdependencies in the compilation of these tables, it is vital that the tables are viewed and

accepted as primary estimates.

Figure 3.3 Structure of the SUTs and the links covered in this Handbook

3.26. Once these six steps are completed, the result is unbalanced SUTs at purchasers’ prices and

basic prices. This represents the start of a balancing procedure which is an iterative procedure

integrating the following aspects:

• Balancing of SUTs at purchasers’ prices

• Compilation of valuation matrices

• Transformation of SUTs into basic prices

Following the terminology used in 2008 SNA (para. 14.2), in this Handbook the expressions “product balance” and

“product flow” methods are used in preference to “commodity balance” and “commodity flow method”, as

reflecting the more recent usage of the word “product” in place of “commodity”. It is noted, however, that the

change in terminology does not indicate a change in methodology.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Compilation of separate use tables for use of domestic output and use of imports of goods

and services

• Balancing of SUTs at purchasers’ prices and at basic prices

• Balancing the production-based GVA and income-based GVA, providing the link to the

institutional sector accounts

All of the above should be balanced with time series in mind to ensure consistent movements of

levels and growth rates.

3.27. One of the key reasons why the SUTs are balanced first at purchasers’ prices is to reflect

as closely as possible the basis of the survey data that is being fed into the use table. The

intermediate uses and final uses, for example, are collected close to the economic reality of the

prices paid by purchasers, in other words, the purchasers’ prices. In addition, no valuation issues

exist with such variables as compensation of employees and other taxes less subsidies on

production.

3.28. These aspects should, however, be viewed alongside the domestic output part of the supply

table reflecting data collected from producers whereby the output is valued at basic prices. Thus a

balance between the two is needed.

3.29. For the SUTs balanced at purchasers’ prices, the two key identities are:

• Total supply of products at purchasers’ prices equals total uses of products at purchasers’

prices.

• Total output of industries at basic prices equals total input of industries at basic prices.

3.30. Balancing is not just necessary in order to achieve the above identities but also makes it

possible to trace inconsistencies of basic data and estimation methods. Ideally, the balancing of

the SUTs system should be done both in current prices and in volume terms simultaneously. In

fact, balancing in this manner means that the process is not complete until the transformation into

basic prices and the separation of the use of domestically produced products from the use of

imported goods and services have been achieved, as these are key steps in producing the SUTs in

volume terms. These steps are in practice interrelated and provide a powerful feedback loop in

terms of quality and validity of the various component estimates.

1. Compilation of SUTs in current prices and volume terms

3.31. The SUTs framework in Figure 3.3, when treated in summary form, can be combined with

the H-Approach to show a simplified version of the compilation schematic configuration when the

SUTs are compiled in current prices and in volume terms. Figure 3.4 illustrates the sequence of

steps involved in the compilation of SUTs, PSUTs, and IOTs. The inner box outlined in red focuses

on the compilation of SUTs. Thus, countries that intend to compile only monetary SUTs can focus

on the steps within the red box and follow the compilation sequence indicated by the arrows in the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

figure. The order of compilation of PSUTs and IOTs in the figure does not imply a compilation

sequence for these tabulations. The compilation of these tables reflects the country’s priority.

3.32. The simplified illustration provided in Figure 3.4 of the compilation of SUTs, PSUTs and

IOTs can be seen in relation, on the one hand, to phase 4, “Process”, when it comes to compiling

unbalanced SUTs and PSUTs and, on the another hand, to phase 5, “Analyse”, when it comes to

compiling balanced SUTs, PSUTs, and IOTs.

Figure 3.4 Compilation of SUTs and IOTs in current prices and in volume terms

PxP: product-by-product

PxI: product-by-industry

IxI: industry-by-industry

3.33. In compiling the seven boxes, in the sequence from box A to box G, a further dimension

of their evolution needs to be reflected. In year 1 of the compilation process, boxes A, B and C

representing current prices are produced in that sequence covering the economy for year (T) for

SUTs and IOTs, and box G covering PSUTs which are linked to the outputs of boxes B and C.

3.34. As mentioned in chapter 2, SUTs in volume terms for one period can be compiled using

SUTs in current prices for one period and deflators. The preferred approach, however, includes a

time-series dimension and boxes D, E and F representing the previous years’ prices should not be

compiled in year 1 as there are no SUTs in current prices for the previous year (T-1).

BOX

Volume terms

SUTs

at purchasers’ prices

PxI

IOTs

PxP and IxI

Assumption

applied

SUTs

at basic prices

PxI

Addition of valuation

and imports matrices

Deflation

processes

Current prices

SUTs

at purchasers’ prices

PxI

IOTs

PxP and IxI

PSUTs

Separation of valuation

and imports matrices

Assumption

applied

SUTs

at basic prices

PxI

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

3.35. It is therefore essential to have two consecutive years of SUTs in current prices to enable

the production of the first year of SUTs in previous years’ prices. If the SUTs are produced less

frequently, say, every five years, it is much more difficult to produce SUTs in volume terms.

3.36. In year 2, boxes A, B and C are produced covering the economy for year (T+1) together

with any revisions to the data for boxes A, B and C for the year (T). In addition, the first set of

SUTs in previous years’ prices can be compiled for year (T+1). In each year thereafter, the process

will extend the availability of SUTs by an extra year in current prices and previous years’ prices,

while also incorporating any revisions to SUTs for earlier periods to ensure consistent time series.

3.37. Figure 3.5 provides a summary of the evolution dimension for the first three years. As time

passes, different challenges will evolve, such as the need to retain an ever-increasing number of

years of SUTs on a consistent basis, the need for a revisions policy, data version and vintage

control, managing the production of consistent levels and growth rates, the organizational

arrangement of resources which may not increase each year, among others. It is thus important to

ensure that this process is properly planned and managed from the start.

Figure 3.5 Evolution of compiling SUTs and IOTs in the first three years

Note: CP: current prices; PYP: previous years’ prices; (r) revised tables.

3.38. Based on the overall strategy for the compilation of SUTs and IOTs, it is possible to provide

step-by-step guidance. This is provided for the first year of compilation, and then subsequent years

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

of compilation, as there are some additional steps that need to be considered in order to ensure a

fully consistent time series of SUTs in current prices and in previous years’ prices.

(a) Step-by-step summary for the first year of compilation

3.39. Figure 3.6 provides a description of the various phases in the compilation of SUTs and

IOTs for year 1 of the compilation, together with the reference to the relevant chapters of the

Handbook. The figure also contains references to boxes A–G of Figure 3.4 and the compilation

stages of Figure 3.2. Although boxes A, B and C form key outputs, there are various intermediate

stages and intermediate outputs as denoted in the separate stages in Figure 3.6.

3.40. Before starting the first year of compilation, as illustrated in Figure 3.6, compilers should

ensure that the overall framework is well in place, comprising: standards, definitions,

classifications and methods, statistical units, business registers and sample frames, census, survey

and administrative data collection.

3.41. The following key features of Figure 3.6 should be noted:

• It is consistent with Figure 3.5 and covers the simplified business process model for

compiling SUTs and IOTs. Each block of work in Figure 3.6 is split according to the type of

work as indicated in the six different stages in Figure 3.2.

• It follows the underlying principles and features of the GSBPM.

• The flow of work is kept as logical and sequential as possible and follows the H-Approach

as covered earlier. As mentioned earlier in this chapter, however, the compilation of SUTs

and IOTs includes several interrelated processes and dependencies which must be reflected

and retained. Furthermore, in some cases, there is more than one approach available, for

compiling trade margins using a supply-side approach or use-side approach or both.

• The flows in Figure 3.6 do not present backward loops, although effective feedback loops

are critical at each phase and improve the process. For example, the compilation and

balancing of PSUTs provides an important feedback loop to the compilation of monetary

SUTs, thus enhancing the quality of physical and monetary SUTs.

• Integrated links bring the PSUTs together with such input data as the prices and quantities

(levels) and material flow accounts alongside the SUTs and IOTs.

• Each phase of work is also linked to the main chapters in parts A and B of this Handbook,

providing much more detail on the compilation – these links provide the key chapters but

not all references.

• The same approach has not been applied to part C (Extensions and applications) of the

Handbook as there are many variations and options.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 3.6 First year of compilation

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

(b) Step-by-step summary for the subsequent years of compilation

3.42. Having completed the year 1 results, Figure 3.7 provides a detailed stage of production for

year 2 with the focus on the SUTs in previous years’ prices, which, as mentioned, can only be

compiled when SUTs have been compiled for the current year and the previous year.

3.43. In compiling SUTs in previous years’ prices for the first time, there may be several years

of SUTs in current prices already available. If so, then the compilation is just an extended version

of the process indicated in Figure 3.7, and it is better in terms of quality and consistency as there

is a time series dimension in place immediately.

3.44. Other features to note in Figure 3.7 are the following:

• The focus of the steps in Figure 3.7 is the right-hand side of the H-Approach of Figure 3.4

and builds on the detail available from the left-hand side – assuming the left-hand side

products are available.

• The deflation approach follows the underlying H-Approach and is covered in chapter 9, on

compiling supply and use tables in volume terms.

• Other approaches are available but this is the recommended approach.

• The compilation of IOTs in volume terms is not essential but is not resource-intensive if all

the other parts are available.

3.45. An additional feature is also achieved whereby GVA in volume terms is arrived at using

the SNA recommended approach, namely, double deflation. However, the results from this

approach need additional quality assurance against other indicators. This is to ensure that the

quality of the GVA estimate in volume terms (and, in turn, GDP) is not reduced as a consequence

of the errors in either the current price estimates of output and intermediate consumption or

inappropriate deflation of these two variables.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 3.7 Second year of compilation

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

2. Summary of the main recommendations, principles and guidelines of the Handbook

3.46. Box 3.1 provides a list of the main recommendations, principles and guidelines relevant

for the compilation of SUTs, IOTs, PSUTs (and EE-IOTs) and related products presented in this

Handbook, covering various aspects such as the organizational or institutional environment,

compilation, data strategy and requirements, and balancing.

3.47. The recommendations and guidelines presented in box 3.1 may be viewed as aspirational

as they provide the model scenario for the compilation and dissemination of SUTs, IOTs and

related products. Countries can gradually implement the recommendations and guidelines in

accordance with their specific situation in terms of such factors as data availability, resource

constraints, legal framework and others, and in line with their priorities.

Box 3.1 Examples of the main recommendations, principles and guidelines

provided in this Handbook

A. Organizational or institutional environment

1. The organization of the economic statistics system should follow an integrated economic statistics approach. The use

of the GSBPM to organize the statistical production process would facilitate the compilation of SUTs, IOTs and related

products.

2. National accounts should have very close links with all its suppliers, in particular, the business register, business

surveys and administrative sources.

3. The compilation of the various components of the SNA framework should be coordinated and integrated in terms of

production processes, such as production schedules, feedback loops, coherence, and other features:

• National accounts (including balance of payments and monetary financial statistics, government finance

statistics)

• SUTs and IOTs together with PSUTs and EE-IOTs

• Environmental-economic accounts to be closely linked with the compilation of SUTs

• Regional accounts

• Prices

• Labour market statistics

4. The compilation of SUTs and IOTs should be performed as part of the regular compilation of the national accounts

and within the core national accounts. This would have the following effects:

• Leads to better quality, coherence and consistency of national accounts, balance of payments and related

statistics

• Creates effective and powerful data quality and coherence feedback loops, which in turn help to address

structural issues and biases and to prioritize resources to targeted improvements

5. The final estimates of the national accounts aggregates should be derived from the balanced SUTs framework and not

the other way around. For example, the SUTs based estimates should be not confined to predetermined estimates or

already published estimates.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

6. The compilation of SUTs and IOTs should reflect stakeholder interests. This can be achieved by organizing regular

meetings with data suppliers and users together with other regular stakeholders.

7. Appropriate internal governance should be exercised to ensure accountability and guidance, supported by programme,

project and process management, including risk management, under a framework reflecting:

• Schedules, timetables and customer-supplier service-level agreements set in place to ensure a regular supply

of source data, briefings and evaluation reviews

• Various standards and policies, such as revision policy, confidentiality and disclosure controls, and others

• Staff recruitment, retention and skill development

8. Skill development needs to take into consideration the following types of training requirements:

• National accounts – Technical skill focus covering national accounts concepts, methods, processes and

guidance, and also such functions as developments, compilation, coordination, balancing, analyses and

dissemination

• Systems – IT systems, programming, data management (standards and principles), data dissemination, also

covering website management, and other systems, including the role of dedicated IT professionals supporting

economic statistics

• Management – staff management, effective leadership, communication, and other aspects

9. For effective and sustainable production of SUTs and IOTs, it is important to have sufficient computing capacity in

place that includes:

• Robust, reliant, structured, quick and well-documented systems

• Database software and hardware, speed, structure, flexibility, statistical functionality, data management and

links to web dissemination

10. It is important that the statistical production process is well documented and kept up to date, reflecting:

• Operational, methodological, system, metadata and recording-specific issues, adjustments, etc. for each

quarterly or annual exercise

11. The compilation of SUTs and IOTs is to be performed with due consideration for the costs and resources available

and also other criteria such as data availability, data quality and time.

B. Compilation

1. SUTs (and IOTs) should be compiled annually and, if possible, on a quarterly basis, following the H-Approach for the

production of SUTs and IOTs in current prices and in previous years’ prices (including valuation and imports matrices).

The application of the H-Approach makes it possible for the volume of GVA to be estimated using a double deflation

method and also ensures greater coherence, linking SUTs to various other parts of the SNA framework.

2. SUTs should be produced first, then IOTs derived from the SUTs, using additional information and assumptions.

3. Rectangular SUTs should be compiled with more products than those provided for by industries:

• The greater the detail, better the quality – while more detail will increase the burden on business, systems and

resources, it can improve the quality of balancing

• Improved matching between prices and values, thereby ensuring better quality of the data in volume terms

• Compilation (and balancing) should be undertaken at the greatest level of detail available – time, quality and

resources permitting. Due to confidentiality-type criteria, however, the level of publication may – or will – be

aggregated to a higher level

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

4. Standard international statis

tical classifications (such as ISIC, CPC, COICOP, and others) should be used at

appropriate detailed levels to ensure international comparability. Within these classifications, greater granularity may be

desired for specific economies.

5. Consistent statistical units should be used throughout the process, from the business register and business surveys

through to the SUTs.

6. Ideally, SUTs and PSUTs (and EE-IOTs, as appropriate) should be based on sound and complete data sources,

reflecting:

• Common concepts, definitions and classifications

• Comprehensive and up-to-date statistical business register

• Wide range of (preferably annual) regular business surveys (including structural detail), household surveys,

administrative data, prices, and other sources

• Benchmarking and reconciliation – preferably conducted annually, reflecting rapidly changing economies

(minimizing the use of fixed factor or stability assumptions)

• Incorporation of labour and capital information ensuring improved coherence for productivity estimates

• Appropriate choice of index number formulae and base year

7. All the data building blocks should be recorded separately, namely, source data, coverage adjustments (including

exhaustiveness), conceptual adjustments, quality adjustments, balancing adjustments, and others

8. A table should be compiled linking the SUTs and the institutional sector accounts, including:

• Goods and services

• Production accounts by industry and by institutional sector

• Generation of income accounts by industry and by institutional sector

• Parts of the use of disposable income account (such as household final consumption expenditure) and parts of

the capital account –

by industry and by institutional sector (such as gross capital formation and its

components)

9. Use of the bottom-up approach should be preferred in the compilation of regional SUTs, which should be reconciled

with national SUTs.

10. For the derivation of IOTs, the following should be the methods most frequently used:

• Model A (product-by-product) IOTs using the product technology assumption

• Model D (industry-by-industry) IOTs using the fixed product sales structure assumption

• Hybrid – mix of technologies usually chosen to avoid having any negatives

11. Comprehensive documentation should be prepared on operational methods and methodology, including appropriate

metadata and revision analysis.

12. Efforts should be made to keep up to date with, and contribute to, internationally evolving and agreed changes to

concepts, methods and systems developments.

C. Data strategy and requirements

1. SUTs are data-hungry and a range of timely, comprehensive, consistent and coherent data sources are needed. The

data strategy should reflect a range of aspects.

2. These should include data-handling aspects such as:

• Data collection (for example, questionnaire design, electronic data capture, receipt of all the data that a

company can provide, etc.)

• Data processing, data editing, metadata and data warehousing

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Data quality frameworks

• Data dissemination and use of SDMX standards

3. The following structural and data-collection issues should be reflected:

• Comprehensive and up-to-date statistical business register used as the sampling frame for all business surveys

• Use of as many data sources as possible, censuses, business and household surveys, administrative data,

company accounts, regulatory accounts, company websites, and other sources

• An international business unit handling all aspects of multinational enterprise groups, from profiling the

business structures to data collection and data reconciliation and feeding coherent data through to the various

statistical domains. In addition, the need to develop links and share data with other national statistics offices

and national central banks for statistical purposes only.

• Frequency of information – monthly, quarterly, annually or five-yearly. The more regular, the better the

information reflects rapidly changing industry structures of sales and inputs, changing patterns of household

consumption, impact of globalization on trade flows, and other factors

• Sufficient, appropriate and relevant, price indices matching the current price values for deflation or use of

suitable volume only indicators where price information may be unavailable

• Strategy for handling, and reviewing, areas where data may be missing

4. The following more general needs should be included:

• Need to minimize the burden on business

• Need to have confidentiality and disclosure testing processes

D. Balancing

1. SUTs should be balanced in current prices and in volume terms, thus leading to:

• A single estimate of GDP incorporating the components of production, income and expenditure approaches

to measuring GDP

• Volume estimates of GVA through double deflation

• Balance between supply of products and use of products and between industry inputs and industry outputs

2. The balancing process should simultaneously encompass:

• SUTs at basic prices and at purchasers’ prices

• SUTs in current prices and in volume terms (preferably, previous years’ prices)

• SUTs links to IOTs, PSUTs and EE-IOTs (as appropriate)

• Link with the institutional sector accounts

3. Balancing should strongly promote integration of the following:

• Goods and services, production account, generation of income account, parts of the capital account and use of

disposable income account

• Incorporation of PSUTs and EE-IOTs (as appropriate)

• Productivity estimates (labour, capital and multifactor)

4. Simultaneous balancing should be preferred to sequential balancing. If this is not possible, sequential balancing (first

in current prices, then in volume terms), with quick and effective feedback loops, should be considered as an alternative.

5. The organization of the balancing function can be set up in different ways across teams. A centralized balancing

approach should, however, be preferred to the decentralized balancing arrangement whereby the balancing of the various

elements related to SUTs and IOTs (such as current and constant prices for a single year and for a time series, links with

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

productivity, regional accounts, among others) is carried out at the same time and within the same unit in order to ensure

the full consistency of all SUTs-related products.

6. The production and balancing of SUTs should enable the identification of source data incoherence. Mechanisms should

be developed to provide feedback to data suppliers and help prioritize areas for improvement and allocation of resources.

7. An annual review and evaluation of the balancing adjustments should be carried out, to identify and address any

evolving biases.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Annex A to chapter 3: Examples of institutional arrangements in countries

A3.1 This annex presents examples of institutional arrangements in selected countries. The

examples cover the cases of centralized and decentralized statistical systems.

A. Centralized production of economic statistics: Canada

A3.2 As a centralized national statistics office, Statistics Canada is accorded the legal mandate

to collect and disseminate a broad range of statistics by a federal act (the Statistics Act). Provisions

in the Act also serve to protect data confidentiality and assure political neutrality and an arms-

length relationship with policymakers.

A3.3 Users are regularly consulted, and the office, through various channels, ensures that priority

requirements are established and met. These channels include national advisory committees,

federal-provincial consultations and regular bilateral meetings with key policy partners such as the

federal finance department and the Central Bank.

A3.4 Statistics Canada produces a full suite of macroeconomic accounts, including:

• National accounts (including financial and wealth accounts)

• Balance of payments

• Government finance statistics

• Productivity measures

• Environmental accounts (natural resource stocks, along with physical flows of energy use,

greenhouse gas emissions and water use)

• Selected satellite accounts covering tourism, culture and pensions

A3.5 The compilation processes are integrated to assure data coherence across components of

the Canadian macroeconomic accounts, and regional SUTs serve as the integrating mechanism for

the production dimensions. The integration is achieved through annual benchmarking and

reconciliation processes with current price measures of GDP income and expenditure, real GDP

by industry and labour and multifactor productivity. Data coherence is a requirement for key policy

applications, such as the input to fiscal formulas for the sharing of sales tax revenues among the

Federal Government and provincial jurisdictions or to formulas for equalizing fiscal capacity

among Canadian provinces.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

A3.6 Economic surveys, along with labour market data, price statistics and international trade

statistics, are produced within Statistics Canada, ensuring the alignment of priority-setting for

feeder programmes to the macroeconomic accounts.

A3.7 All business surveys are linked to a central business register maintained through regular

updates from administrative files. The survey content is harmonized, as are approaches to

collection, processing and estimation within an integrated business statistics programme

framework. The use of administrative data is optimized throughout all stages of the process, and

continuous access to required files is assured through formalized arrangements with data providers,

such as the Canada Revenue Agency and national and provincial regulatory authorities.

A3.8 In recent years, Statistics Canada has made significant progress towards implementing

consistent classification standards across all feeder programmes, thereby facilitating the

compilation of SUTs. The North American Industry Classification System (NAICS) serves as the

basis of industry statistics and North American Product Classification (NAPCS) the basis of

statistics on products. Continuing efforts are being made to ensure compliance and to coordinate

input from the macroeconomic accounts and feeder programme areas into the development of

updated standards.

B. Centralized production of economic statistics: Norway

A3.9 Statistics Norway has overall responsibility for official statistics in Norway, and also for

the conduct of extensive research and analysis activities. Statistics Norway reports to the Ministry

of Finance, with the support of the Statistics Act of 1989. Statistics Norway is a professional,

autonomous organization with the mandate to determine what it publishes, and when and how the

publishing shall take place.

A3.10 Statistics Norway is responsible for the production and maintenance of the business

A3.11 The Department of National Accounts and Industry Statistics comprises nine divisions,

with the following responsibilities:

• National accounts

• Primary industry statistics

• Manufacturing and research and development statistics

• Construction and service statistics

• Transport, tourism and ICT statistics

• Energy statistics

• Natural resources and environmental statistics

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

• Accounting statistics

• Business register

A3.12 The division for national accounts is responsible for the following:

• Quarterly and annual national accounts (including SUTs, IOTs and regional accounts)

• Quarterly and annual non-financial accounts

• Quarterly balance of payments

The balance of payments has been an integral part of the national accounts since the 1950s.

Satellite accounts are also prepared by the division for national accounts but not on a regular basis.

A3.13 The Department of Prices, Financial and External Trade Statistics comprises six divisions,

with the following responsibilities:

• Financial market statistics (including financial accounts)

• Public finance statistics

• Financial corporations

• External trade statistics (in liaison with the Customs Department)

• Price statistics

• Development cooperation

Banking statistics were originally under the responsibility of Statistics Norway and were then

moved to Bank of Norway before being moved back to Statistics Norway.

C. Centralized production of economic statistics: United Kingdom

A3.14 The structure of the United Kingdom statistical system has evolved over many decades,

helped by several reorganizations of statistical departments and changes in legislation,

consolidating the responsibility for almost all economic statistics under the Office for National

Statistics and the Government Statistical Service. The United Kingdom system continues to

evolve, for example, by developing better links and access to administrative data.

A3.15 Currently, the United Kingdom has in place resources, systems and processes for producing

detailed, integrated and timely quarterly and annual economic accounts. The Office for National

Statistics, as an independent statistical body with a central role, is wholly responsible for the

compilation of the national accounts, balance of payments, public sector finance statistics, labour

market statistics and price statistics. The compilation of SUTs is central to the annual national

accounts system. The Office also produces regional accounts, environmental accounts and IOTs

(Mahajan, 2016).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

A3.16 The Office for National Statistics is one of the few national statistics offices with such

centralized responsibility and coverage of economic statistics – this has only been the case since

the late 1980s. Furthermore, since 2011, all the above economic statistics are being produced at

the same location.

A3.17 The independent status of the Office for National Statistics is supported by national

legislation, pursuant to which it reports to the United Kingdom Statistics Authority. The Statistics

Authority, which was established on 1 April 2008 under the Statistics and Registration Service Act

2007, is a non-ministerial department overseen by Parliament and not by a government minister.

Figure 3A.1: Integrated process of compiling national accounts and balance of payments

United Kingdom

Compiled by Sanjiv Mahajan March 2014

D. Decentralized production of economic statistics: Chile

A3.18 Economic statistics in Chile are mainly produced by three institutions:

• Central Bank of Chile, which is responsible for the compilation of most of the

macroeconomics statistics, namely, national accounts (non-financial and financial accounts),

balance of payments and international investment position

• Finance Ministry, which produces the government finance statistics

• National Statistics Office, which undertakes the data collection covering economic and

business surveys and the compilation of price indicators, labour market indicators and socio-

demographic data

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

A3.19 The organizational structure described above allows the Bank to achieve a high level of

consistency between external statistics and national accounts in terms of both the statistics

themselves and the methodology.

A3.20 In the production of macroeconomics statistics, the Bank uses a significant amount of data,

provided mainly by the Tax Revenue Service, the National Customs Service, the General

Comptroller’s Office and the National Statistics Office, the latter being the main provider of

statistics for national accounts compilation. Dependency on the statistics from the National

Statistics Office entails a high degree of coordination between both institutions. To this end, a

framework agreement is in place to ensure that the requirements and conditions for the provision

of statistical products are met. In addition, a committee with members from both institutions

regularly meets to coordinate issues related to data collection and the specific needs of national

accounts.

A3.21 There are strong links between the Central Bank of Chile and the National Statistics Office,

buttressed by a continuous programme to improve the cooperation and the quality of the links and

the data flows between the customer and supplier.

A3.22 Other salient features of the Chilean system include the following:

• The SUTs and IOTs are compiled within the national accounts in the Central Bank of Chile.

• Where the balance of payments is concerned, the Central Bank of Chile collects the data on

international trade in services to supplement the data on the international trade in goods

collected and provided by the National Customs Service.

• The Central Bank of Chile also produces regional GDP figures on an annual basis.

• The national statistics office produces a business register which, in turn, is employed by the

Central Bank of Chile after making improvements and modifications.

• Although the environmental accounts are not produced for Chile, various efforts have been

undertaken by the Minister of the Environment to produce a range of environmental

indicators.

A3.23 The diagram below shows the components of the statistical system in Chile.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 3A.2 Statistical system in Chile

E. Decentralized production of economic statistics: United States of America

A3.24 The United States of America has a highly decentralized statistical system, under which

responsibility for producing a substantial portion of official federal economic statistics is divided

among 13 agencies that have statistical work as their principal mission.

A3.25 There are also numerous other entities that are considered part of the statistical system in

the United States but statistical work is not their principal mission. Most of the country’s primary

economic indicators are produced by one of three main federal statistical agencies, while the

United States Census Bureau conducts economic censuses and surveys. The three main agencies

and their responsibilities are the following:

• Bureau of Economic Analysis: this body relies primarily on data generated by other agencies

to compile the national accounts (non-financial accounts) and the balance of payments.

• Federal Reserve Board (the United States central bank): this compiles the financial accounts

and government finance statistics.

• Bureau of Labor Statistics: this body prepares the labour market statistics and price statistics.

A3.26 The Bureau of Economic Analysis also undertakes a number of business surveys. At the

same time, however, most of the statistics used by the Bureau in preparing GDP and input-output

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

accounts come from non-Bureau sources, including other statistical offices. The Census Bureau

provides most of the other expenditure components of GDP and output and intermediate purchases

within the input-output framework. For the period 1997–2017, the Bureau of Economic Analysis

produced SUTs at basic prices with a transformation to purchasers’ prices, make/use tables at

purchasers’ prices for benchmark years and at producers’ prices annually, together with IOTs.

Annual make/use tables at producers’ prices at a more aggregated level of detail are also available

for the period 1947–1996.

A3.27 The agencies each produce and maintain their own business register, often created using

different sources, as detailed below:

• The Census Bureau’s business establishment list is compiled mainly from federal tax forms

and used as the primary sampling frame for the five-year economic censuses and many of

the economic surveys.

• The Bureau of Labor Statistics business establishment list is based on information collected

in connection with the joint federal and state unemployment insurance programme and used

by Bureau establishment surveys, including the producer price index (PPI) survey.

Figure 3A.3 Overview of the products produced by the main agencies

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Figure 3A.4 Overview of the statistical system structure

A3.28 Data sharing between agencies, as practised in the United States, may significantly improve

data harmonization and eventually lead to savings. In recent years, the Bureau of Economic

Analysis has signed memorandums of understanding with other agencies and ministries to

facilitate the exchange of data, including confidential data.

A3.29 Although there may be differences in the concepts of statistics employed by different

agencies and in their statistical coverage (for example, productivity statistics are published in the

United States by the Bureau of Labor Statistics and then used by the Bureau of Economic Analysis

in its measurement of national accounts), the confrontation of the data themselves or of the data

processing steps raises the data validation to another level and enhances the quality of statistics.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

Chapter 4. Specify needs, design, build and collect stage

A. Introduction

4.1. The objective of this chapter is to describe the activities specific to the compilation of SUTs

and IOTs which take place during the phases of the GSBPM relating to the specification of needs,

designing, building and collecting, referred to together as the “specify needs, design, build and

collect stage”. The chapter will also explore specific elements that need to be considered when

compiling SUTs and IOTs. These phases of the GSBPM is of particular importance in the

compilation process because of its direct implications for all the remaining phases and their

elements need to be continuously reviewed and corresponding adjustments made to the process.

4.2. The following section of the chapter focuses on the “specify needs”, “design” and “build”

phases of this stage. The level of detail of the industry and products in the tables must be carefully

evaluated at the beginning of the compilation process, together with other elements such as the

compilation schedule, the revision policy and others. Section C moves on to the collect phase and

describes the main data sources used for SUTs and IOTs.

B. Specify needs, design and build phases

1. Specify needs

4.3. The identification of user’s needs is a fundamental step in the compilation of any statistics,

as it aims to identify what statistics need to be compiled, in which format, when and for what

purpose. All these elements affect the planning of the compilation process of SUTs and IOTs since

they have implications, for example, for the choice of the level of industry and product detail of

the SUTs. Thus an assessment of the objectives of these tabulations has to take place during the

specify needs phase of the statistical production process and this assessment must be regularly

reviewed in the light of feedback from users, to ensure the relevance of the compiled SUTs and

IOTs. During this phase, consultation with relevant stakeholders, through meetings, workshops

and surveys, is of key importance.

4.4. Other elements of this phase include the identification of the statistical outputs that are

required to meet the user needs and checking the data availability to see if existing data sources

can meet the user requirements, if there are alternative data sources that would be more suitable

for the specific statistics, and if there are data gaps to fill.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

100

4.5. There are many different users – policymakers, analysts, researchers and others – and uses

– planning, modelling, monitoring and so forth – of SUTs and IOTs. It is therefore important to

maintain links with the users to ensure that their needs are met through a single efficient

compilation process. For example, in the context of growing concerns about the environment, if a

specific environmental topic, such as water, energy, fish, or forests, is considered as a key user

need to be addressed, it is important to develop and design a compilation process which includes

these elements from the outset, rather than adjusting ex post facto the statistical output through

modelling based on various assumptions.

4.6. The compilation of any statistical output depends to a large extent on the availability of

appropriate infrastructure for information technology and human resources. The technology has

changed enormously over the last fifty years. Statistics which were once compiled with calculators

are now processed in seconds by modern computers, laptops or even smartphones. This rapid

development has facilitated the work of statisticians and improved the timeliness of their statistical

outputs. When compiling SUTs and IOTs, a variety of software, databases and custom-designed

platforms is available and can be adapted to the specific compilation process in any given country.

4.7. It is therefore important to have a clear understanding of the information technology

requirements necessary for all the phases of the compilation process. In practice, more than one

software package is often required and the ones selected are those which best meet the diverse

functional requirements of the specific phase and which are able to communicate with one another

in an easy, effective and efficient manner. If the links between the software packages are

cumbersome or time-consuming, alternatives should be sought. Many national statistical offices

have developed in-house tailored software to meet their needs – this has advantages but may entail

greater overhead maintenance and training requirements.

4.8. Box 4.1 and Box 4.2 provide examples of custom-made software produced and maintained

by, in the first case, Statistics Netherlands and, in the second, the Institut national de la statistique

et des études économiques (INSEE) – the French national statistical office – and Eurostat – the

statistical office of the European Union.

Box 4.1 Example of in-house custom-built software: Statistics Netherlands

Statistics Netherlands has a long-standing tradition of compiling SUTs both in current prices and in volume

terms, and also IOTs, and therefore has extensive experience in tackling the various challenges associated

with the use of computer systems to produce and maintain long-run series of SUTs and IOTs.

Statistics Netherlands has been publishing SUTs since 1990 (relating to the benchmark revision for the year

1987) and IOTs since the 1950s (for years starting from 1948).

Statistics Netherlands has always used in-house custom-built software for the compilation of the balanced

SUTs and IOTs. There are two separate software tools which are both updated continuously on an SQL

database. These are combined with a graphical user interface in Visual Basic for Applications (VBA) code,

enabling data to be accessed and adjusted by the national accounts staff. Both systems can handle quarterly

and annual data.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

101

The first application includes tools for the transformation of source data to national accounts definitions and

standards, both in current prices and in volume terms.

The second application includes tools for simultaneous balancing of the SUTs in current prices and in volume

terms, for compiling the valuation matrices and for transforming the SUTs to industry-by-industry IOTs.

The present systems are based on a major overhaul carried out in 1995 and have been continuously updated.

A key rebuild took place in 2004/2005, reflecting a new programming language and also including new

features.

Box 4.2 ERETES

ERETES – the acronym for “équilibre ressources-emplois et tableau entrées-sorties” (supply-use

balances and input-output tables) – is a computer system designed to assist national accountants to

compile the SUTs and integrated economic accounts (including sector accounts), which complies with

the principles and guidelines set out by the SNA. ERETES was developed by INSEE and Eurostat. In

2014, it was being used by several countries in Africa and Latin America and the Caribbean. Further

countries are expected to adopt the system. The ERETES system is available in French, Spanish and

English.

Although the objective of ERETES is to generate SUTs and the integrated economic accounts, it can also

be used by countries that have limited data resources. The minimum data requirements are an enterprise

survey and a household budget survey, foreign trade statistics, government accounts

, balance of

payments and banking statistics. With these data, ERETES can help countries to generate estimates of

GDP in current prices. If sufficient price and volume indices are available, then estimates of GDP in

volume terms can also be generated. The compilation of SUTs would also require information on

intermediate consumption and trade and transport margins.

One key advantage of ERETES over other computer systems is that it is supported by a permanent

secretariat that can call on a group of multilingual national accountants and information technology

experts with extensive experience of applying the system in a number of developing countries. ERETES

is regularly updated and improved. ERETES is available at: http://www.eretes.net/EN/index.htm.

4.9. Another successful example of custom-built software produced by one national statistical

office, and then made available for use by other countries under specific terms and conditions, may

be seen in the Norwegian software SNA-NT. Here, Statistics Norway provided both the software

and the associated human resources for training in the use of the applications by, for example,

Malawi, the Czech Republic and Slovakia.

4.10. When choosing the software and hardware to support the compilation of SUTs and IOTs

as part of the national accounts, consideration should be given to various criteria such as the

database environment, in particular its flexibility and structure, its statistical functionality and

diagnostic tools required, the necessary availability of mathematical functions such as matrix

calculations, the resources and costs, the training programme, compatibility with data suppliers,

data management, and the data dissemination platform envisaged.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

102

4.11. Across the world a range of different products are used to this end, such as Oracle, JAVA

programming, MATLAB, SPSS, SQL, SAS, Excel and custom-made software built to meet

specific requirements. For example, the use of Excel or input systems such as an Oracle database

could provide an effective solution in the “collect” and “process” phases, for preparing and

validating the data, while SAS could be used for the further processing of SUTs and IOTs. Tools

such as Excel, an output system and web-tools for dissemination may offer the best means of

validating and balancing, and also of analysing and disseminating the data.

4.12. Skilled and trained human resources are also a fundamental pillar for the compilation of

SUTs and IOTs. It is thus important to recruit and retain skilled and effective staff and develop

and use internal and external training opportunities on the theoretical and practical aspects of the

compilation of economic statistics.

4.13. An important step in this phase is the preparation of a document summarizing the findings

of all the activities mentioned above – namely: needs identification; establishment of output

objectives; checking of data availability and information technology requirements, and others – in

the form of a business case, with a view to securing approval to implement the new or modified

statistical business process. Such a business case would need to conform to the requirements of

the approval body but would typically include elements such as a description of the “as-is”

business process (if it already exists), with information on how the current statistics are produced,

highlighting any inefficiencies and issues to be addressed; the proposed “to-be” solution (with

clear improvements and benefits); detailing how the statistical business process will be developed

to produce the new or revised statistics; an assessment of costs and benefits, and any external

constraints.

2. Design and build phases

4.14. The design phase comprises all the activities undertaken to define the statistical output and

the concepts, methods, collection instruments and operational processes necessary. Accordingly,

this phase includes all the design elements needed to define or refine the statistical output identified

in the previous phase, all relevant metadata, ready for use later in the statistical business process,

and quality assurance procedures.

4.15. These activities make substantial use of international and national standards, in order to

reduce the length and cost of the design process and enhance comparability and usability of

outputs. Organizations are also encouraged to reuse or adapt design elements from existing

processes. In addition, outputs of design processes may form the basis for future standards at the

national and international levels.

4.16. The design of the statistical output for SUTs and IOTs consists of the size and layout of

the tables; the breakdown of industries and products; disclosure control methods; processes

governing access to any confidential information; and the identification of the statistical variables

needed, which is then linked to the data collection phase.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

103

4.17. In the build phase, the production solution is put together and tested to the point where it

is ready for use in a live environment. This phase is broken down into several activities, which

include a review of data sources; the configuration of the workflow from data collection through

to dissemination; and testing of the statistical business process. For statistical outputs produced on

a regular basis, the design and build phases usually occur for the first iteration, and following a

review or a change in methodology or technology, rather than for every iteration.

4.18. During the design and build phases a number of specific issues in the compilation of SUTs

and IOTs should be considered. These include, for example, the choice of the level of detail of

industries and products of SUTs and IOTs at working level and in the dissemination phase; how

to handle confidentiality throughout the compilation process; the schedule for the compilation and

dissemination of SUTs and IOTs; the revision policy and analysis; the resources required to sustain

the compilation; the benchmarking; and the choice of the index formula and the base year. In

addition, it is important to create and maintain documentation for all phases of the compilation

process to serve as a quality control measure of the process. All these activities in the design and

build phases are further elaborated in the following sections.

(a) Level of industry, product detail and size of SUTs and IOTs

4.19. The level of industry and product detail of the published and disseminated SUTs and IOTs

greatly depends on the objectives of the tabulations and their uses. The industries and products

explicitly identified in the disseminated tables reflect to a great extent the users’ needs and the

specific policy concern of interest. For example, if a specific environmental domain is of interest,

such as energy, specific industries and products are likely to be explicitly identified in the

disseminated tables in order to address the specific domain. The ultimate level of aggregation of

the disseminated SUTs and IOTs has an impact on – and at the same time is affected by – the data

availability and the data collection, compilation and balancing procedures.

4.20. The number of products in the SUTs is usually higher than the number of industries, thus

showing more than one primary product for each industry, casing the SUTs to be rectangular. Their

size and shape will have appropriate implications for IOTs, the physical tables and other related

products analyses (for example, productivity).

4.21. The level of detail of industries and products at the working level is generally very

disaggregated and the recommendation is to work with the most detailed level of aggregation

taking into consideration the constraints posed by the available data, resources, and burden on

business. Various aspects need to be considered, including the user needs, the availability of data,

and the level of detail used in national accounts. For example, compilation aspects that influence

the level of detail (since they facilitate the compilation and validation of the data at the working

level) include:

• Distinction between industries which are allowed to deduct VAT and those that are not

allowed to deduct VAT, and different VAT rates per product and categories exempt from

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

104

VAT to facilitate the compilation of the valuation matrices in particular for taxes and

subsidies.

• Distinction between market and non-market activities to facilitate the understanding of the

input structure of GVA.

• Subdivisions of industries according to institutional sectors which have an impact on the

linking table between SUTs and the institutional sector accounts.

• Separate identification of certain industry and product subdivisions to facilitate the

compilation of the trade and transport margin matrices, such as section G and its divisions

of ISIC Rev. 4.

• Links to international industry, product and functional classifications.

• Links between output structures, input structures, price indices and values for deflation,

along with the availability of price statistics to generate estimates in volume terms.

• Links between domestic supply and exports, and also intermediate consumption and

imports, to study the input structure of the industries.

• Links to the environmental accounts.

• Enterprises that trade internationally and have links to global value chains (see extended

SUTs and the OECD trade by enterprise characteristics (TEC) database).

• Availability and quality of source data.

• Size and value of output and value added of the industry: if the industry is too large and

heterogeneous then it should be further broken down. The same can apply to products.

• Benchmarking (annual as opposed to five-yearly) using comprehensive sources and

censuses, since the level of detail in benchmarking years is much larger than that during

regular annual compilation of SUTs and IOTs.

• Annual chain-linking the volume estimates.

• Staff resources, time schedules for production and publication, confidentiality and system

infrastructure.

4.22. An appropriate choice of the level of industry and product detail in the SUTs at working

level will facilitate the compilation and the search for causes of inconsistencies. For many

products, it is possible, by their nature, to identify the industry in which they are used. For example,

fertilizers are mainly used in agriculture, crude oil in oil refineries, concrete in construction, and

so forth. For some products, it may also be possible to identify whether they are used as

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

105

intermediate consumption or final consumption. Haircut services, for example, can be assumed to

be mainly consumed by households and thus recorded in household final consumption expenditure.

The more detailed the classification of products used in SUTs, the easier it is to use expert

knowledge to supplement surveys in allocating products to different uses.

4.23. Linking and matching products to valuation-related elements (taxes, subsidies, trade and

transport margins) make the composition of transactions more transparent and clear-cut and

significantly facilitates their analysis.

4.24. If the products are more detailed, it will also mean that the number of users of a certain

product is greatly reduced. Where there is only one producer and one user of a product, the search

for the cause of inconsistencies would only require investigating two source statistics. When a

product has 20 users, for example, the search becomes more complicated.

4.25. Questions relating to the layout of SUTs tables also arise for the final use part of the SUTs.

It may be useful to integrate the functional classifications in the final consumption data, showing

final consumption by products and by consumption purpose. On the other hand, it may be better

to keep such detail and transformation separate, but this integration will still be needed in some

form or other in the compilation of SUTs and the other accounts of the system.

4.26. In the dissemination phase, the size and breakdown between industries and products shown

in the SUTs (and thus in the IOTs) mainly reflect the user’s needs and the objective of the

tabulations, taking into account confidentiality considerations. Other presentational considerations

include, for example, the size and relative value of output and value added for the industries and

the size and value of supply for the products. Industries or products that are not economically

significant or relevant for a particular economy may be aggregated together, while a more detail

breakdown may be shown for economic activities that contribute substantially to GDP, in order

more effectively to analyse the cost structures and the interdependencies with other economic

activities.

4.27. It should be mentioned that, when compiling consistent annual, or quarterly, SUTs, the

stability of the level of detail of the applied classifications is also important, as many ratios and

proportions will usually be taken as a starting point in the estimates for the following year.

4.28. A higher degree of product detail also supports the use of certain estimation methods, for

example the product flow method of compiling national accounts (namely, balancing the supply

and use of products) by taking into account the relevant differentiation concerning product tax

rates, margin rates and homogeneity in prices. Moreover, it is much easier to distribute

disaggregated products and services across industries and final use categories with the product

flow method than at a higher aggregate level. Detailed product accounts also help in the balancing

procedure, as it is easier to explore and detect the causes of imbalances if the basis is determined

by homogeneous single products rather than aggregate groups of products. The work on a detailed

product level certainly increases the data quality but has resource and systems implications. At

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

106

higher levels of aggregation, problems of imbalance might not even be detected at all and therefore

not remedied.

4.29. Table 4.1 provides an example of the size of SUTs and IOTs compiled by selected

countries. It is worth noting that the internal working level of the industry and product detail used

for compilation and balancing is much higher than which is actually published. For example, in

the United States, the working level in producing the SUTs is over 800 industries, whereas in

Denmark, there are around 2,350 products at the working level but the SUTs are published only at

a level of 64 products by 64 industries.

4.30. It is important to distinguish between the detail required for the compilation and balancing

work at the working level as opposed to the information required for the publication. The in-house

operating detail should be the same or, as in most cases, in greater detail in terms of number of

industries and products than that allowed for disclosure by the publication. For example, many

countries distinguish hundreds or even thousands of products but do not publish at these levels, as

a great deal of confidential information would thereby be released. It should be noted, however,

that countries often allow people outside the national statistical office to have access to more

detailed data, albeit confidential and under signed agreements, for analytical purposes.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

107

Table 4.1 Examples of the size of published and internal working level SUTs and IOTs

Key: PYP – previous year’s prices

P x P – product-by-product

I x I – industry-by-industry

(1) The above table was compiled by Sanjiv Mahajan (Office for National Statistics, United Kingdom) at the time of preparation

of this Handbook. The numbers are indicative for the reference year 2014 (as at December 2016) unless denoted otherwise.

(2) Other differences will exist in terms of comparability, given that the tables are compiled on different bases, in terms of the

frequency of the tables, classifications used, the SNA version, latest reference year, latest benchmark year, valuation of SUTs

(basic prices or producers' prices or purchasers' prices), assumptions underpinning the IOTs, etc.

(3) For Argentina the SUTs reference year is 2004 and the IOTs reference year is 1997.

(4) For Australia the IOTs working level operates at 1268 products and 114 industries.

Volume terms

or PYPs

Number of

products

Number of

industries

Number of

products

Number of

industries

P x P

Tables

I x I

Tables

P x P

Tables

I x I

Tables

Argentina

(3)

271 162 271 162 N n/a 124 n/a 124

Australia

(4)

301 67

n/a

n/a Y

n/a 114 n/a 114

Au stri a 573

135

74 74 P 74 n/a 74 n/a

Belgium

(5)

355

135 64 64 P 135 n/a 64 n/a

Brunei Darussalam

(6)

324 74 74 74 N 74 74 74 74

Canada 490 230

490

230 P

n/a

230 n/a 230

Chile 275 160

180

111 Y n/a 111 n/a 111

Columbia 369 61 61 61 Y 61 61

61 61

Costa Rica

183 146 183

138 Y 183

136 183 128

Czech Republic

252 120 88

88 Y 184 184 82

Denmark

(7)

2 350

117 117 117

Y n/a 117

n/a 117

Estonia

247 98 64 64

Y 64 n/a 64 n/a

Finland 776 179 64

64 Y n/a

179 n/a 64

Germany 86 63 85 63

P 73 n/a 72 n/a

Hungary

(8)

820 242 64 64 Y 88 88 64 64

Iceland

(9)

561 142 n/a

n/a P n/a n/a n/a

n/a

India

(10)

142 126 140 66 N 130 n/a 130 n/a

Indonesia

(11)

244 81 n/a

n/a P 251 n/a

185 n/a

Ireland

(12)

82 82 58 58 Y 82 n/a 58 n/a

Kuw ait

(13)

43 43 n/a

n/a N 43 43

n/a n/a

Mexico

(14)

819 814 262 262 P 814 262 814 262

Netherlands 614 128 85 76

Y n/a 128 n/a

New Zealand

(15)

299 118 201 106 P n/a 106 n/a 106

Norway 860 156 64 64

P n/a 156 n/a

Republic of Korea

1 851 328 384 328 N 1 851 n/a 384 n/a

Saudi Arabia

(16)

59 59 18 18

N 59 59 n/a

n/a

Serbia

216 88 n/a n/a N n/a n/a n/a n/a

Singapore 71 71 71

71 N n/a 71

n/a 71

Slovakia

290 88 64 64 Y 88 n/a 64

n/a

Slovenia

(17)

350 230 64 64

Y 64 n/a 64

n/a

South Africa 104 293 104 62 N n/a 50

n/a 50

Sweden

(18)

403 97

62 64 P 62 n/a 62 n/a

United Kingdom

(19)

112 112 112 112 P

112 n/a 112 n/a

United Republic of Tanzania

(20)

250 59 250 59

P n/a n/a n/a n/a

Unite d States of America

(21)

4 988 819 73 71

P 73 71 n/a n/a

Submissions to European

Commission reflect EU Member

States

(22)

64 64 64 64 P 64 64 64 64

Number of

industries / products

Number of

industries / products

Country

(1),(2)

National Supply and Use Tables

National Input-Output Tables

Current prices

Internal working /

compilation levels

Published levels

Do you

produce such

table s:

Y (ye s)

N (no)

P (plan to)

Internal working /

compilation levels

Published levels

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

108

(5) For Belgium the IOTs reference year is 2010.

(6) For Brunei Darussalam the reference year for SUTs and IOTs is 2005.

(7) For Denmark the number of products can vary from year to year. Detailed data can be made available outside Statistics Denmark

for analytical purposes.

(8) For Hungary the IOTs reference year is 2010 (compiled five-yearly).

(9) For Iceland the SUTs are produced but not yet published. The plans are to publish 64 products x 64 industries.

(10) For India the SUTs reference year is 2012/13 and the IOTs reference year is 2007/08. Industry outputs are published in full

whereas industry uses are aggregated to higher levels.

(11) For Indonesia the SUTs and IOTs reference year is 2010.

(12) For Ireland the SUTs reference year is 2013 and the IOTs reference year is 2011.

(13) For Kuwait the SUTs and IOTs reference year is 2010.

(14) For Mexico the reference is year 2008, the PxP IOTs reference year is 2008 (814 products), updated version for 2012 (259

products), and the IxI IOTs reference year is 2012.

(15) For New Zealand the production of SUTs in volume terms is being planned but there are no plans for publication.

(16) For Saudi Arabia the SUTs and IOTs internal working levels reference year is 2011 and the published SUTs reference year is

2013.

(17) For Slovenia the SUTs reference year is 2013 and the IOTs reference year is 2010. SUTs and IOTs will be published in 2017,

both with a reference year of 2014.

(18) For Sweden the SUTs in PYPs are produced but not yet published. A version is, however, submitted to the Commission

(Eurostat) and the plan is to publish at some stage.

(19) In the United Kingdom the IOTs PxP tables use the hybrid assumption approach

(20) For the United Republic of Tanzania the SUTs reference year is 2007. IOTs for the reference year 2007 are being finalized.

(21) For the United States, in benchmark years, the SUTs are published at the 389 product and 389 industry level of detail. While

IOTs themselves are not currently published, the IxI Leontief Inverse tables calculated from IOTs are published at the 71

industry level of detail, and PxP tables at the 73 product level of detail.

(22) All EU Member States are expected to provide:

Annual SUTs both in current prices and PYPs as well as five-yearly IOTs under the ESA 2010 Transmission

Programme after the expiry of any National Derogations.

The SUTs and IOTs are to be supplied using 64 products and 64 industries.

4.31. Confidentiality is a fundamental principle of official statistics (see United Nations, 2013b).

It ensures that individual data collected by statistical agencies for statistical compilation, whether

they refer to natural or legal persons, are to be strictly confidential and used exclusively for

statistical purposes. It is important therefore that procedures are put in place to ensure the

confidentiality of the information disseminated.

4.32. Countries may apply different criteria to decide whether specific data may be disclosed or

not. This decision is likely to be driven by the legislation in place underpinning the collection of

data from businesses. The decision is normally influenced by the number of enterprises observed

in an industry or by whether the data can be disclosed through a process of deduction. One solution

would be to choose a higher aggregation level with a sufficient number of enterprises in an industry

to overcome any disclosure problems. There may not always be an easy solution for some

industries or the products to which they should be allocated. The price is a loss of information due

to aggregation, resulting in the increased heterogeneity of the SUTs system. Other methods might

therefore be explored or combined, such as creating or redefining products.

4.33. Cases where there are one or two dominant producers in an industry, such as mining,

extraction of crude oil, sugar, pharmaceuticals and others, pose a different challenge. In these

cases, it is recommended, when necessary, that specific permission is sought from the business

when their data are publicly available from other public sources, for example published company

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

109

annual reports and accounts. If permission is not granted, then suppression of the relevant cells in

the SUTs should be considered. The aggregation of industries and products with non-disclosive

industries and products should, however, be avoided, as this results in the loss of useful details for

non-disclosive industries and products.

(b) Schedule of the work programme

4.34. Aligning the monthly, quarterly and annual timetables covering data collection processes,

compilation processes, data supply, validation, balancing and publication for all the accounts and

outputs is a key step in ensuring coherence and consistency. This should include the compilation

of SUTs and other input-output-related products, as appropriate.

4.35. The overall process needs therefore to be split into well-defined blocks of work with clearly

defined processes and linkages between the processes, in such a way that they all fit within realistic

schedules (including contingency planning and risk management) with clear roles and

responsibilities for the staff and management involved. The governance of the programme should

be clear, with regular monitoring and meetings scheduled at key junctures, for example, linked to

key milestones in the process. The project management of the process should ensure that dedicated

resources are attributable to this support function.

4.36. The schedule needs to incorporate deadlines of both data providers and data users, together

with various internal intermediate deadlines. For annual business surveys, the time lag from

changing the questionnaire to incorporating the new results and publication in SUTs and national

accounts could be around three years; thus it is important to retain schedules, which are regularly

reviewed and reflect the incorporation of continual data improvements.

4.37. In general, it is useful to put in place service-level agreements with data providers and also

with data users. Agreements with data providers would cover the types of data to be provided, the

quality criteria, briefings, schedules and the format in which the data will be delivered. Important

elements to consider in such service-level agreements include:

• Clear ownership – senior representatives from both the supplier side and the customer side

• Reasons for data requirement

• Publication of results including disclosure requirements

• Process of data delivery by data provider (for example, format)

• What data are required (need to specify the variables needed)

• Timing of data deliveries

• Required briefing to accompany the data

• Handling of customer queries

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

110

• Quality (covering such criteria as consistency, credibility, revisions, precision and

communication)

• Methodological notes supporting the data (for example, sources, methods, coverage, and

others)

• Development, improvements and consultation

• Arrangements for review of the process, among others

4.38. Service-level agreements with data users, on the other side, would cover, for example,

users’ deadlines (such as those linked to policy agendas, research schedules, and other factors) to

be reflected in the statistical production schedule. For example, the finance ministry or the central

bank or both may require structural updates on the economy to fit with their policy review. Such

user requirements may form part of regular annual schedules for the producers of SUTs.

4.39. In general, there should be a regular review, possibly on an annual basis, of all aspects of

the process (including timetables, data quality, implementation of future changes, and other

considerations) with both data providers and data users, to ensure continuous improvement and to

guide changes as necessary.

4.40. The schedule for the compilation of SUTs naturally depends on the periodicity and

frequency of SUTs and IOTs. In general, it is recommended that SUTs be compiled annually, in

line with the United Nations Statistical Commission recommendations on the scope of the

implementation of the 2008 System of National Accounts.

While it is recommended that a

benchmark system of SUTs based on specific survey results be compiled every five years, rapid

changes in the economy, the external impacts of globalization, the increasing rate of change of

technology and its impact, new products, new industries, the impact of digitization and other

factors may affect the production process of SUTs and the periodicity of benchmarked tables, with

the result that an annual benchmark process may be preferred.

4.41. Schedules setting the frequency of SUTs compilation may be assessed against the revision

policy and the uses of intra-annual sources (such as quarterly and monthly short-term indicators),

with a view to incorporating if necessary the revision guidelines and indicators policy.

4.42. Revisions to time series data are an important part of the production process. Changes to

published data can occur for many different reasons. For example, forecast data may be replaced

by survey data, the reclassification of industries, methodological changes to the way in which data

are estimated or just through the correction of errors. Changes due to the correction of errors should

be identified as corrections and distinguished from revisions that are more commonly associated

See the recommended data set in the report of the Intersecretariat Working Group on National Accounts

(E/CN.3/2011/6), available at http://unstats.un.org/unsd/statcom/doc11/2011-6-NationalAccounts-E.pdf.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

111

with improving estimates as more information is gathered over time (Mahajan, 2015). There is a

conflict between the release of timely estimates and that of accurate estimates. If statistical offices

and central banks waited to publish the most accurate data possible, given the nature of data

collection, there would be a large time lag between the date to which the data refer and the date of

publication.

4.43. Bringing together data for the purposes of compiling SUTs and other input-output-related

products integrated within the national accounts involves aligning production timetables (quarterly

or annual) and schedules, and also revision policies. Ideally, a highly effective revision policy to

ensure that revisions are implemented in a coordinated and coherent fashion across the accounts

should cover the national accounts, balance of payments and government finance statistics, should

include the SUTs and IOTs, and should also extend to primary source data and other domains,

such as regional accounts and environmental accounts.

4.44. The revision policy should reflect appropriate criteria to assess each revision, including

when best to implement the changes, for example, in a quarterly exercise (for example, for short-

term revisions) or in an annual exercise (for example, for long-period revisions). This will have an

impact on how to compile SUTs and IOTs and the revision guidelines should be operated flexibly,

reflecting such issues as economic significance and practical aspects such as their impact on

resources and systems.

4.45. The usual guidelines applied to all the accounts cover:

• Revisions to the latest quarters for an incomplete year (these can affect annual SUTs)

• Revisions to past recent quarters since the last full benchmark year

• Revisions made through the annual process to recent years, say, between three and five

completed years

• Revisions to a longer period, sometimes viewed as a major revision. Many earlier years may

be revised depending on whether they meet certain criteria such as those relating to

methodological improvements, correction of errors and economic significance

4.46. Any changes to back data will also have an impact on the monthly and quarterly seasonally

adjusted estimates.

4.47. In some countries, occasional or major revisions of national accounts are usually carried

out every five years and require more resources if the revision is implemented on the basis of a

large SUTs system. A revision at a more aggregated level is always easier and less demanding.

4.48. The various revision practices have different pros and cons:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

112

• Regular revisions to SUTs and national accounts help to preserve good quality levels and

growth rates. Some users do not welcome regular revisions, however, whereas others do

not welcome the “big bang” approach to revisions.

• Five-yearly revision exercises may mean more significant changes for a number of years

until the next revision window.

• Revisions only at the aggregated level may provide problems with the detail and provide

discontinuities in the long time series of annual SUTs

• Revisions also at the level of IOTs ensures that SUTs and IOTs remain consistent.

4.49. Revisions applied only to part of the accounts and not all the relevant outputs would

generate incoherence across different outputs and not help users, for example, in some countries,

SUTs may be revised but the IOTs are not revised, thus the links between the two products are out

of line. This situation should be avoided.

4.50. Analysis of revisions can provide information about the reliability of estimates and how

they change between the first estimate and final estimate as well as a source for the identification

of any biases, (Mahajan, 2004b). Note that revision analysis does not give information on the

accuracy of an estimate, for example, the final estimate may not be accurate in terms of sampling

error or non-sampling error.

4.51. The knowledge of the source and the reasons for the revisions is key and helps producers,

and users, to better understand the data. Sometimes with major revisions going back in time,

understanding the changes can be quite complicated, such as changes in definitions, classifications

and data. For analytical users of SUTs and IOTs, it is important to understand the reasons and

impact, especially when there is a mix of revisions, such as, for example, with the introduction of

2008 SNA, necessitating the revision of SUTs and IOTs going back 30–40 years or more. The

forthcoming handbook on backcasting methodology (United Nations, forthcoming) will provide

more information on this topic.

(d) Resources

4.52. When considering the resources and human requirements for the full integration of SUTs

and IOTs (and also physical SUTs) in the compilation of national accounts and balance of

payments, it is essential to distinguish between the first compilation of the SUTs, the recurrent

production of SUTs and the process surrounding major backward revisions.

4.53. A substantial amount of resources is required to build up an integrated SUTs framework

for the first time. This work involves establishing all the planning, conceptual and methodological

work, data collection needs, requirements of all the individual industry and product balances, the

development of appropriate techniques for incorporating the primary sources and new software for

handling the SUTs system and the necessary training and investment in staff and systems. The

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

113

investment may lead to considerable changes in working procedures towards a better integration

of activities from data collection through to publication, following an integrated statistics

approach.

4.54. The resources needed to establish an integrated SUTs framework should, however, be

viewed against the way in which the development will evolve, for example, the level of integration

and the organization of roles and responsibilities across the statistical domains. The

implementation of the recommendations and guidelines provided in this Handbook can be carried

out in a gradual manner taking into account countries’ specific situation in terms of the resources

available and their national priorities. Countries’ practices may vary considerably, among such

scenarios as the following:

• Only annual SUTs in current prices

• Annual SUTs in both current prices and in volume terms

• Annual SUTs complemented with a quarterly SUTs system

• Annual and quarterly SUTs in current prices and in volume terms

• SUTs at basic prices or at purchasers’ prices

• Benchmarking annually or, say, at least every five years

• Links between SUTs and IOTs:

o SUTs only

o SUTs and IOTs, both in current prices (possibly also in volume terms)

o Only IOTs and no SUTs – this is clearly the least favoured scenario but it has a

historical legacy in some countries

4.55. Developing a new SUTs and national accounts system poses different challenges involving

the change of existing production systems while maintaining business-as-usual activities.

4.56. Where periodicity is concerned, at least every five years a benchmark system of SUTs

should be compiled which is based on more exhaustive specific survey and administrative data

results. As described earlier in this chapter, however, when considering schedules for SUTs (and

IOTs), with rapidly changing and developing economies, the impact of globalization and

digitization, and other factors, it is recommended that the development of new SUTs systems

should reflect an annual benchmarking and reconciliation process. This will also help to avoid

significant revisions and distortions to the levels of data. Together with annual chain-linking, this

will mean better quality measurement of the growth rates of GDP in volume terms, in particular

for the more recent periods. Appropriate techniques have been developed and the trends of

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

114

structural change (for example, composition of outputs and intermediate consumption) during the

previous years can be used to take forward the structures in SUTs, if no new structural information

is available, for example, beyond benchmark years.

4.57. The experience from countries which have integrated the SUTs framework into the

compilation of their national accounts suggests that the resources needed are similar to the

resources for those countries following a more traditional approach with a separate SUTs

compilation. It is therefore recommended that SUTs are compiled as an integrated process and a

regular part of the compilation of the national accounts. The fact that IOTs are produced with

relatively few additional resources if a SUTs system is in place militates in favour of an integrated

and regular approach for the compilation of IOTs.

(e) Benchmarking, extrapolation using indicators, price and volume information

4.58. Planning a new system of SUTs is generally linked to a benchmark year for which the most

important areas of the economy are covered by censuses and surveys. This is especially important

when policy decisions are based on the levels of the figures, for example, the level of gross national

income (GNI). Some detailed data sources are collected at more or less regular intervals and will

not all be available for the benchmark year. Hence, figures that do not relate to the actual year will

need to be corrected for the changes that have taken place between the reference period of the data

and the period for which they are being used. This can be done using indicators for value or volume

and price indicators.

4.59. When a balanced benchmark SUT exists, the compilation of SUTs for following (or

previous) years will usually be considerably easier. It is possible to use information on the

structures of the benchmark table to fill the gaps between those cells for which no new source data

are available or to extrapolate as appropriate. For example, input structures will change over time

and information on new input structures can then be used as it becomes available, replacing the

extrapolated information. Taking into account that the other estimated structures are subject to

uncertainty, it may be sufficient to review them at intervals of a few years, one at a time. Even

when such structures are reused from the previous year, they will change over time as a result of

balancing the SUTs. This could happen even more rapidly with the impact of globalization and the

development of new products.

(f) Features of GDP in volume terms

4.60. When looking at the change in the economy over time, the main concern is often whether

more goods and services are actually being produced now than at some time in the past. With

productivity, however, the point of interest is whether this output is increasing relative to the

inputs.

4.61. Over time, changes in current price GDP show changes in the monetary value of the

components of GDP. These changes in value can reflect changes in both price and volume, making

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

115

it difficult to establish how much of an increase in the series is due either to increased activity in

the economy (volume change) or to an increase in the price level. For productivity measures, only

volume changes are used. It is therefore useful to measure GDP in volume terms (preferably in

previous years’ prices), meaning that price effects are excluded, including in current prices. In

most cases, the revaluation of current price data to remove price effects (known as deflation) is

carried out by using price indices such as component series of the retail prices index or producer

price index, to deflate current price series at a detailed level of disaggregation.

4.62. At the international level, constant price estimates and chain-linked volume measures are

two common measures for volume change of GDP. Under the constant price method, a certain

year is selected as the base year; constant price volume estimates of GDP for subsequent years are

the aggregation of its components computed by multiplying the price of each component in the

base year by its volume in the current year, and the real growth is derived from the comparison of

constant price volume estimates at different years.

4.63. Under the chain-linked volume measures method, the annually reweighted chain-linking

approach is adopted to compile the volume measures of GDP and its components. First, the volume

estimates of major components of GDP in the current year are revalued at preceding year prices,

which in practice are calculated by deflating the current price values of subcomponents by the

relevant price indices. Second, the short-term volume indices for different years, calculated by

dividing the volume estimate of GDP from the initial step by the current price GDP in the previous

year, are chain-linked to a selected reference year in order to obtain a continuous time series of the

chain volume indices of GDP and its components. The chain-linked volume index series can be

converted into the chained monetary series by multiplying the chain-linked volume index by the

current price value in the reference year.

4.64. For some series, price indices for particular goods and services are used to deflate the

current price series, such as components of the following:

• Consumer price index (CPI)

• Retail price index (RPI)

• Producer price index (PPI)

• Corporate services price index (CSPI)

• Import prices

• Export prices

4.65. The process known as “double deflation” is the preferred method for the estimation of

GVA in volume terms. This is achieved by deflating the value of output and the value of

intermediate inputs separately to get corresponding volume measures, and then subtracting the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

116

latter from the former. This double deflation approach means that an industry’s total output is

deflated by the price of its primary and secondary output, while each intermediate input is deflated

by its own price index.

4.66. This is in contrast to the single deflation method, whereby GVA in current prices is deflated

directly using an output-based deflator to arrive at GVA estimates in volume terms. The single

indicator volume estimates can also be derived in other ways, for example, by deflating output

with output price indices, assuming a constant GVA to total output ratios from the base year, or

using volume indicators directly. This direct price deflation of GVA is not recommended by the

SNA when using a single indicator method.

4.67. The SUTs offer a major advantage and a natural framework that enables double deflation

to be applied in a coherent and consistent manner across the national accounts.

4.68. Chain-linked volume measure series are expressed as index numbers in which the series

are simply scaled proportionately to a value of 100 in the reference year. These index numbers are

volume indices of the so-called “base weighted” or “Laspeyres” form.

4.69. Aggregate price indices are of the “Paasche” or “current-weighted” form. They are

generally calculated indirectly by dividing the current price value by the corresponding chained

volume measure and multiplying by 100. Examples are the GDP deflator and the households’

consumption deflator.

4.70. Value indices are calculated by scaling current price values proportionately to a value of

100 in the reference year. By definition, such a value index, if divided by the corresponding volume

index and multiplied by 100, will give the corresponding price index.

4.71. From the point of view of production, GDP at market prices is at best estimated with

reference to annually compiled SUTs both in current prices and in previous years’ prices. The

SUTs are compiled in previous years’ prices in order to achieve an accurate breakdown of value

changes in subsequent years according to volume and price changes.

4.72. The base-year table provides the specific weights for each industry and product, used in

the index formulae by which the price data are aggregated.

4.73. The great statistical benefit of a system based on previous years’ prices is the fact that the

weights in the index formulae are always up-to-date, thus reflecting the structure of the recent past,

and in turn optimizing the quality of the GDP growth rates in volume terms for more recent periods.

(g) Choice of index number formulae

4.74. In order to calculate price and volume measures, a number of methodological choices have

to be made, for example:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

117

• Which index number formulae will be applied

• Whether a fixed base year or an annually changing base year will be applied

4.75. Different index formulae may be applied using different weighting schemes. It is beyond

the scope of this manual to discuss in depth the theoretical and practical considerations relating to

this choice. Chapter 15 of the 2008 SNA and Eurostat (2016) provide much more detail on the

choice of index formulae.

4.76. Economic theory suggests that an index formula that assigns equal weight to the current

year and the base year is to be preferred. This is one of the reasons why the SNA has a preference,

albeit not a strong one, for the so-called “superlative” indices, like Tornqvist and Fisher.

4.77. Although the superlative indices have a number of attractions, it should be noted they also

have notable disadvantages:

• Superlative indices are demanding in their data requirements and will increase the work

burden significantly.

• Superlative indices are less intuitive than Laspeyres and Paasche indices.

• Superlative indices are not additively consistent, which is a serious constraint when applied

in an accounting framework.

• Values change does not always equal volume change times price change.

4.78. From a practical point of view, a number of requirements can be imposed on the index

numbers:

• The applied index formulae should be a good approximation of the changes obtained by

the superlative indices.

• A change in value must be divided into a price change and a volume change without a

residual.

• Values in volume terms for aggregates should equal the sum of values in volume terms of

constituent parts, applying the same index formulae.

• In addition, it is sensible for the index formulae to be relatively straightforward and easy

to interpret for users.

4.79. The last three requirements can only be met with the application of the Laspeyres volume

index formula and the Paasche price index formula. The formulae underpinning these indices are

shown below

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

118

Laspeyres volume index

Paasche price index









































































































where  is the price and  is the quantity.

4.80. The characteristic for the Laspeyres volume index is that the volume changes of individual

goods are weighted with the value of the transaction concerned in the base year..

4.81. The characteristic for the Paasche price index is that the price changes of individual goods

are weighted with their value of the transaction concerned in the current year. The deflated values

derived with this index formula combination can easily be explained as values in prices of the base

year.

4.82. It can be easily shown that the decomposition of value changes, in terms of volume and

price changes, do not have a residual.

























= 



× 



4.83. The deflation of values in current prices by using a Paasche price index gives:





/



= 



























= 







4.84. This illustrates that the deflated aggregate equals the sum of deflated components, which

means that additivity in the SUTs for volume estimates is assured. The use of Laspeyres volume

indices and Paasche price indices ensures that the current price identities also hold in volume terms.

This also means that, after balancing, for every product, the total supply equals total use, and for

every industry, the total output equals total intermediate consumption plus gross value added and,

in volume terms, this is:

Total supply in volume terms equals Total use in volume terms

Total output in volume terms equals Total intermediate consumption in volume terms

plus Total value added in volume terms

(h) Choice of the base year

4.85. By applying the Laspeyres volume index number formula, the volume changes are

weighted with the values of the concerning transaction in a base year. The question arises which

year should be chosen as the base year. Generally speaking, there is a choice between a fixed base

year and a changing base year.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

119

4.86. With the method of fixed weights for a series of years, the weights are derived from a single

year in the past. An advantage of this method is that, with long time series of values at prices of

the base year, the deflated components of aggregates add up exactly to the deflated aggregate. A

serious disadvantage, however, is that volume changes of aggregates are calculated with outdated

weights. This disadvantage is especially severe when relative prices change rapidly and, as a result,

economic growth can be significantly misrepresented. In addition, the disappearance of products

(such as vinyl records, cathode ray tube televisions, and so forth) or the appearance of new products

(such as mobile telephony, pharmaceutical tablets, iPads, and the like) can lead to notable

distortion in the estimates of economic growth. Even a fixed base year has to be changed every

five years, and then all previously published real growth rates will be revised – this change is often

not welcomed by users.

4.87. Applying a changing base year means that the weights are updated every year and are

usually derived from the previous year. Since those weights are more up-to-date, a better

approximation of the volume changes is obtained than with the method of using fixed weights.

The time series can be obtained by multiplying separately estimated year-to-year volume indices

– this is called “chaining”.

4.88. An important advantage of the chaining method is that the above-mentioned

misrepresentation of growth rates is avoided. In fact, chain-linked volume measures provide a

more reliable measure of volume growth, provided individual prices and quantities tend to increase

or decrease steadily over time. They also have a key disadvantage, however, in that when the time

series are in prices of the previous year, the deflated components of an aggregate then no longer

add up exactly to the deflated aggregate. As a result, mathematical discrepancies will appear that

cannot be removed without distorting the underlying actual volume and price movements.

(i) Documentation

4.89. As the compilation of SUTs is a complex process, a thorough documentation of the basic

data and the methods used, the problems encountered, the solutions applied, and the results

achieved is highly recommended. Such an annual (or quarterly, if appropriate) inventory is not

only worthwhile for purposes of publication but also for internal use in the compilation process

itself and future exercises. When SUTs have to be balanced, information on, in particular, the

sources and methods of estimation for each single supply and use element is needed to evaluate

and analyse industry and product imbalances. The documentation helps to evaluate the data quality

and outline the strategy for balancing. The balancing steps should also be documented, of course,

in order to avoid repeating changes and the destruction of already balanced data.

4.90. Documentation of the various compilation steps can also point to missing data issues and

problems of basic data quality. It is important that such findings are used as feedback to primary

statistics and to identify priorities in improving the compilation methodology. A documentation

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

120

system for the SUTs should be seen in the frame of the overall documentation for the system of

national accounts data.

4.91. Producing a balanced set of SUTs for several years is like solving a puzzle. At first, all

macroeconomic data, survey results, census results and other valid economic data on supply and

use of products in the economy have to be collected. In a second step, missing data have to be

estimated on the basis of harmonized methodologies and documented procedures. In the third and

final stage, the balancing of the SUTs system generates a consistent set of macroeconomic

variables in current prices and in previous years’ prices. Thus the documentation of all stages of

the compilation of hard data in terms of sources and soft data in terms of estimates and adjustments,

ideally for each cell of the SUTs system, is key. In turn, a quality assessment using clear criteria

for each cell can also be developed over time.

4.92. Links between survey data and final national accounts data should be maintained in the

system, in particular separately recording documentation on the survey data, coverage adjustments,

conceptual and valuation adjustments, quality adjustments and balancing and coherence

adjustments. Analyses of these types of adjustments over time and over successive exercises can

also help to highlight any biases, incoherence in data sources, and other shortcomings and, in turn,

help in developing priorities and strategies for further improvements and investment.

C. Collect phase

4.93. The collect phase of the GSBPM (see figure 3.2) consists of all the activities concerning

the collection and gathering of all necessary information (data and metadata), using different

collection modes (including extractions from statistical, administrative and other non-statistical

registers and databases), and their storage in an appropriate structured environment for further

processing. While it may include the validation of dataset formats, it does not include any

transformations of the data themselves, as these are all effected in the “process” phase. For

statistical outputs produced regularly, this phase occurs in each iteration.

4.94. Generally, the compilation of SUTs and IOTs relies on the data sources used for calculating

GDP according to the production, income and expenditure approaches. An overview of the range

of data sources generally used is provided in Box 4.3.

4.95. It should be noted that this situation is reversed in some countries, such as Chile and Japan,

where the requirements of SUTs (and IOTs) are first clearly defined and then the surveys (regular

and ad hoc) are undertaken to collect data to meet these requirements. This may be ideal for SUTs

(and IOTs), but it may also be more costly.

4.96. The compilation of SUTs is data demanding and in principle could be based on many

sources that could help to populate the SUTs. It is strongly recommended, however, for purposes

of timeliness, that more data sources should be regularly available, and that these are also based

on official statistics, preferably using the same business register sampling frame. It is also worth

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

121

noting that many sources can provide data feeding into both the supply table and the use table for

the respective industries and products, as appropriate. For example, annual structural surveys

provide details on both sales and purchases.

Box 4.3 Data sources generally used

Monthly, quarterly, annual, regular and ad hoc business surveys based on the business register sampling frame:

• Monthly business surveys covering production, distribution, construction and service industries and

financial information such as employment and turnover

• Monthly surveys collecting price-related details such as producer (input and output) prices covering

all industries (for example, manufacturing and services), retail and consumer price indices, import

prices, export prices, and earnings and wage prices

• Quarterly business surveys covering areas like capital expenditure, inventories and profits

• Annual business surveys covering structural business statistics for each industry:

o Detailed information covering such variables as turnover, employment, purchases, capital

expenditure, inventories, holding gains, taxes and subsidies, and others

o Detailed sales by type of product

o Detailed purchase data by type of product, covering intermediate-type products and capital-

type product separately

• Range of quarterly and annual surveys covering financial services, in particular financial assets and

liabilities, international trade in services and foreign and direct investment

• Economic censuses every three to five years.

Household-based surveys:

• Living costs and food surveys – collecting details on expenditure by households

• International passenger surveys – expenditure by residents abroad and non-residents’ expenditure in

the domestic economy

• Labour force surveys – collecting details on employment including hours worked

Administrative data:

• Census – population estimates for grossing purposes for use in household-based surveys

• Pay and profits data relating to tax and employment records from collecting government departments,

along with data covering self-employed incomes

• VAT payments data and turnover subject to VAT by industry (and by product where differential rates

exist) from the tax collecting departments

• Imports and exports of goods data collected by customs

Other government departments:

• Agriculture industry data from the agricultural ministry

• Banking industry data from the central bank

• Data on government incomes and expenditures from the finance ministry – expenditure will need to

be split between the individual consumption and collective consumption categories. This source can

also provide the full range of taxes receivable and subsidies payable

Other sources:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

122

• Company annual report and accounts (in general, the direct use is limited, however these are often

used for public utility companies like electricity, gas, water, postal services, telecommunications and

others, in many countries)

• Regulatory bodies’ accounts

• Insurance data from the insurance industry regulators

• Airline data from the airline industry regulators

• Financial details from company websites which supplement the company annual report and accounts

4.97. The compilation of SUTs, whether carried out as an integral part of the compilation of

national accounts (as recommended in this Handbook) or as a separate compilation from that of

the national accounts, is based to a large extent on the same data sources as those used for the

national accounts. Figure 4.1 expands on figure 2.3 in chapter 2 to show the typical data sources

used for the compilation of national accounts, which also feed into the compilation of SUTs and

IOTs. Often the same source can provide data feeding into more than one of the approaches to

measuring GDP. For example, the agricultural data from agricultural departments often feeds into

all three approaches to measuring GDP, thereby ensuring natural consistency and coherence of the

data used in SUTs.

4.98. In general, there is a strong correlation with the level of detail used within the SUTs and

the quality of the product balances and the aggregates. The more disaggregated the level of

industries and products, the higher the degree of matching of individual products in terms of

allocation of uses, prices and other characteristics and hence the better the quality. If the product

level is too aggregated, the individual products may be too broad and heterogeneous, and therefore

of lesser quality.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

123

Figure 4.1 Overview of SUTs and IOTs as part of the SNA compilation

1. Typical data sources

(a) Structural surveys

4.99. Many countries rely on annual structural type surveys which fit in naturally to the needs of

the SUTs framework. One of the advantages of such a set-up is that it allows the collection of a

range of variables from a single source – the statistical unit – thereby ensuring the consistency and

coherence of each variable and across variables. Thus, the employment data will be consistent with

the turnover data and, in turn, with the derived GVA data.

1. non-financial corporations

2. financial corporations

3. general government

4. households

5. NPISH

Accumu-

lation

accounts

Supply and use tables at

basic prices

Sector accounts

Total

economy

Rest of

the world

Product-by-

product

IOTs

Industry-by-

industry

IOTs

Goods

and

services

account

Production

account

Distribution

and use of

income

accounts

Balancing

Supply and

use tables at

purchasers'

prices

(balanced)

Valuation

matrices

(balanced)

Single estimate of GDP

(balanced)

System of national accounts

Supply and use tables

Accounts

Supply and

use tables at

purchasers'

prices

(unbalanced)

Valuation

matrices

(unbalanced)

Production

approach

GDP

(unbalanced)

Income

approach

GDP

(unbalanced)

Expenditure

approach

GDP

(unbalanced)

Business

census

Annual

business

survey

Construc-

tion survey

Household

expendi-

ture survey

Govern-

ment

budget

Financial

statistics

Employ-

ment

statistics

Income

survey

Monetary

statistics

Consumer

price

statistics

Producer

price

statistics

Capital

expenditure

survey

Balance of

payments

Foreign

trade

statistics

Production

statistics

Agriculture

statistics

Population

census

Material

input

statistics

Business register

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

124

4.100. Where annual structural surveys exist, data for a number of key variables are collected with

more detail supplemented from other surveys (for example, detailed purchases data). These

variables can include:

• Sales of goods and services (and an appropriate product breakdown)

• Purchases of goods and services (and an appropriate product breakdown)

• Purchases and sales of goods for resale without any further processing (this helps to

produce the trade margins by type of product)

• Changes in inventories (a split between materials and fuels, work-in-progress and finished

goods not sold is required, as the first category affects intermediate consumption and the

latter two categories affect output)

• Capital expenditure (and an appropriate product breakdown)

• Employment costs (and an appropriate breakdown)

• Taxes on products and production (covering business rates, excise duties and others)

• Subsidies on products and production (covering agriculture, transport and others)

• Areas such as research and development, and international trade in services, which link to

other related surveys

4.101. Often data compiled to respond to statistical regulations are used in the compilation of

SUTs. For example, in the European Union, statistical regulations for member States cover various

business statistics derived from monthly, quarterly and annual surveys providing short-term

indictors and structural business statistics. Data from many of these sources also feed into the

compilation of SUTs. The Handbook on the Design and Implementation of Business Surveys

(Eurostat, 1998), provides extensive detail on the conduct of such surveys.

(b) Administrative data

4.102. Administrative data constitute a key data source for the compilation of both quarterly and

annual national accounts and, in some countries, administrative data may be the main data source.

The use of administrative data is growing under the influence of a number of factors, such as good

coverage, declining resource impact on national statistical offices and lessening of the response

burden.

4.103. Administrative data have statistical strengths and weaknesses vis-à-vis sample surveys.

Apart from the low cost of obtaining administrative data, their major strength is that they

commonly have complete – or nearly complete – coverage of the fields to which they relate. As a

result, there are no sampling errors and any non-sampling errors, such as those arising from an

out-of-date business register and inadequate new business provisions, are scarce and minor.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

125

4.104. The weaknesses of administrative data arise from the fact that they are by-products of

administrative systems which are not generally designed to meet the needs of the national accounts.

Examples of these weaknesses include the following:

• Available data do not meet national accounting definitions (for example, wages rather than

compensation of employees, or a measure of depreciation that differs from the national

accounting concept of consumption of fixed capital).

• Purchases registered in the VAT system will usually include both purchases for

intermediate consumption and for gross capital formation.

• Data are not recorded on an accrual basis (for example, exports and imports from customs

are recorded as they cross the customs frontier and not when they change ownership).

• Data are incomplete (for example, movement of oil rigs in and out of territorial waters are

excluded from customs data).

• Data may not be disaggregated in a desirable way (for example, government expenditures

may not distinguish between wages and intermediate consumption or new motor vehicle

registrations may not distinguish between household and business use).

• Administrative data may be untimely (for example, company tax data).

• Administrative data may undergo change as a result of a change in policy.

4.105. The values of outputs, inputs, gross capital formation and other elements have their

counterparts in business accounts or government accounts, but the concepts used in business

accounting often do not consistently follow national accounts definitions. Further details on this

issue may be found in United Nations (2000b) and Mahajan (2013). A few examples are listed

below:

• Differences between concepts: financial intermediation services indirectly measured

(referred to as FISIM), insurance services, and others.

• Change in inventories: different valuations, correction for holding gains and losses, and

others.

• Distinction between intermediate consumption and capital formation: acquisitions of

machinery and equipment included in current expenses, and others.

• Distinction between intermediate consumption and compensation of employees: fringe

benefits, links to own account production, and others.

• Use in business accounts of a time frame that does not follow the calendar year: often

accounts that are closest to the calendar year can be used as an approximation for the annual

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

126

values but it may be appropriate to correct figures for some big enterprises that have

accounts that cover other periods or may have large seasonal patterns, such as the gas

supply industry.

• Use in government accounts of a fiscal year that differs from the calendar year. It can be

misleading to use the information for the nearest fiscal year as if it were identical to

calendar year data. Annual data can be calculated by weighting together data for the two

fiscal years that overlap the actual year. A better method is, however, to use quarterly or

monthly information to split the data from each fiscal year into the shares belonging to

different calendar years where such information exists.

(d) VAT-based statistics

4.106. The VAT system usually provides statistics on those units that are covered by the business

includes the units that collect VAT on all or a part of their turnover and those that can deduct VAT

on all or part of their purchases (capital or current).

4.107. The VAT-based statistics may exist in published form but, even when not published, they

can usually be obtained from the authorities that are responsible for collection of VAT, with

appropriate service level and confidentiality agreements in place. Typically, this source will

contain information on VAT-liable, zero-rated and VAT-exempt turnover and also on deductible

and non-deductible purchases with a classification by industries.

4.108. The VAT-based statistics tend to be available on both a quarterly and an annual basis, and

are usually available shortly after the reference period. A correction based on the final dates of

payment may be necessary if the statistics show payments instead of accruals.

4.109. There are pitfalls to bear in mind when using VAT-based statistics for national accounts

purposes:

• The concepts used in VAT-based statistics are different from those used in national

accounts. As VAT-based turnover covers sales of products from own production, sales of

traded goods and sales of used capital equipment, the VAT purchases cover purchases of

goods and services for use as inputs, goods intended for resale and also purchases of capital

equipment. Before these figures can be used in national accounts, the VAT purchases must

be split into the different shares based on details payments.

• VAT-based statistics do not show figures for units with activities that are not VAT-liable

and they may not include units with turnover below certain thresholds. Such thresholds

differ from country to country but will often exclude a significant share of the smaller

enterprises. The informal economy can – by its very nature – be assumed not to be included

in the VAT-based register sources.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

127

• The industry classification of the VAT-based units may not be the same as the classification

used in national accounts.

• For practical reasons, the units accepted in VAT-based statistics may be enterprises, kind-

of-activity units, establishments, or even conglomerates of enterprises which are allowed a

joint registration for payment of VAT.

4.110. Despite such caveats, the VAT-based statistics may still be the most reliable data source

for the size of some industries that are poorly covered by other sources. The figures from VAT-

based statistics should, however, be seen as the minimal size of the industries in question. It can

be necessary to add values for units below the threshold values and VAT-exempt units, including

those operating in the informal and hidden economy.

(e) Missing data

4.111. If certain data are not available in the official statistical system, the first option would be

to check whether such data are available outside official statistics. One such example is when

intermediate data on advertising costs are not available as a separate item in the business surveys.

In this case, one possible source could be data from relevant trade associations observing the

advertising market. Despite the fact that the data are often not sufficiently comprehensive, or the

classifications differ from the official classifications, so that the data do not fully conform to the

required concepts, these data certainly give a good indication of the advertising market across the

various industries.

4.112. There may be a full set of data observations for a period, and then again for a later period.

Various modelling techniques exist for generating estimates for the intervening periods to populate

the SUTs, such as the basic Holt-Winters approach (Holt, 1957; Winters, 1960). When balancing

SUTs, however, these estimates should be treated as being of much lower quality than the more

reliable and up-to-date estimates.

4.113. Furthermore, if no specific data are available, the expert advice of chambers of commerce,

trade associations, research institutes or other similar organizations could be useful.

4.114. In certain industries, a single company or a few companies are the big players in that

market. Those could also be specifically contacted for expert input or to request some of their

internal data on a confidential basis. For example, telecommunication companies may provide

their revenue data by type of customer; supermarket chains may be asked to provide data on their

sales by products; major railway companies may provide data on the goods transported, and so on.

4.115. Annual company reports and accounts, publications of regulatory bodies and trade

associations, and internet company websites are very useful sources of financial data for businesses

and households.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

128

4.116. Certain estimates can be based on the identities and coherence of the SUTs framework.

This holds true for the application of the product flow method, where detailed supply data are used

to estimate certain use data. The product flow method basically applies fixed allocations which

will need to be reviewed each year. The method should be applied with great caution in populating

SUTs and will depend on the level of product detail. The collection of primary data from various

sources with data confrontation provides the best approach to the populating of SUTs and the

attainment of good quality results.

(f) Exhaustiveness: methods of grossing up

4.117. Statistical sources usually exclude units with employment or turnover below certain

thresholds, while national accounts data should include estimates for these missing units. The

methods used for grossing up will typically be based on an estimate of employment in the excluded

units, and assumptions on output, input, capital formation and other factors by employee. These

assumptions should as far as possible reflect the conditions in comparable units but when the small

units are not covered by source data, the grossing up procedure will necessarily add to the

uncertainty of the estimated totals. It can also be expected that the structures of outputs and inputs

of small units are somewhat different from those found in the units covered by collected data.

4.118. There is considerable interest in the phenomenon of the non-observed economy. This term

is used to describe activities that, for one reason or another, are not captured in regular statistical

enquiries, because they are underground, illegal or informal, consist in household production for

own final use, or simply because of deficiencies in the basic data collection system. Guidance in

this regard is provided in the handbook on measuring the non-observed economy (OECD, ILO and

Interstate Statistical Commission, 2002) and in the manual on the informal sector (ILO, 2013).

4.119. In countries where a significant share of total output and input is found in the informal

economy, it can be appropriate to conduct specific surveys of this activity. With a view to

confronting data on the supply and use of labour, useful information on this subject may actually

be found in population censuses, household budget surveys or labour force surveys. In this respect,

the SUTs framework, in which available source data are combined and balanced, offers the most

promising means of arriving at exhaustive estimates of economic activity.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

129

Part three

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

131

Chapter 5. Compiling the supply table

A. Introduction

5.1. The first step in the compilation of SUTs and IOTs is the construction of an initial and

unbalanced version of the supply table. The values entered into the tables should reflect, as far as

possible, all available knowledge and data on the product structure of each column, although many

values may need to be changed when the SUTs system is balanced. This applies to estimated totals

and also to the values of supply of specific products.

5.2. Before balancing takes place, the estimates for domestic supply, imports of goods and

services should be checked for credibility, and if necessary adjusted as appropriate. These will

then form the starting point for the balancing process.

5.3. This chapter focuses on the steps and data sources needed to compile this initial,

unbalanced version of the supply table. Section B provides a more detailed overview of the

structure of the supply table. Section C focuses on the compilation of the domestic output table

and the necessary compilation steps and section D focuses on the compilation of the imports of

goods and services. Annex A to chapter 5 provides an example of a questionnaire on the collection

of sales of goods and services, inventories of goods and trade-related data.

B. Structure of the supply table

5.4. The supply table shows the supply of goods and services for a given period of time by type

of product of an economy and distinguishes between the output of domestic industries and imports

by type of product. The supply table is generally compiled first at basic prices, reflecting the

valuation of the data sources. As illustrated in Table 5.1, the supply table at basic prices consists

of two main parts: domestic output and imports of goods and services.

5.5. The domestic output matrix contains information on the supply of products by the different

industries. The column for the imports of goods and services contains information on the total

imports by product. The matrices for domestic output and imports of goods and services have the

same row structure, defined by product categories. This structure allows for the horizontal

aggregation of all the elements and the transition from total supply of products at basic prices to

total supply at purchasers’ prices.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

132

Table 5.1 Numerical example of a supply table at basic prices

Millions of euros

Table based on 2011 figures from Austria

5.6. The supply table at basic prices is then transformed into the supply table at purchasers’

prices, through the addition of valuation adjustments represented by valuation matrices covering

trade margins, transport margins, taxes on products and subsidies on products. Table 5.2 shows

the valuation adjustments which are added to the columns of the supply table at basic prices to

arrive at the total supply of each product at purchasers’ prices.

5.7. The first step in the compilation of an initial version of the supply table therefore involves

the compilation of data for total domestic output at basic prices and imports valued at CIF prices

aggregated to total supply at basic prices. The second step involves the compilation of trade and

transport margins, taxes on products less subsidies on products which are used to convert the total

supply of products at basic prices to the total supply of products at purchasers’ prices.

5.8. The data in the domestic output matrix are valued at basic prices, which is the amount

receivable by the producer from the purchaser for a unit of a good or service produced as output

minus any tax payable, plus any subsidy receivable by the producer as a consequence of its

production or sale. The value of output of goods excludes any transport charges invoiced separately

by the producer.

5.9. Data on imports by product from foreign trade statistics are usually valued at CIF prices.

In the 2008 SNA and BPM 6, however, in which total imports of goods are valued FOB, an extra

row has to be added for the CIF/FOB adjustments on imports, in order to reconcile the different

valuations. These adjustments are shown in row (10) of Table 5.1 and explained in detail in section

Agricul-

ture

Manuf ac-

turing

Construc-

tion

Trade,

transport and

communication

Finance and

business

services

Other

services

Output

at basic

prices

(1)

(2) (3)

(4) (5) (6) (7) (8) (9)

Agriculture (1)

8 782

0 0 0 0 0 8 782 3 271 12 052

Manuf acturing (2) 796 182 982 643 1 808 133 44 186 405 124 590 310 995

Construction (3) 83 961 43 060 734 255 179 45 272 563 45 835

Trade (4) 1 4 773 311 54 204 640 257 60 187 600 60 787

Transport (5) 13 465 66 25 538 128 125 26 335 8 150 34 485

Communication (6) 160 1 781 139 43 912 1 253 982 48 228 6 234 54 463

Finance and business services (7) 29 8 902 698 7 588 106 909 3 381 127 508 7 061 134 569

Other services (8) 3 85 13 1 053 143 74 346 75 643 824 76 467

Total (9) 9 867 199 950 44 931 134 837 109 461 79 314 578 360 151 293 729 653

CIF/FOB adjustments on imports (10) 0 0 0 0 0 0 0 - 97 - 97

Direct purchases abroad by residents (11) 0 0 0 0 0 0 0 6 675 6 675

Total (12) 9 867 199 950 44 931 134 837 109 461 79 314 578 360 157 871 736 230

Total of w hich:

Market output (13) 9 763 195 916 41 462 127 401 88 330 18 116 480 989 0

Output for ow n f inal use (14) 104 4 029 3 468 2 134 19 890 2 670 32 295 0

Non-market output (15) 0 4 0 5 302 1 241 58 528 65 075 0

INDUSTRIES

PRODUCTS

Imports

Total supply

at basic

prices

Adjustments

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

133

D. In addition, a further adjustment is added in the supply table to account for direct purchases

made abroad by residents: this is shown in row (11) in Table 5.1.

5.10. These adjustments in the supply table (rows (10) and (11) in Table 5.1 and Table 5.2) have

corresponding entries in the use table (rows (10) and (11) in table 6.1) under the columns for

exports and final consumption expenditures by households. It should be noted that some countries

do not show these estimates in separate rows but consolidate the values across the product groups

in the respective columns, thereby providing a different product balance.

Table 5.2 Supply table at basic prices, including a transformation into purchasers’ prices

Millions of euros

Table based on 2011 figures from Austria

Agric ul-

ture

Manuf ac-

turing

Construc-

tion

Trade,

transport and

communication

Finance and

business

services

Other

services

(1) (2) (3)

(4) (5) (6) (7) (8) (9)

Agriculture (1) 8 782 0 0 0 0 0 8 782 3 271 12 052

Manuf acturing (2)

796 182 982 643 1 808

133 44 186 405 124 590

310 995

Construction (3) 83 961 43 060

734 255 179 45 272 563

45 835

Trade (4) 1 4 773 311 54 204 640 257 60 187 600

60 787

Transport (5) 13 465 66 25 538 128 125 26 335 8 150 34 485

Communication

(6) 160 1 781 139

43 912 1 253 982

48 228 6 234 54 463

Finance and business services (7) 29 8 902 698

7 588 106 909 3 381 127 508 7 061

134 569

Other services (8) 3 85 13

1 053 143 74 346 75 643 824 76 467

Total (9) 9 867 199 950 44 931 134 837 109 461 79 314 578 360 151 293

729 653

CIF/FOB adjustments on imports (10) 0 0 0

0 0 0 0 - 97 - 97

Direct purchases abroad by

residents

(11) 0 0 0

0 0 0 0 6 675

6 675

Total (12) 9 867 199 950 44 931

134 837 109 461 79 314 578 360 157 871 736 230

Total of w hich:

Market output (13) 9 763 195 916 41 462 127 401 88 330 18 116 480 989 0

Output for ow n final use (14) 104 4 029 3 468

2 134 19 890 2 670 32 295 0

Non-market output (15) 0 4 0

5 302 1 241 58 528 65 075 0

Trade

margins

Transport

margins

VAT

Taxes on

products

Subsidies

on products

Total

(9) (10) (11) (12)

(13) (14) (15) (16)

Agriculture (1) 12 052 1 926 274 329 57 - 107 2 479 14 532

Manuf acturing (2) 310 995 48 838 2 540 13 175 7 866 - 49 72 370 383 364

Construction (3) 45 835 0 0 1 529 13 0 1 542 47 377

Trade (4) 60 787 - 52 341

0 575 11 0 - 51 755 9 032

Transport (5) 34 485 0 - 2 800 558 71 - 448 - 2 620 31 865

Communication (6) 54 463

1 493 9 3 375 217 - 34 5 059 59 522

Finance and business services (7) 134 569

0 - 22 2 706 2 159 0 4 842 139 411

Other services (8) 76 467 85 0 1 201

576 0 1 861 78 329

Total (9) 729 653 0 0 23 447

10 969 - 638 33 778 763 431

CIF/FOB adjustments on imports (10) - 97 0 0 0 0 0 - 97 - 97

Direct purchases abroad by

residents

(11) 6 675 0 0 0 0 0 6 675

6 675

Total (12) 736 230 0 0 23 447 10 969 - 638 40 356 770 009

Total of w hich:

Market output (13) 0 0 0 0 0 0 0 0

Output for ow n final use (14) 0 0 0 0 0 0 0 0

Non-market output (15) 0 0 0 0 0 0 0 0

Adjustments

INDUSTRIES

Output

at basic

prices

Imports

Total supply

at basic

prices

PRODUCTS

Total

supply at

basic

prices

V A LUA TION MA TRICES

Total supply

purchasers'

prices

Adjustments

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

134

5.11. A distinction may be made in the SUTs between the three types of production: market

output; output for own final use; and non-market output. In the domestic output matrix, however,

these three categories of production are usually grouped together in the relevant industries and

shown in three supplementary rows for each industry. Thus government services are distributed in

the system according to the various activities in which the government is engaged: for example,

public administration services, education services, health services, recreation services, social

welfare services, and others, but are shown together with the corresponding market producers. For

example, health services provided by market and non-market producers (within the same industry)

are shown as a total. Furthermore, for some industries, the supplementary rows are useful for the

link with the institutional sector accounts.

5.12. Although the supplementary rows make it possible to split output by industry into the three

categories of output, there is no product dimension. Ideally, each industry should be shown

separately (also reflecting different structures and links between the output and the inputs) or

additional analyses produced for the user.

5.13. Imports of goods and services are classified by type of product. Since this table is designed

to show the total supply by type of products, the valuation of imports of goods should be

compatible with the valuation of the domestic production of goods. Imports by type of product are

therefore valued at CIF prices which are comparable with the domestic output at basic prices.

Although total imports of goods are valued at FOB prices, it is not easy to move the imports of

goods by type of product to CIF prices, as this depends upon the source data and the person

providing the transportation (see 2008 SNA, para. 14.77).

5.14. The addition of the two components – production and imports – gives the total supply of

products at basic prices.

5.15. The supply table at purchasers’ prices is obtained by adding various valuation matrices

(earned on both domestic output and imports) to the total supply at basic prices, thus enabling

movement from one valuation to another. The valuation matrices include:

• Trade margins

• Transport margins

• Taxes on products (with non-deductible VAT treated separately from other taxes on

products)

• Subsidies on products (which are deducted)

5.16. It should be noted that, when the supply table is shown with the final column summing to

purchasers’ prices, it is referred to as the supply table at purchasers’ prices. This is actually just

the supply table at basic prices with the addition of the valuation columns. The production and

import sections of the supply table have not been changed and remain valued at basic prices.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

135

5.17. The task of compiling SUTs is a highly integrated process. This is particularly true for the

estimation of the valuation vectors and matrices, where it is often necessary to rely also on

estimates from the use table side in order to obtain the valuation vectors entered into the supply

table. Figure 5.1 provides an overview of how the valuation matrix in the supply table is linked to

a sequence of valuation matrices in the use table. This figure also demonstrates the

interconnections between the valuation matrices linking the supply table and use table. The

estimation of the valuation matrices, considering both the supply table and the use table, is dealt

with in chapter 7.

5.18. The rest of this chapter focuses on the compilation of the supply table at basic prices.

Figure 5.1 Link between valuation matrices in the supply table and the use table

C. Domestic output

1. Structure of the domestic output table

5.19. The first and most elaborated part of the supply table is the domestic output matrix. This

records data on the production of the economy classified along two dimensions: the rows represent

the type of products (based on CPC Version 2.1) and the columns represent the different industry

groupings (based on ISIC Rev. 4). Thus the rows show the domestic output matrix, a single product

by producing industry, and the columns show all the products produced by a single industry.

Supply table at purchasers' prices

Use table at purchasers' prices

Agricul-

ture

Other

services

Output

at basic

prices

Trade

margins

Transport

margins

VAT

Taxes

product

Subsidies

products

Total

Agriculture

Other

services

Total

Final

consumption

expenditure

Gross

capital

formation

Exports

Total

(1)

(6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

(1)

(6) (7) (8) (9)

(10) (11) (12)

Agriculture

(1)

Agriculture (1)

Manufacturing

(2)

Manufacturing (2)

Construction (3) Construction (3)

Trade (4) Trade (4)

Transport (5)

Communication (6) Communication

(6)

Finance and

business services

(7)

Finance and business

services

(7)

Other services (8) Other services (8)

TOT

(9)

Total

(9)

Compensation of

employees

(10)

Other taxes on production (11)

Consumption of fixed

capital

(12)

Net operating surplus/Net

mised income

(13)

Total

(14)

(15)

Subsidies on products

Agriculture

(1)

Other services

(8)

Total

(9)

Taxes on products

Agriculture

(1)

Other services

(8)

Total

(9)

Value added tax (VAT)

Agriculture

(1)

Other services

(8)

Total

(9)

Transport margins

Agriculture

(1)

Other services

(8)

Total

(9)

Trade margins

Agriculture

(1)

Other services

(8)

Total

(9)

Total use

purchasers

’ prices

Imports

Total

supply at

basic

prices

VALUATION

Total supply

purchasers'

prices

INDUS TRIES

FI NAL USE

GVA

PRODUCTS

INDUSTRIES

PRODUCTS

Total input at basic prices

PRODUCTS

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

136

Although this is consistent with CPC and ISIC, countries may however use different and more

detailed classifications, for example, those which reflect country-specific activities.

5.20. The domestic output matrix reflects the principal and secondary products of industries

including by-products. It is the principal activity of the statistical unit that determines its

classification to a specific industry. In special cases where the domestic output matrix is square

(the number of products being equal to the number of industries), and the sequence of products

arranged to reflect the sequence of the industries (based on their principal activities), the principal

activity of an industry is reported on the diagonal of the matrix, while the secondary activities of

an industry are listed as off-diagonal entries.

5.21. In practice, however, it is common to have more products than industries. For this reason,

the production part of the supply table is usually a rectangular matrix with more rows than

columns, as shown in Table 5.1. This reflects the fact that it may be of more interest to specify, for

example, different kinds of agricultural crops, in the case of agriculture, and of less interest or

practical use to distinguish farms specializing in each possible type of crop. In this case, all the

crops would still form the principal output of agriculture, whereas, for example, the production of

wine or the construction of buildings for own use would be treated as secondary outputs of the

industry. The greater the level of product detail, the more scattered the entries will be around the

principal products. In these cases, it is not possible to observe directly the distinction of principal

products versus secondary products or production in the rectangular domestic output matrix.

5.22. Annex A to this chapter provides an extract of a survey questionnaire collecting data on

sales of goods and services by type of product, in addition to other variables by product, such as

opening stocks (inventories), closing stocks (inventories) and trade margins.

5.23. Even though the industry concept is already being applied in the national accounts, the

existing level of detail or precise delimitation should not be viewed as a constraint when compiling

SUTs and, in particular, when compiling benchmark tables. On the other hand, the way in which

statistical units are defined and classified in the business register and covered in basic statistics

represents a real constraint on the possible choices concerning industries in the SUTs. Even though

industries may, in the process of compiling SUTs, be redefined to some extent or otherwise

modified in terms of their basic statistics, the options are much more limited than the range of

choices available when it comes to deciding what product classification should be applied.

5.24. The choice of the level of detail for industries and products to be used in the SUTs must be

based on a thorough examination of the available statistics and considerations concerning the

advantages of using product details in balancing, in estimating margins and taxes on products by

uses, final uses by purpose, in volume estimates and in other applications. The general

recommendation, however, is to work with as much detail as possible, as any aggregation of basic

statistics will also entail a loss of information that could at some stage have contributed to the

overall quality of the balanced SUTs (on this issue, see also chapter 4).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

137

5.25. It is also necessary to clarify any user requirements about the format of the final table,

including international reporting. In general, it would be an advantage to work at a more detailed

level than that warranted by current uses, in order to extract the maximum information from

available data sources and to be prepared for the emergence of new uses and for transformations

necessary to comply with future changes to economic activity and product classifications.

2. Primary statistics and data sources

5.26. The structure of economic entities varies from small enterprises engaged in one or a few

activities that are undertaken either at, or from, a single geographical location to large and complex

enterprises engaged in many different activities. These enterprises may be horizontally or

vertically integrated, and their activities may be undertaken either at, or from, many geographical

locations. The way in which producer units are defined, measured statistically, broken down or

aggregated is of fundamental importance when compiling SUTs.

5.27. In practice, compilers of SUTs will not deal with individual economic units but only with

the aggregates of units in the form of industries, usually based on current business statistics by

economic activity. To arrive at a full understanding of the role of these statistics in the compilation

of SUTs, it is necessary to assess the delimitation of units that influence the properties of the

industries.

5.28. The most important prerequisite for the collection of basic statistics is the business register

and the types of economic units that it holds. Ideally, business registers will contain two types of

unit: enterprise units and establishments.

5.29. Usually the enterprises form the core units of the business register, as they are easier to

identify and track on a current basis because of their legal status. The number of establishments

created depends on the register policy adopted (in other words, how many enterprises are

partitioned into establishments). Different geographical locations of the production units will be

one of the main criteria for subdividing an enterprise into several establishments.

5.30. In the collection of basic statistics, the enterprise will usually serve as the collection entity

and, to the extent that the enterprise is made up of several establishments, the enterprise will be

requested to report a separate range of statistics for each of those establishments. This has

implications for both the supply table and the use table, as some types of costs can only be reported

at the enterprise level, whereas it is possible to report all regular production costs for the individual

establishments.

5.31. As the large majority of enterprises are small or medium in size, and tend to engage in one

kind of activity only, the enterprise and the establishment units may be identical in these cases.

Large enterprises, however, which often contribute the bulk of the production of an economy, will

often cover different kinds of economic activity and therefore, in formal terms, be made up of

several establishment units.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

138

5.32. It is important to note that many primary sources, such as enterprise surveys and production

surveys, which are used to collect data feeding into the domestic output matrix, often also collect

data at the same time through the same survey questionnaire but feed these into the use table (for

example, data on the industries’ input structures and gross fixed capital formation). This approach

provides the data feeding into the SUTs with a high degree of coherence and consistency.

5.33. The estimates of the domestic output matrix are usually based on two main types of

information sources: enterprise surveys and production surveys. Additional information, such as

administrative sources, company accounts and others, will also be used. Figure 5.2 provides a

simplified view of the different types of information used in compiling the domestic output matrix.

Figure 5.2 Different types of information used in compiling the production matrix

5.34. Starting with the enterprise survey, the principal objective is to supply information on the

main structural characteristics of the different economic activities. The basic unit of this type of

survey tends to be the enterprise. From this source, it is possible to estimate the total production

by activity, starting with its private accounting business systems. On the other hand, the production

surveys allow the estimation of the total production by type of product.

5.35. Combining both sources of information, enterprise statistics and production statistics, it is

possible to combine the data and obtain the production by type of product, by principal activities

of the enterprise and by principal activities of the establishment that belongs to this enterprise.

Consequently, the principal production and the secondary production of a product can be

identified, primarily for industrial products. In many cases, lack of information makes it necessary

Agricul-

ture

Manufac-

turing

Construc-

tion

Trade, transport

and

communication

Finance and

business

services

Other

services

(1) (2)

(3) (4)

(5) (6) (7)

Agriculture

(1)

Manufacturing

(2)

Construction

(3)

Trade, Transp. and comm. services (4)

Finance and business services

(5)

Other services

(6)

Total (7)

= Principal activities

= Secondary activities

Production

type surveys

Data on total

sales or

output by type

of product

Surveys of enterprises or establishments

Collect data on total sales by enterprises or establishments plus a range of other

information, preferably with a product breakdown, such as:

(a) Sales by type of products, which are then allocated by CPC products

(b) Changes in inventories (split by asset type, for each an opening and closing

level)

(d) Other taxes and subsidies on products

(e) Trade activity (by type, for example, wholesale or retail)

(f) Total sales, sales for export (by type of product)

Data collected

under (a), (b), (c)

and (e)

Principal activity

Secondary activity

Ancillary activity

INDUSTRIES

Total

domestic

output

PRODUCTS

will link to

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

139

to use reasonable assumptions about what products are constitute the industries’ secondary

production.

5.36. Many enterprises may perform some construction work, for example, own-account gross

capital formation and minor maintenance and repair work. Enterprises in the manufacturing and

service industries are often involved in either wholesale trade or retail trade or even both. Many

service industry enterprises may provide retail trade services as a secondary activity. Lastly,

activities involving the rental of real estate and leasing of equipment are often secondary activities.

5.37. In basic statistics, output by products will usually be available for goods-producing

industries such as agriculture, mining and manufacturing industries – at least for enterprises and

establishments above a certain threshold – and similarly imports and exports of goods will be

covered in great detail by external trade statistics. For service industries, a breakdown of output

by individual kinds of services (as defined in the CPC classification) is less common, although in

recent years many countries have developed such statistics. If there is a lack of product statistics

for services, the output by the most detailed service activities of ISIC may be used as proxy,

assuming that all output consists of the services characteristic for that particular industry.

Concerning product breakdowns, construction will be placed between these two extremes.

5.38. For manufacturing units below a certain threshold, output statistics by products will usually

be missing, whereas total output will be estimated based on either business surveys or

administrative records. Working at the most detailed activity level available, output from these

small units can be broken down into the products of the system, for example, by assuming that the

composition by product is identical to that which has been observed for the smallest category of

those units for which output statistics by product exist. During the balancing process, this

assumption may be modified and the output redistributed by product.

5.39. The products recorded in the domestic output part of the supply table should be output

valued at basic prices at the time it is completed. For manufacturing industries, it is usually only

sales by product that are given in the surveys and adjustments would need to be made for change

in inventories of finished goods and work-in-progress, in order to move from sales to output.

5.40. When information exists about opening and closing inventories by industry, it can be

assumed that the composition by product is identical to the sales by product. The change in

inventories of finished products and work-in-progress by product can be derived by applying

relevant price indices and assumptions about inventory valuation principles used by enterprises.

As the reliability of these data by product is limited, however, and as, by definition, these data

should always be identical on the supply side and the use side, there is no need to adjust the sales

figures or enter them into the final use category of changes of inventories at this stage. In the

system, it is the actual sales figures that are relevant for the distribution by users, and the estimated

data for change in inventories can therefore be simply imposed on the system after the balancing

has been completed. This is, however, not the case for change in inventories of materials and fuels

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

140

(recorded in the intermediate use part of the use table) and in trade. Changes in inventories of

agricultural products and mining products will usually have to be included in the system from the

outset, as the output data will often refer not to sales but to actual output.

3. Principal and secondary products

5.41. The distinction between principal and secondary production has traditionally played a

prominent role in I-O literature, as the existence of secondary production requires certain

assumptions for IOTs to be compiled. It should be noted, however, that a match between products

and industries (determining in which industry a product is the principal output) is really only

necessary in those cases where the chosen techniques for deriving IOTs as a starting point requires

the SUTs to be aggregated to square tables, cases where the sequence of the aggregated products

is made comparable to the sequence of industries. With regard to other techniques for the

compilation of IOTs and for the purpose of the SUTs, there is no need to match products and

industries. It should thus be noted that, when industry-by-industry IOTs are derived on the

assumption of fixed product sales structures, there is no need first to aggregate the rectangular

SUTs.

5.42. When necessary, the match between product and principal producer can be derived either

theoretically (by identifying for each product the principal producer according to the ISIC

definitions of the principal products of each industry – the correspondence keys are available on

the United Nations Statistics Division classification website at

http://unstats.un.org/unsd/class/default.asp) or empirically (established by observation of the

actual domestic output matrix, the industry being the main producer of each product).

5.43. In principle, the empirical match will be the most precise in the sense that it depicts the

production relationships as they actually exist in this particular economy. The theoretical match

may be the preferred approach when considering time series and international comparisons. These

matching methods also demonstrate that the product classification applied when compiling SUTs

can be chosen completely independently of whether or not it may subsequently be necessary to

derive square SUTs.

5.44. When the domestic output matrix is aggregated to a square matrix and arranged so that the

entries for the primary products fall on the diagonal, the off-diagonal elements show the extent of

secondary production. This refers to that part of a product which is produced by industries other

than the one where it principally belongs either formally according to the industrial classification

(theoretical aggregation key), or according to the industry which is actually the main producer

(empirical aggregation key).

5.45. As the secondary production observed in the domestic output matrix depends on the level

of aggregation both of products and of industries, secondary production does not possess any

observable characteristics of its own. The elusive character of the concept of secondary production

makes it difficult to justify that a product should be of particular interest statistically just because

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

141

it is produced in two or more industries at a certain level of industry or product aggregation. When

the industry and product classifications to be used in SUTs have been decided (inclusive of

possible redefinitions), the principal versus secondary distinction plays no role in the subsequent

elaboration and balancing within the SUTs framework.

5.46. For most countries, the domestic output matrix is characterized by showing secondary

production almost exclusively for manufacturing industries, whereas for most other industries,

practically all the production is found on the diagonal elements (or in the rectangular table – in

what is known as the “diagonal field”). There are three main reasons for this:

• Basic statistics for manufacturing industries have traditionally included detailed product

statistics and thus make identification of secondary production possible.

• For service industries, the diagonal structure is simply due to the fact that, more often than

not, very limited detail has been collected on the type of product breakdown of these service

activities. Thus, the total output from establishments (or even enterprises) must be assumed

to be the primary output of the industries to which the units are classified in the business

• The activities of industries such as agriculture, construction and trade are often defined in a

purer form (the industries covering all their principal products, and only those) in the national

accounts and SUTs than in the business register. In this case, all secondary agricultural,

construction and trade activities in other industries would have been transferred to their main

activities. Alternatively, data for some activities may have been constructed in such a way

that from the outset no secondary production exists, for example, agricultural activity is

measured as the sum of all agricultural products, construction activity as the value of new

construction and repairs, and so forth.

5.47. When the rectangular SUTs have been balanced, there may be a need to aggregate them

into a square system either for dissemination purposes or for compiling IOTs by methods that

require square SUTs. In a square system, the number of product groups must be identical to the

number of industries, and furthermore, the products aggregated in such a way that the resulting

product groups can be given corresponding industry names, indicating the industry of which they

are principal products. If aggregation is made solely for dissemination purposes, the product

aggregation could also be carried out to, for example, higher levels of CPC – which, as mentioned

above, have no direct correspondence with ISIC defined industries.

4. Ancillary activities

5.48. When the production of an enterprise takes place in two or more different establishments,

certain ancillary activities may be carried out centrally for the collective benefit of all the

establishments. If, in such a case, a producer unit is undertaking purely ancillary activities and is

statistically observable, in that separable accounts for the production that it undertakes are readily

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

142

available, or if it is in a geographically different location from the establishments that it serves, it

may be desirable and useful to consider it as a separate unit and to allocate it to the industrial

classification corresponding to its principal activity. Another exception is the case when some

products are used both for own ancillary use and also supplied to another unit (see 2008 SNA,

para. 14.33). In the 2008 SNA, however, it is recommended that statisticians do not make

extraordinary efforts to create separate establishments for these activities artificially in cases where

suitable basic data are not available.

5.49. The fact that establishments, even at a detailed level, are classified to the same activity,

does not mean that they are in all respects identical. Each establishment has its own unique

institutional and organizational characteristics, which may influence the composition of its

purchases as much as the underlying technical production processes. Two establishments

producing identical products may have quite different input structures, depending on the degree of

reliance on purchased semi-fabricated products, outsourcing of certain activities (see also chapter

8, section D, on goods sent abroad for processing), whether it owns the capital equipment and

buildings it uses rather than leasing or renting them, and so forth, and in general, on the degree of

vertical integration of the various stages of the production processes. There is no way that these

institutional characteristics inherent in the original source data should be eliminated from the SUTs

(or subsequently from the IOTs) nor does the SNA expect the compilers to try to do so.

5.50. Institutional arrangements of production not only differ between establishments classified

in the same industry but also across countries and over time. It is obvious that there are serious

limitations to the view that the SUTs (and the IOTs) portray the technical characteristics of a

production system. From a statistical point of view, the achievable elimination of institutional

arrangements is obtained by using establishments as production units (with the possible additional

partitioning of vertically integrated enterprises as discussed above), given that the establishments

are designed with this purpose in mind and there are no official statistics providing production

structures below this level.

5.51. In some countries, the recommended establishment unit approach may not be achievable

in practice, since, for legal, practical or historic reasons, statistics are only collected for enterprise

units. Even though compilers of the SUTs may in this situation try to break down the most

important multi-activity enterprises into their constituent establishment units, there is in general

no feasible alternative to working with the existing data. In this case it is still possible to compile

SUTs, although the overall picture of the productive system will become less precise and to some

extent blurred, which will also have an adverse effect on the resulting SUTs and IOTs.

5.52. It should, however, be recognized that the important objectives of compiling the SUTs may

also be achieved when the data are based on enterprise units (see SNA 2008, para. 14.21), although

some product-flow and common-sense procedures may be more difficult to apply because of the

less stringent definition of industries, as the composition of output from the enterprise units will

also be crossing the borderlines between sections of ISIC. For these reasons, an enterprise-based

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

143

approach will in general require more thorough coverage of statistical source data. It should also

be borne in mind that there are no automatic methods available that can disentangle this dataset

and transform it into SUTs or IOTs with analytical properties comparable with those resulting

from SUTs based on establishment-type units. Depending on the specific circumstances, it may in

such cases be decided to compile SUTs alone and not IOTs.

5. Redefinitions

5.53. Redefinitions refer to adjustments made to the source data relative to the way in which they

are obtained from the primary statistics, in order to obtain “purer” industries, so to speak, for use

in the SUTs. This is an exception to the previously mentioned rule that SUTs compilers should not

attempt to create their own versions of basic statistics. That would not be cost-effective nor would

it be conducive to the comparability of SUTs with other economic statistics or on an international

scale. In practice, deviations from the way in which enterprises or establishments are defined in

the business register and reflected in primary statistics should be limited in scope.

5.54. Such redefinitions may be seen as implementing the SNA recommendation to partition

vertically or horizontally integrated enterprises or establishments that have production in two or

more sections of ISIC Rev. 4 (2008 SNA, paras. 5.52–5.54). Redefinitions are generally carried

out manually, using product-specific input structures based on specific insight into the activities,

leading to results that will come as no surprise nor give rise to negative elements, as might have

been the case had more automatic methods been applied. By reducing secondary production,

redefinitions facilitate the subsequent compilation of IOTs, and compilers of SUTs should be

aware of how the choice of compilation techniques will affect the subsequent calculation of the

IOTs.

5.55. Redefinitions (more background is provided in Box 5.1) are usually concentrated on a few

major activities, such as agriculture, energy, construction and trade, or a few major enterprises,

such as mining operations. Redefinitions affect all those activities from which secondary output is

being removed. For some activities, redefinition-type adjustments may have been carried out

already in the source data, as in the following cases:

• The European Union System of Agricultural Accounts requires that all agricultural activity

is covered by these accounts and there are very limited possibilities for the retention of non-

agricultural secondary production within the system definition of agriculture.

• All rented dwellings are usually grouped together in one single industry (together with

owner-occupied dwellings) independently of the activity of the actual owner.

• Trade activities outside the trade industries (trade as secondary activity), by definition, have

already been separately identified when compiling the national accounts, as only the trade

margins, and not the gross turnover of the traded products, should be counted as output, and

may have been grouped together with trade as primary activity.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

144

• Construction activities are also frequently redefined to form one single, “pure” construction

activity, often because total output has been defined by adding up the values of specific types

of construction output rather than output from building establishments, or alternatively

inputs have been determined from the supply of construction materials. Any of these

approaches will also facilitate the distribution of building materials for intermediate

consumption.

Box 5.1 Redefinitions

In Miller and Blair (2009), on page 141, redefinitions are defined as: “Factoring out the amount of

secondary products produced as well as the inputs used in that production and reassigning both to the

industry for which the product is classified as primary”.

In addition, a distinction is made between specific redefinition and mechanical redefinition (page 215),

where the former is the “by hand” procedure and the latter refers to the various mathematical procedures

that can be applied to eliminate secondary production when producing IOTs from SUTs (covered in

chapter 12 of Miller and Blair).

The specific redefinition or two-step process emerges from the practice in several countries. It is explained

in detail for the United States in Guo J. and others (2002). The Bureau of Economic Analysis paper was

presented at the fourteenth International Conference on Input-

Output Techniques, held in Montreal,

Canada, in 2002. The article also analyses the differences between the resulting tables when redefinitions

are not applied (case 1), and when they are applied (case 2).

The redefinition method is also used in Canada and Denmark, whereas the industry-by-industry IOTs in

Norway are more of the case 1 type, in that they retain the micro-macro link to a maximum degree.

The industry-by-industry IOTs of the Netherlands seems to fit somewhere between case 1 and case 2.

In France, the first step (redefinition) is based on enterprise units and is carried out to an extent that the

supply table becomes diagonal. The use tables thereby also form the IOTs, and the second step (compiling

the IOTs) becomes superfluous.

5.56. Although the redefinitions serve the purpose of creating purer activities and thus facilitate

I-O analysis, their main purpose is to arrive at an activity classification that is applicable for use

in the national accounts, and thus conducive to the integrated compilation of SUTs and national

accounts. Three different situations can be distinguished:

• Case 1: no redefinitions take place in the national accounts, the SUTs and the industry-by-

industry IOTs.

• Case 2: redefinitions have been carried out for all national accounts data and in the SUTs

prior to the calculation of the industry-by-industry IOTs.

• Case 3: redefinitions are not carried out when the current national accounts are compiled but

applied when the SUTs and the industry-by-industry IOTs are compiled.

5.57. In the first two cases, the consistency and comparability of the current national accounts

(tables by industry), and of the SUTs and IOTs classifications are upheld, but that is not so in the

third case. Ideally, the choice of redefinitions should already be introduced in the general

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

145

classification of industries used in national accounts. Not only will the manually prepared

redefinitions be more precise at these earlier stages, but they will also facilitate the balancing of

the system as the need to make a large number of small input entries to many cells of the use table

will be obviated.

D. Imports of goods and services

1. General description and definition

5.58. The second part of the supply table covers the total imports of goods and services. In

national accounts, imports refer to transactions that occur when there are changes of economic

ownership of goods between residents and non-residents, whether or not there are corresponding

physical movements of goods across frontiers.

5.59. International merchandise trade statistics represent the main source of data for imports of

goods. International standards are specified in International Merchandise Trade Statistics:

Concepts and Definitions (IMTS 2010) (United Nations, 2011). For imports of services, the main

source of data is either the details available in the balance of payments statistics or specialized

statistics on international trade in services (for example, business surveys), according to the

international standards given in the 2010 Manual on Statistics of International Trade in Services

(United Nations, European Commission, IMF, OECD and WTO, 2011) in connection with product

classifications.

5.60. Some differences exist, however, between the concepts used in international trade statistics

and the 2008 SNA and BPM 6, and adjustments therefore need to be made to the basic statistics in

order that they can be used in the SUTs. The BPM 6 identifies sources of difference between the

IMTS and the 2008 SNA and BPM 6 concepts of imports that may occur in countries. In this

regard, it recommended that a standard reconciliation table be compiled to assist users in

understanding these differences.

5.61. One major difference is the valuation used to record imports of goods. While IMTS 2010

uses a CIF valuation for imports, the 2008 SNA and BPM 6 use a uniform FOB valuation for both

exports and imports of goods. The 2008 SNA states, in paragraph 3.149:

Imports and exports of goods are recorded in the SNA at border values. Total imports and

exports of goods are valued free-on-board (FOB, that is, at the exporter’s customs frontier).

As it may not be possible to obtain FOB values for detailed product breakdowns, the tables

containing details on foreign trade show imports of goods valued at the importer’s customs

frontier (CIF, that is, cost, insurance and freight), supplemented with global adjustments to

FOB values. CIF values include the insurance and freight charges incurred between the

exporter’s frontier and that of the importer. The value on the commercial invoice may of

course differ from both of these.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

146

5.62. The adjustments for the FOB and CIF valuation of imports are described in more detail in

the next section.

5.63. Another difference is the time of recording. In the 2008 SNA and BPM 6, the time of

recording of imports and exports corresponds to the time that ownership of the goods is transferred.

By contrast, IMTS follows the timing of customs processing. While this timing is often an

acceptable approximation to the change of economic ownership principle, adjustments may be

needed in some cases, such as for items with large values or goods sent on consignment (that is,

dispatched before they are sold). It should be noted that, in the case of goods sent abroad for

processing with no change of economic ownership, the values of goods movements are included

in the IMTS-based recording but are to be excluded from the ownership-based recording in the

national accounts and balance of payments. It is recommended, however, that the values of goods

movements be entered as supplementary items in the balance of payments, to indicate the nature

of these arrangements.

5.64. Other adjustments to IMTS may be needed to bring estimates into line with the change of

economic ownership of goods, either generally or because of the particular coverage of each

country. Possible examples include:

• Merchanting

• Non-monetary gold

• Goods entering or leaving the territory illegally

• Goods procured in ports by carriers

• Goods moving physically but where no change of economic ownership has taken place such

as in cases of operating leases

5.65. To maintain consistency with BPM 6, the 2008 SNA introduced new treatment relating to

merchanting and goods sent abroad for processing. Merchanting is a process whereby a unit in

economy X purchases goods from economy Y for sale in economy Z (sometimes within economy

Y itself). The goods legally change ownership but do not physically enter the economy where the

owner is resident. By convention, the purchases of the goods intended for resale is shown as

negative exports. When the goods are sold, they are shown as positive exports. When the purchase

and sale take place in the same period, the difference is shown as an addition to exports. If the

purchase takes place in an accounting period, the negative export is offset by an increase in

inventories of goods for resale, even though those goods are held abroad.

5.66. The surplus on this item in the foreign trade statistics is by its nature a trade margin and

should be included in the output of the industry. In the main, this activity takes place in the trade

industry. In exceptional cases, this may lead to an overall deficit on the item in the foreign trade

statistics but the trade margin would usually still remain positive (the deficit added to changes in

inventories). As indicated, trade margins from merchanting activity primarily occur in the trade

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

147

industry but can occur in many other industries, unless all trade is redefined to the trade industry.

Given that business statistics provide source data as a starting point for the compilation of SUTs,

merchanting activity can then appear in various industries, for example, oil companies and

pharmaceutical companies.

5.67. The new treatment of goods sent abroad for processing is dealt with in more detail in

chapter 8 of this Handbook.

5.68. A special category within imports is the direct purchases abroad by residents. This item

covers all purchases of goods and services made by residents while travelling abroad for business

or pleasure. Two categories must be distinguished because they require different treatments:

• Expenditure by resident business travellers: this item refers to intermediate consumption of

several industries to which the travellers belong (in the use table) and imports of services (in

the supply table).

• Expenditure by other resident travellers on personal trips: this expenditure is recorded in

final consumption expenditures by households (in the use table) and imports of services (in

the supply table).

5.69. Imports broken down by products in the SUTs do not include direct purchases abroad by

residents. Consequently, these must be included in an adjustment row to obtain the overall value

of imports (row (11) in Table 5.1).

5.70. In tables 5.1 and 5.2, the estimates for CIF/FOB adjustments on imports and the direct

purchases abroad by residents are shown separately in the rows. It should be noted, however, that

some countries do not show these estimates in separate rows but consolidate the values in the

product groups in the respective columns. This situation in turn leads to different product balances

but does not change the imports aggregate total. This is often due to the coverage of the data

sources and, in these cases, appropriate adjustments should be applied to extract the corresponding

entries to generate the separate entries.

5.71. Goods procured in ports by carriers may be included in a similar adjustment row. It should

also be noted that imports and exports of ships and aircraft may have to be given special attention,

as these transactions may follow special recording procedures in the external trade statistics that

are not consistent with the way in which output or gross fixed capital formation should be recorded

in the national accounts.

5.72. Imports of goods and services in SUTs are dealt with in more detail in chapter 8.

2. Valuation for imports: CIF and FOB valuation

5.73. In the 2008 SNA and BPM 6, the total imports of goods are valued FOB. The data on

imports by detailed products from the foreign trade statistics used in the SUTs are usually available

at CIF prices, however, following the International Merchandise Trade Statistics (United Nations,

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

148

2011). To reconcile the different valuations used for total imports of goods and for the product

components of imports, two types of adjustments are needed. These adjustments are presented

below.

(a) Data adjustment

5.74. The first type of adjustment must be made to the data of the balance of payments prior to

entering data from this data source into the SUTs system. This adjustment is necessary in order to

ensure as a starting point a consistent set of data for imports and exports of goods and services that

can be balanced across the SUTs. This adjustment is illustrated in Table 5.3.

5.75. The starting point is the account for the rest of the world, as shown in columns (1) and (2)

of Table 5.3 (which mirrors the balance of payments according to BPM 6), where only the entries

for goods and services are shown and where imports of goods are valued FOB (372 in column (2)

of Table 5.3). This is the value of the goods at the point of exit from the exporter’s economy,

including transport charges and trade margins up to the border point. The CIF value of imports

(382 in column (6) of Table 5.3) of goods measures the value of a good delivered at the point of

entry into the importing economy. The difference between the two values (10 in column (4) of

Table 5.3) is made up of the costs of transportation, insurance and other expenditures between the

point of exit of the exporter’s country and the point of entry into the importer’s country.

5.76. The services linked to the difference between the FOB and CIF values can be delivered by

either resident producers or non-resident producers. To the extent that non-resident producers are

involved, the BPM 6 imports of services must be reduced with their services (7 in column (4) of

Table 5.3) to avoid double counting, as these services are now included in the CIF value of the

imported goods. Adjustment for the services delivered by resident producers (3 in column (3) of

Table 5.3) is a bit trickier, as a service that, according to the BPM 6 definition, is a purely domestic

transaction will now appear as an import of services included in the CIF value of imported goods.

As this import originates from resident producers, it is necessary to introduce a balancing service

export of the same value.

Table 5.3 Data adjustment for external trade of goods and services

Note: In practice there will be a further breakdown of both goods and, in particular, services in the

balance of payments, and therefore also for the adjustments in columns (3) and (4).

Uses (FOB) Resources (FOB) Uses Resources Uses (CIF) Resources (CIF)

(1) (2) (3) (4) (5) (6)

Imports of goods 372 10 382

Exports of goods 462 462

Imports of services 84 -7 77

Exports of services 78 3 81

Total 540 456 3 3 543 459

Balance 84 0 84

SNA/BPM balance of payments

Introducing imports CIF

SUTs balance of payments

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

149

5.77. In Table 5.3, all data adjustments are shown in columns (3) and (4), and the resulting “SUTs

Balance of Payments” in columns (5) and (6). It is noted that the balance of the adjustment items

is zero, and consequently, the surplus on the transactions in goods and services – 84 in column (2)

– is identical in the two alternative ways of presenting the external transactions.

5.78. The “SUTs balance of payments” represents the framework of source data for external

trade for SUTs with the appropriate product breakdowns. The composition by specific services

making up the CIF and FOB difference will usually be available from the working tables of the

balance of payments compilers, as their starting point for the FOB recording of imports will usually

have been imports of goods from the external trade statistics valued at CIF Regular surveys may

also have been carried out to illuminate the CIF and FOB difference and the related service

structure.

5.79. It is important to underline that the above data adjustment is not the CIF and FOB

adjustment often seen as a separate row in SUTs or IOTs. The data adjustment must be made

before starting compiling SUTs. At the detailed product level, the supply and use of the individual

services are adjusted so that they can meaningfully be balanced under the CIF valuation of goods,

and these data adjustments will not be separately identifiable in the completed SUTs.

(b) CIF and FOB adjustment row

5.80. The CIF and FOB adjustment is an ex post facto adjustment made at the macro level to the

totals for exports and imports of goods and services to derive the corresponding totals found in the

SNA (the goods and services account and the rest of the world account).

5.81. In principle, the purpose of this adjustment is to demonstrate that the data in SUTs are

consistent with the rest of the national accounts and to avoid the double-counting of CIF-type

services provided by residents. The CIF and FOB adjustment row has no balancing or other

methodological functions in the SUTs, and it may be omitted from the SUTs and also the IOTs if

there is no special need to maintain the exact conceptual relationship to the national accounts.

5.82. Table 5.4 illustrates the place and content of the CIF and FOB adjustment row in SUTs,

albeit limited here to external trade data.

Table 5.4 CIF and FOB adjustment row

5.83. The “SUT total” row includes the totals for imports and exports of goods and services in

the balanced SUTs system, consistent with the “SUTs Balance of Payments” in Table 5.3.

Goods Services Goods Services

SUT total 382 77 462 81

CIF/FOB adjustment -10 7 -3

BOP total 372 84 462 78

Supply Table

Use Table

Imports

Exports

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

150

5.84. In order to obtain totals for the external transactions identical to those found in the rest of

the national accounts (and the balance of payments), the adjustments shown in the “CIF/FOB

adjustment” row of Table 5.4 are introduced. These adjustments mirror those that were made as

data adjustments in Table 5.3. The two types of adjustments have quite different roles, however:

• Those in Table 5.3 relate in principle to columns of the SUTs and must necessary be carried

out prior to the balancing, and there is no way to avoid this adjustment.

• On the other hand, the CIF/FOB adjustment in Table 5.4 is a kind of “memo” row of the

SUTs that can be added ex post facto, or even omitted if there is no need to demonstrate

consistency with the national accounts.

5.85. It should be noted that, if goods and services are lumped together in SUTs, the CIF/FOB

adjustment row will only include the adjustment item, -3, for both imports and exports.

5.86. From a bookkeeping perspective, the data adjustment for exports of services (3 in Table

5.3) could alternatively be recorded as a negative import, even though this action entails a less

logical explanation of how the domestic output of services are disposed of and also requires the

existing imports of those services to be sufficient to prevent a negative net result.

5.87. With this approach, there would be adjustments in Table 5.3 for imports only, showing

identical numerical changes for goods and services, respectively. The CIF/FOB adjustment row in

Table 5.4 would in this case have entries only for imports (-10 for goods and +10 for services),

and if imports were not shown separately for goods and for services, the CIF/FOB adjustment row

would be empty. Further details covering issues of consistency in the SNA are provided in Box

5.2.

Box 5.2 Consistency issues with the CIF/FOB adjustment

The CIF/FOB adjustment is dealt with in both the Eurostat Manual of Supply, Use and Input-Output

Tables (Eurostat, 2008) and in the supply and use table chapters 14 and 28 of the 2008 SNA.

In the numerical example in Eurostat (2008) (pages 60, 70 and 122), external trade in goods and

services are lumped together and, as explained above, for this case, the CIF/FOB adjustment row

therefore contains identical negative adjustments for imports and exports. In the more extensively

elaborated numerical example (pages 113–115), where comparisons are also made to the treatment

in the 1993 SNA, ex ante data adjustments are mixed up with the ex post CIF/FOB adjustment in a

complicated manner.

The CIF/FOB adjustment table has both a column and a row dimension, and the final outcome is

incorrect because the ordinary exports of services linked to exports of goods are included in the

adjustment.

The exposition of the CIF/FOB adjustment in the 2008 SNA is unclear because it starts out from the

assumption that the SUTs have been balanced using inconsistent data, namely imports of goods

valued CIF, and services as defined as in the BPM 6 based on imports being valued FOB. This

shortcoming can obviously not be remedied by ex post adjustments to columns and rows, as any new

column data would require a new balancing of the SUTs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

151

The final outcome in the 2008 SNA are SUTs with a CIF/FOB column in the supply table (table

14.12 of the 2008 SNA), in addition to the CIF/FOB adjustment row with adjustments for imports

only (the resident producers’ delivery of services linked to imports CIF being treated as negative

imports).

If the CIF/FOB adjustment column in the supply table (table 14.12 of the 2008 SNA) is added to the

column for imports of services, services as defined in the SUTs balance of payments are obtained,

so that in principle this could be taken to indicate the ex ante data adjustment. This is not easily

understood from the exposition, however, and, to compound the lack of clarity, the CIF/FOB

adjustment is being distributed by user (in table 14.15 of the 2008 SNA), a step for which there is no

explanation.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

152

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

153

Annex A to chapter 5: Sample questionnaire collecting sales of goods and

services, inventories of goods and trade-related data

A5.1 The extract shown in figure A5.1 is from a business survey questionnaire from the

Statistical Office of Serbia. Data are collected for each industry and by product covering the

following areas:

• Sales of goods produced by the enterprise

• Closing stocks of products and work-in-progress

• Sales of merchandise

• Trade margin

• Closing stocks of goods for resale

A5.2 Additional tables collecting data on the sales of industrial and non-industrial services

provide the full coverage of goods and services needed to calculate the industry totals. An extract

of these tables may be seen in figure A5.2. These data make it possible to calculate the industry

output by product and trade margins required to populate the domestic output part of the supply

table and the trade margins column, as shown in Table 5.2. In some countries, opening stock values

are also collected.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

154

Figure A5.1 Extract of questionnaire covering sales of goods, inventories of goods and

trade activity

No. Code

Product description

Sales of

goods

produced by

the

enterprise

(group of

accounts 61)

Closing

stocks of

products and

work in

progress

(group of

accounts 10

and 11)

Sales of

merchandise

(group of

accounts 60)

Trade

margins

amount

rate %

Closing

stocks of

goods for

resale

(group of

accounts 13)

(1)

(2) (3) (4)

(5)

(6) (7) (8)

1000

TOTAL

AGRICULTURAL PRODUCTS, RAW AND UNPROCESSED PRODUCTS OF PLANT

AND ANIMAL ORIGIN

1001

01.11.1–01.11.4

Cereals, all kinds (except rice), cereal seeds

1002

01.11.6, 01.11.7 Green leguminous vegetables (beans, peas, lentils and other)

1003

01.11.8 Soya beans, groundnuts (row) and cotton seed

1004

01.11.9 Other oil seeds–sunflower, sesame, flax, etc.

1005

01.11.12 Rice, not husked

1006

01.13 except

01.13.7

Vegetables, raw and seeds

1007

01.13.7 Sugar beet and sugar beet seed

1008

01.13.8 Mushrooms and truffles

1009

01.15 Unprocessed, raw tobacco

1010

01.16

Fibre crops (flax, cotton, hemp and other, used in textile industry)

1011

01.19.1

Forage crops and vegetative matter for livestock feeding unprocessed form

1012

01.19.2 Flower and flower seeds

1013

01.21 Grapes

1014

01.22, 01.23 Tropical and subtropical fruits, all kinds (including citrus, figs, etc.)

1015

01.24, 01.25

except 01.25.3

Other fruits, tree and bush fruits, except nuts (apples, pears, cherries, berries, etc.)

1016

01.25.3 Nuts (almonds, hazelnut, walnuts, etc.)

1017

01.26

Olives, coconuts (raw, unprocessed)

1018

01.27 Coffee beans, tea leaves, cocoa beans, not roasted

1019

01.28 Spices, aromatic, drug and pharmaceutical crops

1020

01.11.5, 01.14,

01.19.3, 01.29,

01.3

Vegetables and fruit seeds, other seeds; grass, unprocessed straw and other

residues of cereals; seeds for trees and seedings; planting materials, sugar cane and

other raw, unprocessed and untreated products of plant origin not elsewhere

1021

01.4. except

01.45.3 & 01.49.3

Live animals and animal products (unprocessed milk, eggs, natural honey; seeds and

embryos of animals, except raw skins, shorn wool and skins, etc.)

1022

01.45.3, 01.49.3

Raw fur skins, shorn wool, skins (excluding products of slaughterhouses and

industrial meat production, see 1036)

1023

01.49, part

Other agricultural animal origin products, raw, unprocessed and untreated, not

elsewhere classified

1024

01.7 Hunting and trapping products, raw, unprocessed

PRODUCTS OF FORESTRY

1025

02.2

Wood in the rough–logs, fuel wood and other raw products of forestry, odds and ends

included

1026

02.1, 02.3

Forest trees and seeds, wild growing edible products; natural cork, varnish, balsams

and other naturals gums and resins and other raw products of forestry not elsewhere

classified

FISH AND OTHER FISHING PRODUCTS, UNPROCESSED AND UNTREATED

1027

Fish, sea food and other fishing products; aquaculture products (raw, unprocessed

and untreated)

MINING AND QUARRYING PRODUCTS, UNPROCESSED; CRUDE AND NATURAL

GAS

1028

05.1, 05.2

Coals, hard coal and lignite (coal for heating included)

1029

06.1

Crude petroleum, bituminous or oil shale and tar sands. Note, petroleum products

entered in row 1082

:::::::

MANUFACTURING INDUSTRY PRODUCTS

Food products and other processed products of plant and animal origin; used as

reproduced material

:::::::

Production of electricity and manufactured gas (excluding natural gas extraction and

petrol gases in refineries); trade and distribution of electricity and manufactured gas

:::::::

Construction

1162

41, part Development of building projects

1163

41, part Construction works of residential and non-residential buildings

1164

42 Construction and construction works of civil engineering

1165

43 Specialized constructions works

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

155

Figure A5.2 Extract of questionnaire covering sales of industrial and non-industrial

services

No. CPA code Code and service description

Sales of

services

(groups of

accounts 61

and 65, part)

(1)

(2)

(3)

(4)

2000 TOTAL

Support services directly linked with the production of goods and services

2001

01.6 part Support agricultural services to crop production

2002

01.6 part

Support services to animal production (animal farming ; veterinary services excluded (row 2059)

2003

02.10.2, 02.4 Support services to forestry (cultivation and logging of trees, excluded)

2004 09

Mining support services, services to petroleum and natural gas extraction

2005

13.3 Textile finishing services–bleaching, dyeing, printing etc.

2006 16.10.9

Drying, impregnation or chemical treatment services of timber and product of wood; support services in the

processing of wood and wood products not elsewhere classified

2007 25.5

Forging, pressing, stamping and roll-forming services of metal

2008

25.6 Treatment and coating services of metals; machining

2009

24.5

Casting services of metal and steel

Subcontracted services in industry and construction, trade services and other intermediation commissions.

Note: enter only the value of the services, value of materials of goods excluded

2010 14, part

Subcontracted operations in textile industry (excluding value of materials)

2011 15, part

Subcontracted operations in footwear and leather production industry (excluding value of materials)

2012 16, part

Subcontracted operations in production of processed wood and wood products (value of materials, excluded)

2013

25, part

Subcontracted operations as part of machine industry–processing and finishing materials services (value of

materials, excluded)

2014 –

Other subcontracted operations in production of goods of other enterprises (value of materials, excluded),

please specify

2015 46.1

Trade commissions

2016

–

Other intermediation commissions

please specify

Repair, maintenance, installation services; conversion, reconstruction and fitting out of transport equipment

2017 33.1

Repair and maintenance services of fabricated metal products, machinery and equipment, except motor

vehicles

2018 95.1 Repair services of computers and communication equipment

2019 95.2 Repair services of personal and household goods

2020 45.2 Maintenance and repair services of motor vehicles

2021 33.2

Installation services of industrial machinery and equipment

2022 29.20.4, 29.20.5

Reconditioning, assembly, fitting out and bodywork services of motor vehicles, except installation, maintenance

and repair services

2023

30.11.9, 30.20.9,

30.30.6

Conversion, reconstruction and fitting out services of other transport equipment, except installation,

maintenance and repair services

Transportation services

Note include transportation equipment rental services with driver and removal services

2024 49.1 and 49.3

Land transport services–passengers, taxi include

2025 49.2 and 49.4 Land transport services–freight

2026 50.1 Water transport services–passengers

2027 50.2 Water transport services–freight

2028 51.1

Air transport services–passengers

2029 51.2

Air transport services–freight

2030 52.2

Support services for transportation (loading, unloading, hauling, towing, parking service, etc., transportation

excluded)

Other services

2031 18

Printing services and services related to printing (newspaper printing, pre-press, binding and related services,

reproduction services of recorded media)

2032 35.30 Steam, hot water, air conditioning supply services

2033 36

Natural water, water treatment, supply and distribution services

2034

37 Sewerage services, removal and treatment services

2035 38

Waste collection, treatment and disposal services

:::::::

Donations and state subsidies (group accounts 64), lease of intangible assets and income from fees and

charges.

Note: 2089 and 2090 positions are not entered

2086

Donations and other unconditioned transfers in cash or in kind by resident legal and natural persons (account

640 and 641)

2087 –

Donations and other unconditioned transfers in cash or in kind by foreign legal and natural persons (account

640 and 641)

2088 – Subsidies, grants, donations and transfers of state and local government bodies (account 640 and 641)

2089 –

Income from fees for usage of public non-produced assets (this is filled out only by budgetary units–account

741500)

2090 – Income from administrative and legal fees (this is filled out only by budgetary units–account 742200)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

156

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

157

Chapter 6. Compiling the use table

A. Introduction

6.1. This chapter primarily deals with the construction of an initial, unbalanced version of the

use table. The values entered into the tables should, as far as possible, reflect all available

knowledge and data on the product structure of each column, although many values may need to

be changed when the SUTs system is balanced. This applies to estimated totals and also to the

values for specific products. Before balancing takes place, the estimates for intermediate

consumption, final uses and GVA, and the components of GVA (if available at this stage of the

process), should be checked for credibility and, if necessary, adjusted as appropriate. These will

then form the starting point for the balancing process.

6.2. In section B, this chapter provides an overview of the structure of the use table and

describes the main blocks of the table. Section C focuses on the intermediate consumption; section

D on the GVA part of the table; section E on the final consumption expenditure; section F on the

gross capital formation; and section G on the exports of goods and services. The chapter has two

annexes: annex A set out an example of a questionnaire for the collection of data on the purchase

of goods and services for intermediate consumption and annex B provides a description of the

impact of the change in treatment of research and development according to the 2008 SNA.

B. Structure of the use table

6.3. The use table shows the use of goods and services by product and by type of use for

intermediate consumption by industry, final consumption expenditure, gross capital formation and

exports. The use table also shows the components of GVA by industry for compensation of

employees, other taxes less subsidies on production, consumption of fixed capital, and net

operating surplus and net mixed income. Table 6.1 illustrates the structure of the use table.

6.4. The use table has two main objectives:

• The columns show the cost structure of each industry and the product structure of each type

of final use.

• The rows show the distribution of each product and primary input (labour and capital) by

uses.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

158

6.5. It is customary to compile the use table, at least initially, at purchasers’ prices. This

valuation relates most closely to the basis of the data collected via business and household surveys

and is known by the purchasers of the products.

Table 6.1 Use table at purchasers’ prices

Millions of euros

Table based on 2011 figures from Austria

6.6. The upper part of the use table (rows (1)–(9) in Table 6.1) shows how the use of goods and

services is distributed as intermediate consumption by industry, final consumption expenditure,

gross capital formation and exports. The rows of this part of the table correspond to the same rows

of the supply table. Each row in the upper part of the SUTs represents a product balance for each

product.

6.7. In the lower left part of the use table (rows (14)–(19) in Table 6.1), the components of

GVA are shown below intermediate consumption for each industry. If the industry output is given

and the intermediate consumption of products determined in the use table, GVA of an industry can

be estimated, in the first instance, as a residual variable. If, however, the income measure

components of GVA (compensation of employees, other net taxes on production, consumption of

fixed capital) are known, the residual value is net operating surplus and net mixed income. Net

operating surplus can also be directly estimated using business accounts providing an alternative

for data confrontation with the residual approach – the linking between business accounts and

national accounts is covered in chapter 2. For each industry, the sum of intermediate consumption

at purchasers’ prices and GVA at basic prices will equal the value of output at basic prices shown

as column totals in the supply table.

6.8. The columns of the use table cover the following categories:

Households

NPISH

General

government

(1) (2) (3) (4)

(5) (6) (7)

(8) (9) (10)

(11) (12) (13) (14)

(15)

(16)

Agriculture (1) 2 583 6 570 16 371 34 49

9 623 3 595

180 0 - 27 1 161

4 909 14 532

Manufacturing (2) 2 205 107 190 12 441 16 874

6 015 8 797

153 522 71 438

0 3 180 26 756 2 183 3 034 123 252

229 842

383 364

Constructi o n (3) 105 2 440 9 528

2 446

3 907 1 604 20 029 1 667

0 0

25 155

0 - 38 563 27 348 47 377

Trade (4) 33

1 883 119 2 240 259 308 4 842 3 325 0 0 67 45 0

753 4 189 9 032

Transport (5) 14 4 386 267 8 399 822 321 14 208 5 833 0 3 370 0 0 0 8 453 17 656 31 865

Communication (6)

2 563 299 9 359 5 919 1 833 20 008 26 444

0 121

5 976

0 67

6 905 39 514

59 522

Finance and business services (7) 457

13 578 4 736 20 359 29 166 9 134 77 430

38 838

0 1 006

11 170 0 - 178 11 145

61 981

139 411

Other services (8)

382

59 1 171

415 1 794 3 829 14 923

5 416 53 373 113

107 1 567 74 500

78 329

Total at purchasers’ prices before

adjustments

(9) 5 440

138 991

27 466 61 219 46 538 23 839

303 492 166 063 5 416 61 050 69 418

2 335 2 859 152 800 459 939 763 431

CIF/FOB adjustments on exports (10)

0 0 0

0 0

0 0 0 0 0

0 0

0 - 97 - 97 - 97

Direct purchases abroad by

residents

(11) 0 0 0 0

0 0

0 6 675 0

0 0

0 0 0 6 675

6 675

Purchases in the domestic territory

by non-residents

(12)

0 0 0 0 0 0

0 - 12 945 0 0 0 0 0

12 945 0 0

Total at purchasers’ prices

(13) 5 440 138 991 27 466 61 219 46 538 23 839 303 492 159 792

5 416 61 050 69 418 2 335 2 859 165 648

466 517 770 009

Compensation of employees (14) 551 30 679 10 239 37 906 22 997 41 971 144 343

0 0 0 0 0 0 0 0 0

Other taxes less subsidies on

production

(15) - 1 627 1 077 546 1 755

2 004 1 103 4 858 0 0 0 0

0 0 0 0 0

Consumption of fixed capital (16) 1 845 12 750 1 542

10 917 18 934 7 480 53 469 0 0 0 0 0 0 0 0 0

Net operating surplus/net mixed

income

(17) 3 658 16 453 5 138 23 040

18 989 4 921 72 198 0 0 0 0 0 0 0 0 0

Gross operating surplus/gross

mixed income

(18)

5 503

29 203 6 680 33 957 37 923 12 401

125 667 0

0 0 0 0 0

0 0

GVA (19) 4 427

60 959 17 465 73 618

62 923 55 475 274 868

0 0 0

Total input at basic prices (20) 9 867 199 950 44 931 134 837 109 461 79 314 578 360 0 0 0

0 0 0 0 0 0

Empty cell by construction

Total use at

purchasers’

prices

PRODUCTS

Final consumption expenditure

Total

Exports

Change s in

inventories

Change s in

valuables

Other

services

Finance and

business

services

Total

Trade, transport

and

communication

Constructi o n

Manufacturing

Agriculture

Gross fixed

capital

formation

Adjustments

GVA

INDUSTRIES

FINAL USE

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

159

• Industries (columns (1)–(6) in Table 6.1). Intermediate consumption will be shown with the

same breakdown (number of columns) as for the industries’ domestic production in the

supply table.

• Final consumption expenditure (columns (8)–(10) in Table 6.1), which consists of final

consumption of households, NPISHs and general government, The latter is typically broken

down into individual and collective consumption.

• Gross capital formation (columns (11)–(13) in Table 6.1), which is broken down in its

components of total value of the gross fixed capital formation, changes in inventories and

acquisitions less disposals of valuables. This can be further broken down by types of assets

or industries.

• Exports of goods and services (column (14) in Table 6.1), which may be shown as a single

column or as goods and services in separate columns. These can be further broken down as

columns for export of domestically produced products and re-exports. Exports can also be

broken down by countries or geographical or market groupings of countries.

6.9. As described in chapter 4, the actual number of rows and columns in the SUTs will depend,

among other aspects, on resources and availability of source data.

6.10. The use table also includes a number of rows (in particular, rows (10)–(12) in Table 6.1)

which contain adjustments. In particular, these rows contain the adjustment for the valuation of

exports (CIF/FOB adjustments on exports), direct purchases abroad by residents, and purchase in

the domestic territory by non-residents. In the SUTs, total imports and exports are valued FOB.

Data on detailed flows of imports from foreign trade statistics are valued, however, at CIF prices.

To reconcile the different valuations used for total imports FOB and the imported products CIF, a

total CIF/FOB adjustment row on imports is added to the supply table. The same negative entries

are shown in the CIF/FOB adjustment row for exports. More details on the CIF/FOB issues may

be found in chapter 5.

6.11. The adjustments for direct purchases abroad by residents and purchases in the domestic

territory by non-residents have to be made because the final consumption expenditure of

households, as broken down by product, includes direct purchases by non-residents in the domestic

territory and these must be treated as exports. Similarly, direct purchases by residents abroad must

be treated as imports and thus included in the total final consumption expenditure of households.

6.12. The purchases by residents abroad are treated as both imports and final consumption

expenditure of households. Thus an appropriate positive amount has to be entered in the imports

column of the supply table and, at the same time, as a positive entry in the column of final

consumption expenditure of households in the use table. The purchases in the domestic territory

by non-residents are treated as exports and deducted from households’ final consumption

expenditure. Thus the corresponding amount entered in the exports column with a positive value

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

160

is deducted in the column of final consumption expenditure of households. The balance of the row

is zero.

6.13. The SUTs provide a framework enabling the supply and use of all products to be balanced,

and the total outputs and inputs of each industry to be balanced. In this stage of the compilation

process, however, it is recommended that the SUTs be populated using data based on the best

possible data sources before the balancing process takes place. Preferably, regular (quarterly and

annual) business surveys based on a high-quality comprehensive business register, along with

household-based surveys and the use of administrative data, should be used.

6.14. A simple method for drawing up product balances – traditionally referred to as the “product

flow” method – is to distribute the value available for domestic use based on the characteristics of

each product. This may work well for products that have specific uses, for instance, as input in a

particular industry or as gross fixed capital formation. Most of the products in the SUTs are,

however, broad categories of goods or services that may have several different uses. Estimates of

the use side that are solely based on the supply of specific products can, however, be used where

information from the use side is not available.

6.15. For compiling the use table, two general options are available: the input approach and the

output approach. In the input approach, the cost structures of industries and input structures of

final use categories are compiled on the basis of specific survey results, while in the output

approach the allocation of goods and services is determined with the product flow method. As the

input approach is based on collected data, it is the recommended approach for populating the

intermediate use part of the use table. The output approach is an alternative, providing a cross-

check and forming the basis of the balancing process.

6.16. There is no absolute rule determining whether to give priority to columns or rows of a use

table. This depends on basic surveys and the specific country practices of national accounts, and

also on such indicators as quality and coverage of the data. It is, however, recommended that the

compilation process be started by column, because data received from basic sources will then be

fully reflected. At the same time, this method is consistent with the institutional approach to

identifying the input structures for industries by intermediate consumption and GVA, and for the

categories of final use (consumption, gross capital formation, exports) by product. A distinction

must be made, however, between population of the tables with a tendency by column and the

balancing of the tables with a stronger row dimension.

6.17. The prime objective of the use table is to identify the cost structures of industries and the

input structure of final uses. The input approach can be implemented if survey results are available

which identify the main cost structures – the survey approach to the collection of input data is

recommended. The main types of sources for the input approach are the establishment survey, the

consumer expenditure survey, the government expenditure survey, and the capital expenditure

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

161

survey. At the same time, the use table identifies the use of products and primary inputs. The main

sources of the output approach are production statistics and foreign trade statistics.

6.18. The output approach (product flow method) is highly dependent on survey results from

production statistics and foreign trade statistics.

6.19. Only in the absence of any surveys on the cost structure of industries or any information

from input methods should the product flow method be considered. The product flow method can

be useful for compiling the rows in a first stage, even if later on they are changed during the

balancing process. If product flows are compiled at a very detailed level, it will be possible to

break down intermediate consumption between industries even in the absence of complete and

direct information on cost structures. Only in the case of specific products, for example, ships,

military aircraft, nuclear fuels, and others, should the product flow method be preferred.

6.20. The output approach is often used to compile use tables. It is a valuable complement of the

input approach. The product flow method facilitates identifying the output structure of goods and

services. The more homogeneous goods and services are, the easier it will be to allocate the use in

specific industries or categories of final uses. The product flow method is widely applied for

rectangular systems of products and industries in which the number of products is much larger

than the number of industries. Provided that survey results are available, the input approach is the

best option to identify cost structures. The product flow method is a valuable cross-check for the

input approach. It makes possible the identification of a more refined structure of intermediate and

final uses in terms of specific products. The product flow method is also a powerful tool when it

comes to balancing the system.

6.21. Nevertheless, the first stage is always to compile the totals of industries in terms of output,

intermediates and GVA. This takes place in the production accounts of the system. Then the

categories of final uses are added which were derived from specific surveys and statistics and

product flow accounts.

1. Three-dimensional presentation of the SUTs

6.22. Trade and transport margins, taxes less subsidies on products (except VAT) and non-

deductible VAT must be distributed by products in the valuation table (sometimes referred to as

the “bridge column”), shown as part of the supply by products at purchasers’ prices. It is

recommended that, where the supply values from the different levels of the valuation table are

distributed by uses in “valuation layers”, the matrices should have the same size and format as the

upper part of the use table at purchasers’ prices.

6.23. Figure 6.1 shows how the layers can be shown, stacked one above the other, in a three-

dimensional representation of the use table. In this way, it will be possible to look at each product

balance as a vertical slice, going from left to right, from the SUTs, where the supply is shown at

basic prices, while the uses can be seen as a table showing how purchasers’ prices are transformed

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

162

into basic prices by removal of trade and transport margins, net taxes on products excluding VAT

and non-deductible VAT.

6.24. The exact distribution of trade and transport margins and net taxes on products by uses

cannot usually be observed in the data sources and need therefore to be estimated on the basis of

whatever data are available, and also on common-sense assumptions., Further details of this

approach may be found in chapter 7. Hence the establishment of the full three-dimensional system

will require some additional resources. It can, however, be very useful as a tool for keeping track

of the use of trade and transport margins and net taxes on products by uses, which are also needed

for estimation of SUTs in volume terms and IOTs (both in current prices and in volume terms).

Figure 6.1 Three-dimensional view of SUTs

C. Intermediate consumption part of the use table

6.25. This section describes how to put together an initial unbalanced version of the intermediate

consumption part of the use table and the data sources that can be used.

6.26. Intermediate consumption consists of the value of the goods and services consumed as

inputs by a process of production, excluding fixed assets whose consumption is recorded as

consumption of fixed capital (2008 SNA, para. 6.213). It thus includes all non-durable goods and

services with an expected life of less than one year which are used up in the process of production

by industries, thus excluding any goods purchased for resale without any further processing. The

bought and not-consumed goods are entered in changes in inventories. Goods paid for by

employers for the benefit of their staff can be regarded as remuneration in kind entered in

compensation of employees.

6.27. The compilation methods for intermediate consumption vary across countries, depending

on the data sources available. The recommended approach is to have regular (for example, annual)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

163

data collection on input structures, an approach that is even more advisable with globalization and

new technologies contributing to rapid structural change. The most commonly used approach,

however, consists in starting with total intermediate consumption by industry – namely, the “Total”

row (row (9) in Table 6.1). Then there is a balancing process with the amounts which are available

for intermediate use of the various products (see the following sections). Lastly, an equilibrium is

obtained between the sum of the “Total” row and the sum of the “Total” column.

6.28. Nevertheless, in some countries, in particular those in which accounts rely on data sources

on enterprises (institutional approach), intermediate consumption may be initially known at a high

level. Thus the main problem consists in distributing this total intermediate consumption among

industries.

6.29. Intermediate consumption can be broken down by industries according to ISIC and by

institutional sectors. As for the supply table, it is also possible to distinguish between market

producers, non-market producers and producers for own final use and between producers in the

formal and the informal economy. The columns in the use table correspond to the same

classifications used in the supply table. When estimating the input structure, however, the number

of columns does not necessarily need to be identical, although it is recommended that, where

appropriate, the industry columns headings in the supply table and the use table are the same.

6.30. The separated columns of an ISIC category (for example, market and non-market) can have

significantly different input structures. Trade margins on their inputs may also differ because of

the use of different trading channels and discounts and taxation rules may vary, such as those

relating to the deductibility of VAT.

6.31. When a distinction is made between market producers, non-market producers and

producers for own final use, the inputs in each of these categories of units could be shown as

separate submatrices. This would, however, leave many input columns empty or almost empty. A

practical solution could be to separate only market and non-market producers within the same ISIC

category into different columns where both have a significant size, for example, in the case of

health and education services. Similarly, production on own account or informal activity may be

shown in separate columns, if it is of special interest, for instance within agriculture, construction

or trade.

1. Initial, unbalanced version of the intermediate consumption part of the use table

6.32. The information used for the construction of an initial set of estimates for the intermediate

consumption part of the use table will be drawn from various source data.

6.33. For some industries, the source data may consist of a complete picture of outputs and inputs

by products. These estimates will typically combine information on physical volumes and prices

and may also use information from accounting data. The estimates may be carried out outside the

actual SUTs compilation process. A similar situation obtains for the grossing up of survey results

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

164

to cover the entire industries in question, for example, agriculture, forestry and fishing. From the

perspective of the national accounts and SUTs compilation, the values of all inputs are assumed to

be already grossed up and ready to be entered into the SUTs framework.

6.34. It is important to ensure consistency between estimated outputs, inputs and change in

inventories within each industry. Products delivered between units within the industry should

appear with the same value for sales and purchases of the industry, except for the costs of change

of ownership.

6.35. If different data sources are used to draw up the output and input sides of an industry, it is

important to make sure that the units behind the data are defined in a similar way. Otherwise there

is a danger that the industry’s GVA could be overestimated or underestimated. It is also important

to ensure that all the inputs are covered and that less important inputs are not missing from the

input structure.

6.36. As mentioned before, the most common method used in filling the initial and unbalanced

version of the intermediate consumption part of the use table is, first, to estimate the values for

total intermediate consumption by industry; second, to enter in the table known values of

intermediate consumption by product and by industry when available; and, third, to use additional

information on cost structures to estimate all other values in this part of the use table.

(a) Total intermediate consumption by industry

6.37. The data sources used to populate the use table so as to obtain the value of total intermediate

consumption by industry (see chapter 4 for more detail on this) include statistical surveys,

accounts-based statistics and VAT-based statistics. It should be noted that the same source for the

industry providing data feeding into the supply table often also forms the source for the data

feeding into the use table.

6.38. For some industries, all inputs are provided by the data supplier and total intermediate

consumption is calculated as the sum of these inputs. For most industries, however, total

intermediate consumption needs to be estimated from annual and/or quarterly business surveys

based on information from business accounts, government accounts and other sources, such as

annual reports and financial statements of the economic units themselves.

(b) Inputs by products for each industry

Known values for specific cells, rows or columns

6.39. Some values for cells of the intermediate consumption part of the use table may be known

from source data, or they may already have been calculated within the framework of independent

subsystems, such as the use of energy, FISIM or insurance services by industries. For some

industries, the entire columns of initial inputs can be drawn up in subsystems from which they can

be transferred to the SUTs framework. Entire rows and columns may be filled in this way. The

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

165

source data used to calculate such values may not necessarily be available at purchasers’ prices

and they may need to be converted using the best possible assumption on trade and transport

margins, net taxes on products and non-deductible VAT.

Input structures

6.40. Surveys on cost structures and other information on the input structure could be used for

estimating the input structure in the intermediate consumption part of the use table.

6.41. Ideally, detailed purchases of goods and services should be collected annually covering all

industries through surveys of cost structures. These annual surveys typically capture various

aspects such as:

• Year-on-year structural changes

• Technological innovation and change

• Contracting out

• Company restructuring, mergers and takeovers

• Non-consolidation of businesses and industries

• Technological product and import substitution

• Economies of scale

• Inventory control

• Price changes

6.42. The collection of regular purchases details also makes possible the better measurement of

sudden economic change, due for example to the swine flu crisis (affecting agricultural and

slaughtering industries), storm damage (affecting construction and insurance industries), growth

of Internet activity, or other natural disasters which may affect the supply and price of certain

intermediate products with an impact on the production chains and other consequences.

6.43. Annual surveys of cost structure would also considerably facilitate the production of good

quality SUTs in volume measures, which has a key annual focus. Annex A to this chapter provides

an example of the type of costs and inventories questions, by type of product covering all goods

and services that could be included in a questionnaire.

6.44. The information from surveys based on business accounts may contain some useful

subtotals, for example, purchases of goods or indirect production costs, but it would usually be

insufficient to provide a full breakdown by product and by industry. Similarly, government

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

166

accounts may contain supplementary information on the input of products, but generally not with

sufficient detail for full input structures.

6.45. Known totals and subtotals must be supplemented by other information. Surveys on the

use of raw-materials and services are typically used to serve the needs of national accounts,

specifically the use table. The coverage of such surveys has traditionally been limited to

manufacturing and other industries that have a significant input of raw materials. The cost

structures of construction, distribution and service industries should also be surveyed, however.

The level of detail should preferably make it possible to aggregate the collected data into the

products used in the SUTs but, for practical reasons, it may equally be the product categories from

surveys that need to be split into more detailed categories used in the SUTs – the latter approach

generates lower quality estimates. A survey may only exist for earlier or later years than the year

for which the SUTs are being compiled, in which case the revaluation of the values to prices of

the year in question should be considered.

6.46. Surveys of cost structures may not have total coverage. They may exclude units below a

threshold or they may be based on small samples. When results are grossed up to industry totals,

uncertainty is added to the figures. Input structures of small establishments that are not covered by

the survey will probably be unlike the structures found in the survey. Furthermore, surveys may

collect data for enterprises rather than establishments. Enterprise-based data will include inputs of

establishments classified in other industries, and therefore reflect some inputs which should not be

included here but in another industry.

6.47. Cost structures typically also include some acquisition of capital equipment originally

treated as current expenditure in business accounts.

6.48. If annual surveys are not available, the use of older SUTs, where a more detailed survey

may have been conducted, can be used to form an initial structural base for a more recent period.

Where such periodic surveys exist, different processes can be applied to generate more recent time

series and meaningful data using more recent control totals applied to old structures.

6.49. Lastly, respondents do not always know all their inputs, and categories whose descriptions

begin with “Other” or finish with “not elsewhere classified (n.e.c.)” will usually be overstated in

such surveys. It may be a good idea to apply some common-sense corrections to the survey data

before they are used to create input structures. The grossed-up values of inputs by products

calculated from surveys are probably, despite their inaccuracy, often the best possible initial

estimates of input structures, but they should be used with some caution.

6.50. The availability of statistics on cost structures can vary greatly among countries. For

example, there are countries where surveys on cost structure are conducted yearly for all industries;

there are countries where only manufacturing industries are surveyed annually and most other

industries are surveyed with regular intervals; and there are countries where such surveys are

scarce, outdated or altogether missing.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

167

6.51. It is advisable that the need for new surveys is considered well in advance when new

benchmark SUTs are planned, as it may take several years to plan and implement and obtain

useable results from such surveys. Where no statistics are available, the possible existence of

alternative data sources should also be investigated.

6.52. In some countries, it is common for annual reports and accounts of enterprises to include

detailed descriptions of the use of inputs, for example, supporting purchase ledger details. The

information is, however, typically shown in a rather unsystematic way. To make this detail useful,

the data must be categorized in a way that corresponds to the product classification used in SUTs.

A similar approach can be used for inputs in general government, if the accounts for central and

local government contain detailed information that can used to create input structures.

6.53. If nothing is known about inputs for some industries, it may be possible to use input

structures from other industries that are assumed to be similar, but with some modifications based

on expert opinions. As a last resort, input structures from neighbouring countries or similar

activities in other countries could be considered, in particular, if those structures are based on

actual source data, and account should also be taken of possible differences in the extent of

processing and other factors. The initial inputs that are based on these kinds of approaches are

uncertain, and they are more likely to be adjusted – within reasonable limits – in the balancing

process.

6.54. In countries where the informal economy forms a considerable share of output in specific

industries, the structure of inputs used in informal units can be expected to differ from the structure

found in surveys of formal businesses. Within agriculture and related activities, the input in

informal units may be covered by agricultural statistics. Otherwise it may be appropriate to make

a separate estimation of the input structure in the informal economy. Information on the use of

inputs in informal activities may be found in household budget surveys or labour force surveys, as

mentioned above, if not in special studies of the informal activities. Such data may give an

incomplete picture of the outputs and inputs and it may be necessary to add some supplementary

assumptions based on expert knowledge before they can be used in SUTs.

6.55. Information on what are referred to as the “known” values and information on the input

structure can be put together in order to fill an initial version of the values of intermediate

consumption by product of an industry. Box 6.1 provides a numerical example of how all the

available information can be combined.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

168

Box 6.1 Example of a calculation of the values of an input column

In this example the total input, 2,500, is supposed to be known. A survey-based input structure sums

to 100 per cent.

The values of input of products 2 and 5 are known as 150 and 300 respectively. The known values are

treated as predetermined and subtracted from the total value that is going to be distributed by products.

The residual total value of inputs, 2,500 - 450 = 2,050, is now distributed proportionally with the

“survey-based input structure” excluding the “known values”.

Finally, the known and the calculated input values are added together to form the complete initial

column of intermediate consumption for the industry in question.

(d) Grossed-up data versus the data collected by surveys

6.56. Putting together information from different kinds of sources in the SUTs framework

generally will result in an unbalanced set of SUTs. The total uses of a product will therefore differ

from the total supply for most products. In the final, balanced version of the use table such

differences are removed either by manual adjustments or by automatic methods.

6.57. The corrections that are necessary to remove the differences between the first estimates of

supply and use should as far as possible retain those values that are considered to be reliable

statistics. For this purpose it is useful to be able to distinguish between:

• Inputs by product that have actually been reported as primary statistics by respondents and

values that are found in annual reports, government accounts or other reliable sources

• Initial inputs by product that are the result of grossing up to the estimates for total input in

each industry

6.58. In most cases, the difference between the two will represent the value that can be removed

during the balancing process when total initial use exceeds total initial supply for a product. It can

be useful if the reliable parts of the input values are shown together with the grossed-up values in

the tables presented to the people working on the manual balancing of SUTs.

Survey based

input

structure

Known

(predetermined)

values

Survey-based input

structure excl. known

values

Inputs estimated from

survey-based structure

Result:

Input column

% Value % Value Value

(1) (2) (3) = (1) * 100.0/80.0 (4) = (3) * (2500-450)/100

(5) = (2) + (4)

16.0 0 20.0 410 410

6.0 150 0.0 0 150

7.0 0 8.8 179 179

44.0 0 55.0 1 128 1 128

14.0 300 0.0 0 300

8.0 0 10.0 205 205

5.0 0 6.3 128 128

100.0 450 100.0 2 050 2 500

Input in an

industry

Products

Product 5

Product 6

Product 7

Total

Product 1

Product 2

Product 3

Product 4

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

169

6.59. An example, of part of an industry input structure (column) with supplementary reliable

values at purchasers’ prices in the rightmost column is shown in Table 6.2. Columns (6) and (7)

show values from the two approaches, which in turn should be investigated to achieve a plausible

industry input structure and, in turn, an agreed estimate for each product. Apart from wood-wool,

all the other products show significant differences, which call into question the reliability of the

data feeding into the SUTs for the input structure for this industry.

Table 6.2 Intermediate consumption of selected inputs

into “Manufacture of rubber and plastic products”

P.2: Intermediate consumption.

6.60. An example showing part of a product balance (row) with supplementary reliable values

in the rightmost column is shown in Table 6.3. Here the reported values on the supply side refer

to basic prices, while the reported values on the use side refer to purchasers’ prices. Although the

basic price estimates are in balance (allowing for rounding differences), columns (6) and (7) show

significant differences at purchasers’ prices between the estimated value and the reported value.

Transaction Basic price Margins

Net taxes

on products

(excl. VAT)

VAT

Purchasers'

price

Reported

values

(1) (2) (3) (4) (5) (6)

(7)

.. .. .. .. .. .. ..

Wood-w ool P.2 22 0 0 0 22 20

Wood in logs or roughly cut P.2 5 625 607 0 0 6 232 2 340

Plyw ood, laminated w ood P. 2 10 286 1 024 0 0 11 310 10 539

Packaging material, w ood P. 2 20 085 580 0 0 20 665 12 816

Other w ood products P.2 20 352 1 854 0 0 22 206 20 353

Paper in rolls and sheets P.2 3 027 46 0 0 3 073 1 329

… .. .. .. .. .. .. ..

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

170

Table 6.3 Sample product balance for "Gelatine and gelatine derivatives"

P.1: Output

P.2: Intermediate consumption

P.6: Exports of goods and services

P.52: Changes in inventories

D. GVA part of the use table

6.61. Once the intermediate consumption part of the use table has been estimated, it is possible

to calculate the GVA for each industry. The GVA at basic prices is estimated as total output at

basic prices from the supply table minus total intermediate consumption at purchasers’ prices from

the upper part of the use table. The GVA can be broken down into the following components:

• Compensation of employees

• Other taxes less subsidies on production

• Gross operating surplus and gross mixed income

6.62. Gross operating surplus can be further split into net operating surplus and consumption of

fixed capital on gross operating surplus. In addition, gross mixed income can be further split into

net mixed income and consumption of fixed capital on gross mixed income. If information is

available, these breakdowns could be shown in the Use table as follows:

Transaction Basic price

Reported

values

(1) (2)

(3)

Oher food products, n.e.c. P.1 12 12

Paints and soap, etc. P. 1 230 779 230 779

Imports P. 1 136 245 136 244

Gelatine amd gelatine derivatives Total 367 036 367 035

Use

Transaction Basic price Margins

Net taxes on

products

(excl. VAT)

VAT

Purchasers'

price

Reported

values

(1) (2) (3) (4) (5) (6)

(7)

Meat products P.2 14 029 4 749 0 0 18 778 3 498

Fish products P.2 932 31 0 0 963

Dairy products P. 2 9 925 7 958 0 0 17 883

Bakery products P.2 133 110 0 0 243 134

Other food prod. n.e.c. P. 2 109 577 22 735 0 0 132 312 67 143

Paints and soap, etc. P. 2 34 113 7 170 0 0 41 283 26 765

Pharmaceuticals, medicine P. 2 64 658 9 516 0 0 74 174 7 315

Rubber and plastic products P.2 17 812 3 754 0 0 21 566 10 436

.. .. .. .. .. .. ..

Change in inventories, materials P. 5 2 736 141 0 0 877

Change in inventories, goods for resale P.52 736 141 0 0 877

Exports of domestic production P.6 98 487 357 0 0 98 844 97 197

Re-exports P.6 10 287 35 0 0 10 322 10 150

Gelatine amd gelatine derivatives Total 367 035 56 697 0 0 0 222 638

Supply

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

171

Gross value added

Compensation of employees

Other taxes on production

Other subsidies on production

Gross operating surplus

Consumption of fixed capital on gross operating surplus

Net operating surplus

Gross mixed income

Consumption of fixed capital on mixed income

Net mixed income

6.63. Each of the categories of GVA is described below, together with the relevant data sources.

1. Compensation of employees

6.64. Compensation of employees is defined as the total remuneration, in cash or in kind, payable

by an enterprise to an employee in return for work done by the latter during the accounting period

(2008 SNA, para. 7.5). Compensation of employees has two main components: wages and salaries

payable in cash or in kind; and social insurance contributions payable by employers (actual and

imputed) (2008 SNA, para. 7.42). Generally, statistics drawn from business accounts and

government accounts show values for wages and salaries and possibly also other costs related to

employment. In both cases, there may be conceptual differences from the national accounts

concepts, due, for example, to the different treatment of particular issues such as fringe benefits,

employers actual and imputed social contributions, and other factors. In addition, the information

usually needs to be grossed up to cover the part of each industry that is not covered by the statistics

(for example, non-exhaustive business register) and values that are only available for enterprises

which would need to be distributed by establishments before use.

6.65. Information from tax-collecting authorities can provide data on compensation of

employees that will also cover industries not fully covered by accounts statistics or surveys. This

information may contain a distribution by industries that can be more or less consistent with the

industry classification used in other statistical sources. If a business register is used to classify data

from the various sources, it is likely that the figures from various sources will be classified in the

same way. As the structure of many economic units will change over time, the same units may

nevertheless be classified differently in different data sources. It should be borne in mind that data

collected for administrative purposes may refer to units that are neither enterprises nor

establishments, thus for the purposes of national accounts further alignment may be necessary for

consistency.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

172

6.66. In some countries, data for compensation of employees in a number of industries is

estimated based on labour force surveys, household budget surveys or occasional industrial

censuses in combination with population censuses. This is of particular relevance to cases where

a considerable share of economic activity takes place in the informal economy.

2. Other taxes less subsidies on production

6.67. The other component of GVA at basic prices consists of other taxes less subsidies on

production, which may be shown separately. Other taxes on production consist of all taxes except

taxes on products that enterprises incur as a result of engaging in production (2008 SNA, para.

7.97). Similarly, other subsidies on production consist of subsidies except subsidies on products

that resident enterprises may receive as a consequence of engaging in production. There are

different types of taxes and subsidies on production. They may include taxes or subsidies on

payroll or workforce, subsidies to reduce pollution, recurrent taxes on land, buildings or other

structures, and others.

6.68. Generally, government accounts contain information on the total taxes and subsidy on

production for each type of taxes and subsidies as they cover the various taxes and subsidies

covered by the legislation.

6.69. The distribution of the taxes and subsidies by industries may be available from source data,

but when this is not the case, the total taxes and subsidies on production have to be allocated to

relevant industries (proportionally to the items to which they relate). The amounts should be shown

at an accrual basis.

3. Operating surplus and mixed income

6.70. The value of “gross operating surplus and gross mixed income” is obtained as a residual

when compensation of employees and other taxes and subsidies are subtracted from GVA at basic

prices by industry. Estimates of these industry totals are usually available early in the process of

compilation of the use table and should be checked for credibility before balancing SUTs. As

shown in chapter 2, however, direct estimates of mixed income and gross operating surplus can be

estimated using administrative sources in addition to company accounts. In doing so, a

complementary estimate can provide data confrontation with the residual estimate.

6.71. The consumption of fixed capital on operating surplus and mixed income is usually based

on the perpetual inventory model method. The corresponding values for net operating surplus and

net mixed income are usually calculated in one of the final steps in the compilation of SUTs as the

calculation of consumption of fixed capital requires finalized data on gross fixed capital formation

broken down by industries, by institutional sector and by type of asset.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

173

E. Final consumption expenditure part of the use table

6.72. Final consumption expenditure is the amount of expenditure on consumption goods and

services (2008 SNA, para. 9.7). Final consumption expenditure can be disaggregated between

individual consumption expenditure and collective consumption expenditures. The first consists

of expenditure on individual consumption goods or services that are acquired by a household and

used to satisfy the needs or wants of members of that household. The latter consists of the

expenditures for collective consumption services which are services provided simultaneously to

all members of the community or to all members of a particular section of the community, such as

all households living in a particular region.

6.73. Final consumption expenditure is disaggregated in the use table in final consumption

expenditure by households, NPISHs or general government. Table 6.4 shows the structure of the

final consumption expenditure in the use table.

Table 6.4 Categories of final consumption expenditure

6.74. The manner in which the submatrix of the use table is compiled, showing the use of

products for final consumption, is similar for each of the three types of final consumer (households,

NPISHs and general government) but starts from a different classification for each of them,

reflecting the way (and the basic functional classifications) in which the basic data are collected.

1. Household final consumption expenditure

6.75. Household final consumption expenditure consists of the expenditure, including

expenditure whose value must be estimated indirectly, incurred by resident households on

individual consumption goods and services, including those sold at prices that are not

economically significant and including consumption goods and services acquired abroad (2008

SNA, para. 9.113).

6.76. Information on consumption by households usually starts from household surveys. In these

surveys, household expenditure is classified according to COICOP, as shown in Box 6.2. It is

therefore recommended to underpin the compilation of this part of the use table with a table linking

data on the final consumption expenditures by purpose and by products. This will greatly improve

the quality of the data and will ensure that the different analyses of household final consumption

expenditure are consistent and coherent with the balanced SUTs. It will also ensure homogeneity

in the deflation process, thus ensuring better quality volume data.

Products

Households NPISHs

General

government

Total

Product N

Total

FINAL CONSUMPTION

Product 1

Product 2

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

174

Box 6.2 Classification of individual consumption according to purpose

COICOP is an integral part of the SNA, but it is also intended for use in three other statistical areas: household

budget surveys, consumer price indices and international comparisons of GDP and its component expenditures.

The purposes defined in COICOP are based on the classifications of consumer expenditures which national

statistical offices have developed for their own use to serve a variety of analytic applications. Although COICOP

is not strictly linked to any particular model of consumer behaviour, the classification is designed to broadly

reflect differences in income elasticities. In 2018 the United Nations issued the first revision of COICOP,

COICOP 2018 (United Nations, 2018), to reflect users’ need for more detail and several other issues that required

a revision of the classification. There are 15 divisions in COICOP 2018:

01 - Food and non-alcoholic beverages

02 - Alcoholic beverages, tobacco and narcotics

03 - Clothing and footwear

04 - Housing, water, electricity, gas and other fuels

05 - Furnishings, household equipment and routine household maintenance

06 - Health

07 - Transport

08 - Information and communication

09 - Recreation, sport and culture

10 - Education services

11 - Restaurants and accommodation services

12 - Insurance and financial services

13 - Personal care, social protection and miscellaneous goods and services

14 - Individual consumption expenditure of non-profit institutions serving households (NPISHs)

15 - Individual consumption expenditure of general government

The first 13 divisions add up to total individual consumption expenditure of private households. The last two

identify those parts of consumption expenditure by non-

profit institutions serving households and general

government that are treated as social transfers in kind. Together all 15 items represent actual final consumption

by households.

6.77. Table 6.5 shows the table that is used to cross-classify data of final consumption

expenditures by purpose (COICOP classes) and by products (CPC classes). The list of products in

this table is the same as that used for the supply table and the intermediate consumption part of the

use table. Of course, the higher the level of disaggregation, the better the quality and precision of

allocation to products and the greater the homogeneity for deflation purposes.

6.78. The compilation of Table 6.5 relies on different data sources and is often based on

household budget surveys which directly collect details of expenditure on goods and services by

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

175

households and also on retail trade surveys (although some adjustments are needed). These data

sources are discussed in the following sections.

Table 6.5 Table linking final expenditures by purpose (COICOP) and product (CPC)

6.79. Some adjustments may be needed to ensure that the final consumption expenditures by

households reflect the final consumption expenditures of resident households. This means that, if

the starting point is the final expenditures that took place in the territory by households,

adjustments are needed to remove the expenditures in the country by non-resident households and

include the final expenditures of resident households abroad. When distributions by COICOP

groups of these adjustment items are unknown, they can be placed in one or two supplementary

columns with a positive and a negative value as appropriate.

6.80. Very few products are exclusively used for household final consumption expenditure. For

example, some domestic supply of typical consumer-related goods (for example, food) is used as

intermediate consumption in restaurants, transport services and government institutions, as

individual consumption of NPISHs and general government or – in the case of durables – as fixed

capital formation. Accordingly, the amounts spent by households (this is considered as final

consumption expenditures by households) cannot be determined from the supply of such products

without knowledge of the other uses.

6.81. For some products, full information on household final consumption expenditure can be

provided by subsystems established outside the SUTs framework. The underlying data may have

the form of physical volume and price information, such as energy. Administrative sources can

provide a rich source of detail, for example, covering purchases of motor vehicles, school fees and

other outlays such as education, prescription medicine, and other expenses on health services.

6.82. Table 6.6 provides a numerical example of the link between household final consumption

expenditure of households in the use table and COICOP by product breakdown. In this example,

the largest expenditure items are housing, followed by transport, restaurants and food.

Food

Non-alcoholic

beverages

…

Other services Total

Products

01.1 01.2 … 13.9

01.1-13.9

Total

Consumption

COICOP

groups

Household consumption (COICOP)

Product 1

Product 2

Product N

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

176

Table 6.6 Final consumption expenditure of households (by COICOP headings)

Table based on 2011 figures from Austria

Note: the products “Trade”, “Transport” and “Communication” of table 6.1 are presented together in table 6.6. At the

time of drafting, this table was compiled in accordance with COICOP (United Nations, 2000a).

6.83. COICOP also distinguishes household final consumption expenditure according to the

following product classes: services (S), non-durables (ND), semi-durables (SD) and durables (D).

This supplementary classification provides data for other analytic applications, such as assessing

household stocks of goods and the cyclical variation in consumer demand. For example, in dealing

with the stock of capital goods held by households, goods in COICOP classes that are identified

as durables provide the basic elements for such estimates. Box 6.3 provides a description of

durable, semi-durable, non-durable goods and services.

Box 6.3: Non-durable, semi-durable and durable goods

Non-durable goods are defined as goods that can be used only once, for example: food, non-

alcoholic beverages, alcoholic beverages, tobacco, materials for the maintenance and repair

of dwellings, pharmaceuticals, fuels, energy, garden plants, flowers, pets, newspapers and

stationery.

Semi-durable goods differ from durable goods in that their expected lifetime of use, although

more than one year, is significantly shorter, and their purchase price is typically less than

that of durable goods. For example: clothing, footwear, household textiles, motor vehicle

spare parts, recording media, games, toys, books, and electrical appliances for personal care.

Durable goods are those goods which can be used repeatedly or continuously over a period

of more than a year, for example: furniture and furnishings, carpets, major tools, vehicles,

telephone equipment, computers, photographic equipment, jewellery, clocks and watches.

Services include cleaning and hire of clothing, actual and imputed housing rental, repair

services, domestic services, outpatient and hospital services, transport services, post and

telecommunication services, recreational and cultural services, education, catering,

accommodation, hairdressing, insurance and financial services.

(a) Household budget surveys

6.84. Household budget surveys provide a good source for expenses classified by purpose and

by product according to internationally agreed standards. Such surveys may also include

Food and

non-

alcoholic

beverages

COICOP 01

Alc oholic

beverages,

tobacco

and

narcotics

COICOP 02

Clothing and

footw ear

COICOP 03

Housing,

w ater,

electricity,

gas and

other f uels

COICOP 04

Furnishings,

household

equipment and

routine household

maintenance

COICOP 05

Health

COICOP 06

Transport

COICOP 07

Commu-

nication

COICOP 08

Recreation

and culture

COICOP 09

Education

COICOP 10

Restaurants

and hotels

COICOP 11

Misc ella-

neous

goods and

services

COICOP 12

PRODUCTS

(1)

(2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)

Agricultural products

(1) 2 240 60

0 434 0 0 0 0 860 0 0 0 3 595

Manufactured products

(2)

14 016 5 537 9 749

7 090 10 198 2 254 12 809 2 548 2 767 0 0 4 470 71 438

Construction

(3) 0 0

1 667 0 0 0 0 0 0 0 0 1 667

Trade, transport, communication

(4) 0 0 0 0 0 0 7 827 3 361 4 407 0 20 008 0 35 602

Financal and business services

(5)

0 0 0 26 218 747 0 1 339 0 2 590 0 0 7 944 38 838

Other services

(6) 0

194 77 212 3 730 235 133 4 204 1 221 0 4 918 14 923

Total

(7)

16 257 5 597

9 943 35 487 11 157 5 984 22 209 6 041 14 827 1 221 20 008 17 332 166 063

CLA SSIFICA TION OF INDIV IDUA L CONSUMPTION BY PURPOSE (COICOP)

Total use at

purchasers'

prices

COICOP

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

177

information on a wider range of household activities and living conditions, and sometimes may be

named “living standard surveys”. Household budget surveys are generally used to compile or

update the weights for the basket of goods used in the CPIs and to collect information on household

income, possession of assets, type and equipment of dwellings, outlays for repair and maintenance,

production for own use and other informal economic activity. This supplementary information is

often useful for national accounts and PPPs and may actually have been collected with this purpose

in mind, and also for satellite accounts such as social accounting matrices. Household budget

surveys can contain information relating to consumption of goods produced for own consumption

and services from owner-occupied dwellings which in some countries may not be available from

other data sources.

6.85. Household expenditure surveys frequently use COICOP as the basis for the collection of

household expenditure information. The results are reallocated to products classified by CPC and

then used to estimate the vector of household final consumption expenditure by product for the

SUTs.

6.86. Household budget surveys often provide good initial estimates feeding into household final

consumption expenditure. Attention should be paid, however, to the coverage of the survey in

order to ensure that the survey results can be used for final consumption expenditures by

households. For example, some household budget surveys may not cover the year for which the

SUTs are being compiled, in which case the survey data should be referenced to prices of the actual

year and, if possible, corrected for the development in volumes from the surveyed period. In

addition, since household final consumption expenditure refers to the total resident population, the

statistician must ensure that the results of the household budget surveys are grossed up to cover

the total population.

6.87. If no household budget survey exists for the SUTs reference year, an alternative approach

may be to use the structure of expenditure from the last household budget survey and prorate this

structure to the estimate for total expenditure of the reference year. This clearly assumes a fixed

basket of spending on goods and services (this does not allow for relative price changes or changes

due to new products or products not anymore consumed). The balancing process will, however,

generate changes to this structure, for example reflecting changes from the supply side.

(b) Retail trade surveys

6.88. Household final consumption is linked to turnover of retail trade after adjusting for sales

to businesses and non-residents. Consumers buy most of their goods from retail outlets. Retail

trade statistics provide data on turnover broken down into product groups. Retail trade statistics,

however, do not include imputed transactions, such as imputed rentals of owner-occupied

dwellings and FISIM, which are included in household final consumption expenditure. These are

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

178

compiled using different sources and affect other parts of the national accounts, not just household

final consumption expenditure.

6.89. Retail sales statistics that are used to feed into household final consumption expenditure

need to undergo appropriate adjustments. For example, although consumers purchase most of their

goods and services from retail outlets, they also purchase goods and services from units not

classified to the retail industry, for example, directly from manufacturers and service industries.

On the other hand, sales by retail outlets are not all consumed by resident household consumers

but purchases from these retail outlets are also made by non-residents (for example, visitors) and

these are categorized as exports made by businesses and, in turn, treated as intermediate

consumption.

6.90. Accordingly, the turnover of retail trade and some services disaggregated by detailed

industries can provide valuable information on household final consumption expenditure

exclusively by broad categories of products and there is no one-to-one correspondence between

retail trade turnover by industries and household final consumption expenditure by COICOP

groups.

6.91. It should also be borne in mind that informal activities may contribute significantly to

household final consumption expenditure. In the cases like retail trade, it would be particularly

useful to extrapolate the structure of the column totals from already existing SUTs to form initial

estimates for subsequent years.

6.92. It is often possible to obtain detailed data on products that are subject to regulation, taxation

or subsidization, since this information is available from the responsible authorities.

6.93. Motor vehicles, alcohol and tobacco are typical examples of products subject to regulation.

It may, for example, be possible to use information on motor vehicle registration to determine

household final consumption expenditure of motor vehicles. Information on the use of alcohol or

tobacco (for example, related to duties paid) could be used to determine consumption of products

that are used for household final consumption expenditure, taking into account that these products

could also be used for hospitality by businesses or as an input used by restaurants, in which case

they would be treated as intermediate consumption and thus excluded from final consumption

expenditures by households.

2. Final consumption expenditure of NPISHs

6.94. Final consumption expenditure of NPISHs consists of the expenditure, including

expenditure whose value must be estimated indirectly, incurred by resident NPISHs on individual

consumption goods and services and possibly on collective consumption services (2008 SNA,

para. 9.115).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

179

6.95. Similar to the final consumption expenditure of households, it is useful to cross-classify

the consumption expenditures of NPISHs by products (according to CPC) and by purpose. The

reference classification of these expenditures by purpose is COPNI (United Nations, 2000a), see

Box 6.4. by convention, all consumption expenditures of NPISHs are treated as individual

consumption (see 2008 SNA, para. 9.107). Thus, all consumption expenditures of NPISHs are

described in COPNI, and also in division 13 of COICOP.

6.96. Table 6.7 shows the matrix that links the final consumption expenditures by NPISHs by

purpose (according to COPNI) and by product (according to CPC).

6.97. In some countries, NPISHs may produce most of the services within education, health or

social protection, while in other countries such services may mainly be produced by general

government and private enterprises. It may be appropriate to provide separate columns for COPNI

divisions in the use table. Where the activity of NPISHs is negligible, its detailed breakdown can

be considered to be less relevant.

6.98. Various data sources will be used to cover the details of NPISHs by industry and by

product. It is recommended, for example, that a business survey is used based on a sample of

NPISHs selected from the business register. The grossing-up methodology for a sample survey

would need to reflect that NPISHs are non-market bodies and not market bodies, in other words,

output is the sum of costs and not related to turnover. Other sources may also include company

accounts, regulatory bodies and collective group accounts covering, say, a group of trade unions

or religious bodies.

Box 6.4 Classification of the purposes of NPISHs

The main use of COPNI is to classify expenditures by NPISHs in a manner consistent with the purposes of

the individual consumption expenditures of households and general government in order to obtain the SNA

aggregate of actual final consumption of households.

COPNI can also be used to facilitate international comparisons of the activities of NPISHs. In many countries,

activities of these institutions are an important complement to government activities in terms of supplying

education, health and social protection services to the population. In some countries, NPISHs are also

becoming prominent in non-traditional areas, such as environmental protection, the protection of human rights

and the defence of minority groups.

Nine divisions are distinguished in COPNI:

01 – Housing

02 – Health

03 – Recreation and culture

04 – Education

05 – Social protection

06 – Religion

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

180

07 – Political parties, labour and professional organizations

08 – Environmental protection

09 – Services n.e.c.

Note that the nine divisions of COPNI correspond to division 14 of COICOP 2018, which classifies the

individual consumption expenditure of NPISHs (see box 6.2).

Table 6.7 Table linking final consumption expenditures of NPISHs

by purpose (COPNI) and by product (CPC)

3. Final consumption expenditure of general government

6.99. General government final consumption expenditure consists of expenditure, including

expenditure whose value must be estimated indirectly, incurred by general government on both

individual consumption goods and services and collective consumption services (2008 SNA, para.

9.114).

6.100. Final consumption expenditures of general government can be classified in several ways.

For example:

• According to whether the goods or services have been produced by market or non-market

producers.

• According to whether the expenditures are on collective services or individual goods or

services.

• By function or purpose according to the Classification of the Functions of Government

(COFOG).

• By type of good or service according to CPC.

6.101. The column of final consumption of general government in the use table is usually

underpinned with a matrix linking the final consumption of general government by product and by

purpose. The reference classification of final consumption of general government by purpose is

Housing Health …

Other services

n.e.c.

Total

Products

01 02 … 09

01-09

COPNI

Division

Final consumption of NPISHs (COPNI)

Product 1

Product 2

Product N

Total

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

181

COFOG (United Nations, 2000a) (see Box 6.5). Data from government accounts are usually

classified by COFOG groups. This classification may be more or less detailed but should in most

cases make it possible to distinguish between individual consumption and collective consumption.

Individual consumption corresponds to group 15 of COICOP.

Box 6.5 Classification of functions of government

A major use of COFOG is to identify consumption expenditures that benefit individual households and that

are transferred to division 15 of COICOP 2018 in order to derive the SNA aggregate of actual final

consumption of households (or actual individual consumption). The divisions, groups and classes covering

these expenditures are clearly indicated in the classification. COFOG also permits trends in government

outlays on particular functions or purposes to be examined over time.

COFOG is used in the analysis and presentation of statistics on government finance. COFOG consists of

ten divisions:

01 – General public services

02 – Defence

03 – Public order and safety

04 – Economic affairs

05 – Environmental protection

06 – Housing and community amenities

07 – Health

08 – Recreation, culture and religion

09 – Education

10 – Social protection

Each class of COFOG is clearly identified as collective services or individual services. In general, all of

classes 01–06 are collective services, as are sections 07.5 and 07.6 of health, sections 08.3–08.6 of

recreation, culture and religion, sections 09.7 and 09.8 of education, and sections 10.8 and 10.9 of social

protection. These sections cover expenditures on general administration, regulation, research that is not

recorded as capital formation and so on. The remaining sections of health, recreation, culture and religion,

education and social protection (which dominate each of the classes) are individual services (2008 SNA,

para. 9.100).

6.102. Table 6.8 shows the table linking consumption expenditures of the general government by

purpose (COFOG) and by product (CPC). An alternative would be to have column headings in

terms of industries (ISIC) and the row headings by product (CPC).

6.103. The transformation of the data based on type of function or purpose to industry is very

important, for example, as many regulatory and administrative services are classified to the public

administration and defence function or purpose but should be classified to industries such as health

and education. Again the level of detail is determined by its importance in the country. A split by

level of government could depict different characteristics in terms of COFOG categories, inputs

and outputs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

182

Table 6.8 Table linking final consumption expenditure of general

government by COFOG and CPC

6.104. Individual consumption of general government consists of two parts which may be shown

as separate columns in the use table:

• Goods and services produced by general government as a non-market producer

• Goods and services purchased by general government from market producers for onward

transmission to households either free or at prices that are not economically significant.

These goods and services are not included in the output of general government.

6.105. Data sources for general government primarily rely on central government and local

government administrative data, and are mainly provided by the finance ministries and local

government bodies. These are often supplemented with specific survey data such as very detailed

local government expenditure. Furthermore, some estimates are based on models such as the

perpetual inventory model estimating the consumption of fixed capital for both the central

government and local government sectors. All these sources may use a COFOG basis or an

industry basis – in both cases, they will need a CPC product breakdown.

F. Gross capital formation part of the use table

6.106. Gross fixed capital formation is measured by the total value of a producer’s acquisitions,

less disposals, of fixed assets during the accounting period plus certain specified expenditure on

services that adds to the value of non-produced assets (2008 SNA, para. 10.32). Gross capital

formation is measured by the total value of the gross fixed capital formation, changes in inventories

and acquisitions less disposals of valuables (2008 SNA, para. 10.31). In the use table, GCF is

usually at least broken down into three separate columns to display its components separately as

shown in Table 6.9. These are discussed in the following sections.

Collective services Individual services

General

public

services

Defence …

Soc ial

protection

Total

Health

Recreation

culture and

religion

Education

Soc ial

protection

Total

Products

01 02 … 10

Products

Total

Divisions

(COFOG)

Individual consumption general government (COFOG)

Product 1

Product 2

Divisions

(COFOG)

Collective consumption general government (COFOG)

Product 1

Product 2

Product N

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

183

Table 6.9 Categories of gross capital formation

1. Gross fixed capital formation

6.107. Gross fixed capital formation is measured by the total value of a producer’s acquisitions,

less disposals, of fixed assets during the accounting period plus certain specified expenditure on

services that adds to the value of non-produced assets (2008 SNA, para. 10.32). Fixed assets are

produced assets that are used repeatedly or continuously in production processes for more than

one year (2008 SNA, para. 10.11). Gross fixed capital formation is also described as capital

investment, fixed investment or capital expenditure.

6.108. One approach to the compilation of this part of the use table is the demand-based approach,

which requires detailed information on investment. Under this approach a matrix is compiled

linking gross fixed capital formation by industries (according to ISIC), by type of asset (see Box

6.6) and by products (according to CPC). This matrix is often referred to as the “investment

matrix”. In order to develop such a matrix by industry, and by institutional sector, for each type of

asset there should be an allocation to the appropriate product. In some cases, there may be a one-

to-one relationship between the asset and product but in the case of machinery, for example, there

are many one-to-many relationships.

Box 6.6 Gross fixed capital formation by type of asset

Gross fixed capital formation is usually shown by type of asset. The types of assets distinguished in the 2008

SNA are the following (see 2008 SNA, chapter 10, table 10.2).

Gross fixed capital formation by type of asset:

Dwellings

Other buildings and structures

Buildings other than dwellings

Other structures

Land improvements

Machinery and equipment

Transport equipment

ICT equipment

Other machinery and equipment

Products

Gross fixed

capital

formation

Changes in

inventories

Aquisitions less

disposals of

valuables

Total

Gross capital f ormation

Product 1

Product 2

Product N

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

184

Weapons systems

Cultivated biological resources

Animal resources yielding repeat products

Tree, crop and plant resources yielding repeat products

Costs of ownership transfer on non-produced assets

Intellectual property products

Research and development

Mineral exploration and evaluation

Computer software and databases

Computer software

Databases

Entertainment, literary or artistic originals

Other intellectual property products

The output and capital formation of research and development are particularly difficult to measure. In theory,

the value of the output of research and development is equal to the value of discounted future benefits that

businesses get from their research and development investment. These future benefits

are difficult to

estimate. Furthermore, most research and development is produced on an own-account basis. For that reason,

the sum-of-costs approach to the valuation of output will usually be applied. More detail on the impact of

capitalizing research and development costs on SUTs and IOTs is provided in annex B to this chapter.

6.109. Table 6.10 shows the structure of the matrix linking gross fixed capital formation by type

of asset, by product and by industry. Typically this matrix is based on surveys of capital

expenditure, which tend to focus on institutional sectors, industries and assets; supplementary data

sources are often also needed, however, such as, for example, specifically designed surveys which

collect and provide investment detail by type of product. Detailed information on investments is

of particular importance for gross fixed capital formation, which tends to be an erratic series and

cannot be modelled easily: for instance, not all businesses buy vehicles every year, and the length

of their use will vary across, and within, industries.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

185

Table 6.10 Table linking gross fixed capital formation by industries, assets and products

6.110. In countries where such level of detail data is not available, the product flow approach may

be applied, using assumptions linking the output of a product to the destination of the product in

terms of its purpose. This is a less optimal approach but allows output and demand to be matched.

6.111. It is important to note that gross fixed capital formation – like other product-based variables

in the use table – is valued at purchasers’ prices. It is recorded, however, by including any non-

deductible VAT and excluding any deductible VAT. This will have an impact on those industries

and products to which exemption applies and will be consistent with the valuation of the

intermediate inputs for the corresponding industries.

6.112. In its simplest version, the use table may show gross fixed capital formation as a single

column, and this would also fulfil the requirements for compiling SUTs and some users’ needs.

The single column approach may be preferred if information on gross fixed capital formation is

Gross fixed capital formation by type:

Gross fixed capital formation by industries and type of assets

GFCF

Dwellings

Other

buildings

and

structures

Machinery

and

equipment

Weapon

systems

Cultivated

biological

ressources

Intellectual

property

products

Total

Agriculture

Forestry

and

logging

Fishing and

aquaculture

. . .

Activities of

households

employers

Total

PRODUCTS

PRODUCTS ISIC 01 ISIC 02 ISIC 03 . . . ISIC 97

N N

Total

Agriculture

Forestry

and

logging

Fishing and

aquaculture

. . .

Activities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 02 ISIC 03 . . . ISIC 97

Agriculture

Forestry

and

logging

Fishing and

aquaculture

. . .

Activities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 02

ISIC 03 . . . ISIC 97

Agriculture

Forestry

and

logging

Fishing and

aquaculture

. . .

Activities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 02

ISIC 03 . . . ISIC 97

E.T.C

Gross fixed capital formation

GFCF, Dwellings

GFCF, Other buildings and structures

GFCF, Machinery and equipment

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

186

missing or incomplete. It should, however, be possible to distinguish between broad groups of

assets based on the product classification used in the SUTs framework.

6.113. The quality of the product breakdown is greatly enhanced, however, by the greater level of

detail linking industries, institutional sectors, assets and products. The disaggregation of gross

fixed capital formation by industries and institutional sectors is also needed to calculate the

consumption of fixed capital by industry, and in turn, the value of non-market producers’ output.

The breakdown can be done by columns that correspond fully to the columns for output and

intermediate consumption. It may, however, be feasible to limit the number of columns in such a

breakdown if precise information from source data is lacking.

6.114. If all combinations of gross fixed capital formation by types, industries and institutional

sectors of a detailed matrix were shown as columns in the use table, these columns would

completely dominate the presentation. Furthermore, a disproportionate share of the resources

needed to balance the SUTs might be required to distribute products between the gross fixed capital

formation columns, and finalization of the SUTs might be delayed unnecessarily.

6.115. One practical solution could be to show columns for a few broad categories of industries.

Another could be to show only columns for different types of capital formation within the SUTs

framework. The breakdown by industries could instead take place outside the central SUTs

framework in a subsystem of investment matrices. Here, gross fixed capital formation by product

from the final balanced version of the use table could be allocated to specific industries and

institutional sectors as a separate process.

6.116. Estimates of gross fixed capital formation by industries have, however, an important role

in the preparation of initial column totals for gross fixed capital formation for the use table.

Furthermore, a preliminary version of gross fixed capital formation by products and industries can

provide the starting point for the gross fixed capital formation columns of the use table.

6.117. Table 6.11 illustrates the gross fixed capital formation by industry and product link to the

gross fixed capital formation column in the use table. In essence, this identifies the producers of

capital goods in the rows and the investing industries in the columns.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

187

Table 6.11 Gross fixed capital formation by investing industry

Table based on 2011 figures from Austria

Note: the products “Trade”, “Transport” and “Communication” of table 6.1 are presented together in table 6.11

6.118. The gross fixed capital formation matrix has further roles to play. The calculation of capital

stock data and the calculation of a valuation matrix for non-deductible VAT require an assessment

of gross fixed capital formation by product (by producing industry) and investor (by investing

industry). In the investment matrix, the user concept of capital – and not the owner concept of

capital – should be reflected.

6.119. The assessment of consumption of fixed capital as a component of GVA should be based

on empirical capital stock data. The human-made capital stock is derived from cumulative

investment of the past in buildings, machinery and transport equipment, based on the actual

lifetime of a capital good and allowing for retirements and obsolescence by using the perpetual

investor method (see OECD, 2009).

6.120. Consumption of fixed capital is calculated at the current replacement cost of the net capital

stock. The net capital stock is defined as the financial value of the gross capital stock still in use.

Clearly, it will be easier to estimate these types of matrices on the basis of a rectangular SUTs

system with many disaggregated homogeneous products.

(a) Sources for gross fixed capital formation

6.121. For industries that are covered by business statistics, the source data will usually include

information on purchases and sales of capital equipment. It is normally possible to distinguish

between gross fixed capital formation in buildings, structures, vehicles and machinery and

equipment. Investment in intellectual property products is usually also shown but it is not

necessarily classified by categories that can be used for national accounts. For example, these

sources will often include purchased software and may also show a value for larger software

projects but not all projects produced on own account. The measurement process is further

complicated with the inclusion of acquisitions of patents, franchises and goodwill and the fact that

research and development may or may not be wholly or partially capitalized in the source data and

not necessarily recorded in line with the requirements of national accounts.

Agriculture Manuf acturing Construction

Trade,

transport,

communication

Financial

and

business

services

Other

services

PRODUCTS

(1) (2) (3) (4) (5) (6) (7)

Agricultural products

(1) 128 2 0 4 36 10 180

Manuf actured products

(2) 1 223 7 225 664 5 893 9 124 2 626 26 756

Construction

(3) 828 1 752 224 3 822 15 995 2 534 25 155

Trade, transport, communication

(4) 12 1 206 189 2 280 1 672 684 6 043

Financal and business services

(5) 124 4 331 82 1 324 3 065 2 245 11 170

Other services

(6) 0 0 0 0 0 113 113

Total

(7) 2 314 14 516 1 160 13 323 29 892 8 212 69 418

INV ESTING INDUSTRIES

Total at

purchasers'

prices

INDUSTRIES

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

188

6.122. The distinction between intermediate consumption and gross fixed capital formation in

business statistics sourced from business accounts may be different from that recorded in the

national account. For example, items such as purchased computer software may be recorded as

current expenditure in business accounts but as gross fixed capital formation in the national

accounts, thus an equal and opposite adjustment is required. There are other conceptual differences

between company accounts and national accounts, such as the following:

• Business accounts may not fully show gross fixed capital formation according to economic

ownership. Some financially leased assets may be included in investment by their legal

owners.

• Sales of existing assets should be treated as a negative gross fixed capital formation valued

at the actual prices obtained by the seller. When the sale takes place between two resident

producers, the positive and negative investment will cancel out for the economy as a whole,

except for costs of change of ownership. In business accounts, the figures for disposals of

assets will often be shown at historical cost while the corresponding cumulated depreciation

is shown as a separate item. The difference is the bookkeeping value of the sold asset. If the

actual price obtained differs from this value, the residual is included in secondary income,

and should be reviewed and adjusted for as appropriate. In practice, it can usually be assumed

that the difference between the negative gross fixed capital formation and the bookkeeping

value of the sold asset is insignificant but there may be important exceptions where figures

from company accounts are misleading.

• In business accounts, it is common practice to treat minor or regular purchases of equipment

as current expenses. Such acquisitions may not always be identifiable in the accounts. For

some big corporations, the threshold for classifying purchases as investment can actually be

high, say, $10,000 or more, but practices may vary between countries owing to differences

in legislation and taxation rules.

• Own account production of capital goods may be capitalized in business accounts. Even if

this is not the case, the accounts may contain information on the value of own account

production. The value shown may not, however, be at basic prices as it may only include the

direct cost of raw materials and wages and salaries attributed to its production, in which case

a correction for indirect costs and gross operating surplus may be appropriate.

• Own account production of intellectual property products may not be directly identifiable in

company accounts. Production of software, databases, research and development and

literary, artistic or entertainment originals may sometimes have been capitalized as

intangible assets in company accounts but may be treated as current expenses. For some

intangibles, gross fixed capital formation may be covered by business surveys. In the absence

of further detail, it is recommended that gross fixed capital formation is estimated based on

the wages and salaries paid for this kind of work with an appropriate mark-up for other

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

189

expenses and typical gross operating surplus. More details on this matter may be found in

the Handbook on Deriving Capital Measures of Intellectual Property Products (OECD,

2010).

6.123. Information on gross fixed capital formation within general government can usually be

found in government accounts. Many of the above issues are also relevant to investment within

general government. Accounts of central and local government will usually include a level of detail

that reveals the distinction between intermediate consumption and gross fixed capital formation.

Extra-budgetary units within general government may, on the other hand, provide less information

on the nature of their costs as may also be the case for NPISHs.

6.124. Special care should be taken when projects are partially or wholly financed by capital

transfers from outside, for instance from international organizations. In such cases, the accounts

may show values that are net of financing from outside. In national accounts, gross fixed capital

formation should record the full value of such projects.

6.125. The value of investment covered by business accounts will often give an incomplete picture

of total gross fixed capital formation because some industries are only partially covered or lack

information on some types of investment. The initial estimates of gross fixed capital formation

may be prepared within an investment matrix framework that shows investment by industries,

institutional sectors, types and products. In the event that uncertain data and so-called

“guestimates” are used in many cells within such a framework, the framework can indicate the

cells that are badly covered by source data but the cells should definitely contain values.

(b) Gross fixed capital formation by products

6.126. It is recommended that regular business surveys should be used as the key source for gross

fixed capital formation by product, in particular since gross fixed capital formation is an erratic

time series and cannot be modelled easily. If possible, these surveys should be linked to those

collecting details on purchases of goods and services for intermediate consumption, in order to

avoid double-counting or missing expenditure. Surveys of the product structure of gross fixed

capital formation may exist but may not cover all industries. As for intermediate consumption, it

may be possible to find detailed information in the annual reports of enterprises.

6.127. Government accounts contain information providing much more detail than a simple

distribution by main types of investment. This information can, including for investment products,

typically appear in an unsystematic form and it will need to be coded by product categories in

SUTs before it can be used.

6.128. For many industries available information on the product dimension of investment is

limited to a few categories or even non-existent. Initial estimates will therefore require some

common-sense decisions.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

190

6.129. In the end, the product structure of gross fixed capital formation will to a large extent be

determined by the availability of investment products within the SUTs framework. In the simplest

case, where other information is unavailable, gross fixed capital formation will alone be

determined by the supply of typical investment products that are largely not used for other

purposes.

6.130. Very few products are used exclusively for gross fixed capital formation. In most cases,

the distribution between intermediate consumption, household final consumption expenditure and

gross fixed capital formation that can be estimated using information from business surveys and

company accounts is uncertain, in particular if the rows of the SUTs represent broad categories of

products. The disaggregation of products can reduce this uncertainty, when the necessary source

data are available. For some products, the distribution may be based on other types of information,

as in the following examples:

• If an official register of motor vehicles is accessible, it may be possible to identify the

changes from year to year in the numbers of different types of cars and lorries by age, size

and ownership. Combined with typical prices for the various groups of vehicles, the register

may be used to estimate household final consumption expenditure together with gross fixed

capital formation by industry of new and used vehicles. Registers may also provide

information to help distribute registration taxes by purpose.

• Similarly, it may be possible to use register information to follow the capital stock, and

change in capital stock for other types of transport equipment.

• Official registers of buildings may also include information on type, size, use and owners

and may be used for the distribution of buildings by purpose. For dwellings and private

commercial buildings that are not completely covered by company accounts or business

surveys, register information can be used in the estimation of the value of investment.

• Information on investment in buildings, transport equipment or specific types of machinery

and investment may be collected by business surveys. Some countries carry out quarterly

and annual business surveys covering gross fixed capital formation and underlying details.

On the other hand, some countries occasionally carry out comprehensive industrial censuses

that may contain information on the use of capital equipment for units that are not covered

by other kinds of statistics.

• Household budget surveys can include information on investment in new dwellings and

capital repairs. In countries with a large informal economy, household budget surveys may

be the most important source for the estimation of investment in buildings, machinery and

equipment in small farms (which may also, however, be included in agricultural censuses),

small retail trade and repair workshops.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

191

6.131. Furthermore, the product structure of investment tends to be more volatile than the

structure of input cost structures. Nevertheless and despite its uncertainties, an initial version of

investment by products estimated from the uses side is usually preferred to an estimated

distribution based solely on the supply of specific products.

2. Changes in inventories

6.132. In order to define changes in inventories, it is useful to define first what is covered by

inventories. Inventories are produced assets, consisting of goods and services, that came into

existence in the current period or in an earlier period and that are held for sale, use in production

or other use at a later date (2008, SNA, para. 10.12). Changes in inventories are measured by the

value of the entries into inventories less the value of withdrawals and less the value of any recurrent

losses of goods held in inventories during the accounting period. Some of these acquisitions and

disposals are attributable to actual purchases or sales but others reflect transactions that are internal

to the enterprise (2008 SNA, para. 10.118).

6.133. The change in inventories column in the use table should be underpinned with a matrix

linking, as column headings, the classification of industries (for example, ISIC), and, as row

headings, the product grouping (for example, CPC) as appearing in the SUTs for each type of

asset. Changes in inventories can be analysed by industry and by types of assets, which need to be

linked via CPC in the SUTs as illustrated in Table 6.12.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

192

Table 6.12 Table linking change in inventories industries, assets and products

6.134. The change in inventories should separately distinguish the following types of assets:

• Materials and supplies, which consist of all products that an enterprise holds in inventory

with the intention of using them as intermediate inputs into production.

• Work-in-progress, which consists of output produced by an enterprise that is not yet

sufficiently processed to be in a state in which it is normally supplied to other institutional

units.

• Finished goods, which consist of goods produced as outputs that their producer does not

intend to process further before supplying them to other institutional units.

Changes in inventories used for correction from sales and purchases to

output and intermediate consumption

Total

PRODUCTS

Agriculture

Forestry

and logging

Fishing and

aquaculture

. . .

Ac tiv ities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 0 2 ISIC 03 . . .

ISIC 97

Agriculture

Forestry

and logging

Fishing and

aquaculture

. . .

Ac tiv ities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 0 2

ISIC 03 . . .

ISIC 97

Agriculture

Forestry

and logging

Fishing and

aquaculture

. . .

Ac tiv ities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 0 2 ISIC 03 . . . ISIC 9 7

Agriculture

Forestry

and logging

Fishing and

aquaculture

. . .

Ac tiv ities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 0 2

ISIC 03 . . . ISIC 97

Agriculture

Forestry

and logging

Fishing and

aquaculture

. . .

Ac tiv ities of

households

employers

Total

PRODUCTS

ISIC 01 ISIC 0 2

ISIC 03 . . . ISIC 97

Changes in inventories, Retail trade

Changes in inventories, Materials and supplies

ASSET

TY PE

Changes in inventories

Finished

goods

Work-in-

progress

Goods for

resale,

w holesale

Goods for

resale, retail

Changes in inventories, Finished goods

Changes in inventories, Work in progress

Changes in inventories, Wholesale trade

Materials

and

supplies

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

193

• Military inventories, which consist of single-use items, such as ammunition, missiles,

rockets, bombs and other items, delivered by weapons or weapons systems. Some single-use

items such as ballistic missiles highly destructive may be classified as fixed assets in the

sense that they provide deterrence services against aggressors.

• Goods for resale, which are goods acquired by enterprises, such as wholesalers or retailers,

for the purpose of reselling them to their customers.

6.135. Although change in inventories for all asset types appear in the final uses part of the use

table, they also play a role across other parts of the supply table and use table:

• For each industry, intermediate consumption can be calculated as purchases of goods and

services less change in inventories of materials, fuels and raw materials.

• For each industry, output can be calculated as sales plus change in inventories of work-in-

progress and finished goods. Producers of services may actually have inventories of work-

in-progress in the form of projects lasting for more than one accounting period, such as films,

advertising campaigns, legal contracts, and so forth.

6.136. The output value in trade can be calculated as sales less purchases less change in

inventories of goods purchased for resale without any further processing. If a distinction between

wholesale and retail trade is made in SUTs, then the calculation of the output value for each of

these industries requires separate values for the change in inventories of goods for resale.

6.137. For estimating output and intermediate consumption, changes in inventories are usually

estimated by industry and institutional sector. These detailed breakdowns are, however, seldom

shown in the use table.

6.138. It must be noted that the changes in inventories reported by the survey respondents should

be adjusted to comply with national accounts definitions: thus the changes in inventories should

not include any holding gains or losses. Source data may be presented as output and intermediate

consumption instead of sales and purchases. In practice, the data collected may nevertheless

contain values of sales and purchases. These data may be replaced by a correction that uses

inventories that appear in the company accounts. If such changes include any holding gains or

losses, the correct treatment is to remove the original correction for inventory changes and replace

it by a correction that uses inventory changes according to national accounts definitions.

6.139. Statistics based on company accounts usually contain information on inventory changes

and also on stocks of inventories. The value shown for change in inventories will usually include

holding gains and losses and will be misleading if significant price changes take place during the

year. A correction for holding gains and losses can usually be based on the nominal values of

opening and closing stocks, or the book value levels. It will require the availability of adequate

information on price changes during the year. The correction can typically be carried out by the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

194

inflation or deflation of opening and closing stocks to the average prices of the year before

calculation of the difference. To calculate constant price values of stock of inventories (and

changes in current prices), these prices values should be broken down by products for which price

indices or volume indicators can be found.

6.140. Annual company accounts may not contain information of opening stocks, in which case

it might be necessary to use values of closing stocks from the previous year as a measure of opening

stocks. As coverage changes and establishments are reclassified, the observed differences between

closing stocks and opening stocks may then need some adjustments at the industry level.

6.141. Given the links between sales, purchases and inventories, along with other variables, it is

recommended that the data on inventories are collected via the same survey questionnaires to

ensure coherence across the variables being collected with the same source and at the same point

in time.

6.142. The Eurostat-OECD Compilation Guide on Inventories (Eurostat and OECD, 2017)

provides more detail relating to compilation issues and guidance.

(a) Estimation of changes in inventories by product

6.143. To fill in the columns for changes in inventories, the totals for this item must be broken

down into changes in inventories for the products used in the SUTs and must be completed as part

of generating the estimates of output and intermediate consumption.

6.144. For some products, the values of changes in inventories may be calculated based on

knowledge of physical opening stocks and closing stocks and information on the development in

prices, for example, for crops and livestock in agriculture and for energy products. The calculation

of changes in inventories should be an integral part of subsystems used to provide the complete

product balances for such products.

6.145. The inventories of most industries include a broad selection of products that are usually

not known from statistical sources. The totals of opening stocks and closing stocks used to

calculate changes in inventories need to be distributed by products based on assumptions on the

product structure for each total. The following are typical examples:

• Inventories of finished goods and work-in-progress can be distributed proportionally with

those outputs of each industry that can appear in inventories of goods. Caution will be needed

on service products.

• Inventories of raw materials and fuels can be distributed proportionally with the use of inputs

of each industry that can appear in inventories.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

195

• Inventories of goods for resale can be distributed using various proxies for the product

structure, for example, output or input in specific industries, household final consumption

expenditure in specific COICOP groups or gross fixed capital formation of specific types.

• Values of inventories of specific products may already be known from other calculations.

Such values can be retained as predetermined values.

• Some products, for instance electricity, are not likely to appear in inventories. Services

should only appear in inventories as work-in-progress.

6.146. Calculating the distribution of changes in inventories by products in this way is of course

uncertain and it may be adjusted during the balancing of SUTs. There is also a danger that errors

in source data will not be detected if changes in inventories are used as balancing items or not

subjected to adequate quality assurance. The total of changes in inventories should, at least, only

be adjusted in exceptional cases, where this is believed to be the most realistic solution to a

balancing problem.

6.147. It should be noted that any balancing adjustment to the asset composition of changes in

inventories, or to the total changes in inventories, will affect different parts of the supply table and

use table and will have the same total impact on each of the production, income and expenditure

approaches to measuring GDP. This is because the various components of changes in inventories

feed into the estimation of output and intermediate consumption and, in turn, GVA and final uses,

thus the impact of any adjustment will be equal on production, income and expenditure. It is more

likely that quality adjustments will be made to source data, where such changes will, and should,

change the intermediate consumption and output and, in turn, GVA of the corresponding

industries.

3. Acquisitions less disposals of valuables

6.148. Valuables are produced goods of considerable value that are not used primarily for

purposes of production or consumption but are held as stores of value over time (2008 SNA, para.

10.13). Valuables include precious metals and stones, antiques and other art objects and other

valuable items. Acquisitions less disposals of valuables capture these alternative forms of

investments. Capital formation in valuables usually needs to be based on the domestic supply of

specific goods, that is, imports less exports plus the margin. Valuables are by nature difficult to

distribute across industries (based on establishments) as they share certain properties with financial

assets, and the industry breakdown does not reflect those valuables held by, for example,

households.

6.149. Trade data by detailed product provide a good source for identifying such items. Whereas

the margin or fee type data by detailed product may be collected via questionnaires sent, for

example, to auctioneers. For SUTs, the product breakdown is key: thus the relevance of imports

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

196

less exports by type of product. In terms of the purchaser, this is less important, as the purchaser

could be industry, government or households.

G. Exports

6.150. Exports are shown in the use table by product. Depending on the specific user’s need, an

additional breakdown by column could be provided to show exports by destination. It should be

noted that the treatment, issues and sources of data applied to imports of goods and services are

also applicable to exports of goods and services. More detail on the imports of goods and services

may be found in chapters 4, 5 and 8, including such issues as the new treatment of goods sent

abroad for processing.

6.151. In the use table, exports are valued FOB at the point of exit from the exporter’s economy.

This value includes the cost of transport from the exporter’s premises to the border of the exporting

economy. FOB price includes:

• Value of goods at basic prices

• Trade and transport services to the border

• Taxes minus subsidies on products; there is no VAT on exports

6.152. Since data on exports on goods are collected on a FOB basis, no further transformation is

needed.

1. Data sources

6.153. Most countries have comprehensive foreign trade statistics for goods. Data are generally

collected according to the Harmonized System, valued at FOB and often available with a high

level of detail, say by six-digit or eight-digit Harmonized System codes. Thus the only adjustment

needed to the basic data is the conversion between Harmonized System codes and CPC. It is

usually possible to convert the data from the Harmonized System classification using a

correspondence table from the United Nations website

(http://unstats.un.org/unsd/class/default.asp), supplemented by a conversion table that defines the

SUTs products as aggregates of CPC classifications.

6.154. Various adjustments would be needed to move the foreign trade statistics on to a balance

of payments basis in line with BPM 6, such as the change in economic ownership and the

difference crossing the border.

6.155. Foreign trade statistics will also usually include the distribution of exports of goods by

countries for all products.

6.156. Enterprises with exports below certain threshold values can be allowed to report their

foreign trade without a distribution by products, for example with survey-based external trade

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

197

statistics. In this case, the values of exports by products will need to be grossed up to cover total

exports. The difference between grossed-up and reported values is uncertain, and may need to be

corrected during the balancing of SUTs. As for inputs, information on the reported values can be

shown together with the grossed-up values in the tables presented to those working on the manual

balancing of SUTs before any automated balancing process.

6.157. The main source for data on exports of services is the balance of payments-based data and

the sources used to produce these data. The classification according to the EBOPS 2010 (United

Nations et al. 2011 and IMF, 2009) will usually provide sufficient detail for conversion into the

classification used in the SUTs. Failing this, access may be available to statistics that show imports

and exports by industries. A conversion into detailed SUTs-based products can be established

based on the coding in balance of payments and the information on the industry classification of

exporting units.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

198

Annex A to chapter 6: Sample questionnaire collecting purchases of goods and

services for intermediate consumption

A6.1 The extract shown in Figure A6.1 is from a business survey questionnaire from the

Statistical Office of Serbia. The data are collected for each industry and by product covering the

following:

• Cost of materials

• Closing stocks of materials and fuels

A6.2 Full coverage of goods and services consumed as intermediate consumption to calculate

the industry totals is achieved via further tables collecting data on the costs of industrial and non-

industrial services, an extract from which is shown in figure A6.2. These data make it possible to

calculate the intermediate consumption by product required to populate the intermediate use part

of the use table, as shown in table 6.1.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

199

Figure A6.1 Extract from questionnaire covering costs and closing stocks of raw materials

and other material inputs

No.

Code

Product description

Cost of

materials

(group of

account 5.1)

Closing stocks

(group of

accounts 10)

3 4

3000

TOTA L

A GRICULTURA L PRODUCTS, RA W A ND UNPROCESSED PRODUCTS OF PLA NT A ND A NIMAL ORIGIN

3001

01.11.1 - 01.11.4 Cereals, all kinds (except rice), cereal seeds

3002

01.11.6 - 01.11.7 Green leguminous vegetables (beans, peas, lentils and other)

3003

01.11.8

Soya beans, groundnuts (row ) and cotton seed

3004

01.11.9

Other oil seeds - sunflow er, sesame, lin etc.

3005

01.11.12 Rice, not husked

3006

01.13 except 01.13.7

Vegetables, raw

3007

01.13.7

Sugar beet and sugar beet seed

3008

01.13.8 Mushrooms and truff les

3009

01.15 Unmanuf actured tobacco

3010

01.16 Fibre crops (lin, cotton and other f ibre crops, used in textile industry)

3011

01.19.1 Forage crops and vegetative matter for livestock feeding unprocessed form

3012

01.19.2

Flow ers and flow er seeds

3013

01.21

Grapes

3014

01.22 - 01.23

Tropical and subtropical fruits (citrus, figs etc.)

3015

01.24, 01.25 except 01.25.3

Other fruits, tree and bush fruits, except nuts (apples, pears, cherries, berries etc.)

3016

01.25.3 Nuts (almonds, hazelnut, w alnuts etc.)

3017

01.26

Olives, coconuts (row , unprocessed)

3018

01.27

Coff ee beans, tea leaves, cocoa beans, not roasted

3019

01.28 Spices, aromatic, drug and pharmaceutical crops

3020

01.11.5, 01.14, 01.19.3,

01.29, 01.3

Vegetables and fruit seeds, other seeds, grass, unprocessed straw and other residues of cereals, seeds

for trees and seedings, planting materials, sugar cane and other raw , unprocessed and untreated products

of plant origin n.e.c.

3021

01.4. except 01.45.3 &

01.49.3

Live animals and raw animal products (unprocessed milk, eggs, natural honey, except raw skins, shorn

w ool and skins, see line 3022, etc.)

3022

01.45.3, 01.49.3

Raw fur skins, shorn w ool, skins (excluding products of slaughterhouses and industrial meat production,

see 1036)

3023

01.49. part

Other animal products, raw , unprocessed an untreated

3024

01.7

Hunting and trapping products, raw

PRODUCTS OF FORESTRY

3025

02.2

Wood in the rough - logs, f uel w ood and other row products of forestry

3026

02.1, 02.3

Forest trees and seeds, w ild grow ing edible products; natural cork, varnish, balsams and other raw

products of f orestry n.e.c.

FISH A ND OTHER FISHING PRODUCTS, UNPROCESSED A ND UNTREATED

3027

Fish and other f ishing products; aquaculture products (raw , unprocessed and untreated)

MINING A ND QUA RRY ING PRODUCTS; UNPROCESSED

3028

05.1, 05.2

Coals, hard coal and lignite

3029

06.1

Crude petroleum, bituminous or oil shale and tar sands. Note petroleum products - fuels are entered in the

row 3118

3030

06.2

Natural gas, processed (Manufactured gas distributed though mains, heating gas and petroleum gases f rom

ref ineries should be reported in row s 3116 and 3119)

3031

07.1

Iron ores

3032

07.2

Other metal ores

3033

08.1 Stone, sand, and clay and other raw materials f or construction, industrial and craft activities

3034

08.9 Other mining and quarrying products n.e.c.

MA NUFA CTURING INDUSTRY PRODUCTS

Food products and other processed products of plant and animal origin; used as reproduced material

3035

10.11 except 10.11.4 &

10.12.5

Meat (red meat, including frozen) except live animals and unprocessed and untreated products of animal

origin (goes to row s 1021 -1024); raw offal and edible fat and oils

:::::::

Electricity, ref ined petroleum products for energy purposes, gas (excluding natural gas), steam, hot w ater,

air conditioning (including energy products use for heating)

Account 513

3115

35.11 Electric ity costs

3116

35.22

Manufactured gas for industrial purposes and for heating - gas distributed through mains (excluding natural

gas, see 3030, petroleum gas f rom ref ineries, see 3119 and industrial and medical gases, see 3069)

3117

35.30 Steam and hot w ater, air conditioning supply services

3118

19.2

Refined petroleum products - motor, engine and other fuels

3119

19.2 Petroleum gases - propane, butane etc. (excluding natural gas and industrial and medical gases)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

200

Figure A6.2: Extract from questionnaire covering costs of industrial and

non-industrial services

No. CPA code Product description

Costs of

services

Account code

1 2 3 4 5

4000 TOTA L

Support services directly linked w ith the production of goods and services

4001

01.6 part Support agricultural services to crop production

4002

01.6 part Support services to animal production; veterinary services excluded (row 4044)

4003

02.10.2, 02.4 Support services to forestry

4004

09 Mining support services; services to petroleum and natural gas extraction

4005

13.3 Textile finishing services–bleaching, dyeing, printing etc.

4006

16.10.9

Drying, impregnation or chemical treatment services of timber and product of w ood; support services in the

processing of w ood and w ood products n.e.c.

4007

25.5 Forging, pressing, stamping and roll-forming services of metal

4008

25.6 Treatment and coating services of metals, machining

4009

24.5 Casting services of metal and steel

Subcontracted services in industry and construction, trade services and other intermediation commissions.

Note: enter only the value of the services, value of materials of goods excluded

4010

14, part Subcontracted operations in textile industry (excluding value of materials)

4011

15, part Subcontracted operations in footw ear and leather production industry (excluding value of materials)

4012

16, part Subcontracted operations in production of processed w ood and w ood products (excluding value of materials)

4013

25, part

Subcontracted operations as part of machine industry–processing and finishing materials services

(excluding value of materials)

4014

41, 42, 43 Subcontracted operations in construction

4015

–

Other subcontracted operations in production of goods of other enterprises (excluding value of materials),

please specify

4016

46.1 Trade commissions

4017

– Other intermediation commissions,

please specify

Transportation costs, postal and courier services

4018

Land transport of freight, taxi operation services including rental services of land transport vehicles w ith

operator

4019

50 Water transport

4020

51 Air transport

4021

52.2

Support services f or transportation (loading, unloading, hauling, tow ing, parking services, etc.,

transportation excluded)

4022

53 Postal services under universal obligation

4023

53 Other postal and courier services

Repair, maintenance, installation services; conversion, reconstruction and fitting out of transport equipment

4024

33.1

Repair and maintenance services of f abricated metal products, machinery and equipment, except motor

vehicles

4025

43 Repair and maintenance services of buildings and electrical, plumbing, heating and similar installations

4026

45.2 Maintenance and repair services of motor vehicles

4027

95.1 Maintenance and repair services of computers and communication equipment

4028

95.2 Repair services of personal and household goods

4029

33.2 Installation services of industrial machinery and equipment

4030

29.20.4, 29.20.5

Reconditioning, assembly, f itting out and bodyw ork services of motor vehicles, except installation,

maintenance and repair services

4031

30.11.9, 30.20.9, 30.30.6

Conversion, reconstruction and fitting out services of other transport equipment, except installation,

maintenance and repair services

Rentals, rents on land, w arehousing and storage services

4032

68.2, part Rental costs on buildings and off ice space ow ned by legal persons (except rents on land)

4033

68.2, part Rental costs on buildings and off ice space ow ned by natural persons (except rents on land)

4034

68.2, part Rents on private land

4035

68.2, part Rents on public or state land

4036

Rental and leasing services of motor vehicles, machinery, equipment and tangible goods (excluding real

estate and financial lease)

:::::::

Expenditure related to the intellectual property (royalties, licence f ees, rights of usage, publication,

reproduction, transmission, broadcasting and the like), other services n.e.c. Note: services as

expenditures; if not capitalised

4089

58,1 Royalties for the publication of books, magazines, new spapers, etc.

4090

58,2 Royalties and similar payments for usage of softw are

4091

59 Royalties for publishing (music and movies, TV series)

4092

60 Royalties (broadcast rights, etc.) in the production and broadcast of radio and television programs

4093

71,2 Certification of products and processes

4094

77,4 Royalties and fees for the use of intellectual property if it is not capitalised

4095

96,01 Laundry and w ashing of textile and fur

4096

– Other industrial and non-industrial services not elsew here specified

please specify

539 or 559

532

533

530

530, 539

531

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

201

Annex B to chapter 6: Impact of capitalizing the costs of research and

development in SUTs and IOTs

A. Research and development as fixed capital formation

B6.1 The 2008 SNA introduced changes relating to the treatment of research and development.

Research and development is creative work undertaken on a systematic basis in order to increase

the stock of knowledge, including knowledge of people, culture and society, and to enable this

stock of knowledge to be used to devise new applications. The 2008 SNA does not treat research

and development activity as an ancillary activity and it recommends that a separate establishment

should be distinguished for research and development when possible.

B6.2 The output of research and development should be capitalized as “intellectual property

products” except where it is clear that the activity does not entail any economic benefit to its

producer (and hence owner), in which case it is treated as intermediate consumption.

B6.3 The 2008 SNA now includes expenditures for both bought-in and own-account research

and development as gross fixed capital formation and the depreciation of these assets as

consumption of fixed capital.

B6.4 Table B6.1 shows a summary of the impact of these changes using a simplified hypothetical

example simply to demonstrate the impact of capitalization. The example is simplified in the sense

that it shows only two years and the capitalization of research and development (R&D) is assumed

to be introduced in year 1, when there is no increase in the consumption of fixed capital. In reality,

the capitalization costs will be spread over several successive years and the consumption of fixed

capital will occur in year 1 and throughout the lifetime of the asset.

B6.5 Under 1993 SNA, own-account R&D activity was treated as an ancillary activity and no

separate output was estimated in the system, and expenditures for this purpose were not separately

identified. Only in those cases, usually unimportant, where R&D services were purchased from

outside specialist producers, classified in the research and development activity (ISIC Rev. 4,

division 72) or imported, do R&D services appear in the SUTs as intermediate consumption.

B6.6 With the capitalization of R&D expenditures under the 2008 SNA, the output of own-

account R&D is separately estimated and allocated to GCF or exports. In practice, the introduction

of own-account R&D output in the system has just resulted in additional output with the

intermediate inputs being left unaffected, as the intermediate inputs needed to produce the own-

account R&D were already included. With the 2008 SNA change of treating research and

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

202

development not as an ancillary activity, the question for the compilation of SUTs and IOTs is

whether the R&D output from own-account R&D activity should be seen as a secondary product

from that particular branch or the principal product of the activity “Scientific research and

development” ISIC Rev. 4, division 72. This issue of reclassification of own-account R&D activity

is considered in the following section.

B6.7 The challenging methodological and practical problems related to the actual estimates of

output of own-account research and development are not being dealt with at this point; reference

is made to the Manual on Measuring Research and Development in ESA 2010 (Eurostat, 2014a),

the Handbook on Deriving Capital Measures of Intellectual Property Products (OECD, 2010) and

specific country documentations. In the following, only some problems of special interest for the

compilation of SUTs and IOTs will be highlighted.

Table B6.1 Summary of the impact of the capitalization of research and development in the

new 2008 SNA

For own account value of production is 1000. For non-market producers, consumption of fixed capital in

Year 2 is 200.

For bought-in, value of intermediate consumption is 500. For non-market producers, consumption of fixed

capital in Year 2 is 100.

Year 1 estimates. Estimates in brackets to relate to consumption of fixed capital to impact in Year 2 for

non-market producers.

B. Implications of valuation of output as sum of costs

B6.8 Output for own final use should be valued at the basic prices at which the goods and

services could be sold if offered for sale on the market. When reliable market prices cannot be

obtained, a second best procedure must be used, in which the value of the output of the goods or

services produced for own final use is deemed to be equal to the sum of their costs of production,

that is, as the sum of: intermediate consumption; compensation of employees; consumption of

fixed capital; a net return to fixed capital; and other taxes (less subsidies) on production. By

Ow n

account

Bought-in

(1) (2)

Intermediate consumption 0 -500 0 -500

Non-market consumption of f ixed capital 0 0 (+200) (+100)

Gross operating surplus +1 000 +500 (+200) (+100)

GVA +1 000 +500 (+200) (+100)

Total output +1 000 0 (+200) -500 (+100)

Output f or ow n fnal use +1 000 0 +1 000 0

General governemnt or NPISHs final

consumtpion expenditures

0 0 -1 000 (+200) -500 (+100)

Gross fixed capital formation +1 000 +500 +1 000 *500

Impact on GDP +1 000 +500 0 (+200) 0 (+100)

(3)

(4)

Indicative impact

Change in the treatment of Research and Development

Market producers

Non-market producers

Ow n account

Bought in

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

203

convention, no net return to capital is included when own-account production is undertaken by

non-market producers (2008 SNA, para. 6.125).

B6.9 The calculation of the output of own-account R&D from the sum of costs approach means

that the cost structure for this particular type of output will be separately specified; this is a

departure from the usual situation where intermediate and primary inputs used for various types of

outputs will be indistinguishably lumped together. Based on the known cost structure, it would in

principle be possible to create separate establishments for the own-account R&D activity, and if

further on this activity was seen as secondary, to reclassify these establishments from the original

activity to which they are classified, for example, pharmaceuticals, electronics and so forth, to the

specialist activity for scientific research and development (ISIC Rev. 4, division 72).

B6.10 For a number of analytical, methodological and practical reasons, however, such a

reclassification is not recommended. These reasons include, primarily, the following:

• There is an analytical interest in keeping track of those economic activities that actively

involve R&D and this information would be lost through such a reclassification.

• As each R&D output is uniquely defined, the own-account research and development is

usually not suitable for delivery outside the producing unit, and a reclassification would not

reflect the economic reality of the activity.

• Given the lack of any other information, it is assumed that the producing unit is also the

owner of the resulting R&D capital stock and the associated consumption of fixed capital –

this assumption is often not valid.

• The cost structures calculated to derive the estimate of R&D output will not usually have

product details corresponding to the SUTs product requirement, and in general, will only

exist as internal worksheet exercises not intended for a wider audience.

• If all R&D were reclassified to ISIC Rev. 4, division 72, in many developed countries this

division would increase to the same size or even larger than the agricultural sector, and

seriously distort the relative proportions, in particular between manufacturing industries and

service industries. It would also make industries active in R&D (such as the pharmaceutical

industry) rather meaningless truncated residuals compared with the usual perception of the

size and structure of these industries and thus the reclassifications would be

counterproductive from the point of view of users’ needs and a wide-range of analytical

purposes.

C. Own-account research and development as principal or secondary output

B6.11 Research and development services, division 81 of CPC Version 2.1, also existed prior to

the capitalization of R&D, and were mainly made up of the services actually sold in the market by

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

204

enterprises classified in division 72 of ISIC Rev. 4, “Scientific research and development”. This

was generally small, however, when compared with the total value of own-account R&D estimated

in connection with the capitalization.

B6.12 In division 81 of CPC Version 2.1, the subclasses are organized according to the type of

research (for example, chemistry, biotechnology, and others), and not according to the economic

activities carrying out the R&D, and all R&D services are indicated as characteristic products of

ISIC Rev. 4 subclasses 7210 and 7220. This follows logically from the fact that CPC Version 2.1

was not designed for a situation where the overwhelming share of R&D services comes into

existence as estimated own-account output in ISIC industries other than ISC Rev. 4, division 72.

B6.13 Under the system of estimated own-account output of R&D services, it would be more

appropriate to introduce a CPC structure for R&D similar to the structures for “Maintenance, repair

and installation (except construction) services” (CPC Version 2.1, division 87) and for

“Manufacturing services on physical goods owned by others” (CPC Version 2.1, division 88)

where the subclasses (four digits) are made up industry specific outputs, each corresponding to a

characteristic ISIC class (four digits). This means, in particular, that outputs of these services in

industries which are active in R&D form principal outputs. When this approach is followed for

own-account R&D services, there will be as many subclasses of CPC division 81 as there are

industries with R&D activities, and own-account R&D will formally change from a secondary to

a principal activity of the producing industries. The adoption of this approach will also have

important implications when deriving IOTs from the SUTs, as it will prevent major structural

differences between industry-by-industry IOTs and product-by-product IOTs, and avoid truncating

research and development-intensive product-adjusted industries in the product-by-product IOTs.

B6.14 In practice, specialized R&D departments of enterprises with major own-account R&D

activities may, for various reasons (legal, tax-related, and other), have already been classified in

ISIC Rev. 4, division 72, in the business register, and thus be included in business statistics with

this activity, rather than the principal activity (for example, pharmaceutical, electronic and so

forth) of the parent enterprise.

B6.15 In such cases, the flows between the R&D department classified in ISIC Rev. 4, division

72, and the parent enterprise, and also the applied valuation principles, should be carefully

assessed. In this connection, it is important to realize that business accounting practices will

usually not follow the principle of capitalizing R&D expenditures. Thus the total output (however

estimated) from an R&D department classified in ISIC Rev. 4, division 72, may, in the business

accounts, reappear as intermediate consumption in the accounts of the parent enterprise.

Depending on the circumstances, one solution might be to reclassify the R&D department back to

the activity of the parent enterprise. In national accounts, sometimes legal structures may be

overruled if they are found not to reflect economic realities. Alternatively, the intermediate

consumption of R&D services could simply be removed from the parent enterprise and instead

treated as gross fixed capital formation, but this may leave a truncated enterprise of little analytical

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

205

interest, as noted above, and it would still be necessary to deal with the valuation of reported output

of the R&D department.

D. Balancing supply and use of research and development services

B6.16 Assuming that the output of R&D services (market, non-market and for own-use) by

industry is available from the current national accounts calculations, allocation by user should, in

principle, be fairly straightforward, as these services under the 2008 SNA treatment of R&D

should be allocated to gross fixed capital formation. There are, however, two problem areas worth

mentioning:

• Some research and development purchases are still to be treated as intermediate

consumption.

• Foreign trade in research and development services must be taken into account when

balancing the R&D services.

B6.17 When market R&D services are purchased by an own-account producer of R&D from a

commercial R&D producer (usually, although not necessarily, classified in ISIC Rev. 4, division

72) or imported, it must be decided whether this is an acquisition of an asset or an intermediate

product used as an input into the own-account production of R&D. As there is usually insufficient

information available on the individual transactions for an informed decision to be made, it is

recommended by the Eurostat Manual that, in the absence of any strong evidence to the contrary,

all purchases by own-account R&D producers from units classified in ISIC Rev. 4, division 72,

(and also purchases by other units in ISIC Rev. 4, division 72) should be treated as intermediate

consumption. This assumption will also ease the distribution in cases where no product statistics

for the output from ISIC Rev. 4, division 72, are available, as some output may be non-research

and development services that would nonetheless usually be allocated to intermediate

consumption.

B6.18 The above-mentioned case involving a specialized R&D department classified in division

72 of ISIC Rev. 4 may, however, interfere with this solution, and it is recommended that the SUTs

compilers coordinate closely with the compilers of the R&D estimates made for the national

accounts.

B6.19 In the 2010 Manual on Statistics of International Trade in Services (United Nations, IMF,

OECD, European Union, UNCTAD, UNWTO and WTO, 2011) and Extended Balance of

Payments Services Classification (EBOPS, 2010), there are special entries on R&D services,

which are explicitly separated from transactions on the results of R&D (for example, royalties and

license fees paid for use of patented entities).

B6.20 It would therefore appear that the balancing of the R&D services, also taking into account

imports and exports, would be straightforward. The transactions registered in the balance of

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

206

payments are, however, actual economic transactions where the prices may be quite different from

the cost-based valuation of the domestic own-account output of R&D services, and the delimitation

of the R&D concept may also deviate. Further balance of payments transactions in R&D may

include significant elements of transfer pricing and trade with subsidiaries in low-tax jurisdictions.

When the aim is to achieve consistency with existing balance of payments data, the final balancing

of the R&D services may be quite difficult, even though the capitalized R&D services allows trade

in “used” R&D so that gross fixed capital formation may in principle (though no very realistically)

become negative. More detail on these issues may be found in chapter 7 of the guidance on the

impact of globalization on national accounts prepared by the Economic Commission for Europe

(UNECE, 2011).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

207

Chapter 7. Compiling the valuation matrices

A. Introduction

7.1. The compilation of valuation matrices is a fundamental step in the process of compiling

SUTs. These matrices are necessary to bridge the different valuation concepts of the product flows.

This chapter covers the main concepts and methods of compiling matrices for trade margins,

transport margins, taxes on products and subsidies on products. In particular, in section B, the

chapter starts with an overview of the valuation concepts in the 2008 SNA and of how the valuation

matrices fit within the SUTs presented in chapters 5 and 6. Sections C–E elaborate on each

component of the valuation matrices and describe the main compilation steps. The annex to this

chapter provides further details on how to compile trade margins for the SUTs from survey data

based on a country practice.

B. Valuation of product flows

7.2. Transactions are valued at the actual prices agreed upon by the purchasers and sellers.

Market prices are thus the main reference for the valuation of transactions in the SUTs system, in

line with 2008 SNA. In the absence of market transactions, the valuation is made according to

costs incurred (for example, for non-market services produced by government) or by reference to

market prices for analogous goods and services (for example, for services of owner-occupied

dwellings).

1. Valuation concepts in the 2008 SNA

7.3. More than one set of prices may be used to value outputs and inputs depending on how

taxes and subsidies on products, trade and transport margins are recorded. The 2008 SNA

distinguishes three main valuation concepts of the flows of goods and services: the two main

recommended valuations being basic prices and purchasers’ prices and the lesser used producers’

prices.

7.4. The basic price is the amount receivable by the producer from the purchaser for a unit of a

good or service produced as output minus any tax payable, and plus any subsidy receivable, by the

producer as a consequence of its production or sale. It excludes any transport charges invoiced

separately by the producer (2008 SNA, para. 6.51).

7.5. The producers’ price is the amount receivable by the producer from the purchaser for a unit

of a good or service produced as output minus any VAT, or similar deductible tax, invoiced to the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

208

purchaser. It excludes any transport charges invoiced separately by the producer (2008 SNA, para.

6.51).

7.6. The purchasers’ price is the amount paid by the purchaser, excluding any VAT or similar

tax deductible by the purchaser, in order to take delivery of a unit of a good or service at the time

and place required by the purchaser. The purchasers’ price of a good includes any transport charges

paid separately by the purchaser to take delivery at the required time and place (2008 SNA, para.

6.64).

7.7. The difference between these valuation concepts for a product relates to trade margins,

transport margins and taxes on products and subsidies on products. The relationship between the

three types of prices is as follows:

Basic prices + Taxes on products excluding invoiced VAT

- Subsidies on products

= Producers’ prices

+ Wholesalers’ trade margins

+ Retailers’ trade margins

+ Separately invoiced transport charges

+ VAT not deductible by the purchaser

= Purchasers’ prices

7.8. The basic price measures the amount retained by the producer and, therefore, the price most

relevant for the producers’ decision-making and is often reported in business surveys. For imported

products, taxes on products include import duties. When the relationship between basic prices and

purchasers’ prices is compiled for the total economy, the transport charges and trade margins will

cancel out because they form only a reallocation of value across products.

7.9. The concept of producers’ prices does not form any of the main valuations. The preferred

valuation of output and GVA in the SNA is made at basic prices and for intermediate consumption

at purchasers’ prices. It is worth recognizing that source data from business surveys for sales may

be valued at producers’ prices. In these cases, data should be adjusted to a basic price valuation

before they are entered into the SUTs. If this step is not completed, then a different recording of

taxes on products and subsidies on products must be established, and GVA by economic activity

would be partly at market prices, a practice which is not recommended by the SNA.

7.10. It is important to note that the relationship between basic price and purchasers’ price does

not describe a process over time for an identifiable product. In this case, the difference between

basic prices and purchasers’ prices is likely to contain an element of holding gains and losses while

the product is with the producer and with wholesale and retail traders (2008 SNA, para. 3.148).

The SNA value concepts are consistently defined in such a way that holding gains and losses do

not become part of GVA and GDP. Hence a trade margin is relative to the replace price of the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

209

good “at the time it is sold”, and the price of intermediate consumption relates to what the producer

would have to pay to replace it “at the time it is used”.

7.11. The source data used to fill the cells in the SUTs may have different valuations:

• Production and output data are valued at basic prices.

• Intermediate consumption and final uses are valued at purchasers’ prices.

• Imports are valued at CIF prices. This is the valuation of a good delivered at the point

of entry into the importing economy or the price of a service delivered to a resident,

before the payment of any import duties or other taxes on imports or trade and transport

margins within the country. In the SUTs framework, for national SUTs, the CIF value

is taken to be the basic price of imports of goods.

• Exports are valued at FOB prices. This is the valuation of a good at the point of exit

from the exporter’s economy or the price of a service delivered to a non-resident,

including transport charges and trade margins up to the point of the border, and

including any taxes less subsidies on the goods exported. In the SUTs framework, the

FOB value is taken to be the purchasers’ price of exports of goods.

7.12. It is recommended that the different valuation components of the product flows are

separated to ensure that the SUTs framework is balanced in a fully coherent and consistent manner.

One of the purposes of the valuation matrices is to bridge the difference between the valuation at

basic prices and the valuation at purchasers’ prices and to come up with SUTs at basic prices.

Figure 7.1 illustrates how the valuation matrices link the supply table with the use table. They

comprise all flows that are related to the supply and use of trade margins and transport margins

and to taxes and subsidies on products.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

210

Figure 7.1 Schematic representation of the valuation matrices in the SUTs

7.13. In the supply table, the valuation matrices consist of columns (the “bridge column” in

Figure 7.1) which transform the supply of each product from basic prices to purchasers’ prices,

and, in turn, match the product values in the use table compiled at purchasers’ prices.

7.14. In the use table, the valuation matrices consist of product-by-industry matrices of trade

margins, transport margins, taxes on products and subsidies on products which allow for the

transformation of the values of the use table from purchasers’ prices into basic prices. The

availability of such matrices makes possible the balancing of SUTs at basic prices and purchasers’

prices and, as recommended in the Handbook, both valuations should be balanced simultaneously.

7.15. Although it is not strictly necessary that a balanced SUTs framework ends up with tables

showing valuation both at purchasers’ and at basic prices, this is recommended for several reasons.

For analytical purposes, the SUTs data must have the same valuation, and usually the basic price

version is the most appropriate. This is also the case for the process of transforming the SUTs into

IOTs and for the volume estimates in a consistent SUTs framework leading to the estimation of

GVA in volume terms, using double deflation.

7.16. For these input-output based analytical purposes, a valuation as uniform as possible of the

cells in a row of the use table is essential. The values at purchasers’ prices in the different uses will

usually be affected by differences in trade and transport margins, and by differences in taxes on

products and subsidies on products, according to the specific user. The uniformity requirement is

therefore best fulfilled by values at basic prices, although the cells valued at basic prices may still

show user specific price variation: this is the most uniform valuation concept that in practice can

be achieved.

Final use

Purchasers' prices

Non-deductible VAT

Other net taxes on products

Trade and transport margins

Use of industries

Basic prices

Total use

Exports

Changes in inventories

Changes in valuables

Purchasers' prices

Gross fixed capital formation

Non-deductible VAT

Final consumption expenditure of government

Other net taxes on products

Final consumption expenditure of NPISH

Trade and transport margins

Final consumption expenditure of households

Industries

Intermediate consumption

Basic prices

Total use

Output

USE

SUPPLY

Products

Total supply

Imports

Bridge

column

for

valuation

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

211

2. Valuation matrices in the SUTs framework

7.17. The valuation matrices comprise information on trade margins, transport margins, taxes on

products and subsidies on products. Valuation matrices can be established for the supply table

(supply-side valuation matrices) and the use table (use-side valuation matrices). In a balanced

SUTs system, the use-side valuation matrices and the supply-side valuation matrices will add up

to the same totals. In this section, the full set of valuation matrices is described.

(a) Supply-side valuation matrices

7.18. The supply-side valuation matrix consists of a set of columns added to the supply table at

basic prices to derive the supply at purchasers’ prices. These columns comprise trade margins,

transport margins, VAT, taxes on products and subsidies on products. Table 7.1 shows the

structure of the supply table at basic prices, including a transformation into purchasers’ prices. The

table corresponds to table 5.2 of chapter 5 and is reproduced here for ease of reference. The left

part of this table starts with the domestic output of the various industries by products at basic

prices. The inclusion of the imports valued at CIF prices by products generates the total supply by

products at basic prices, as shown in column (9).

7.19. In the supply table, the output at basic prices of trade services (of which trade margins form

the major part) is included in row (4) and that of transport services in row (5). To arrive at

purchasers’ prices for each product, the trade margin and transport margin shares of this output

have to be reallocated from trade margins and transport services to the traded and transported

products. Columns (10) and (11) of Table 7.1 contain the allocation of trade margins and transport

margins respectively, with positive entries (+) in the rows of the traded and transported products

and negative entries (-) in the rows of trade services and transport services. The totals by row of

columns (10) and (11) of trade and transport margins respectively are always zero.

7.20. The columns of taxes less subsidies of products – columns (12)–(14) of Table 7.1 – are

also added to total supply at basic prices in order to arrive at the total supply at purchasers’ prices.

Taxes on products comprise VAT-type taxes, taxes and duties on imports and other taxes on

products. Similarly, subsidies on products comprise import subsidies and other subsidies on

products. Taxes and subsidies on products should be compiled separately although they may be

shown as a single column.

7.21. The addition of columns (10)–(14) to total supply at basic prices of column (9) gives total

supply at purchasers’ prices in column (16). Columns (10) and (14) thus provide the bridge needed

to compare and balance total supply with total use when both sides are valued at purchasers’ prices.

7.22. Both trade margins and transport margins can be produced by any industry outside the trade

and transport industries. The bulk of the output of trade margins, however, is generally produced

by the trade industries and the bulk of transport margins by the transport industries, as illustrated

in rows (4) and (5) of Table 7.1.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

212

Table 7.1 Supply table at basic prices, including a transformation into purchasers’ prices

Millions of euros

Table based on 2011 figures from Austria

(b) Use-side valuation matrices

7.23. The use-side valuation matrices consist of a sequence of matrices – mirroring the shape of

the intermediate use and final use parts of the use table – for each component of the valuation,

namely, for trade margins, transport margins, taxes on products and subsidies on products.

7.24. Table 7.2 illustrates the use table at purchasers’ prices. This table corresponds to table 6.1

of chapter 6 and is reproduced here for ease of reference. It shows the structure of the table,

comprising the following three sub-matrices:

Agriculture Manufacturing Construction

Trade,

transport and

communication

Finance and

business

services

Other

services

(1)

(2) (3)

(4)

(5) (6)

(7) (8) (9)

Agriculture

(1) 8 782

0 0 0 0 0 8 782

3 271

12 052

Manufacturing (2) 796 182 982 643 1 808

133

44 186 405 124 590

310 995

Construction (3) 83

961 43 060 734 255 179

45 272 563 45 835

Trade (4) 1 4 773 311 54 204 640 257 60 187

600 60 787

Transport (5) 13 465 66 25 538 128 125 26 335 8 150 34 485

Communication (6) 160 1 781 139 43 912 1 253 982 48 228 6 234 54 463

Finance and business services (7) 29 8 902 698 7 588 106 909 3 381 127 508 7 061 134 569

Other services (8) 3 85 13 1 053 143 74 346 75 643 824 76 467

Total (9) 9 867 199 950 44 931 134 837 109 461 79 314 578 360 151 293 729 653

CIF/FOB adjustments on imports (10) 0

0 0

0 - 97 - 97

Direct purchases abroad by residents (11) 0 0 0 0 0 0 0 6 675 6 675

Total (12)

9 867 199 950 44 931 134 837

109 461 79 314 578 360

157 871 736 230

Total of w hich:

Market output (13) 9 763

195 916 41 462 127 401 88 330 18 116

480 989 0

Output for ow n final use (14)

104 4 029

3 468 2 134 19 890

2 670 32 295

Non-market output (15) 0 4 0 5 302

1 241 58 528 65 075

Trade margins

Transport

margins

VAT

Taxes on

products

Subsidies

on products

Total

(9) (10) (11)

(12) (13) (14) (15) (16)

Agriculture (1)

12 052 1 926 274 329

57 - 107 2 479

14 532

Manufacturing (2) 310 995 48 838 2 540 13 175

7 866 - 49 72 370 383 364

Construction (3) 45 835

0 0

1 529

13 0 1 542 47 377

Trade (4)

60 787 - 52 341 0

575 11 0

- 51 755 9 032

Transport (5) 34 485

0 - 2 800 558 71 - 448 - 2 620

31 865

Communication

(6) 54 463 1 493 9 3 375 217 - 34 5 059 59 522

Finance and business services (7) 134 569 0 - 22 2 706 2 159 0 4 842 139 411

Other services (8) 76 467 85 0 1 201 576 0 1 861 78 329

Total (9) 729 653 0 0 23 447 10 969 - 638 33 778 763 431

CIF/FOB adjustments on imports (10) - 97

0 0 0 0

0 - 97 - 97

Direct purchases abroad by residents (11) 6 675 0 0 0 0 0 6 675

6 675

Total (12)

736 230 0 0 23 447

10 969 - 638 40 356 770 009

Total of w hich:

Market output (13)

0 0 0

0 0

Output for ow n final use (14) 0 0 0 0 0

0 0 0

Non-market output (15)

0 0 0 0 0 0 0 0

Adjustments

INDUSTRIES

Output

at basic

prices

Imports

Total supply

at basic

prices

PRODUCTS

Total supply

at basic

prices

V A LUA TION MA TRICES

Total supply

purchasers'

prices

PRODUCTS

Adjustments

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

213

• Intermediate consumption matrix showing intermediate consumption for each industry by

product

• Final uses matrix showing final uses by type of final use and by product

• GVA matrix showing the components of GVA for each industry

Table 7.2 Use table at purchasers’ prices

Millions of euros

Table based on 2011 figures from Austria

7.25. The matrices covering intermediate consumption and final uses are valued at purchasers’

prices, thus they include trade margins and transport margins, along with taxes on products less

subsidies on products. Accordingly, the sum of total intermediate consumption at purchasers’

prices (column (7) of Table 7.2) and total final use (column (15) of Table 7.2) gives the total use

by product at purchasers’ prices, which, in a balanced system, is equal to the total supply by

products at purchasers’ prices in column (16) of the supply table in Table 7.1.

7.26. Table 7.3 shows the use-side valuation matrices covering trade margins (split between

wholesale and retail trade margins), transport margin, VAT, taxes on products and subsidies on

products. These matrices have the same structure and dimension as the intermediate consumption

and final uses sub-matrices of the use table at purchasers’ prices. They show the allocation of the

trade margins, transport margins, taxes and subsidies on products to each element of the use table

at purchasers’ prices. They represent the amounts that must be deducted from each element of the

use table at the purchasers’ price in order to arrive at the use table at basic prices.

Households NPISH

General

government

(1) (2)

(3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture

(1) 2 583 6 570 16 371 34 49 9 623 3 595 0 0 180 0 - 27 1 161 4 909 14 532

Manufacturing (2) 2 205 107 190 12 441 16 874 6 015 8 797 153 522 71 438 0 3 180 26 756 2 183 3 034 123 252 229 842 383 364

Construction (3) 105 2 440 9 528 2 446 3 907 1 604 20 029 1 667 0 0 25 155 0 - 38 563 27 348 47 377

Trade (4) 33 1 883 119 2 240 259 308 4 842 3 325 0 0 67 45 0 753 4 189 9 032

Transport (5) 14 4 386 267 8 399 822 321 14 208 5 833 0 3 370 0 0 0 8 453 17 656 31 865

Communication (6) 34 2 563 299 9 359 5 919 1 833 20 008 26 444 0 121 5 976 0 67 6 905 39 514 59 522

Finance and business

services

(7) 457 13 578 4 736 20 359 29 166 9 134 77 430 38 838 0 1 006 11 170 0 - 178 11 145 61 981 139 411

Other services (8) 8 382 59 1 171 415 1 794 3 829 14 923 5 416 53 373 113 107 1 567 74 500 78 329

Total at purchasers’ prices

before adjustments

(9)

5 440

138 991 27 466 61 219 46 538 23 839 303 492 166 063 5 416 61 050 69 418 2 335 2 859 152 800 459 939 763 431

CIF/FOB adjustments on

exports

(10)

0 0 0 0 0 0 0 0 0 0 0 0 - 97 - 97 - 97

Direct purchases abroad by

residents

(11)

0 0 0 0 0 0 0 6 675 0 0 0 0 0 0 6 675 6 675

Purchases in the domestic

territory by non-residents

(12)

0 0 0 0 0 0 0 - 12 945 0 0 0 0 0 12 945 0 0

Total at purchasers’ prices (13) 5 440 138 991 27 466 61 219 46 538 23 839 303 492 159 792 5 416 61 050 69 418 2 335 2 859 165 648 466 517 770 009

Compensation of employees (14)

551 30 679 10 239 37 906 22 997 41 971 144 343 0 0 0 0 0 0 0 0 0

Other taxes less subsidies

on production

(15) - 1 627 1 077 546 1 755 2 004 1 103 4 858 0 0 0 0 0 0 0 0 0

Consumption of fixed capital (16) 1 845 12 750 1 542 10 917 18 934 7 480 53 469 0 0 0 0 0 0 0 0 0

Net operating surplus/Net

mixed income

(17) 3 658 16 453 5 138 23 040 18 989 4 921 72 198 0 0 0 0 0 0 0 0 0

Gross operating

surplus/gross mixed income

(18) 5 503 29 203 6 680 33 957 37 923 12 401 125 667 0 0 0 0 0 0 0 0 0

GVA (19) 4 427 60 959 17 465 73 618 62 923 55 475 274 868 0 0 0 0 0 0 0 0 0

Total input at basic prices (20) 9 867 199 950 44 931 134 837 109 461 79 314 578 360 0 0 0 0 0 0 0 0 0

PRODUCTS

Adjustments

GVA

Final consumption expenditure

Gross f ixed

capital

formation

Total use at

purchasers’

prices

Agricul-

ture

Manufac-

turing

Construc-

tion

Trade,

transport and

communication

Finance and

business

services

Other

services

Total

Changes in

valuables

Changes in

inventories

Exports

Total

INDUSTRIES

FINA L USE

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

214

Table 7.3 Use-side valuation matrices

Table based on 2011 figures from Austria

7.27. It should be noted that the use-side valuation matrices in Table 7.3 relate to the supply-side

valuation matrices in Table 7.1 as follows:

Households NPISH

General

government

(1)

(2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture (1) 31 440 2 63 4 6 547 326 10 4 165 506 1 052

Manufacturing (2) 164 6 450 1 415 1 879 326 1 049 11 284 5 941 560 2 718 15 265 8 995 18 493 29 777

Construction (3)

Trade (4) - 196 - 6 910 - 1 420 - 1 997 - 367 - 1 082 - 11 972 - 6 464 - 569 - 2 776 - 15 - 273 - 9 232 - 19 329 - 31 301

Transport (5)

Communication (6) 0 20 3 55 37 27 141 197 9 48 4 72 330 472

Finance and business services (7)

Other services (8)

Total (9) 0 0 0 0 0 0 0 0 0 0 0 0 0

Agriculture (1) 3 0 13 0 1 17 856 856 873

Manufacturing (2) 19 56 20 339 38 90 562 16 781 477 1 142 99 18 499 19 061

Construction (3)

Trade (4) - 19 - 86 - 25 - 431 - 104 - 150 - 815 - 18 363 - 511 - 1 168 - 184 - 20 226 - 21 040

Transport (5)

Communication (6) 0 27 5 79 66 59 236 725 34 26 785 1 021

Finance and business services (7)

Other services (8) 85 85 85

Total (9) 0 0 0 0 0 0 0 0 0 0

Agriculture (1) 7 201 1 7 0 1 217 44 2 1 11 57 274

Manufacturing (2) 20 1 125 191 127 30 64 1 557 303 21 144 2 26 486 983 2 540

Construction (3)

Trade (4)

Transport (5) - 27 - 1 321 - 191 - 135 - 31 - 65 - 1 771 - 347 - 21 - 146 - 2 - 27 - 487 - 1 030 - 2 800

Communication (6) 0 0 1 1 0 4 3 0 2 0 1 5 9

Finance and business services (7) 0 - 5 - 1 - 1 0 0 - 7 - 3 0 - 2 0 0 - 10 - 15 - 22

Other services (8)

Total (9) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Agriculture (1) 0 0 0 0 0 3 4 324 1 325 329

Manufacturing (2) 18 71 18 68 200 942 1 317 10 624 368 734 132 11 858 13 175

Construction (3) 0 1 7 4 242 185 439 265 825 1 090 1 529

Trade (4) 3 16 4 20 17 39 100 467 2 7 476 575

Transport (5) 0 1 0 9 8 17 36 487 34 521 558

Communication (6) 0 1 0 16 207 150 374 2 888 10 103 3 001 3 375

Finance and business services (7) 0 3 3 28 561 713 1 308 1 235 163 1 398 2 706

Other services (8) 0 0 0 1 17 44 63 921 209 8 1 138 1 201

Total (9) 22 93 32 147 1 252 2 093 3 639 17 210 621 1 830 147 19 807 23 447

Agriculture (1)

0 50 0 1 0 0 51 5 1 0 0 6 57

Manufacturing (2) 62 834 179 1 068 212 544 2 898 4 272 6 284 5 7 393 4 968 7 866

Construction (3) 0 1 3 1 1 0 5 0 7 8 13

Trade (4) 0 5 0 4 0 0 10 1 0 0 1 11

Transport (5) 0 7 1 8 11 3 30 26 0 14 40 71

Communication (6) 0 6 1 14 47 5 72 130 0 0 0 15 145 217

Finance and business services (7) 9 88 18 142 172 37 467 936 755 1 691 2 159

Other services (8) 0 0 0 0 0 0 1 574 1 0 575 576

Total (9) 71 991 202 1 238 443 590 3 535 5 944 8 1 048 6 7 422 7 434 10 969

Agriculture (1) 0 - 89 0 0 0 0 - 89 - 2 - 5 0 - 11 - 18 - 107

Manufacturing (2) 0 - 16 - 2 - 3 - 1 - 1 - 24 - 9 0 - 2 0 0 - 14 - 25 - 49

Construction (3)

Trade (4)

Transport (5) 0 - 26 - 2 - 33 - 6 - 10 - 77 - 300 - 71 - 1 - 371 - 448

Communication (6) 0 0 - 34 - 34 - 34

Finance and business services (7)

Other services (8)

Total (9) - 1 - 131 - 5 - 36 - 6 - 11 - 190 - 344 - 71 - 7 0 0 - 25 - 448 - 638

Agriculture (1) 38 644 3 83 4 8 781 1 226 12 5 176 1 419 2 200

Manufacturing (2) 203 7 631 1 626 2 345 393 1 204 13 403 23 025 1 058 4 004 117 291 9 481 37 975 51 378

Construction (3)

Trade (4) - 215 - 6 996 - 1 445 - 2 428 - 471 - 1 233 - 12 786 - 24 827 - 1 080 - 3 944 - 199 - 273 - 9 232 - 39 555 - 52 341

Transport (5) - 27 - 1 321 - 191 - 135 - 31 - 65 - 1 771 - 347 - 21 - 146 - 2 - 27 - 487 - 1 030 - 2 800

Communication (6) 1 47 8 136 104 86 381 925 44 75 4 72 1 121 1 501

Finance and business services (7) 0 - 5 - 1 - 1 0 0 - 7 - 3 0 - 2 0 0 - 10 - 15 - 22

Other services (8) 85 85 85

Total (9) 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Agriculture (1) 0 - 38 0 1 0 3 - 34 327 - 3 0 - 10 313 279

Manufacturing (2) 80 888 194 1 133 411 1 485 4 191 14 888 374 1 016 137 7 379 16 800 20 992

Construction (3) 0 1 10 5 243 185 444 265 833 1 098 1 542

Trade (4) 3 21 4 24 17 40 109 468 2 7 477 586

Transport (5) 0 - 18 - 1 - 15 13

10 - 11 214 - 36 14 191 180

Communication (6) 0 6 1 30 253 155 446 2 984 10 104 0 15 3 112 3 558

Finance and business services (7) 9 91 21 171 733 750 1 775 2 171 918 3 090 4 865

Other services (8) 0 0 0 2 17 44 63 1 495 210 8 1 713 1 777

Total (9) 92 952 229 1 349 1 689 2 672 6 984 22 810 557 2 870 152 7 397 26 794 33 778

Transport margins

Value added tax (VAT)

Taxes on products (excl. VAT)

Subsidies

Trade and transport margins

PRODUCTS

Other

services

Finance and

business

services

Total

Trade,

transport and

communication

Construc-

tion

Manufac-

turing

Agr ic ul-

ture

PRODUCTS

Taxes less subsidies on products

Final consumption expenditure

Exports

Changes

inventories

Changes

valuables

Gross f ixed

capital

formation

Total use at

purchasers’

prices

Total

INDUSTRIES

FINA L USE

Wholesale trade margins

Retail trade margins

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

215

• The sum of the totals in column (16) in Table 7.3 for “Wholesale trade margins” and “Retail

trade margins” must be equal to “Trade margins” in the supply-side valuation matrix, namely

column (10) of Table 7.1.

• The total in column (16) of Table 7.3 for “Transport margins” must be equal to the

“Transport margins” in the supply-side valuation matrix, namely column (11) of Table 7.1.

• The total in column (16) of Table 7.3 for “Value added tax (VAT)” must be equal to the

“Value added tax” in the supply-side valuation matrix, namely column (12) of Table 7.1.

• The total in column (16) of Table 7.3 for “Taxes on products (excl. VAT)” must be equal to

the “Taxes on products” in the supply-side valuation matrix, namely column (13) of Table

7.1.

• The total in column (16) of Table 7.3 for “Subsidies” must be equal to the “Subsidies on

products” in the supply-side valuation matrix, namely column (14) of Table 7.1.

7.28. In general, the information needed to construct the trade and transport margins matrices of

Table 7.3 is available only to a limited extent, and some balancing between the supply and use of

trade margins and transport margins is necessary. In addition, depending on available data and

whether a benchmark or a current SUT is being compiled, it must be assessed whether it is best to

start from the supply or the use side when estimating total trade margins and transport margins by

products. In the cases when trade and transport margins are estimated first from the supply side,

they will serve as a constraint when allocating the supply of trade margins and transport margins

to the various use categories.

7.29. There is one type of tax on product, namely VAT, for which it is not possible to start with

the supply-side estimates. In the VAT system as set out in the SNA, only non-deductible VAT is

recorded as a tax on product, and there is no means by which the actual VAT payers (VAT

collectors) can obtain information about the final users and their ability to deduct VAT or not. The

structure of VAT by products has therefore to be estimated from the use-side by identifying all

user categories not exempted from the VAT system and the appropriate effective tax rate has to be

applied to all their purchases of products. One increasingly problematic area is the treatment of

digital intermediation platforms in terms both of trade and transport margins and also of the taxes

(in particular, VAT) paid to foreign governments.

7.30. Once all the matrices in Table 7.1 are compiled, the next step is to deduct the trade margins,

transport margins and taxes less subsidies on products from the use table at purchasers’ prices to

arrive at the use table at basic prices as shown in Table 7.4. In addition, it is necessary to reallocate

the deducted trade margins and transport margins to the specific trade and transport service

products distinguished in the product classification applied, and the taxes on products to a separate

row. After these steps, the use table is transformed into a valuation at basic prices as shown in

Table 7.4.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

216

7.31. It should be noted that all the entries from rows (1)–(9) in Table 7.4 are at basic prices.

Row (10) of Table 7.4 contains the net taxes on products which bridge the total use at basic prices

with the total use at purchasers’ prices, which is shown in row (11) of Table 7.4. This latter

coincides with row (9) of the use table at purchasers’ prices in Table 7.2.

Table 7.4 Use table at basic prices

Millions of euros

Table based on 2011 figures from Austria

C. Trade margins

7.32. This section deals with the compilation of the valuation matrix for trade margins. The

amounts involved can vary significantly by type of product and can be of great magnitude in total.

In most countries, the trade margins may vary between 10 and 25 per cent of the total domestic

supply of goods and services.

7.33. The data sources needed for compiling the trade margin estimates by product for the SUTs

are special in the sense that they are not necessary when compiling the current national accounts

GVA by industry and, for that reason, may not be readily available on an annual basis.

7.34. Whereas information on trade turnover and purchases of goods for resale by industry is

usually included in the general business statistics or estimated in order to compile the annual GVA

by industry, the situation is different for information on trade turnover and margins by product.

These data are generally not as readily available for a number of reasons: most often they are

collected at intervals of several years, if at all; or information on trade margin ratios may be

available only for a limited number of products and provided by government agencies dealing with

Households NPISH

General

government

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture (1) 2 545 5 964 13 287 29 38 8 877 2 042 0 0 170 0 - 32 996 3 176 12 052

Manufacturing (2) 1 922 98 670 10 621 13 397 5 211 6 108 135 928 33 525 0 1 749 21 736 1 929 2 737 113 392 175 067 310 995

Construction (3) 105 2 439 9 518 2 441 3 664 1 419 19 585 1 402 0 0 24 323 0 - 38 563 26 250 45 835

Trade (4) 245 8 857 1 560 4 644 712 1 501 17 519 27 684 0 1 080 4 008 238 273 9 985 43 267 60 787

Transport (5) 41 5 724 459 8 549 840 376 15 990 5 967 0 3 427 146 2 27 8 926 18 495 34 485

Communication (6) 33 2 510 290 9 194 5 562 1 592 19 181 22 535 0 68 5 797 0 63 6 818 35 281 54 463

Finance and business services (7) 448 13 492 4 716 20 189 28 433 8 384 75 662 36 669 0 1 006 10 254 0 - 177 11 156 58 907 134 569

Other services (8) 8 381 59 1 169 398 1 750 3 765 13 429 5 416 53 163 113 14 1 567 72 702 76 467

Total at basic prices (9) 5 348 138 038 27 236 59 870 44 849 21 167 296 507 143 252 5 416 60 492 66 548 2 182 2 852 152 403 433 145 729 653

Taxes less subsidies on

products

(10) 92 952 229 1 349 1 689 2 672 6 984 22 810 557 2 870 152 7 397 26 794 33 778

Total at purchasers’ prices

before adjustments

(11) 5 440 138 991 27 466 61 219 46 538 23 839 303 492 166 063 5 416 61 050 69 418 2 335 2 859 152 800 459 939 763 431

CIF/FOB adjustments on

exports

(12) 0 0 0 0 0 0 0 0 0 0 0 0 0 - 97 - 97 - 97

Direct purchases abroad by

residents

(13) 0 0 0 0 0 0 0 6 675 0 0 0 0 0 0 6 675 6 675

Purchases in the domestic

territory by non-residents

(14) 0 0 0 0 0 0 0 - 12 945 0 0 0 0 0 12 945 0 0

Total at purchasers’ prices (15) 5 440 138 991 27 466 61 219 46 538 23 839 303 492 159 792 5 416 61 050 69 418 2 335 2 859 165 648 466 517 770 009

Compensation of employees (16) 551 30 679 10 239 37 906 22 997 41 971 144 343 0 0 0 0 0 0 0 0 0

Other taxes less subsidies on

production

(17) - 1 627 1 077 546 1 755 2 004 1 103 4 858 0 0 0 0 0 0 0 0 0

Consumption of fixed capital (18) 1 845 12 750 1 542 10 917 18 934 7 480 53 469 0 0 0 0 0 0 0 0 0

Net operating surplus/Net

mixed income

(19) 3 658 16 453 5 138 23 040 18 989 4 921 72 198 0 0 0 0 0 0 0 0 0

Gross operating surplus/Gross

mixed income

(20) 5 503 29 203 6 680 33 957 37 923 12 401 125 667 0 0 0 0 0 0 0 0 0

GVA (21) 4 427 60 959 17 465 73 618 62 923 55 475 274 868 0 0 0 0 0 0 0 0 0

Total input at basic prices (22) 9 867 199 950 44 931 134 837 109 461 79 314 578 360 0 0 0 0 0 0 0 0 0

Total use

at basic

prices

Agricul-

ture

Manufac-

turing

Construc-

tion

Trade,

transport and

communication

Finance and

business

services

Other

services

Total

Changes in

valuables

Changes in

inventories

Exports

Total

INDUSTRIES

FINA L USE

PRODUCTS

Adjustments

GVA

Final consumption expenditure

Gross f ixed

capital

formation

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

217

price control or monopoly surveillance; or they are limited to what can be derived from current

price statistics by, for example, comparing wholesale prices collected for the wholesale price index

(WPI) with the prices collected for the consumer price index (CPI).

7.35. It is recommended that a special survey be conducted, in particular for the purpose of

compiling benchmark SUTs, that covers all trading activity (both as primary and secondary output)

by industry and by product at a level sufficiently detailed to match the level and classification of

products applied in the SUTs. The annex to chapter 7 provides an example of a questionnaire used

for a survey of this kind, along with a template for the optimal use of the collected data in compiling

the SUTs trade margin matrices.

7.36. Even when less than ideal source data are available, it is still necessary to estimate the trade

margin matrices. In these cases, however, the results will become more dependent on the

assumptions made to populate the trade margin columns in the supply table and the trade margin

matrices in the use table. In principle, the same type of trade margin data table (shown in Table

7.5 and in the annex to this chapter) must be generated, irrespective of the coverage or quality of

primary source data on trade and trade margins.

1. Definition of trade margins

7.37. Even though wholesalers and retailers actually buy and sell goods, the goods purchased are

not treated as part of their intermediate consumption as they are resold with only minimal

processing such as grading, cleaning and packaging. Wholesalers and retailers are treated as

supplying services. Their output is measured by the total value of the trade margins realized on the

goods they purchase for resale and some non-margin trade services. This notwithstanding, actual

trade turnover is an important supporting variable when compiling the trade margin matrices in

the SUTs.

7.38. The 2008 SNA provides the following definition of a trade margin:

“A trade margin is defined as the difference between the actual or imputed price realized

on a good purchased for resale and the price that would have to be paid by the distributor

to replace the good at the time it is sold or otherwise disposed of” (2008 SNA, para. 6.146).

7.39. With regard to valuation, the SNA goes on to state as follows:

“Goods purchased for resale should be valued excluding any transport charges invoiced

separately by the suppliers or paid to third parties by wholesalers or retailers: these

transport services form part of the intermediate consumption of the wholesalers or

retailers” (2008 SNA, para. 6.148).

7.40. This valuation principle means that there can be no transport margins linked to purchases

of goods for resale. This follows from the fact that, in the national accounts definition of output

from trade activity, the traded goods are not seen as actually flowing in and out of the trade activity.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

218

The trade activity only adds services to the goods that are seen as flowing directly from the

producer or importer to the user. As a result, there is no flow to which a transport margin could be

attached.

7.41. In practice, the output of wholesalers and retailers is derived as the difference between the

trading sales and the costs of goods purchased for resale, adjusted by changes in inventories (see

2008 SNA, para. 6.147):

Value of output = value of sales

+ value of goods purchased for resale but used for intermediate

consumption, compensation of employees, etc.

- value of goods purchased for resale

+ value of additions to inventories of goods for resale

- value of goods withdrawn from inventories of goods for resale

- value of recurrent losses due to normal rates of wastage, theft or

accidental damage

7.42. In order to derive trade margins, either for trading activities of single goods, trading

activities of a statistical unit industries or total economy, data on trading sales (trade turnover),

data on goods purchased for resale without further processing, and data on inventories of goods

for resale at the beginning and at the end of the period must be available. As a rule, business

surveys or specialized trade surveys can collect and provide data at the level of trade industries.

Trading is also an important secondary activity in many non-trade industries, and trading activities

in the system are measured by trade margins, regardless of whether the activity is conducted by

traders as their main activity or by other industries as part of their secondary outputs.

7.43. Even though distributive trade is defined as purchases of goods for resale without any

transformation, certain operations usually associated with distribution are included in the

definition, such as sorting, mixing, breaking bulk and repackaging for distribution into smaller

lots. Other services may also be included, if not separately invoiced, such as installation in situ.

7.44. Trade services should be separated at least into two main categories: wholesale and retail.

Wholesale is the resale (sale without transformation) of new and used goods to retailers, industrial,

commercial, institutional or professional users or to other wholesalers and export. Retailing is the

resale (sale without transformation) of new and used goods, mainly to the general public for

personal or household consumption or non-resident visitors or for a minor part to business

(intermediate consumption and fixed capital formation). This separation is essential for the correct

and transparent estimation of the trade channels and the cumulative trade margins for products

passing through both the wholesale and the retail trade links.

7.45. The services provided by the trade industry include both margin and non-margin services.

Margins services are those related to the trade activity of resale. Non-margin services are other

services provided by such trade establishments as repair and installation services. Some trade

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

219

margins may, however, best be estimated and treated as non-margin services, in other words,

should not be calculated as a percentage of the basic value of the underlying goods. This is the

case when the underlying goods do not appear in the system, such as used household goods, in

particular cars; or when the value of the underlying goods makes up a very small and fluctuating

share of the total selling price like scrap materials; and when the underlying goods by convention

are not specified by type (as is the case for merchanting, where the goods involved are not covered

by the merchandise trade statistics but only appear as a net item in the balance of payments – with

acquisitions shown as negative exports, as in SNA 14.73 and 26.21).

7.46. It should be noted that merchanting, and thus trading output, also includes the case of so-

called factoryless goods production, where a principal has completely outsourced the

transformation process but does not own the input materials (see para. 5.81 of the Guide to

Measuring Global Production (UNECE, 2015), although, in paragraphs 2.69-113, an alternative

view is explored that treats factoryless goods production units as manufacturers). The treatment as

non-margin services means that this part of trading output will not be shown in the trade margin

matrices. Where the above-mentioned cases are concerned, this will facilitate the use of the trade

margin matrices in volume estimates and analytical uses of the resulting IOTs. With the CIF

valuation of imports, there are no imports of trade margins.

7.47. Trade activity is classified in the industrial classification ISIC Rev. 4 under division 45,

“Wholesale and retail trade and repair of motor vehicles and motorcycles”, division 46,

“Wholesale trade, except of motor vehicles and motorcycles”, and division 47, “Retail trade,

except of motor vehicles and motorcycles”. The following should be noted:

• Division 45 is a mixture of both wholesale and retail trade activities, and also of non-trade

activities. In the context of SUTs and IOTs, division 45 must be subdivided into repair

activities (group 452), and trade activities (groups 451 and 453).

• Group 454 “Sale, maintenance and repair of motorcycles and related parts and accessories”

has to be dealt with appropriately, depending on the significance of this activity in the

country. This is a heterogeneous group and if the GVA contribution is significant, then the

sale, maintenance and repair activities should be separated, as the respective margin ratios

are very different.

• Division 46 also covers wholesale on a fee or contract basis which is not part of the trade

margin activity. It also covers some merchanting activity.

• Division 47 includes the resale (sale without transformation) of new and used goods, mainly

to the general public for personal or household consumption or use, by shops, department

stores, stalls, mail-order houses, hawkers and peddlers, consumer cooperatives and other

outlets.

7.48. Trade services are classified in CPC Version 2.1 in division 61, “Wholesale trade services”,

and division 62, “Retail trade services”. It should be noted that, even though many industries have

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

220

output of trade services as either principal or secondary products, the product classification used

in the system may distinguish only a few groups such as wholesale and retail trade services.

2. Compilation of trade margin matrices

7.49. The compilation of the trade margin matrices in the SUTs may in principle be started either

from the supply side or from the use side. In general, especially when compiling a benchmark

SUT, the preferred approach is to start with estimating trade margins in the supply side resulting

in data on the total amount of trade margins by products. These estimates will then represent the

restrictions with which the use-side trade margin estimates have to comply.

7.50. Business surveys or special trade surveys usually provide data on the total output of trade

margins by industries (including subdivisions of the trade industries), which then need to be

transformed into margins by products if special surveys do not provide this information.

7.51. Starting the compilation of trade margins from the use-side means that estimates are made

on the share of trade margins included in each element of the use table at purchasers’ prices. The

effective trade margin associated with each cell depends both on the typical trade channels for this

particular use and on the typical product margin ratios for those parts passing through the

wholesale and retail links. Normally, such information cannot be gathered by asking purchasing

enterprises and establishments and other users, as they will only be aware of the last step in the

distribution channel (where they have purchased the product) but clearly not of the previous steps.

Even for the last step, they do not know the margin implicitly invoiced to them. Thus, plausible

assumptions both on distributive channels and margin ratios have to be made and, for this purpose,

it is useful to have the supply-side trade margin estimates at hand.

7.52. In the case of benchmarked estimates, it is necessary to calculate the full range of supply-

side and use-side trade margin matrices by exploiting all available data sources and thus also to

establish a basis for subsequent calculation, where the trade margin ratios of the previous years

can be taken as the point of departure, and depending on available data, it may in this case be best

to start from the use-side, as this represent the most detailed set of trade margin ratios.

7.53. The compilation of trade margins can be organized in three steps. The first two steps can

be seen as dealing with estimating the supply-side trade margins by product, while the third step

deals with the estimation of the use-side trade margin matrix. Box 7.1 gives a general summary of

the three compilation steps for trade margins.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

221

Box 7.1 Compilation process for trade margins

Supply-

side

trade

margins

Step 1

Estimation of turnover and output (trade margins) by principal

and secondary trade margin producers.

Absolute constraint for totals by

producing industries

Output estimates entered into the

domestic output part of the supply

table

Step 2

(a) Estimation of trade turnover matrices (dimension: product

by margin producing industry).

(b) Estimation of product specific trade margin ratios.

product.

(d) Balance step 2 (c) against step 1.

Relative constraints for margins

by product.

Result entered into the trade

margin columns of the supply

table

Use-

side

trade

margins

Step 3

The starting point is step 2 (d) and the use table, either

balanced or unbalanced, and either at purchasers’ prices or at

basic prices.

Main objective:

If use table at purchasers’ prices: To compile the full matrices

for trade margins to derive basic price values in each cell, and

the corresponding rows for the margin producing industries.

If use table preliminary estimated at basic prices: To compile

the full matrices of trade margins in order to adjust the

preliminary estimates of the rows for margin producing

industries (often the case in an already established SUTs

system)

Estimated trade margin matrices are eventually balanced

against the constraint in step 1 (absolute constraint) and in step

2 (d) (relative constraint)

Common problems:

- Question of distribution

channels.

- No direct information of

trade margin ratios by user

category.

- Extensive use of assumptions

necessary.

(a) Step 1

7.54. Step 1 of the compilation process for trade margins covers the compilation of trade services

by industry that must already be part of the current annual national accounts calculations of GVA

by industry. The national accounts estimates of output usually do not include a distinction between

primary and secondary output. In the case of trade output, however, the situation is different, since

trade output must be derived separately as the difference between sales and purchase of goods for

resale. The data needed to fill in the domestic production part of the supply table should therefore

be readily available, and these calculations are not unique to the SUTs.

7.55. When compiling the SUTs, and in particular benchmark SUTs, there may be a need to take

a closer look at existing calculations and to supplement them with more detail, such as, for

example, through the introduction of the distinction between wholesale and retail output, if it has

not already been done.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

222

7.56. Trading services are an important secondary output in many industries other than trade,

such as manufacturing enterprises, barber shops, museums, hotels, recreational and sporting

activities, and so forth. Accordingly, as a first step, it should be assessed if current data fully cover

all secondary output of trade activity, as some data sources such as industrial statistics may exclude

trading, whereas data sources for service industries may not specify trade turnover separately, and

some existing estimates may need adjustments. Thus, for certain industries only total sales may be

known, without any distinction between trade sales and other sales. Estimates based on plausible

assumptions should be made to achieve data on total trade services of the economy.

7.57. Trade turnover should be separated into wholesale trade turnover and retail sale trade

turnover, and wholesale trade margins and retail trade margins recorded as different products in

the domestic output part of the supply table at basic prices. In business statistics, wholesale and

retail trade turnover and margins are often available for the trade industries separately (ISIC Rev.

4, divisions 45, 46 and 47). For industries with trade activity as a secondary activity this breakdown

may not be available and, even in cases where trade turnover has been reported separately for

wholesale and for retail sales, the value of goods purchased for resale may not be thus subdivided.

In this case, wholesale and retail trade margins cannot be derived directly but have to be estimated

on the basis of plausible assumptions.

7.58. If no direct separate information on the type of trade (wholesale or retail) is available from

surveys, the subdivision can be based on the kind of primary economic activity. Thus all ISIC Rev.

4, division 46, may be assumed to carry out wholesale trade, and all ISIC Rev. 4, division 47, retail

trade, whereas ISIC Rev. 4, division 45, must be further broken down, as indicated above. The

subdivision of trade for secondary trade producers into wholesale and retail trade has in this case

to be based on assumptions. For example, it can be assumed that the trade turnover of restaurants

and hotels, hairdressers, cinemas and theatres will probably be retail trade turnover, whereas trade

activities of advertising agents will more likely be wholesale trade. Manufacturing industries will

often trade in products similar to those that they produce or in complementary products and such

sales will usually be of a wholesale type, although some may be sold directly to consumers.

Manufacturing industries may also be trading in similar imported goods, and such trade is again,

likely to be classified as wholesale trade. There may also be industries where a grouping by size

might be relevant for the correct identification of the type of trade activity performed, for example,

the trade activity of small bakeries would normally be retail sale, whereas the trade activities of

the larger ones would probably be wholesale trade.

7.59. For the purpose of the following calculation steps, these estimates should be made also at

the most detailed level of classification of the trade industry available in source statistics. Although

these more detailed estimates would not be shown in the supply table at basic prices, they will be

very useful when estimating trade margins by products in cases where this information is not

available from existing surveys.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

223

7.60. In the discussion above, it has been assumed that the secondary output of trade services

remains in the industries of the secondary producers. As explained in chapter 5, however, the SNA

recommends the partitioning of horizontally-integrated enterprises that have production in two or

more sections of the ISIC Rev. 4 (these sections are broad activity groups such as agriculture and

fishing, mining, manufacturing, construction, trade and others) and the creation of new

establishments to be classified together with the primary producers of the secondary product if that

has not already been done in basic statistics. Such reclassification of secondary output is called

“redefinition” and is typically carried out for trade activities in many countries. A redefinition

implies that there will be trade activity and output of trade margins only from divisions 45–47 of

ISIC Rev. 4. This will significantly simplify the calculations and will also facilitate the estimates

of input structures and the calculations of industry-by-industry IOTs. It will not, however, affect

the basic methodology outlined in this section and in the annex to this chapter.

7.61. Step 1 results in an estimate of total output of trade services forming an absolute constraint.

This value is then disaggregated by products in step 2, as described below.

(b) Step 2

7.62. In step 2, the product dimension is in focus, in particular in the allocation of total wholesale

and retail trade margins to the products to which the margins apply.

7.63. The output of both wholesale and retail trade services must be first separated into the output

of margin activities (trade margins) and the output of non-margin activities (non-margin trade

services), such as trade services related to used goods, waste and scrap, and to merchanting. The

output of non-margin trade services do not form part of the valuation matrices (see para. 7.45).

7.64. Table 7.5 illustrates how the trade data needed in the supply table are related to the survey

data (or primary data obtained in other ways, where necessary by relying on plausible estimates).

The yellow shaded cells are data that will feed into the rows for wholesale and retail trade services

in the domestic output part of the supply table. They are obtained as explained under step 1 above.

Step 2 deals with the problem of distributing the values in the yellow shaded cells to the various

products, whereupon the grey shaded column for the total trade margins will form the trade margin

columns of the supply table (column (10) of Table 7.1). It should be noted that the trade turnover

data appear as supporting variables only; they are not moved forward to the supply table.

7.65. In the situation where data on margins by product are already available, either from current

business surveys or from special surveys conducted in connection with the SUTs, the task is limited

to aligning the survey results with the product classification used in the SUTs, and to grossing up

the results to make them consistent with the output data for trade dealt with in step 1 above.

7.66. In the case where no trade data by product are available, it is necessary to subdivide the

margin trade turnover by each industry into turnover and trade margins by the products traded.

This should be done separately for wholesale trade turnover and retail trade turnover.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

224

Table 7.5 Trade turnover and trade margins for wholesale and retail trade margins

7.67. The subdivision by products of trade turnover of each trade service-producing industry will

result in the two trade turnover matrices, each of which as the dimensions: “Products (traded) by

producing industry (output)”: one matrix for wholesale trade turnover and one matrix for retail

trade turnover. These matrices are presented together in Table 7.5. The availability of source data

for this subdivision may vary considerably across countries. Even in cases where no direct survey

information is available, scattered information may be available, including from administrative

sources and related and older surveys, although ad hoc information on specific units may not be

representative for the total branch. In this case, it is of particular importance to use data from the

most detailed level of subdivisions of the trade industry, as these will indicate the types of products

traded.

7.68. Even with the availability of data for subdivisions of the trade industries, the estimation of

trade turnover by products is not straightforward. This is the case, for example, with non-

specialized trade divisions such as supermarkets and department stores where a wide range of

goods are traded (even though, in some cases, it may be possible to get access to computerized

cash transactions with detailed information about the goods sold and purchased data from

businesses). It is generally easier to make estimates for specialized retail divisions, as their

turnover by product is known from everyday life and more uniform. Estimating turnover by

product for the wholesale divisions is more difficult because of the range of product mix, although

there are branches with a clear concentration on one or a few product groups such as, for example,

wholesale trade of motor vehicles and energy products.

7.69. For trade as secondary activity, plausible assumptions must be made about the products

traded. For example, for hairdresser trade in cosmetic articles, hotel trade in souvenirs,

newspapers, journals, food and beverages, and museum trade in books, multimedia products, and

so on, the share of each identified product group in turnover must also be determined. In

manufacturing, it can be difficult to estimate trade turnover structures by goods traded without any

specific information, as the level of specialization may be very high. Information could be obtained

on ad hoc basis by asking selected units with significant trade turnovers. Making these estimates

Acti vi ty IS IC

. . . . . .

Trade turn-

over

Trade

margin

Trade turn-

over

Trade

margin

. . .

Trade turn-

over

Trade

margin

Trade turn-

over

Trade

margin

Trade turn-

over

Trade

margin

. . .

Trade turn-

over

Trade

margin

Product CPC

. . . . . .

Total wholesale

Total retail

SUM

Trade turn-

over

Trade

margin

Indicate W or R

Wholesale (W)

Retail [R]

Indicate W or R

Ec. activity 1

Ec. activity 2

Trade 45

Trade 46

Trade 47

Ec. activity n

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

225

on the basis of plausible assumptions at the highest level of detail may provide acceptable results

for the SUTs aggregates, even if only rudimentary information is available.

7.70. Having compiled the two trade turnover matrices in Table 7.5, it is possible to check the

wholesale and retail trade turnover against the supply of the goods (domestic production and

imports) from the supply table. At this stage, the comparison cannot be done at completely

comparable prices, as the supply will be at basic prices and the trade turnover will be inclusive of

either wholesale or retail trade margins, or both, but such checks should ensure that the trade

turnover estimates are plausible in relation to the supply of the goods. For instance, there should

not be much retail trade turnover of intermediate and capital goods; wholesale trade turnover

should normally not be much higher than domestic production plus imports (plus wholesale

margins); retail trade turnover of consumer goods should not be much higher than household

expenditure for these goods. There may, however, be exceptions to these general rules for certain

products, such as those that are traded twice within the same chain of distribution. This arises, for

example, when one wholesaler imports a product or purchases it from many small producers (such

as agriculture) and subsequently resells it to another wholesaler.

7.71. From the two trade turnover matrices, the trade margin matrices of the same dimensions

must be derived. This is formally done by multiplying the trade turnover matrix by the assumed

product margin ratios as described below.

7.72. The margin ratios are defined here as the share of a trade margin relative to the trade

turnover. Margin ratios can be defined at the level of industries, which would show the average

margin obtained by margin producing industry (the necessary information is already available

from the Step 1 calculations), or at the level of products, where information is usually less readily

available, although some countries may conduct regular surveys on percentage trade margins by

products, classified by CPC or COICOP. It must in general be assumed that margin ratios are more

closely connected with the products traded than with the industry carrying out the trading activity

as either primary or secondary production.

7.73. It is obvious that even a single benchmark survey of product-specific trade margins would

contribute greatly to the overall quality of the SUTs.

7.74. It is important to be aware of the basis for the calculation of percentage trade margins.

When reported by enterprises, the trade margins will often appear as a percentage of the total

selling price, including excise taxes and VAT, whereas the compiler of the SUTs will usually need

the trade margin as either a percentage of the basic price (when estimating margins in the supply

table) or a percentage of the purchasers’ price excluding taxes on products (when estimating

margins in the use table). The source data on percentage trade margins must therefore be adjusted

to the appropriate basis before being applied in the system.

7.75. When specific survey information is missing, alternative sources for product-specific

margin ratios must be explored. One possible approach might be to compare the prices observed

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

226

for the CPI and for the wholesale price index (WPI) for identical products. The same could be

achieved by comparing PPIs with WPIs, which could provide proxies for wholesale margin ratios.

In the case of regulated prices, the price levels in the different distribution channels may be

available, and more generally price information available from the monopoly and price control

agencies could be used. The margin ratios of specialized trade branches may be used as proxies

for the related product margin ratios. Thus the margin ratio of the retail trade branch selling shoes

could be taken as the typical retail margin for shoes. In practice, the usefulness of this approach

would depend on the level of detail in the product classification applied, and the availability of

data by detailed sub-branches of wholesale and retail activity, which would facilitate linking

branches and products.

7.76. Having established a set of product-specific margin ratios, the multiplication of the trade

turnover matrix could then be performed on the assumption that these product-specific margin

ratios are valid in all industries trading in that product (as primary and as secondary). Next the

resulting wholesale and retail trade margins by producing industries must be compared with the

total trade margins by industries determined in step 1. The reasons for any differences could be

inaccuracies in the trade turnover matrices, in the subdivision between wholesale and retail

margins, and in the assumed or derived product margin ratios. These differences must be

eliminated either by proportional adjustments or, if appropriate, by more refined methods.

7.77. It should be noted that difficulties in determining trade margins are attributable not only to

the often weak data sources but also to the continuing changes in the structure of the trade

industries, such as the following:

• Changing forms of supply of trade services

• Concentration in retail trade branches

• Growing size of shops

• Increasing importance of internet trade

7.78. These developments also affect the validity of benchmark estimates that may relatively

quickly become outdated.

7.79. Step 3 relates to the calculation of the use-side trade margins matrices. In the previous

steps, the question of trade channels has not been dealt with, as the data sources have been either

survey data for the trading activities or estimates based on administrative or other indirect sources.

When compiling the use table trade margin matrices, however, trade channels become important,

as individual users may purchase their goods from different levels of the distribution system, or

even directly from producers.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

227

7.80. Direct data sources are more limited for the use table margins than for the supply table

based margins. This is because buyers of the goods do not know what share the trade margins

constitute in the price that they have paid. Sometimes they do not even know whether they have

purchased the good from a trader or not. In cases where the goods have been purchased in a retail

shop or from a wholesaler, the buyer will only know that the price paid includes some trade margin

but not the full amount of the margin. This is because the distribution channels before the final

seller are usually unknown to the buyer.

7.81. Figure 7.2 illustrates in a schematic way the possible distribution channels for goods from

the producer of market output and imports of goods to the user. The distribution can go directly

from the producer to the user (represented by box A); through wholesale trade (box B) in which

case on wholesale margins are applicable; through wholesale and retail trade (box C) in which case

both on wholesale and retail margins are applicable; or only though retail trade (box D), in which

case only retail margins are applicable.

Figure 7.2 Alternative distribution channels of goods

7.82. The calculation of the use table trade margin matrices must therefore be based on plausible

assumptions and eventually balanced with the estimated total supply of the trade margins by

products. In principle, the following types of information are necessary:

• For each single cell of the use table, the share of total purchases that has been channelled

through trade activities (for all involved steps in the distributive channel)

• The margin ratios to be applied for the products actually traded in the particular intermediate

or final use part of the use table

7.83. Usually specific knowledge about the distributive channels for the goods in the individual

cells is missing. The same holds true for the possible variation of the actual margin ratio across

Channel A: From the producer directly to the user (without margins)

Channel B: From the producer to the user through a wholesale trader (wholesale margins)

Channel C: From the producer to the user through a wholesale trader and a retail trader (wholesale and retail margins)

Channel D: From the producer to the user through a retail trader (retail margins)

Market output and imports of goods

Wholesale trade

Retail trade

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

228

users. For that reason, plausible assumptions have to be made. Concerning the use of wholesale

and retail trade channels, the following assumptions may reasonably be made:

• For intermediate consumption, mostly wholesale trade margins and in very few cases retail

trade margins are relevant. Retail trade margins for intermediate consumption can be

relevant, for example, when stationery and handicrafts materials are being bought by smaller

shops and small-scale enterprises.

• For household final consumption expenditure, retail trade margins are mostly relevant, but

also with some exceptions when consumers have direct access to wholesale channels , thus

generating wholesale trade margins, or are able to buy directly from the producer of the

goods (for example, farmers, bakeries, tailors and so forth), thus not involving trading

services at all.

• Wholesale services are also connected with household final consumption expenditure as

some of the products bought in retail trade may have been delivered from wholesalers, and

thus also include wholesale trade margins.

• For gross fixed capital formation, the wholesale channel is the most important and a very

small measure of importance also attaches to the retail sale channel, for example, valuables

and smaller equipment.

• In changes in inventories, it may reasonably be assumed that only wholesale margins and

not retail trade margins can be involved (although the involvement of the latter is

theoretically possible). Furthermore, wholesale trade margins can only be allocated to input

stocks and trading stocks but not to output stocks, finished products and work-in-progress.

• For exports, it may reasonably be assumed that only wholesale margins may be involved

(allowing for retail trade margins allocated to non-residents expenditure).

7.84. For much intermediate consumption, fixed capital information in equipment and

machinery, and exports it is possible that no trade margins are involved at all, as bigger enterprises

in particular will deal directly with one another. Imported goods are more likely to be bought

through wholesalers than domestically produced goods, again depending on the size of the

enterprises. Small enterprises will be more inclined to buy certain goods through retail traders.

7.85. The allocation of the trade margins to the individual cells of the use table has to be done in

a step-wise manner. VAT and other net taxes on products must first be removed from the use table

at purchasers’ prices. The remaining value of the cells comprises only basic values and trade

margins and is called the “residual” use table at purchasers’ prices. It may be better first to estimate

the retail trade margins for consumption of households and deduct those amounts in order to get

all use data, including household final consumption expenditure, in a more uniform valuation,

including only wholesale trade margins, making the allocation of these wholesale trade margins

easier and of slightly higher quality. That notwithstanding, the retail margins and the wholesale

margins are determined as follows:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

229

• Based on the relationship between the total supply of each product and the trade turnover

determined in step 2, it is possible to estimate the average share of the total supply of each

product passing through wholesale and retail trade links.

• At this stage, it is necessary to make assumptions about the distributive pattern for each cell,

on the proviso that the total trade turnover (separately for wholesale and retail trade) is

known from step 2. In this way the absolute amount of each cell passing through either

wholesale or retail trade is determined.

• If no specific information is available, it can be assumed that the average wholesale or

retail trade margin ratio for the specific product (known from Table 7.5 in step 2) should be

applied to the share of the total value passing through this trade link as determined in step

7.86. When the two trade margin matrices have been compiled, they can be deducted from the

“residual” use table at purchasers’ prices to obtain the use table at basic prices.

7.87. The resulting use table and trade margin matrices should be checked for overall

plausibility, regarding both the relationship between allocated wholesale and retail trade margins

and that between the use table data at purchasers’ prices and the allocated trade margins. In this

process, the previously estimated trade margins by products of step 2 may also be reallocated.

7.88. The procedures outlined in this section for estimating the trade margin matrices is one that

typically has to be applied when estimating benchmark SUTs. As mentioned above, the approach

may be somewhat different when compiling annual tables on a current basis, as it may be better in

this case to start from the use table side, taking as the starting point the effective trade margin ratios

(in principle the proportion passing through this trade channel multiplied by the actual trade margin

ratio) of the previous year as this detailed set of trade margin ratios is less prone to aggregation

errors than the application of the average margin ratios by product that can be derived from the

supply table.

7.89. The annex to this chapter provides a numerical example where the principles outlined in

this section are illustrated in a complete template for the derivation of trade margin matrices, and

which clarifies the importance of the distinction between wholesale and retail trade margins in

getting the correct results.

D. Transport margins

7.90. Transport margins are another valuation component relating to the delivery chain of the

products from the producer to the final user. Transport margins represent freight services of

products when invoiced separately by the seller. Transport margins are transport charges paid

separately by the purchaser to take delivery at the required time and place. They are included in

the use of products at purchasers’ prices but not in the basic price of a manufacturer’s output or in

the trade margins of wholesalers or retail traders.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

230

7.91. Transport margins include in particular:

• Transport of goods arranged by the manufacturer, the wholesale or the retail trader in such

a way that the purchaser has to pay separately for the transport costs even when the

transport is done by the manufacturers, wholesale or retail traders themselves.

• Transport of goods from the place where they are manufactured or sold to the place where

the purchaser takes delivery of them, in the event that the manufacturer or trader pays a

third party for the transport, if this amount is invoiced separately to the purchaser.

7.92. This definition of transport margins means that the transport has to be arranged by the seller

(producer or trader). This also means that transport arranged directly by the purchaser (and thus,

of course, also directly paid for by the purchaser) is not included in the transport margins.

7.93. The existence of a transport margin is thus related to the way in which the transport costs

are paid. This means that transport margins cannot be derived from the output of the respective

transport services but that information on the payments between the two related parties of the seller

and the buyer is required.

7.94. Since transport margins only occur when transport services are separately invoiced, this

means that no partitioning of transactions is necessary because the transport service is already

treated as a separate product and necessarily known to the purchaser. (see 2008 SNA, para. 14.130)

7.95. Thus, from a statistical point of view, transport margins should therefore be much easier to

deal with and to estimate than the trade margins because they could be surveyed directly, based on

information available in the bookkeeping department of the purchaser and because it is not

necessary to break down the purchasers’ price on the basis of assumptions. In reality, however, it

may not be easy to obtain the data, for example, with large companies that keep their bookkeeping

departments offshore.

7.96. Based on the definition of transport margins, which is logically connected to the definitions

of basic prices, purchasers’ prices and trade margins, Box 7.2 provides examples of transport costs

which are not recorded as transport margins because they do not contribute to the valuation

difference between basic prices and purchasers’ prices.

7.97. The value concepts and the consequent transport margins are defined in the 2008 SNA in

such a manner as to reflect the way in which the transport costs are treated in business accounts

and thus in the source data. When the cost of transport that is arranged by the purchaser is included

in the price of the intermediate consumption or final use in the source data, it should be treated as

transport margins but not when it is recorded as a separate cost item. Hence the recording of

transport costs in the source data will influence the actual delimitation of transport margins, in

particular when source data which are not SNA-compatible are not being adapted (by the compiler

of the national accounts) to bring them into line with the SNA definition. Such general adaptations

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

231

of existing source data will in general not be feasible, however, and it may even be questioned

whether such adaptation should be attempted at all. Without such adaptation, the concepts of basic

value and transport margins would deviate somewhat from the recommended concepts but the

overall properties of the system would not be compromised.

7.98. Contrary to the treatment of trade margins, imports of transport margins can exist. This

happens when a foreign carrier transports freight into, within, or out of the domestic territory. This

would be the case of road, water (only inland waterways), and air (inland) transport. Pipelines

within the domestic territory are normally run by a resident enterprise.

Box 7.2 Examples of transport costs which do not form transport margins

Transport margins are not the same as actual transport costs. Since these two concepts are often

confused, examples of activities which are not recorded as transport margins because they do not

contribute to the valuation difference between basic prices and purchasers’ prices are listed below:

• If the manufacturers or traders transport the goods themselves and do not invoice the transport

separately, these transport costs will be included in the basic price of the manufacturers’ output

or traders’ output. This transport represents an ancillary activity and the individual costs of

transport will not be identifiable as transport costs.

• If the manufacturer arranges for the goods to be transported by a third party without a separate

invoice for the transport services, the transport costs will be included in the basic prices of the

manufacturers’ output. These transport costs will be identifiable and recorded as part of the

manufacturers’ intermediate consumption.

• If wholesale and retail traders arrange for goods to be moved from the point where they take

delivery of them to that where another purchaser takes delivery, these costs will be included in

the trade margin if no separate charge is made for transport to the purchaser, where these costs

will be part of the intermediate consumption by the wholesale trader and retail trader.

• If a household buys goods for final consumption purposes and arranges for transport by a third

party, these transport costs are recorded as household final consumption expenditure on transport

services and not included in transport margins.

• If a domestic carrier transports goods from country A to country B through the domestic territory

(transit transport), this will also not be considered as a transport margin as it does not relate to

goods that forms part of domestic supply and use. These transport services will be recorded under

exports of services.

• Transport services of domestic carriers outside the domestic territory (merchanting) are not part

of the transport margins but form exports of services.

• Freight of used goods, scrap and waste, earth and similar freight connected with construction

projects are also not part of transport margins as these goods are not considered as products. This

also includes the transport of goods in connection with removals.

7.99. According to the specific modes of transport (such as road, railway, water, air, and

pipeline), several kinds of transport margins need to be distinguished, provided they are classified

as separate products in the system. In addition, the services of forwarding agencies also form part

of the transport margins when paid separately by the buyer. Transport insurance services must also

be considered under the same terms as the general definition of transport margins.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

232

7.100. Compared with trade margins, transport margins are of a much lower magnitude in

accordance with the definition in the 2008 SNA, which is more restricted than that used in the

system based on the 1968 SNA. The transport margin is much more complex, however, not only

because of the different kinds of transport margins but also because of the definition itself.

7.101. Furthermore, the data situation gives rise to numerous practical problems. The relationship

between the supply of goods and the transport margins connected with them is looser than in the

case of trade margins. Thus, transport costs are usually not related to the value of the goods

transported; much transport is carried out as an ancillary activity and the manner in which transport

costs are paid might differ from product to product and from transaction to transaction.

7.102. According to the 2008 SNA, total imports and exports of goods are to be valued FOB. For

the purpose of compiling SUTs, however, total imports will be valued CIF and an appropriate

adjustment item should serve for the transition between both valuation concepts. A CIF valuation

means that transport costs up to the border of the importing country are included in the CIF-based

value.

7.103. Transport services between the border of the importing country and the domestic location

of the buyer are thus to be considered as transport margins (if paid for by the buyer and separately

invoiced by the seller). By corollary, transport services between the domestic location of the seller

and the border in the case of exports are also to be considered as transport margins (if paid for by

the buyer and separately invoiced by the seller). Transport services delivered outside the domestic

territory by resident producers will never become transport margins but are exports of trade

services. Non-resident carriers can also provide transport services within the domestic territory for

resident or non-resident buyers.

1. Compilation of transport margin matrices

7.104. Before embarking on the task of estimating transport margins, the compiler should

carefully study the instructions provided in the business questionnaires used to collect data on sales

and purchases and any other sources for these data. The instructions should be studies with a view

to determining the extent to which the collected data meet the conditions for the existence of

transport margins, and, if that is the case, how closely such transport margins correspond to the

SNA definition.

7.105. In particular, the exact way in which the (non-margin) transport costs directly collected in

the business surveys or available from other sources have been defined should be examined, in

order to clarify if such cost items could possibly include those transport costs that, according to

the SNA definition, are to be considered as transport margins. Only after this examination, can a

decision be made on the existence or the exact delimitation of the transport margins to be

estimated.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

233

7.106. Trade margins make up the bulk of the total output of trade services (the exceptions being

only the trade services on used goods, waste and scrap, and trade relating to goods in transit and

merchanting), and virtually all trading activity in the economy is covered by the total output of

trade services identified in the system. The situation for transport margin activity is quite different.

A significant part of all transport activity in an economy takes place as ancillary activity in non-

transport industries and is therefore not identified in the system. The intermediate consumption

related to the ancillary activity is lumped together with the intermediate consumption related to

the principal and secondary activities of the industry. Only the transport services carried out by the

transport industries and, if statistically identified, a very minor output of transport services as

secondary output in non-transport industries are shown explicitly in the system.

7.107. If transport margins are estimated, it is therefore not possible to assess their importance

relative to the total transport activity in the economy, and the estimated transport margins should

not be mistakenly perceived to reflect the actual physical freight activities carried out in the

economy, comparable to the activities customarily covered by specialized transport statistics.

Within the SUTs, the only way to assess total freight activity, ancillary and marketed, is by means

of the distribution of those inputs typically used for transport, such as fuel, auto repair, and current

taxes on motor vehicles (other taxes on production). On the other hand, these inputs will often –

and ideally – have been estimated on the basis of a distribution of all motor vehicles by type and

size across industry.

7.108. The size of the transport margins, and even of the total output of freight services, is usually

much smaller, relatively speaking, than that of the trade margins. In some cases, the imbalance

between the supply and use of a product might even be bigger than the transport margins of that

specific product. It is therefore recommended that a careful review be made of those products

where important transport services are involved, such as, for example, agricultural and forestry

products, energy products, iron and steel products and products related to construction. This

situation will vary across countries.

7.109. If no secondary transport activity is shown in the supply table at basic prices, this means

that the transport of goods has been arranged by the manufacturer, the wholesale trader or the retail

trader in such a way that the purchaser has to pay separately for the transport costs even when the

transport is done by the seller .

7.110. As noted above, transport margins could in principle be surveyed directly, on the basis of

the purchasers’ bookkeeping information. In practice, however, this information is not collected,

as coverage of this kind would involve not only the total transport margins paid by the enterprises

being surveyed but also their distribution by product and by kind of transport. In addition, even if

this information does exist, respondents may have to go back to the individual invoices and collate

the data required in order to obtain it. Thus, even though it would be possible to establish such

special surveys, for example related to the compilation of benchmark SUTs, it is generally not

conducted as it would entail such a large burden for the businesses concerned. When such

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

234

information is available and sufficiently representative, then this is all that is needed to compile

the transport margins of the system.

7.111. Box 7.3 provides four options to consider in the absence of any direct information on

transport margins. The argument is support of option 4 is that the matrix of wholesale trade margins

will in any event be based on inadequate information, and that it is not possible to ascertain whether

the margin associated with a particular cell is the, so to speak, “pure” wholesale trade margin or if

it also includes some transport margin. As trade margins will be much higher (by some 10–25 per

cent) than any contribution from possible transport margins (around 0.5 per cent), the additional

uncertainty introduced by choosing a joint wholesale and transport margin will therefore be rather

limited. The input structure of wholesale trade will, however, be somewhat distorted.

Box 7.3 Options to consider where no data exists on transport margins

In the absence of any direct information on transport margins, there are basically four ways to proceed:

Option 1

: In complete absence of any information on transport margins, decide that transport

margins are insignificant, given the way in which output and intermediate consumption values are

defined, and therefore need not be estimated at all.

Option 2: Concentrate on those products where important transport services are involved – normally

agricultural and forestry products, energy products, iron and steel products and products related to

construction – and collect ad hoc information about transport arrangement from selected enterprises.

Option 3: Decide to establish a full matrix of transport margins based on general assumptions about

total transport margins and their distribution by products and uses.

Option 4: Reroute transport margins by product and by use through wholesale trade. This can be

done by estimating for each type of transport output the share being transport margins. Then record

this as input into wholesale trade. The output of wholesale trade should be increased by the same

amount. This rerouting via wholesale trade recognizes the existence of transport margins but their

actual distribution is hidden in an untraceable way in the wholesale trade margin matrix.

Where options 3 and 4 are concerned, the total transport margins by type of transport output could in

principle be determined residually as the difference between total supply and the identified uses of each

type of transport service. This approach would, however, require a very high degree of confidence in the

preliminary estimates of transport costs entered into the use table. As previously noted, transport margins

are expected to make up only a very minor share of total transport services. This residual would be highly

unreliable, and probably reflect the statistical uncertainty of the estimated output and use data, rather than

the actual level of any transport margins.

7.112. On the basis of the supply table alone, it is not possible to distinguish transport services

paid for by the seller from those invoiced to the purchaser. Starting from the output of transport

services (principal or secondary) in the industries, only total output can be calculated. From this

total output, some non-margin services can be clearly deducted. These are the transport services

related to transit transport and merchanting, and to used goods, waste and scrap. It will also be

possible to deduct some statistically identified transport services paid for by the seller and not

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

235

invoiced separately or directly paid for by the purchaser. This could still leave a residual, however,

that would be much higher than any reasonable estimate of total transport margins.

7.113. Transport costs are usually surveyed in current business statistics, at least as a cost item.

By definition, these transport costs relate to the goods produced or traded. If the purchaser arranges

the transport, these costs may be incorporated in intermediate consumption. Based on the structure

of the output and the products traded, an estimate can be made, based on the structure of the

products for which the transport costs have been paid. It is implausible to assume, however, that

the transport costs are to some extent proportional to the value of the products produced or traded.

Such assumptions and the subsequent estimates will be of limited use.

7.114. The various estimation steps to be followed in calculating the transport margin matrices

should as far as possible be separated into the different modes of transport (for example, road,

railway, water, air, pipeline, forwarding) and transport insurance; the available data might not be

broken down in this way, however, and different estimation methods would need to be applied.

7.115. Given the limited availability of data, it may be better to concentrate on the products with

large transport margins and to allocate margins to the remaining products according to certain

plausible assumptions. As only a part of all transport services are transport margins, it is difficult

to check the resulting data for plausibility. The supply and use of transport margins should of

course be equal but estimation of the one side is not independent from estimation of the other side.

7.116. For the forwarding agents’ services, the same estimation problem exists as for the transport

itself. The forwarding agents’ services are much more closely related, however, to the transport

costs, and estimates could be based, if available at this stage, on the structure of the transport

margins. That said, not all transport is organized by forwarding agents. Forwarding agents are

usually engaged in cross-border transport rather than in domestic transport. In evaluating the

practical problems connected with the correct estimates of forwarding agents' margin matrices,

consideration could be given to the treatment of these services as not being part of the transport

margins.

7.117. Transport insurance services are usually a very small part of the transport margins. Here

too they may be more important for cross-border transport than for domestic transport. A key

difference is that the insurance premiums depend on the value of the goods transported, rather than

on the actual transport costs of the freight. Similarly to the forwarding agents’ services, and in

view of the practical implementation of such services and their limited scope, it could be decided

to treat them also as ordinary services outside the margin system.

7.118. Once the use table-based transport margin matrices, whenever relevant, have been

estimated, these matrices must be deducted from the use table at purchasers’ prices, and the total

transport margins by intermediate and final uses are then allocated to the transport service products

of the applied product classification.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

236

E. Taxes on products and subsidies on products

7.119. Taxes on products and subsidies on products are the other major valuation component in

addition to the trade and transport margins. Compared with the margins matrices, the elaboration

of the matrices of taxes on products and subsidies on products is less complicated because the data

situation is usually more favourable and the delimitation and calculation of taxes and subsidies

forms an integral part of the regular compilation of national accounts and not just an aspect of the

SUTs. Thus the main task arising with regard to taxes on products and subsidies on products when

compiling SUTs is the need to establish the relationship between the different kinds of taxes and

subsidies and the product flows.

7.120. The matrices for taxes on products and subsidies on products are usually derived by

separate calculations for each of those taxes and subsidies and related to the intermediate use and

final use parts of the use table. Contrary to the trade and transport margins, no specific information

on distribution channels or transport deliveries is needed here; only the relations between the

product classification and the individual taxes and subsidies are needed.

7.121. A tax on a product is a tax that is payable per unit of some good or service. The tax may

be a specific amount of money per unit of quantity of a good or service (the quantity units being

measured either in terms of discrete units or in continuous physical variables such as volume,

weight, strength, distance, time, and so forth), or it may be calculated ad valorem as a specified

percentage of the price per unit or value of the goods or services transacted. A tax on a product

usually becomes payable when it is produced, sold or imported, but it may also become payable

in other circumstances, such as when a good is exported, leased, transferred, delivered, or used for

own consumption or own capital formation. An enterprise may or may not itemize the amount of

a tax on a product separately on the invoice or bill that it submits to its customers (2008 SNA,

para. 7.88).

7.122. VAT is a special type of tax on products collected in stages by enterprises but ultimately

charged in full to the final purchasers. It is described as a deductible tax, because producers are

not usually required to pay to the government the full amount of the tax that they invoice to their

customers, being permitted to deduct the amount of tax that they have been invoiced on their own

purchases of goods or services intended for intermediate consumption or fixed capital formation.

VAT is usually calculated on the price of the good or service, including any other tax on the

product. VAT is also payable on imports of goods or services in addition to any import duties or

other taxes on the imports (2008 SNA, para. 7.89). General sales and turnover taxes give rise to

many of the same compilation problems as VAT.

7.123. A subsidy on a product is a subsidy payable per unit of a good or service. The subsidy may

be a specific amount of money per unit of quantity of a good or service, or it may be calculated ad

valorem as a specified percentage of the price per unit. A subsidy may also be calculated as the

difference between a specified target price and the market price actually paid by a buyer. A subsidy

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

237

on a product usually becomes payable when the good or service is produced, sold or imported, but

it may also be payable in other circumstances, such as when a good is transferred, leased, delivered

or used for own consumption or own capital formation (2008 SNA, para. 7.100).

7.124. Three main categories of taxes on products are distinguished:

• VAT-type taxes

• Taxes and duties on imports excluding VAT

• Taxes on products, except VAT and import taxes

7.125. Similarly, there are three main categories of subsidies on product:

• Import subsidies

• Export subsidies

• Other subsidies on products

7.126. The 2008 SNA gives further definitions and lists typical examples for all these different

types of taxes and subsidies on products. It should be noted that profits of fiscal monopolies which

are transferred to the State are treated as taxes on products, and that losses of government trading

organizations and subsidies to public corporations and quasi-corporations may have to be treated

as subsidies on products.

7.127. Taxes on products should be recorded on an accrual basis in the national accounts, that is,

when the activities, transactions or other events occur that create the liabilities to pay taxes. The

amounts to be recorded in the system are determined by the amounts due for payment only when

evidenced by tax assessments, declarations or other instruments which create liabilities in the form

of clear obligations on the part of taxpayers. The system does not impute missing taxes not

evidenced by tax assessments.

7.128. Subsidies on products are recorded when the transaction or the event (production, sale,

import, and others) which gives rise to the subsidy occurs.

7.129. The recording in the SNA of transactions related to taxes on products and subsidies on

products does not mirror the way in which those involved view them. The system contains no

transactions between economic units that are the actual payers (collectors) of taxes on product or

the actual receivers of the subsidies on products and government. In the SNA, taxes on products

and subsidies on products are recorded only at the level of the total economy and are not payable

out of GVA of domestic producers. They are also not split by institutional sector.

7.130. In the context of SUTs, this has the important implication that it is never necessary to

consider the actual payment flows related to these taxes and subsidies but only to identify the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

238

products to which they relate. It is therefore also irrelevant at which stage in the turnover sequence

(producer, wholesaler or retailer) the tax is actually being collected or the subsidy paid out.

1. Compilation of taxes on products and subsidies on products matrices

7.131. Generally, the compilation of the taxes on products and subsidies on products matrices

consists in three main steps. The first compilation step is the allocation of taxes and subsidies on

products by the products of the supply table corresponding to columns (12), (13) and (14) of Table

7.1. The second compilation step is to allocate taxes and subsidies on products to the relevant

entries of the use table, as shown in Table 7.3. The third compilation step covers the specific task

relating to VAT necessary to calculate non-deductible VAT.

7.132. The allocation of the taxes on products and subsidies on products would be easier when

SUTs are compiled at a level of product detail where there is a one-to-one relation between the

product classification item and the specific tax and subsidy. Furthermore, in cases where the tax

or subsidy was linked to the physical quantities, such additional information might be necessary.

7.133. In the first compilation step, the amounts of the different and specific taxes and subsidies

– usually taken directly from the respective government revenue accounts – are allocated to

specific products in the SUTs. If these data are not already on an accrual basis, then they must be

adjusted from a cash basis to an accrual basis, which can often be done by summary time-

adjustments. No further compilation steps would be needed to arrive at the required column of

taxes on products (exclusive of VAT) less subsidies on products for the supply table at purchasers’

prices, as in columns (13) and (14) of Table 7.1. The allocation of non-deductible VAT depends

upon the user and, in general, can only be derived on the basis of the use table, and is covered later

in this section.

7.134. The second compilation step refers to the allocation of taxes and subsidies on products to

the entries of the use table (intermediate use and final uses) at purchasers’ prices and to the

separation between “other taxes on products” and “subsidies on products”, as in Table 7.3. For

those product categories for which the tax or subsidy has been allocated, the share of the tax or

subsidy component in the purchasers’ price must be calculated. This step needs to be based on the

appropriate taxation basis according to tax legislation, and Table 7.3 is the result of the appropriate

calculations for each single kind of taxes on products and subsidies on products.

7.135. In order adequately to allocate the taxes and subsidies to the use table elements, not only

must the appropriate tax rates be explored but also the share of the use flows at which the tax rate

is to be applied. A certain product classification category might include not only flows that are

taxed but also other types of products not taxed, and certain products may be free of taxes for

certain users. Thus, an effective rate may need to be estimated: for example, the mineral oil tax

may not only have different tax rates for the different mineral oil products but some of them might

also have a tax rate of zero (for example, aviation fuel) and some users, such as the agriculture

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

239

industry, may be exempt from tax. As mentioned, this problem may be alleviated by having a

sufficiently detailed product classification.

7.136. There may be cases where no rates or limited data are available for the allocation of taxes

and subsidies to the use table element. In these cases, a pro rata approach may need to be applied:

for example, the total value of tobacco excise duty received by government may need to be

prorated against all industries’ purchases of tobacco except the principal industry and including

components of final uses. This implicitly assumes that all purchasers of tobacco pay the same

proportion of duty in relation to the value of their purchase. This is clearly a sub-optimal approach

but it achieves an allocation constrained to the corresponding total value in the supply table at

purchasers’ prices.

7.137. Usually the taxes on products and subsidies on products are restricted to only a small

number of products. Furthermore, only a small number of taxes on products cover the bulk of the

total value of taxes on products. This situation is even more prevalent with subsidies on products.

7.138. The third compilation step covers a specific tax on products, VAT, which requires separate

handling. According to the 2008 SNA, VAT is to be recorded net in that:

• The output of goods and services and imports are valued excluding invoiced VAT.

• Purchases of goods and services are recorded inclusive of non-deductible VAT.

7.139. VAT is recorded as being borne by the purchasers, not the sellers, and then, only by those

purchasers who are not able to deduct VAT. This applies to both intermediate consumption and

gross capital formation.

7.140. Accordingly, the overwhelming part of non-deductible VAT will be recorded as being

levied on final uses, mainly on household final consumption expenditures. A small part of VAT,

however, is levied on enterprises and institutions that are exempt from VAT.

7.141. According to the definition of purchasers’ prices, only the non-deductible part of VAT is

included in the purchasers’ prices. Thus, the rows (products) in the use table at purchasers’ prices

include non-deductible VAT. In order to balance supply and use for each product, the non-

deductible VAT by products has to be estimated and either included in column (13) of Table 7.1

or deducted from the use table.

7.142. In general, VAT exemptions are related to products or activities. If an industry has only

exempted activities, this causes no problem. In the case of an industry that has exempted and non-

exempted activities, however, additional assumptions on the estimation of non-deductible VAT

are necessary. In such cases, the ratio of exempted activities and total activities must be applied to

intermediate consumption in the estimation of VAT. The exemptions may differ across countries

and be dependent on the respective countries’ taxation policies. For estimation of VAT by type of

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

240

product, and by type of use, details should be sought from the relevant tax authorities and should

be reviewed annually.

7.143. In order to calculate non-deductible VAT, it is necessary to identify those industries and

final users that are exempted from VAT, and therefore not permitted to deduct VAT from their

purchases, and to relate the VAT rates (explicit rate or an effective rate depending upon the product

mix) to the product classification used. Both steps need to be based on the actual VAT legislation.

This calculation will be further complicated where there is more than one VAT rate in operation,

as some product items in the product classification applied might be mixed in terms of their VAT

tax rates. In this case, additional breakdowns of those product groups would be desirable or an

effective rate should be calculated, using a weighting of lower level product detail and VAT rates.

7.144. A certain part of an industry might be VAT-exempt, in which case appropriate subdivisions

might be helpful. It could also be the case that certain VAT-exempt industries are normal VAT

payers for their secondary outputs. The VAT legislation may also have specific rules for very small

enterprises that have to be considered, for example, thresholds for being registered in the VAT

system. It is important to note, for VAT exempt industries, that non-deductible VAT must be

calculated both for intermediate consumption and also for gross capital formation.

7.145. Examples of exempt-type industries, dependent upon individual countries’ tax legislation,

tend to cover such industries as postal services, newspapers, housing, banking, insurance, some

business services, education services and health services. In addition, small producers below the

VAT threshold may also be exempt.

7.146. Total calculated non-deductible VAT derived by using the official tax rates and the

purchasers’ values of the relevant cells in the use table will generate a theoretical VAT estimate,

which should exceed VAT revenue (on accrual basis) received by the government. This is because

there will always be some degree of missing VAT due to evasion, cash transactions and fraud

involving products that, based on other statistical sources, are included in SUTs. Using the official

VAT rates may lead to an over-estimation of theoretical VAT; for example, where the tax law

allows reductions for losses on debtors or in countries where there are high thresholds for

registration for VAT. It can also be appropriate to lower the rate if local tax authorities are known

to be inefficient.

7.147. In the balancing process of the VAT vector and matrices, the theoretical VAT must be

adjusted to the revenue received (due to be paid) by government. This adjustment process should

be based on information from the tax authorities, such as industries and products where VAT is or

is not likely to be paid: for example, general government is likely to be fully compliant whereas

households are less so.

7.148. In producing SUTs, it is also important that the rates are reviewed each year to allow for

changes in VAT rates, schemes and legislation. For example, if the effective VAT rate on a product

changes in mid-year, an appropriate weighted estimate for the period will need to be established.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

241

Annex A to chapter 7. Example for deriving trade margins in SUTs based on

survey data

A7.1 This annex provides an illustration of how to calculate trade margin matrices using survey

data. The example is based upon the survey data obtained through the questionnaire shown in

figure A7.1, which is used by the Statistical Office of Serbia. Similar information can be obtained

with other forms of questionnaires.

Figure A7.1 Extract of questionnaire

Source: Structure of income and expenditure of economic subjects in the republic of Serbia 2011. Statistical office of the

Republic of Serbia, 2013.

Note: Data were collected for about 250 goods and 50 services according to the Classification of Products by Activity

(CPA) used in the European Union, varying from two-digit to four-digit groups. The CPA is consistent with the CPC.

The same questionnaire was used for all non-financial market enterprises.

A7.2 The objective of the calculation is to populate table A7.1 with available data in order to

obtain the trade margin matrices. If the available data sources are less complete, the results will of

course be more dependent on the assumptions used to populate the trade margin columns (in the

supply table) and the trade margin matrices (underpinning the use table). Irrespective of the

coverage or quality of source data on trade and trade margins, it is important to generate the table

illustrated in table A7.1 below.

No. Code Product description

Sales of

goods

produced by

the

enterprise

(group of

accounts 61)

Closing stock

of products

and work in

progress

(groups of

accounts 10

and 11)

Sales of

merchandise

(group of

account 60)

Trade margin

amount or

rate %

Closing

stocks of

goods for

resale (group

of accounts

(1) (2) (3) (4) (5) (6) (7) (8)

1072 14.3 Outerwear, knitted or crocheted; socks, sweaters, vest

1073 15.1

Tanned or dressed leather; luggage, handbags, saddlery and harness;

dressed and dyed fur

1074 15.2 Footwear

Wood and products of wood and cork, articles of straw and plaiting

1075 16.1

Cut and treated wood for further processing

1076 16.2

Wood and products of wood and cork, except furniture (see 1147); articles

of straw and plainting

Paper and paper products

1077 17.1

Pulp, paper and cardboard for further industrial processing and printing

1078 17.2

Articles of paper and paperboard for industrial use–boxes, containers and

packing material

1079 17.2

Articles of paper and paperboard for personal use–paper towels, napkins,

toilet paper, cleaning items and deletion of the pulp and paper

1080 17.2

Paper stationary and articles of paper and paperboards (notebooks,

binders, forms etc..)

Coke and refined petroleum products - manufacturing (columns 4–5;

trade and wholesale (columns 6–8)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

242

A7.3 In collecting data from the above questionnaire, the following considerations are followed:

• For all economic activities, trade turnover and either purchase of goods for resale or trade

margins (either absolute or as a percentage of either purchasing or selling price), data are

collected by type of product.

• The product specification is at least as detailed as the product classification applied in the

SUTs, and either uses the same classification or a version that is easily transformed into the

SUTs product classification.

• For each product, data are sought on either opening or closing stocks of merchandise to

facilitate the calculation of changes in inventories of merchandise needed in the event that

the total trade margin is derived from the difference between sales and purchases.

• Surveys usually include total coverage for enterprises above a certain threshold (based on

either turnover or employment) and samples for the smaller enterprises. It is assumed that

the survey results have been grossed up to cover the whole population.

Based on this information it is possible to compile table A7.1.

A7.4 As shown in table A7.1, the vast majority of trade turnover and output of trade margins

(trade services) originates from the three trade activities (ISIC Rev. 4, divisions 45–47), whereas

many other industries generate relatively small amounts of trade output as their secondary

production. In table A7.1, estimates both of grossed-up trade turnover and trade margins are

shown, as this information is needed later in the compilation process.

Table A7.1 Trade data from survey: trade margins identified separately

for wholesale and retail trade margins

A7.5 At this stage, it is important to introduce the distinction between wholesale and retail trade

margins. In this example, it is assumed that this distinction is not made directly in the survey results

(although it can be established from survey returns). It is therefore necessary to make decisions on

the type of margin associated with the various combinations of economic production activities and

products.

A7.6 First, it may be reasonably assumed that ISIC Rev. 4, division 46, wholesale, produces

mainly wholesale trade margins and, similarly, that ISIC Rev. 4, division 47, retail trade, produces

. . . . . .

Trade

turnover

Trade

margin

Trade

turnover

Trade

margin

. . .

Trade

turnover

Trade

margin

Trade

turnover

Trade

margin

Trade

turnover

Trade

margin

. . .

Trade

turnover

Trade

margin

Trade

turnover

Trade

margin

Trade

turnover

Trade

margin

. . . . . .

Product: CPC

1 10 1 15

2 0 0 750 90 950 230

10 3 800 100 1 000 250

0 0 10 4 0 0 1 000 100 700 300 20 8 1 100

100 800 400

Total W 5 200 1 000

9 000 0 0 11 000

Total R 0 50 2 000 0 13 000 25 17 000

SUM: all W

SUM: all R

Indicate W or R

Wholesale (W)

Reta il [R]

Indicate W or R

Ec. activity 1

Ec. activity 2

Trade 45

Trade 46

Trade 47

Ec. activity n

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

243

mainly retail margins, whereas ISIC Rev. 4, division 45, produces a mix of margins, which must

be decided on the basis of the individual products.

A7.7 For trade carried out as secondary activity, it may be assumed, for example, that trade

turnover of restaurants and hotels is probably retail trade turnover; the same holds for the trade

turnover of such suppliers as hairdressers, cinemas and theatres. On the other hand, trade activities

of advertising agents are more likely to be wholesale trade. Manufacturing industries often trade

in products similar to those that they produce or in complementary products and the majority of

such trade sales are usually of the wholesale type, although some may be sold directly to

consumers. These industries may also be trading in similar imported goods and such trade is again

likely to be classified as wholesale trade.

A7.8 There will be some products where a trade margin may not be applicable and additional

survey detail may be collected that makes it possible to improve the specific nature of these

decisions.

A7.9 Once these decisions have been made, the row and column totals for wholesale margins

and retail trade margins can be calculated and, for the row totals, also the turnover by product

broken down by wholesale and retail turnover. Table A7.1 illustrates how the results of these

decisions are fitted into the system. It should also be noted that, at this stage, only absolute and not

percentage margins are being processed.

A7.10 As explained in chapter 7, the source data for trade activity and trade margins may in

practice be available in a variety of forms and with a varying degree of detail. Different

assumptions may therefore be needed to establish the dataset shown in table A7.1, which is

essential for deriving the trade margins needed in both the supply table and the use table.

A. Supply table

A7.11 The trade activity and the trade margin entries needed in the supply table consist of rows

for output of margin activities by economic activity in the “domestic output at basic prices” part

of the supply table, and of the columns for trade margins needed to transform the values by product

from basic prices to purchasers’ prices.

A7.12 It will become increasingly clear throughout this process, and beyond, that it is essential to

retain the distinction between wholesale trade margins and retail trade margins derived in

connection with table A7.1.

A7.13 Table A7.1 resembles the format (product by industry) of the supply table. All necessary

information on the output of trade services and trade margins from table A7.1 can be transferred

to table A7.2.

A7.14 As explained in chapter 5, many countries may choose to redefine secondary output of

trade services in the supply table so that secondary output is classified together with the output of

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

244

the primary producers. Such a redefinition would mean that there will be trade activity and output

of trade margins services only from ISIC Rev. 4, divisions 45–47. In this example, it is assumed

that redefinition has taken place. This simplifies all the calculations in this example and facilitates

the estimation of the input structures and the calculations of IOTs. The basic methodology outlined

in this annex will not be affected, however.

A7.15 It should be noted that the output of trade products may also contain some non-margin

items (for example, commissions, fees, margins on second-hand sales, and others), so that total

supply from trade may still be positive after deduction of trade margins in the trade margin columns

– this reflects actual output produced.

Table A7.2 Supply table

A7.16 Before compiling the trade margin matrices associated with the use table, it is useful to

derive a number of ratios from tables A7.1 and A7.2, which are shown in table A7.3.

Table A7.3 Distribution channels and percentage trade margins

A7.17 The question of trade channels (namely, how big a share of the supply of a given product

passes through the wholesale or retail channels or both) is central when compiling the use table

trade margin matrices as shown in Figure 7.2. Fortunately, there is enough combined information

in tables A7.1 and A7.2 to tackle this issue.

Economic

acti vity 1

Economic

acti vity 2

. . .

Trade 45 Trade 46 Trade 47 . . .

Economic

acti vity n

Output at

basic

prices

Imports

Supply at

basic

prices

Wholesale

trade

margins

Retail

trade

margins

VAT

Taxes on

products

Subsidies

products

Supply at

purchasers'

price

1 1 000 100 250 150 20 1 520

2 2 000 100 400 200 - 10 2 690

Wholesale 5 200 1 000 9 000 0 11 000 - 11 000 0

Retail 0 50 2 000 13 000 25 17 000 - 17 000 0

Total 0 0

PRODUCTS

Basic value

plus wholesale

margins

Basic value

plus wholesale

and retail

margins

Trade

turnover

wholesale

Trade

turnover

retail

Wholesale Retail Wholesale Retail Wholesale Retail

1 1 100 1 350 800 1 000 72.7 74.1 100 250 12.5 25.0

2 2 100 2 500 1 100 800 52.4 32.0 100 400 9.1 50.0

Wholesale

Retail

Total

Average percentage trade

margin for traded goods

out of comparable prices

PRODUCTS

Total supply at purchasers'

prices made comparable to

trade survey turnover

(from Table A7.2)

Survey turnover,

grossed up

(from Table A7.1)

Average percentage of

supply passing through

wholesale and retail trade

channels

Absolute trade margins

(from Table A7.1)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

245

A7.18 To calculate these shares, the total supply in the supply table must be made comparable to

the turnover concept used for the survey data (assumed to be turnover exclusive of VAT and net

of taxes on products). The supply table value concept comparable for the wholesale trade turnover

is basic value plus wholesale trade margin (1100 for product 1). For retail trade turnover, the

comparable concept is basic value plus both wholesale and retail trade margins (1350 for product

1).

A7.19 The average percentages of supply passing through each of the wholesale and retail trade

channels can now be calculated as shown in table A7.3. For product 1, the shares are 72.7

(800/1100) and 74.1 (1000/1350), respectively. It is further possible to calculate the average

percentage trade margin for actual traded goods from comparable purchasers’ prices. For product

1, the percentage trade margins are 12.5 (100/800) and 25.0 (250/1000), respectively.

A7.20 It should be noted that the percentage trade margins are calculated as percentage of sales

prices, as required for the estimates in the use table (and not the usually applied survey percentages

from the traders’ buying price). It should be noted that the VAT column does not need to be

completed to calculate these memo items.

B. Use table

A7.21 The use table is initially valued at purchasers’ prices and this table is the starting point for

determining the valuation matrices that will permit the gradual transition of the use table from

purchasers’ prices to basic prices.

A7.22 The first step is to estimate the VAT matrix, and then to deduct it from the use table at

purchasers’ prices. In the next step, the matrix for other taxes on products must be determined and

deducted, and the matrix for subsidies on products determined and added.

A7.23 The elements in the residual use table resulting from these procedures will consist of only

basic values and trade margins, as illustrated in table A7.4, and the task is now to separate each

element into its basic value and the possible wholesale and retail trade margins. In order to

illustrate the restrictions and sum conditions, it is assumed that only those economic activities

specified (1, 2 and n) have intermediate consumption.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

246

Table A7.4 Use table after removal of net taxes on products

A7.24 The product-by-product procedure uses the information in table A7.3, as illustrated in

tables A7.5, A7.6 and A7.7. If a product passes through both a wholesale channel and a retail

channel, the retail trade margin comes on the top of the wholesale trade margin, corresponding to

the trade margin percentages calculated in table A7.3. Accordingly, the retail trade margins must

be estimated first, followed by the wholesale trade margins.

Table A7.5 Product 1: retail margins

A7.25 The estimates of the retail trade margins are illustrated for product 1 in table A7.5. From

table A7.3, it is known that 1,000 of the 1,350 passes through the retail trade, and the total retail

trade margin on this product is 250. The knowledge of these totals provides a good starting position

but it is still not known to which of the individual uses (or part thereof) the retail turnover is linked,

and it is therefore necessary to decide on (or to make assumptions regarding) the figures to be

entered in row (2) – based on which specific knowledge may be at hand, and on common sense,

to comply with the restriction that their sum must be 1,000. For example, final consumption

expenditure of households is usually assumed to include a high share of the available retail trade

margins, whereas intermediate consumption and gross capital formation may have very narrow

retail margin, and exports none at all, as non-resident expenditure is a summary adjustment item,

and the related margins will be included in the domestic consumption concept.

A7.26 When row (2) in table A7.5 has been determined, the distribution of the retail trade margin

can be determined either by distributing the 250 proportionally to the values in row (2), or by

applying the percentage retail trade margin of 25 per cent from table A7.3 to the values in row (2).

In row (3), the effective percentage retail trade margins relative to the elements of the use table are

calculated. These are the percentages that will be used to recalculate the retail trade margin table

after changes made to the original data during the balancing.

Economic

activity 1

Economic

activity 2

. . .

Economic

activity n

Total

intermediate

consumption

Final

consumption

expenditure of

households

Final consumption

expenditure of

general government

Gross fixed

capital

formation

Changes in

inventories

Exports

Total use at

purchasers’

prices

1 100 50 150 300 700 50 150 50 100 1 350

Wholesale

Retail

Total

PRODUCTS

Economic

acti vity 1

Economic

acti vity 2

. . .

Economic

acti vity n

Total

intermediate

consumption

Final consumption

expenditure of

households

Final consumption

expenditure of

general

government

Gross fixed

capital

formation

Changes in

inventories

Exports

Total use at

purchasers’

prices

1. Starting row (from Table A7.4)

100 50 150 300 700 50 150 50 100 1 350

2. Selected values with retail margin

50 150 200 700 100 1 000

3. Retail margin distributed

13 0 38 50 175 0 25 0 0 250

4. Average percentage retail margin

12.5 25.0 25.0 0.0 16.7 0.0 0.0 18.5

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

247

Table A7.6 Product 1: wholesale margins

A7.27 A similar procedure is used to determine the distribution of wholesale trade margins in

table A7.6. The first row in this table is the first row in table A7.5 minus the estimated retail trade

margins. When the estimated wholesale trade margins are deducted from row (1) in table A7.6,

the row at basic prices in table A7.7 below is obtained, and thus the desired use table at basis prices

has been derived. It should be noted that, in the use table at basic prices, the rows for wholesale

and retail products will be made up of the column totals of the two trade margin matrices and, in

addition, that they will include any non-margin trade output.

Table A7.7 Product 1: row in use table at basic prices

A7.28 Following the outline of these procedures, it is clear why it is essential to distinguish

between wholesale and retail trade margins. If this is not done, it will not in practice be possible

to manage the problem of successive trade channels. Thus the cumulative trade margin on

household consumption of (175+62.5)/463 = 51 per cent of the basic value total could not have

been derived directly from the survey results if simply aggregated.

Economic

activi ty 1

Economic

activi ty 2

. . .

Economic

activi ty 2

Total

intermediate

consumption

Final consumption

expenditure of

households

Final consumption

expenditure of

general

government

Gross fixed

capital

formation

Changes in

inventories

Exports

Total use at

purchasers’

prices

Product 1 at basic prices

81.3 50.0 100.0 231.3 462.5 50.0 112.5 50.0 93.8 1 000.0

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

249

Chapter 8. Compiling the imports use table and domestic use table

A. Introduction

8.1. This chapter describes the disaggregation of the use table into the imports use table and the

domestic use table. The first table, the imports use table, contains information on the use, in the

national economy, of imported products (by product) for intermediate consumption and final uses.

The second table, the domestic use table, provides information on the use of domestically produced

products (by product) for intermediate consumption and final use. The compilation of these two

tables primarily consists in the estimation of the imports use table, since the domestic use table is

obtained by subtracting the imports use table from the use table. This chapter therefore focuses

mainly on the compilation of the imports use table.

8.2. The compilation of an imports use table is embedded in the SNA and it is important to

balance the supply and use of products for the domestic economy, accurately to deflate components

of GDP by linking imported intermediate products with appropriate import price deflators, and to

ascertain the correct distribution of the changes in the volume of GVA by industry and industry

contributions to GDP growth.

8.3. Historically, the compilation of the imports use table was mainly considered as an

intermediate step towards the compilation of IOTs (although not an essential step). The imports

use table is becoming increasingly important in its own right, however, for analytical purposes.

With the globalization of economic activities, exports and imports are growing more rapidly than

GDP and the GVA chains in production are becoming more complex and more international. It is

therefore very important for the national accounts to provide a sectoral disaggregation of

macroeconomic data for both domestic production and imports.

8.4. Over time, many domestic economies have seen significant changes in the import share of

domestic supply that can be attributed to changes in international trade, and in particular, trade in

goods for processing and other intermediate material inputs. In addition, many multinational

enterprises, and previously large domestic businesses, have shifted their production processes

around the world, taking advantage of lower production costs and thereby increasing their

competitiveness and profitability.

8.5. The set of SUTs at basic prices – both in current prices and in volume terms – that should

be compiled includes the following tables:

• Supply and use table at basic prices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

250

• Domestic use table at basic prices

• Imports use table at basic prices

8.6. Since direct information for compiling the use table for imported products is generally rare

and available only in exceptional cases, the recommendation is to work on a highly detailed level

of product group (which means rectangular SUTs with a detailed product specification). A detailed

level will help in identifying the likely users of a specific imported product.

8.7. Chapter 5 provides a detailed description of the concepts and definition of imports of goods

and services, where emphasis is placed on the imports by products as part of the total supply of

products. In the supply table, imports are only shown as a vector of products covering goods and

services. In practice, however, it may be desirable to subdivide the import vector by regions so as

separately to identify imports within and outside particular regions. Furthermore, the import vector

could show separate columns, for example, displaying goods and services separately. Further

elaborations that are very useful for the analysis of global value chains and globalization include

splitting residents’ expenditure abroad into individual products; separately identifying the

transport and insurance margins included in CIF estimates of goods; and separately identifying

imports of manufacturing services provided under goods for processing arrangements (ideally, also

with complementary information showing the underlying value of goods processed).

8.8. This chapter focuses on the compilation of the imports use table. In particular, Section B

describes the structure of the imports use table, provides a numerical example and describes how

to obtain the domestic use table. Section C focuses on the compilation of the imports use table and

potential issues that arise during the compilation process.

B. Structure of the imports use table and domestic use table

8.9. Imports consist of purchases of goods and services by residents from non-resident

producers and suppliers. In the SNA, total imports are valued FOB. Data on detailed flows of

imports by product from foreign trade statistics, however, are usually valued on a CIF basis. To

reconcile the different valuations used for total imports and the product components of imports, a

global CIF/FOB adjustment on imports is required, and this needs to be allocated in accordance

with the type of goods involved. More details on the CIF/FOB adjustment may be found in chapter

8.10. The supply of imports shown in the supply table has to be allocated in the imports use table

to the different use categories of intermediate uses and final uses. The general structure of the

imports use table is shown in Table 8.1. The table shows the total use of imported products, goods

and services, by products and by industries and by final use categories. In the columns, the table

has the same format as the use table. It distinguishes two main sub-matrices, one for the

intermediate use and one for the final uses of products. The total use of imports must be equal to

the total supply of imports of the supply table. This equality is given for each of the products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

251

distinguished in the SUTs. Table 8.2 shows a numerical example of the imports use table as part

of the SUTs system.

Table 8.1 Structure of the imports use table

Table 8.2 Numerical example of the imports use table

Millions of euros

Table based on 2011 figures from Austria

8.11. Once the imports use table is compiled, the domestic use table is obtained by deducting the

imports use table from the use table. As shown in Table 8.3, the structure and size of the domestic

use table are the same as those of the use table, except for an additional row in the primary inputs

section to reflect the sum of the columns in the imports use table. The body of the domestic use

table does not include direct or indirect imports of goods and services. Some countries compile

and reconcile both the imports use table and domestic use table concurrently, instead of compiling

the use table first, followed by compilation of the imports use table and domestic use table. This

happens, for example, where there are very good quality data on both imports and domestic use,

separately available.

8.12. Table 8.4 provides a numerical example of the domestic use table.

Agriculture Manufacturing

…

Services Total

Final

Consumption

Gross capital

formation

Exports Total

Agriculture

Manuf acturing

…

Other services

Total

Intermediate consumption by industry

Total final uses by category

Imported products f or intermediate consumption

at CIF values

Total imported

products for

intermediate

consumption

Imported products f or final uses at CIF values

Total imported

products for

final uses

Total use

at basic

prices

Imported

total use

Industries

Products

Industries

Final uses

Households NPIS H

General

government

(1) (2) (3)

(4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture

(1)

191 1 680 5 170 14 17 2 077 1 079 47 9 58 1 194 3 271

Manufacturing (2) 706 55 898 4 365 5 621 1 126 2 985 70 702 20 894 1 422 12 310 807 1 344 17 112 53 888 124 590

Construction (3) 255 197 68 38 5 563 563

Trade (4) 257 0 274 30 39 600 600

Transport (5) 10 1 300 95 2 181 265 75 3 926 139 9 59 1 6 4 011 4 223 8 150

Communication (6) 4 860 65 2 449 1 267 248 4 893 447 17 686 22 169 1 342 6 234

Finance and business service

(7) 8 1 786 106 1 566 2 654 322 6 443 145 473 618 7 061

Other services (8) 14 1 110 23 127 275 384 47 118 549 824

Total (9) 919 62 051 4 834 12 439 5 417 3 819 89 479 23 087 0 1 495 13 575 926 1 381 21 350 61 814 151 293

INDUSTRIES

FINA L USE

PRODUCTS

Total use

at basic

prices

Manufac-

turing

Agricul-

ture

Other

services

Finance and

business

services

Trade,

transport and

communication

Total

Construc-

tion

Gross f ixed

capital

formation

Final consumption expenditure

Exports

Changes in

inventories

Changes

valuables

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

252

Table 8.3 Structure of the domestic use table

Table 8.4 Numerical example of a domestic use table

Millions of euros

Table based on 2011 figures from Austria

Agriculture Manufacturing … Services Total

Final

Consumption

Gross capital

formation

Exports Total

Agriculture

Manuf acturing

…

Other services

Total at basic prices

Use of imported products, CIF

Taxes less subsidies on products

Adjustments

Total at purchasers' prices

Compensation of employees

Other net taxes on production

Consumption of fixed capital

Net operating surplus/net mixed income

Value added at basic prices

Total inputs at basic prices

Empty cells

Total value added by industry

Total input by industry

Industries

Products

Industries

Final uses

Final uses at basic prices

Total imported products for final uses

Intermediate inputs at purchasers' prices

Total use at

basic prices

Total use by

product

Domestic intermediate inputs at basic prices

Domestic products f or intermediate consumption

at basic prices

Total imported products for intermediate consumption

Net taxes on products f or intemediate consumption

Value added by component and by industry

Domestic products f or final uses

at basic prices

Net taxes on products f or final use

Adjustments on final uses

Final uses at purchasers' prices

Households NPIS H

General

government

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture (1) 2 354 4 284 8 117 15 21 6 800 963 123 - 42 938 1 982 8 782

Manufacturing (2) 1 216 42 772 6 256 7 776 4 085 3 123 65 227 12 631 327 9 426 1 122 1 393 96 280 121 178 186 405

Construction (3) 105 2 184 9 321 2 373 3 625 1 414 19 021 1 402 24 323 - 38 563 26 250 45 272

Trade (4) 245 8 601 1 560 4 370 682 1 462 16 919 27 684 1 080 4 008 238 273 9 985 43 267 60 187

Transport (5) 31 4 424 364 6 368 575 301 12 063 5 828 3 418 87 2 21 4 916 14 271 26 335

Communication (6) 29 1 651 226 6 745 4 295 1 343 14 289 22 088 51 5 111 40 6 649 33 940 48 228

Finance and business services

(7) 439 11 706 4 611 18 623 25 779 8 062 69 219 36 524 1 006 9 781 0 - 177 11 156 58 289 127 508

Other services (8) 8 367 58 1 060 375 1 622 3 490 13 045 5 416 53 116 113 - 105 1 567 72 153 75 643

Total at basic prices (9) 4 429 75 987 22 402 47 431 39 431 17 348 207 028 120 165 5 416 58 997 52 973 1 257 1 471 131 053 371 332 578 360

Imports (10) 919 62 051 4 834 12 439 5 417 3 819 89 479 23 087 1 495 13 575 926 1 381 21 350 61 814 151 293

Taxes less subsidies on

products

(11) 92 952 229 1 349 1 689 2 672 6 984 22 810 557 2 870 152 7 397 26 794 33 778

Total at purchasers’ prices (12) 5 440 138 991 27 466 61 219 46 538 23 839 303 492 166 063 5 416 61 050 69 418 2 335 2 859 152 800 459 939 763 431

CIF/FOB adjustments on

exports

(13)

- 97

- 97 - 97

Direct purchases abroad by

residents

(14) 6 675 6 675 6 675

Purchases on the domestic

territory by non-residents

(15) - 12 945 12 945

Total at purchasers’ prices (16) 5 440 138 991 27 466 61 219 46 538 23 839 303 492 159 792 5 416 61 050 69 418 2 335 2 859 165 648 466 517 770 009

Compensation of employees (17) 551 30 679 10 239 37 906 22 997 41 971 144 343

Other taxes less subsidies on

production

(18) - 1 627 1 077 546 1 755 2 004 1 103 4 858

Consumption of fixed capital

(19) 1 845 12 750 1 542 10 917 18 934 7 480 53 469

Net operating surplus/net

mixed income

(20) 3 658 16 453 5 138 23 040 18 989 4 921 72 198

Gross operating surplus/gross

mixed income

(21)

5 503 29 203 6 680 33 957 37 923 12 401 125 667

GVA (22) 4 427 60 959 17 465 73 618 62 923 55 475 274 868

Total input at basic prices (23) 9 867 199 950 44 931 134 837 109 461 79 314 578 360

Empty c ells

PRODUCTS

Adjustments

GVA

Manufac-

turing

Agricul-

ture

Changes

inventories

Exports

Total

Total use

at basic

prices

INDUSTIES

FINA L USE

Other

services

Finance and

business

services

Trade,

transport and

communication

Construc-

tion

Total

Final consumption expenditure

Gross f ixed

capital

formation

Changes

valuables

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

253

1. Input table for imports

8.13. Table 8.5 shows a numerical example of an input table for imports of goods and services

at basic prices. This table is either a product-by-product table or industry-by-industry table but is

not an IOT, as imports form an input and not an output.

8.14. It should be noted that the sub-matrices for final uses and the row totals for products are

the same in the imports use table and the IOT table of imports. An input table for imports, as

already noted, may also be a step in the process of compiling IOTs but not a necessary step. This

is covered in more detail in chapter 12 (see box 12.3) on the transformation of SUTs into IOTs,

where the transformed imports use table can be applied in two different ways.

8.15. The only way in which it differs from the imports use table in Table 8.2 is that Table 8.5

shows the intermediate use of the imports in a product-by-product format (or this could be an

industry-by-industry format). The final use part is unchanged. Chapter 12 provides more details

on how the imports use table, and in turn the input table for imports, can be used to produce IOTs

where, for imports of goods and services, this is only an input table.

Table 8.5 Example of an input table for imports at basic prices

Millions of euros

Table based on 2011 figures from Austria

2. Input-output table for domestic output at basic prices

8.16. If the imports use table is subtracted from the use table at purchasers’ prices, the

corresponding domestic use table can be derived which shows only consumption of domestic

produced output. A further step, however, is to subtract and reallocate the trade and transport

margins and to deduct the taxes less subsidies on products, in order to achieve the SUTs at basic

prices.

8.17. Table 8.6 shows the IOT for domestic output and this forms the basis for input-output

analyses. It should be noted that, in this table, the use of imported goods and services is only shown

in an aggregated form in one row. More detail on the transformation of SUTs into IOTs may be

found in chapter 12.

Households NPISH

General

government

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture (1) 176 1 722 3 148 14 15 2 077 1 079 47 9 58 1 194 3 271

Manufacturing (2) 618 55 846 4 392 5 506 1 398 2 941 70 702 20 894 1 422 12 310 807 1 344 17 112 53 888 124 590

Construction (3) 265 204 47 44 4 563 563

Trade, transport and communication (4) 9 2 095 150 5 150 1 678 337 9 419 586 26 745 1 28 4 179 5 565 14 984

Finance and business services (5) 7 1 531 97 1 527 2 974 308 6 443 145 473 618 7 061

Other services (6) 10 0 108 29 127 275 384 47 118 549 824

Total (7) 811 61 469 4 846 12 485 6 136 3 731 89 479 23 087 1 495 13 575 926 1 381 21 350 61 814 151 293

PRODUCTS

Agriculture

Manufacturing

Construction

Trade,

transport and

communication

Total

Total use

at basic

prices

Total

Final consumption expenditure

Gross f ixed

capital

formation

Changes

valuables

Changes

inventories

Exports

PRODUCTS

FINA L USE

Finance and

business

services

Other

services

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

254

Table 8.6 Input-output table for domestic output at basic prices

Millions of euros

Table based on 2011 figures from Austria

8.18. It should be noted that table 8.6 does not contain any of the adjustment rows shown in table

8.4. The adjustment items are the following:

• CIF/FOB adjustments on exports (recorded as part of imports)

• Direct purchases abroad by residents (recorded as part of exports)

• Purchases in the domestic territory by non-residents (recorded as part of exports)

8.19. In this form, as shown in table 8.6, the IOTs always show the correct GDP; the totals for

household final consumption expenditure, exports and imports in the IOTs differ, however, from

the totals in the SUTs. Given that all the omitted adjustment items relate to final uses, the GDP

calculated from the expenditure side (308,647) is still correct and identical to the results shown in

connection with box 2.10. It is always possible to include the corresponding adjustment items in

the final IOTs to arrive at the correct totals for household final consumption expenditure, exports

and imports, as illustrated in Table 8.7, which is consistent with tables 8.2, 8.4 and 8.6.

8.20. Ideally, the adjustment items should be included in the IOTs for the purpose of consistency

with the national accounts framework and the SUTs, and for complete coverage of the economy

in analytical uses, as illustrated by table 8.7 (which, to illustrate the point, also shows a net export

presentation – this presentation is covered in more detail in chapter 12). Different practices

regarding these adjustment items have, however, been emerged across countries and international

organizations. Thus the Austrian IOTs shown here replicate the tables contained in the Eurostat

database. On the other hand, the IOTs in the OECD input-output database include all adjustment

items, to ensure full consistency with national accounts data.

8.21. The problem of how to deal with the adjustment items necessarily arises when compiling

empirical IOTs adjustment rows. For ease of exposition and in order not to overload the

Households NPISH

General

government

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture (1) 2 316 4 344 4 101 15 19 6 800 963 123 - 42 938 1 982 8 782

Manufacturing (2) 1 091 42 919 6 362 7 534 4 369 2 951 65 227 12 631 327 9 426 1 122 1 393 96 280 121 178 186 405

Construction (3) 73 1 883 9 927 1 969 3 890 1 279 19 021 1 402 24 323 - 38 563 26 250 45 272

Trade, transport and communication (4) 239 13 805 2 109 18 364 5 909 2 846 43 272 55 600 4 549 9 207 239 334 21 550 91 479 134 750

Finance and business services (5) 370 9 320 4 530 17 653 29 781 7 564 69 219 36 524 1 006 9 781 0 - 177 11 156 58 289 127 508

Other services (6) 6 286 51 1 066 453 1 629 3 490 13 045 5 416 53 116 113 - 105 1 567 72 153 75 643

Total (7) 4 094 72 557 22 984 46 687 44 418 16 288 207 028 120 165 5 416 58 997 52 973 1 257 1 471 131 053 371 332 578 360

Imports (8) 811 61 469 4 846 12 485 6 136 3 731 89 479 23 087 1 495 13 575 926 1 381 21 350 61 814 151 293

Taxes less subsidies on products (9) 78 862 226 1 333 1 839 2 646 6 984 22 810 557 2 870 152 7 397 26 794 33 778

Total at purchasers’ prices (10) 4 983 134 889 28 056 60 506 52 393 22 665 303 492 166 063 5 416 61 050 69 418 2 335 2 859 152 800 459 939 763 431

Compensation of employees (11) 411 25 857 10 216 38 422 28 962 40 475 144 343

Other taxes less subsidies on produ

(12) - 1 446 717 545 1 762 2 267 1 014 4 858

Consumption of fixed capital (13) 1 620 11 519 1 422 10 172 21 759 6 977 53 469

Net operating surplus/net mixed

income

(14) 3 214 13 423 5 032 23 889 22 127 4 512 72 198

Gross operating surplus/gross

mixed income

(15) 4 834 24 942 6 455 34 061 43 886 11 489 125 667

GVA (16) 3 799 51 516 17 216 74 245 75 115 52 978 274 868

Total input at basic prices (17) 8 782 186 405 45 272 134 750 127 508 75 643 578 360

PRODUCTS

GVA

Agricul-

ture

Manufac-

turing

Construc-

tion

Exports

Total

Output at

basic

prices

Other

services

Total

Final consumption expenditure

Gross f ixed

capital

formation

Changes

valuables

Changes

inventories

PRODUCTS

FINA L USE

Trade,

transport and

communication

Finance and

business

services

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

255

presentation of the SUTs and IOTs, these additional rows are not included in the numerical

examples in this Handbook. Their absence does not mean that they have been distributed by

products and thus included in the upper part of the SUTs and IOTs, an arrangement which, for

example, some analytical users of IOTs would prefer.

Table 8.7 Input-output table for domestic output at basic prices, net exports with

adjustment items

Millions of euros

Table based on 2011 figures from Austria

C. Compilation of the imports use table

8.22. Compilation of the imports use table may be a difficult process because direct information

for the estimates of imported products by industry and by final use may not be available or only

available in limited cases. As a result, direct information has to be supplemented by reasonable

assumptions and indirect techniques. As noted earlier, in a large rectangular SUTs system, many

homogenous products can be identified which have to be imported from abroad. Thus, the

allocation of goods and services in the use table for domestic output and the use table for imports

is easier if a large rectangular SUTs system is available. The two main approaches to the

compilation of Imports Use tables are presented below: the first is based on the availability of

directly collected data and the second on assumptions regarding the imports.

1. Using directly collected data

8.23. There are two major sources for direct information for the imports use table:

House-

holds

NPISH

General

govern-

ment

(1) (2) (3) (4) (5) (6) (7)

(8) (9) (10) (11) (12) (13) (14) (15) (16) (17)

Agriculture (1) 2 492 6 065 8 248 29 34 8 877 2 042 170 - 32 996 - 3 271 - 95 8 782

Manufacturing (2) 1 708 98 765 10 754 13 040 5 768 5 893 135 928 33 525 1 749 21 736 1 929 2 737 113 392 - 124 590 50 477 186 405

Construction (3) 73 2 148 10 131 2 016 3 934 1 282 19 585 1 402 24 323 - 38 563 - 563 25 687 45 272

Trade, transport and communication (4) 248 15 900 2 258 23 514 7 586 3 183 52 690 56 185 4 575 9 951 240 363 25 729 - 14 984 82 060 134 750

Finance and business services (5) 377 10 851 4 627 19 180 32 755 7 872 75 662 36 669 1 006 10 254 - 177 11 156 - 7 061 51 846 127 508

Other services (6) 6 297 51 1 174 482 1 756 3 765 13 429 5 416 53 163 113 14 1 567 - 824 71 878 75 643

Total at basic prices (7) 4 905 134 027 27 830 59 173 50 554 20 019 296 507 143 252 5 416 60 492 66 548 2 182 2 852 152 403 - 151 293 281 852 578 360

Taxes less subsidies on products (8) 78 862 226 1 333 1 839 2 646 6 984 22 810 557 2 870 152 7 397 26 794 33 778

Total at purchasers' prices (9) 4 983 134 889 28 056 60 506 52 393 22 665 303 492 166 063 5 416 61 050 69 418 2 335 2 859 152 800 - 151 293 308 647 612 138

CIF/FOB adjustments (10) - 97 97

Direct purchases abroad by

residents

(11) 6 675 - 6 675

Purchases on the domestic territory

by non-residents

(12) - 12 945 12 945

Total (13) 4 983 134 889 28 056 60 506 52 393 22 665 303 492 159 792 5 416 61 050 69 418 2 335 2 859 165 648 - 157 871 308 647 612 138

Compensation of employees (14) 411 25 857 10 216 38 422 28 962 40 475 144 343

Other taxes on production (15) - 1 446 717 545 1 762 2 267 1 014 4 858

Consumption of fixed capital (16) 1 620 11 519 1 422 10 172 21 759 6 977 53 469

Net operating surplus (17) 3 214 13 423 5 032 23 889 22 127 4 512 72 198

Gross operating surplus (18) 4 834 24 942 6 455 34 061 43 886 11 489 125 667

GVA (19) 3 799 51 516 17 216 74 245 75 115 52 978 274 868

Total input at basic prices (20) 8 782 186 405 45 272 134 750 127 508 75 643 578 360

INDUSTRIES

FINA L USE

Total

output at

basic

prices

Agricul-

ture

Manu-

facturing

Construc-

tion

Trade,

transport

and commu-

nication

Finance and

business

services

Other

services

Total

Total consumption expenditure

Gross

fixed

capital

formation

Changes

in valu-

ables

Changes

in inven-

tories

Exports

Less

imports

Total

PRODUCTS

ADJUSTMENTS

VALUE ADDED

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

256

• Business surveys which could be developed further, for example, for each industry, more

product detail of imports of goods and services by type of product and more information on

imports of services

• Trade surveys which provide extensive details of imports of goods. Traditionally, the

customs department collects foreign trade statistics.

8.24. The data linking of units’ data from trade and business registers provides another source,

also ensuring some degree of coherence between the two sources.

8.25. Business surveys (annual or for benchmarked years) generally collect details such as sales

by type of product and purchases by type of product. These surveys could be expanded to include

additional questions useful for the imports use table, such as the value of purchases of imports of

goods and the value of purchases of imports of services.

8.26. For certain industries, it is important to ask specific questions on the imports of goods and

services for certain specific products. For example, for the sugar refining industry it may be

important to have information on the purchases of sugar beet separate from sugar cane. The

economy may have little, or no, domestic production of one or both of these products and would

have to rely on imports, which at the two-digit level would appear in the same product

classification. This approach of asking for specific details may apply also to other products such

as, for example, tobacco and tobacco leaf.

8.27. This use of the business survey data would help to provide an industry total of direct

imports. These values could then be developed and matched with imports of goods from the trade

data suppliers by product, to help develop the body of the imports use table in terms of intermediate

use. This works for direct imports but less so for indirect imports, such as imports sold to

manufacturers via resident distributors such as wholesalers. Imports by retailers could, however,

be assumed in the main for final use categories. Again, this would need scrutiny, for example small

items (not purchased in bulk) like stationery, may be purchased by businesses from retailers. For

imports of services, indirect imports should not be an issue, because by their nature services cannot

be resold – in other words, they are used only once.

8.28. International trade surveys provide much of the detail by product. Further work is often

required, however, to identify the importing industry or industries. For imports of goods, for

example, very detailed international trade data from the Harmonized System can be more easily

used to link imports to specific products that are used by industries as intermediate consumption

and those products that are components of specific categories of final use. These data could be

developed with the data collectors to identify the industry to which the importer is classified and

the value of imports by product.

8.29. In terms of imports of services, a product-by-industry matrix should be generated for each

of the 12 components forming trade in services (applicable to both imports and exports), as

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

257

illustrated in Box 8.1, which would also highlight various improvements required to imports of

services in the balance of payments. For business services, a separate matrix can be generated

using imports of services data from business surveys providing details for other variables such as

sales and purchases.

8.30. There are various existing sources of data used for imports of services by product, such as

the following:

• International trade in services – which collects data from businesses covering their imports

and exports of services by product. This is a statistical survey which has advantages over

administrative data.

• International passenger survey – which collects expenditure data by product by travellers, at

the point of entering or exiting the resident economy (for example, airports and ports). There

is a need to separate the expenditure by business travellers (recorded as intermediate

consumption) and expenditure by households (recorded as households’ final consumption

expenditures).

• Specific sources capturing imports of services such as shipping, air transport, financial

services, and others.

• Development of non-traditional survey-type sources such as credit card data and

international microdata-sharing, for example between national statistical offices.

8.31. For some of the standard services components, however, there are specific issues which

need careful handling, for example, disbursements, freight costs, royalties, and so forth.

Box 8.1 Standard services components of BPM 6

1 Manufacturing services on physical inputs owned by others

2 Maintenance and repair services not included elsewhere (n.i.e.)

3 Transport

4 Travel

5 Construction

6 Insurance and pension services

7 Financial services

8 Charges for the use of intellectual property not included elsewhere (n.i.e.)

9 Telecommunications, computer, and information services

10 Other business services

11 Personal, cultural, and recreational services

12 Government goods and services n.i.e.

Source: IMF (2009)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

258

2. Alternative approach

8.32. As noted, there may be a lack of source data (unless surveys are designed to collect detail

data and flows from the trade industry) to estimate so-called “indirect” import use, that is, the

ultimate destination of the use of imported goods, and not, say, the wholesaler acting as an

intermediary. This situation will often require strong assumptions and indirect techniques to

allocate the use of imports by product for each industry and by category of final use by product.

8.33. A widely used approach in estimating import use by product across using industries and

categories of final uses is the application of the import proportionality or comparability

assumption. This assumes that imports are used in the same proportion across all industry

intermediate inputs and final uses (except exports and allowing for imports for re-export purposes).

This is often a two-step procedure in which the ratio of imports to domestic supply is first

calculated and is then applied to each product that is used by industries as intermediate inputs to

production and by categories of final uses (except exports). For example, if imports of semi-

conductors represent 50 per cent of the domestic supply of semi-conductors, then it is assumed

that each industry that purchases semi-conductors purchases 50 per cent of them from foreign

sources. This procedure results in the same distribution of imported products across a given row

in the use table, thus providing another reason to work at the most detailed level of products

available within the SUTs system, where there are likely to be fewer users of very specific

products. Thus this procedure works much better with large numbers of products (for example,

10,000) as opposed to, say, fewer than 100 products. With foreign trade data, there tends to be

much more data available by products relative to other data sources. The procedure should,

preferably, be applied at basic prices.

8.34. Certain products can be very straightforward to allocate. For example, there are very few

users of imports of crude oil in the domestic economy, whereas imports of food are used by many

industries and households, or refrigerators may be used by households or in gross fixed capital

formation. Nevertheless, the main task remains, namely, to attribute allocation ratios and

percentages for each category of imported products across using industries and categories of final

uses.

8.35. For this procedure, BEC, or a national variant of BEC, can also be used when applying the

import proportionality assumption. BEC allocates imports of goods into categories of intermediate

goods, consumer goods and capital goods. The elements of BEC are the subclasses of SITC, which

are defined in terms of the Harmonized System. Only broad use categories are distinguished,

however, and these are of less help for the intermediate uses by specific industry. Nonetheless,

BEC may be helpful in categorizing the products for the imports use table.

8.36. It is worth noting that it is time and resource-consuming, especially for the first time, to

generate detailed allocation ratios and percentages covering intermediate use and final uses for

each category of imported products.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

259

8.37. In defining the allocation percentages, there is a need also to keep in mind that, as a

consequence of secondary output, products are also used in industries where they might not be

typical inputs. This may have already been addressed in the use table.

8.38. It is important that the import proportionality assumption or related ratio procedures be

used only after direct information about import use has been compiled. Lastly, once the

proportionality assumptions have been applied, it is essential to evaluate the generalized results

for reasonableness, and to adjust these percentages in the light of an understanding of how the

specific economy operates with regard to production chains and purchases of products by final

use.

8.39. The product imbalances, and the balancing process, can often be used to correct for

implausible results from an initial allocation based on proportions.

8.40. Although the proportionality approach is not time-consuming, the allocation percentages

can generally also be applied for other years without any large changes. Usually a great share of

total imports will fall under only a few specific product classifications (for example, manufacturing

products purchased by manufacturing industries) and efforts should be concentrated on those as,

to a large extent, they determine the quality of the resulting imports use table and thus the accuracy

of GDP in volume terms and the distribution of GVA by industry.

8.41. A difficult category of final use in respect of the allocation of imports of goods is changes

in inventories. First, it is often assumed that the import share of semi-finished and finished products

is zero. Second, the category of changes in inventories is a balancing item between the inventories

at the end of the period minus the inventories at the beginning of the period without knowing the

inflows and outflows over the period. As a result, the sign of the estimate can be positive or

negative and, in the latter case, this needs to be handled with caution, otherwise it could lead to

implausible values.

8.42. Inventories of finished goods should, however, be treated separately from imports of

finished goods for resale without further processing. The latter are goods likely to be held mainly

by distributors.

8.43. Lastly, in the process of balancing it must be expected that the procedure may need to be

repeated in order to achieve plausible results. As noted, the allocation shares might need to be

corrected on the basis of implausible results, which may include negative use elements. Naturally,

the allocation of imports to a use element with a zero entry is not permissible and may indicate a

wider problem with the level of aggregation.

D. Specific issues in the compilation of the imports use table

8.44. There are specific issues that need careful consideration when compiling the imports use

table. These include the recording of goods sent abroad for processing, investment goods repaired

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

260

abroad, imports for re-export purposes, and direct expenditures by residents abroad. These specific

issues are briefly discussed in this section.

8.45. Other issues include: arrangements within multinational enterprises including transfer

pricing; contract manufacturing and manufacturers; factoryless goods production and processors;

foreign direct investment relationships; intellectual property products – ownership and

cross-border use; international labour movement and remittances; Internet trading; limitations of

national data collections; merchanting of goods and services; ownership of property abroad;

special-purpose entities; and toll processing and processors. The guide to measuring global

production (UNECE, 2015) provides much more detail on how handle these types of issues.

1. Goods sent abroad for processing

8.46. Sending materials or partly finished goods to another affiliate or non-affiliate enterprise for

processing is an established practice which has become more common with the drop in transport

costs, specialization among enterprises and the emergence of new production sources. The

enterprise processing the items may be resident in the same country as its client or it may be abroad.

8.47. The procedure of sending material for processing is called “goods sent abroad for

processing”. This practice is very common in industries such as wearing apparel (clothing);

chemicals and the manufacture of electronic and metal goods. One variant of particular interest for

the national accounts and balance of payments is goods sent abroad for processing, where the unit

in country A (the principal) makes a contract with the unit in country B (the contractor) under

which B transforms in a substantive way raw materials or semi-processed goods sent by country

A. Throughout the process the principal maintains legal ownership of the raw materials and semi-

processed goods, and also of the processed goods. The principal pays the contractor a fee for the

processing.

8.48. The issue of how to record the goods sent abroad for processing in national accounts,

including SUTs and balance of payments, has been the subject of extensive discussions in

connection with successive versions of the SNA and the BPM. The central question has been

whether to impute a value of the goods when sent abroad for processing and subsequently for the

processed goods when returned to the legal owner and, in this approach, to impute transactions

even though no change of ownership has taken place (the gross principle), or just to record the

processing fee as a separate service delivered from the processor to the principal (the net principle).

8.49. In the following sections, the different implications of applying either the gross or the net

method are illustrated by numerical examples, both for goods sent for domestic processing and for

goods sent for international processing. Against this background, the international

recommendations established by the 2008 SNA and BPM 6 are explained, with the inclusion of

possibilities to deviate from these recommendations to reinforce desirable properties of the SUTs.

The Eurostat Manual on Goods Sent Abroad for Processing (Eurostat, 2014b) provides further

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

261

details in this regard. Lastly, related measurement problems in current economic statistics are also

considered.

(a) Domestic processing

8.50. In the example in Table 8.8, a principal unit classified in Industry 1 sends semi-processed

goods (Product A) for further processing to a contractor unit classified in Industry 2. The contractor

does not pay for the material received from the principal unit. The value of the goods sent for

processing is 100. The value of the goods after processing, assumed to be finished goods requiring

no further processing (Product B) is estimated at 180. Processing fees in this example are set, for

the sake of simplicity, at 80. In this example, Industry 1 and Industry 2 could also be interpreted

as two units belonging to the same industry but, for maximum clarity of the exposition, a two-

industry case is chosen.

Table 8.8 Processing within the country

8.51. Under the gross treatment, transactions of the values of 100 and 180 are imputed, the 100

being the output of Industry 1, and the 180 being the output of Industry 2. As the 180 is assumed

to consist of finished goods, they are not, as it were, “returned” to the owner industry but enter the

product balance as an output of Product B from Industry 2, even though this industry is not the

legal owner. From the point of view of Industry 1, it would be goods for resale which are therefore

not recorded as a flow from Industry 1 in the system, except if held in inventories at the end of the

period. The processing fees do not appear separately, as they are included in the output of Product

8.52. Under the net treatment, the processing fee of 80 is the only output of Industry 2 and it is

used as intermediate consumption by Industry 1. The processing fee will be classified as a service

and not a good. The output of Industry 1 will be 180. As processing fees can usually be found in

current industrial statistics, there are no imputations associated with implementing the net

treatment.

8.53. It is noted that the GVA in the two industries (30 and 50 respectively) are identical for the

two alternative treatments but the input structures are quite different. For Industry 1, the gross

Gross recording Net recording

Industry 1 Industry 2 Imports Industry 1 Industry 2 Exports Industry 1 Industry 2 Imports Industry 1 Industry 2 Exports

Product A 100 100 Prod A

Product B 180 Prod B 180

Processing fees Processing fees 80 80

Other intermediate

consumption

70 30

Other intermediate

consumption

70 30

GVA 30 50 GVA 30 50

Total output 100 180 0 100 180 0 Total output 180 80 0 180 80 0

Note: Industry 1 is the principal and Industry 2 is the contractor.

Supply Table

Use Table

Supply Table

Use Table

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

262

treatment results in a much higher GVA share in output than the net treatment, and vice versa for

Industry 2.

(b) International processing

8.54. In Table 8.9, the gross and net treatments are illustrated for goods sent abroad for

processing. The numerical example is essentially identical to that presented for domestic

processing, except that now it is assumed that the principal is a resident of Country I, and the

processor a resident of Country II, and also that the two-country case involves export and import

transactions, either actual or imputed.

Table 8.9 Goods sent abroad for processing

8.55. Under the gross treatment, the 100 output of semi-processed goods (Product A) from

Industry 1 in Country I is the exports of goods from Country I and imports of goods into Country

II, where it is used as intermediate consumption in Industry 2. The output of 180 from Industry 2

(Product B) in Country II is exported to Country I. As the 180 is assumed to consist of finished

goods, they are not returned to the owner industry but enter the balance of Product B in Country I

as imports of goods. As there is no change of ownership related to the 100 and the 180, the values

of these transactions must be imputed, but as later noted these cross-border movements of goods

will usually be included in and valued for the external merchandise trade statistics.

8.56. Under the net treatment the semi-processed goods (Product A) disappear, and the processed

goods (Product B) will appear as produced in Country I, as actual output from Industry 1. Only

processing fees will appear in international trade, under services. As international processing fees

are usually covered both by current industrial statistics and by statistics on international trade in

Gross recording (1993 SNA)

Principal (Country I) Contractor (Country II)

Industry 1 Industry 2 Imports Industry 1 Industry 2 Exports Industry 1 Industry 2 Imports Industry 1 Industry 2 Exports

Product A 100 100 Product A 100 100

Product B 180 Product B 180 180

Processing fees Processing fees

Other intermediate

consumption

Other intermediate

consumption

GVA 30 GV A 50

Total output 100 0 180 100 0 100 Total output 0 180 100 0 180 180

Net recording (2008 SNA)

Principal (Country I) Contractor (Country II)

Industry 1 Industry 2 Imports Industry 1 Industry 2 Exports Industry 1 Industry 2 Imports Industry 1 Industry 2 Exports

Product A Product A

Product B 180 Product B

Processing fees 80 80 Processing fees 80 80

Other intermediate

consumption

Other intermediate

consumption

GVA 30 GV A 50

Total output 180 0 80 180 0 0 Total output 0 80 0 0 80 80

Supply Table

Use Table

Supply Table

Use Table

Supply Table

Use Table

Supply Table

Use Table

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

263

services, there are no imputations associated with the implementation of the net treatment of goods

sent abroad for processing.

8.57. As for domestic processing, the GVA remains the same for the two approaches but the

input structure is significantly different.

8.58. It should be noted that these numerical examples are highly stylized, in order to emphasize

the main characteristics of the gross and net treatments. In practice, the difference between the

value of the finished goods and the semi-processed goods may not be equal to the processing fee

paid, either because the prices have changed over the processing period or because part of the

increase in the value of the finished goods reflects the embodiment of intellectual property or

trademarks owned by the principal. It may also be that the processed goods require further

processing by the principal, in which case an additional entry of intermediate consumption (180 in

the example following the gross treatment) would be necessary, and output increased accordingly.

It should be noted that these issues will only arise under the gross treatment. The net treatment

automatically resolves them.

8.59. In practice, the situation may be much more complicated and prove a significant

measurement challenge. For example, goods are often not really sent abroad; for example, they

may be purchased abroad. On the other hand, goods do not necessarily return after processing;

they can be shipped immediately to a third country for final use.

8.60. The treatment of goods sent for processing, as laid out by the 1993 SNA and BPM 5, was

quite complex but the main recommendations were that domestic processing should be based on

the net treatment (except when transactions take place between two establishments belonging to

the same enterprise, in which case the gross treatment should be used), and that international

processing should be based on the gross principle. It was indicated in the 1993 SNA that this

treatment of goods for processing was designed to facilitate input-output analysis.

8.61. The question raised leading up to the discussion of 2008 SNA was whether there was still

a valid reason to record goods for international processing on a gross basis or if the advent of

globalization and the growing volumes of goods processed abroad suggest that a change in practice

would be appropriate. In response to this discussion, the recommended treatment of goods for

processing was changed in the 2008 SNA and BPM 6, in that the change in economic ownership

principle was prioritized over the actual movements of goods and physical technology, so that all

goods sent for processing (both domestically and internationally) should be treated according to

the net principle. This much simplified recommendation is, at the same time, more in line with

company accounts and principally avoids imputations but may still require adjustments to ensure

that the domestic activity is consistent with that recorded in the imports and exports data.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

264

8.62. The new treatment of goods for processing potentially leads to larger variation in input-

output coefficients and it is important to put this change into perspective. In the 1993 SNA, the net

treatment was already applied if the goods were sent for processing to a non-affiliated domestic

processor. Moreover, input structures change for a number of reasons: for example, because of

changes in product mix, more or less use of semi-finished products, changes in capital and labour

intensities, and outsourcing of services. The change in goods sent abroad for processing merely

adds to the other changes in the 2008 SNA, which in turn change the input-output coefficients.

8.63. The effects on the input structure of the alternative treatments of goods sent for processing

are illustrated in Table 8.10 for Industry 1. The input structures when using the gross or net

treatment are taken from Table 8.8. To get an idea of what the input structure would have looked

like if no goods had been sent for processing, the processing fee has been decomposed into “other

intermediates” and “GVA”, using the input structure of Industry 2, and the components added to

the “net treatment” column to obtain the result in the last column. When looking at the GVA to

output ratios of the three input structures, they are obviously quite different. From the standpoint

of input-output technology, the structure in the last column would be preferred as neither of the

first two columns would be a good approximation. In practice, when following the 2008 SNA, the

input structure of an industry sending goods for processing would probably represent a weighted

average of the structures in the second and the last column, as the industry will probably also have

some primary output that has not been sent for processing. When the share of goods sent for

processing changes, the overall input structure of the industry will change even though no

technological change has taken place.

Table 8.10 Industry 1 – Alternative input structures

8.64. As already emphasized, Table 8.10 provides a stylized example. There are various business

activity models, and mixed models, not just within an industry but within an individual business –

with the impact of globalization, this is becoming more prevalent. For example, the same oil

company can cover the following three activities:

• Extract the crude oil, refine it and then sell the refined petroleum. This activity has the

full input structure as indicated in the final column of Table 8.10.

• Extract the crude oil, sell the crude oil for processing to another company and then

purchase the refined petroleum from that company for resale without any further

Gross

treatment

Net treatment

Processing fee

breakdow n

Structure w hen

no processing

Processing fees

80 -80

Other intermediates 70 70 30 100

GVA 30 30 50 80

Output 100 180 180

GVA/Output 0.30 0.17 0.44

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

265

processing. Note, in this case, that there is a change in economic ownership and the oil

company undertakes no refining.

• Extract the crude oil, pay another company a processing fee for refining the crude oil

(note, no change of economic ownership), and then sell the refined petroleum. Again,

the oil company undertakes no refining. This activity is indicated in the second column

of Table 8.10.

8.65. If the three activities are separable, then they would reflect different input structures. This

illustrates, however, that an individual company, and an industry, can reflect a mix of activities

and input structures from both the point of view of the outsourcing industry and the point of view

of the processing industry. This makes the data collection, measurement and interpretation of the

industry input structures much harder.

8.66. In paragraphs 28.18 and 28.19, the 2008 SNA outlines two ways to proceed that would

retain the technical interpretation:

• Option 1: split the economic activity into two: one processing on own account, and one for

goods sent for processing.

• Option 2: use the gross recording approach.

8.67. It should be emphasized that option 1 is the recommendation of the 2008 SNA and, for

goods sent abroad for processing, BPM 6. Option 2 is shown as a supplementary presentation that

may be adopted for reasons of continuity with past practices. Option 1 more accurately reflects the

economic processes taking place, while option 2 focuses on the physical transformation process

(2008 SNA, para. 28.20). The idea – when seen in the perspective of a published IOT – is rather

theoretical, however, as only in very exceptional cases would it be realistic to have the tables

include such dual branches, and furthermore this way of reasoning could be extended to cover all

other reasons for differences in input structures and thus lead to an expansion of the number of

economic activities ad infinitum. In these cases, where the individual producer units represent

blends of traditional production and contracting-out, this approach would anyway not be feasible.

8.68. Even in gross recording, the input structures of principals and contractors would probably

be quite different from the input structure when processing on own account. Specific adjustments

may, however, be appropriate in certain cases. If, for example, the share of oil refining made on

contract basis (as a contractor) varies drastically over time, it might be justified to apply the gross

treatment in this case, while on the other hand an increasing trend in goods sent abroad for

processing should not be counteracted. Practical data problems will also limit the options. Whereas

it may be perfectly realistic to implement the gross treatment in the above oil refinery case, it

would require industrial insight and data beyond any realistic possibility to adjust for the latter

general trend.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

266

8.69. The increasing activity by businesses involving the sending of goods for processing,

whether domestically or internationally, will in any event affect the input-output coefficients (both

for the principal and the contractor), and imputing sets of data that are completely detached from

the actual economic transactions and their statistical recording is not a viable way of dealing with

the complications arising from the institutional changes taking place in the economy. The growth

of outsourcing under the globalization of markets means that these inherent institutional changes

are more rapid and more significant, and this is a phenomenon with which the input-output

compilers and analytical users will have to accommodate. This will not be done by making an

artificial world of their own that denies these structural changes, rather than exposing them.

(d) Data and balancing issues

8.70. There are three main data sources involved in preparing the industry and product estimates

for goods sent for processing in the SUTs:

• Industrial statistics in which manufacturers provide information on receipts for doing work

to the orders of others (as contractor) and subcontracting expenses (as principal) but in

which manufacturers are not asked to estimate values for the materials received for

processing or for the processed goods when returned to the principal

• Merchandise trade statistics in which estimated values for the goods sent abroad for

processing – and returned from processing abroad – are included as a border crossing

principle, rather than following a change of ownership principle

• Statistics on international trade in services in which in-going and out-going processing fees

are recorded

8.71. If data from all three data sources are just taken at face value and entered into the SUTs,

major imbalances will obviously result.

8.72. Under the net treatment, the merchandise trade related to goods sent abroad for processing

will have no counterpart in the data recorded by domestic industrial statistics and must therefore

be removed from the merchandise trade data. This will often be possible, where these types of

goods have been given a special code in the customs procedure. If, however, they are

indistinguishably included, a more comprehensive approach may be needed, based on specific

information on industrial practices and processes. Under the net approach, the processing fees

recorded in the statistics on international trade in services should be directly applicable, and in

principle consistent with (but not equal to) the processing fees recorded in domestic industrial

statistics, as these also include payments for domestic processing.

8.73. In order to obtain qualitative data on the fees paid and received for international processing,

detailed comparisons of different data sources at the firm level may be required. Survey data on

the import and export of services may provide the basis for the fee paid and received but may be

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

267

incomplete in terms both of respondents and of product detail. These data must be supplemented,

and reconciled, with data on industrial production and international goods flows under processing.

They must also be consistent with firm data on turnover and costs, ensuring that the domestic

account and rest of the world account are consistent.

8.74. Under the gross treatment, the imbalance between supply and use of goods would be

removed by imputing additional inputs and outputs for domestic industries corresponding to the

merchandise imports and exports related to processing. As the processing fees are in this case

embedded in the value of the processed goods, they should be removed from the statistics on

international trade in services before these data are entered into the SUTs.

8.75. In order to keep track of the conceptual and data-linked complexities of goods sent for

processing, both domestically and internationally, it is recommended that a subsystem be

established in which all the related goods and services balances are separately set out and analysed,

so as to secure full coverage and consistency in the SUTs before the data are entered into the full

system.

2. Investment goods repaired abroad

8.76. Investment goods which are sent abroad for major repair result in substantial amounts of

value being created in the reconstruction. Both the export and the re-import are part of the import

and export flows. In the case of minor repair, maintenance or servicing, however, the flows

concerned are not to be recorded under imports and exports. In the case of major repairs, similar

problems of recording in the SUTs framework occur. Thus, for practical reasons, it may be

assumed that it usually would be only a minor repair. Furthermore, it could be assumed that the

cross-border transport of investment goods for repair is quite rare and thus negligible (with the

exception of products like aircraft and ships, where the activity and values involved are often very

significant).

3. Imports for re-export purposes

8.77. Another challenge in the compilation of the imports use table is posed by re-exports. Re-

exports are transactions of goods which were previously imported with a change in economic

ownership and then exported without any substantial transformation. These re-exports are included

as exports in foreign trade statistics. In the case of products that are not produced domestically,

any exports of these products could easily be identified as re-exports. It should be noted that – with

the exception of transport margins – re-exports of services are, by their very nature, not possible,

thus the identification of re-exports is only a problem for goods.

4. Direct expenditures by residents abroad

8.78. A further problem which may also be of some importance concerning the data involved is

posed by direct purchases abroad by residents in connection with tourism. These direct purchases

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

268

abroad by residents should cover all purchases of goods and services made by residents while

travelling abroad for business or pleasure. Such purchases are part of the import flows and need to

be estimated on a product basis. As a result, these purchases must be allocated to intermediate use

in the case of business travellers and to household final consumption in the case of private

travellers.

5. Transit trade

8.79. These are goods admitted under special customs procedures that allow the goods to pass

through the territory. They are excluded from the general merchandise of the territory of transit.

The issue of transit trade takes various forms, including quasi-transit trade and others, and poses

challenges in terms both of statistics and of measurement. For some countries, these may be

significant, in particular with large ports. Transit trade, also referred to as “simple transit”, and

quasi-transit trade flows do appear in merchandise trade statistics and should be excluded from

national accounts and balance of payments.

E. Enhancements to the imports use table for analytical uses

8.80. In addition to those enhancements described above, the analytical potential of the imports

use table for users is considerably enhanced if the table also shows supplementary classifications

such as the distinction between “competitive imports” and “complementary imports”, or imports

subdivided by regions, such as country or region of origin.

8.81. Competitive imports are products that are also domestically produced and thus are

consequential in estimating an accurate domestic use table. Complementary imports (also referred

to as “non-competitive imports”) are products that are not domestically produced.

8.82. This distinction is of analytical interest as both types of imports can be expected to have a

different relationship with and importance for the national economy. Competitive imports may be

the subject of economic analysis concerning substitution policies and effects. Complementary

imports, as products not produced in the national economy, are sometimes vital and analyses may

focus on the impact of changes in their prices or volume.

8.83. In theory, the distinction between competitive and complementary imports seems to be

clear. In practice, however, a number of borderline cases need to be solved. For the validity of this

distinction, the product level of disaggregation is of utmost importance. Even at a very detailed

product level, it is sometimes difficult to classify the products as competitive or complementary.

Furthermore, this classification may not be stable over time.

8.84. Compared to the distinction between competitive and complementary imports, a

geographical breakdown of the use table of imports is easier to compile as there may be data

problems but no basic conceptual problems. Where goods are concerned, information on the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

269

geographical origin of the imports is usually available in foreign trade statistics. For services, the

data situation concerning the geographical breakdown of imports is less favourable.

8.85. The main problem in compiling use tables on imports with a geographical breakdown is

how to allocate a single product imported from two geographical regions to the respective use

categories: should it be allocated proportionally to the assumed users? Similar questions already

arise when compiling the use table of imports without geographical breakdowns: is the import

share of an imported product the same in all use categories? In consequence, a geographical

breakdown might need such additional assumptions to be made, in order to allocate the imports.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

271

Chapter 9. Compiling SUTs in volume terms

A. Introduction

9.1. One of the major objectives of national accounts is to provide comprehensive and coherent

data which can be used for analysing and evaluating the performance of an economy. Data on the

real growth of major economic flows such as production, household consumption, capital

formation and exports serve as inputs for formulating economic policy. Furthermore, the national

accounts data play a key role in helping to investigate the causal mechanisms within an economy.

9.2. Estimation of the parameters for macroeconomic models by applying econometric methods

requires consistent time series of national accounts data with a focus on annual changes. The

decomposition of annual current price changes into price changes and volume changes is therefore

an important aspect of the compilation of national accounts.

9.3. Contrary to data in current prices, much of the data in volume terms cannot be directly

observed. They must be derived from current price data combined with appropriate price and

volume indicators, meaning that estimates in volume terms are derived more through modelling or

based on proxies rather than estimates used when formulating data in current prices. In addition,

the choice of index formulae influences the result of the estimates in volume terms.

9.4. The calculation of price and volume changes for the transactions of goods and services in

the national accounts is ideally supported through the use of the SUTs framework. When

established in an accounting framework, the volume indices and deflators of several variables at

different levels of aggregation are interrelated in a systematic way. By applying an appropriate

combination of price and volume index formulae, all the identities and relationships of SUTs in

current prices are maintained in the SUTs in volume terms.

9.5. When balanced both in current prices and in volume terms, the SUTs ensure coherent and

consistent deflation of the components of the production and expenditure approaches to measuring

GDP, together with coherent and consistent estimates of price and volume indices. Another

advantage of compiling price and volume measures within the SUTs framework is that price and

volume measures can be derived for such important balancing items as GVA and GDP through

applying the so-called “double deflation” approach – this is the recommended SNA approach to

estimating GVA in volume terms.

9.6. This chapter focuses on the compilation of SUTs in volume terms. In section B it is

acknowledged that there are alternative approaches to compiling SUTs in volume terms. Section

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

272

C describes the compilation of SUTs in volume terms using the H-Approach and sections D and

E present more details on how to deflate the various components of the SUTs. Lastly, section F

presents some considerations for compilation of IOTs in volume terms. The balancing of SUTs in

current prices and in volume terms (and also at basic prices and at purchasers’ prices) is covered

in chapter 11.

B. Recognition of alternative approaches

9.7. Based on issues like the availability of data, resources and time, it is important to note that

there are different ways of deflating SUTs, and also different sequences in so doing. In turn, these

variations will also generate different balancing schemes and, most important, different degrees of

quality for the detail and the aggregates.

9.8. This chapter focuses on a recommended approach, although alternative approaches are

available. One such alternative is the deflation of uses at purchasers’ prices as inputs to generate

the SUTs in volume terms and balance the rest of SUTs, which may be deflated at basic prices (or

producer prices), for example, output. This approach may be easier to implement but has in-built

incoherence and inconsistency reflected through balancing different valuations. Albeit balanced

out, the impact on the quality of the aggregates is implicit (and not explicit), and, in addition, for

some areas like gross fixed capital formation is clearly sub-optimal, as appropriate price indices at

purchasers’ prices are not usually available.

9.9. In addition, SUTs in volume terms for one period can be compiled using SUTs in current

prices for one period and deflators. The recommended approach, however, includes a time-series

dimension.

9.10. Another alternative approach is to remove taxes, subsidies, trade and transport margins and

imports, to deflate domestic output at basic prices (with weighting for exports using export price

indices) and to deflate imports using import price indices. Trade and transport margins and taxes

and subsidies in volume terms are estimates using rates of the previous year and the volume change

at basic prices.

9.11. The underlying principles of this approach form the recommended approach covered in

detail in this chapter (referred to as the H-Approach), an overview of which is given in Figure 9.1.

This is a transparent, coherent and consistent approach compiling SUTs both in current prices and

in previous years’ prices and ensuring the better matching of the valuation of current price values

with appropriate price indices.

9.12. There are many benefits of the H-Approach as, for example, it allows for the data

confrontation to take place in both current prices and in volume terms and also ensures consistent

deflation across the national accounts. In addition, the incorporation of high-quality alternatives

helps to improve the quality of the SUTs in volume terms, for example, by deflating household

final consumption expenditure using CPIs. Using either a dual approach or all the information

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

273

available, the result is conducive to the production of better quality estimates of household final

consumption expenditure but these will be different from those purely based on the CPIs. There is

a variety of reasons for the use of CPIs but it does have some drawbacks: for example, CPIs may

not allow for discounts or bulk purchases, or fully meet national accounts requirements. Thus, the

implicit household final consumption expenditure deflator from the SUTs would be, and should

be, different from the CPI. For such areas as capital formation, however, it is unlikely that there is

any collection of direct prices at purchasers’ prices which can be used to deflate components like

gross fixed capital formation, thus alternative proxies, such as producer prices, may be used

instead. In these cases, the use of the H-Approach provides a much better basis and results in a

higher quality volume estimate.

9.13. The H-Approach presumes SUTs in current prices for the year t, SUTs in current prices for

the year t-1 and appropriate deflators are available. This is not, however, a necessary precondition,

as SUTs in volume terms can be created with just SUTs in current prices for the year t and

appropriate deflators. At the same time, this approach does not reflect the time series dimension

and the use of margin and tax rates in the previous year.

C. Overview of the steps in the H-Approach with a focus on volumes

9.14. This section provides an overview of the compilation process, where Figure 9.1 shows a

framework for estimating SUTs simultaneously at purchasers’ prices and at basic prices and in

current prices and volume terms. This section supplements the description of the H-Approach

provided in chapters 2 and 3.

9.15. On the left-hand side of the “H”, the current price data are presented, while the right-hand

side of the “H” presents the volume data in previous years’ prices. In the middle, the links between

SUTs in current prices and in volume term (through the process of deflation) are shown, where the

SUTs on both sides are valued at basic prices. The two legs below the middle cover the IOTs in

current prices and in volume terms. The compilation flow can be from left-top to left-middle then

through deflation to the right-middle and then to right-top. As explained in chapter 3, if better final

use-based purchasers’ price deflators exist, such as CPIs, then a compilation flow starting from the

right-top can be incorporated, reversing the process and ensuring a balanced approach at each

stage. The H-Approach allows for the use of such deflators and volume-based indicators. In terms

of system design, however, there are different options of how better quality price deflators can be

used but these need to be clearly set out at the outset. For example, it may be agreed that the

deflation of household final consumption expenditure using CPIs gives a better measure of the

volume of such expenditure in practice than deflation at basic prices, thus these deflators could be

incorporated, with the objective of retaining these results and balancing back to the basic price

valuation of household final consumption expenditure.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

274

9.16. This framework is shown as a summary in figures 3.4 and 3.5 covering a simplified version

of the outputs in the compilation schematic layout for SUTs and IOTs in current prices and in

previous years’ prices. Application of this compilation scheme will generate consistent and

coherent estimates of volume and price indices for all entries of the SUTs linking the different

valuations and the valuation matrices, together with the IOTs.

9.17. Figure 9.1 provides an overview that underpins the annexes covering the sequence and the

stage of production processes compiling SUTs in both current prices and in volume terms,

provided in chapter 4.

9.18. As stated in the introduction, many of the above transactions in volume terms cannot be

directly observed. As a result, volume estimates must be derived from current price data combined

with information on price or volume changes. As a consequence, the starting point for volume

estimates are the SUTs in current prices as shown in the top left corner of Figure 9.1.

Figure 9.1 Overview of the compilation schematic layout linking SUTs

in current prices and in volume terms

Compiled by Sanjiv Mahajan June 2009

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

275

1. Step 1

9.19. Once current price SUTs at purchasers’ prices (i.e. top left-hand side of Figure 9.1) have

been established, the separation of the valuation matrices covering taxes on products, subsidies on

products, trade margins and transport margins, and the separation of the imports use table are used

to derive a domestic use table at basic prices (i.e. middle of the left-hand side of Figure 9.1). The

compilation of each of these matrices and tables is covered in chapters 7 and 8 of this Handbook.

9.20. At this stage (i.e. middle of the left-hand side of Figure 9.1), the tables for imported goods

and services and domestically produced goods and services form the starting point for the

compilation of the IOTs in current prices (i.e. bottom of the left-hand side of Figure 9.1) but also

the first step in the deflation phase for compiling SUTs in volume terms.

2. Step 2

9.21. For intermediate consumption at purchasers’ prices in the use table, appropriate price

indices are mostly not available. The output and import price indices might be used as an

approximation (see below); the optimal process, however, is to compile the use table at basic prices

by deduction of the valuation matrices from the use table at purchasers’ prices, shown on the left-

hand side of Figure 9.1. in order to apply the most appropriate price indices, the use table at basic

prices should be split between the uses of imported goods and services (imports use table), separate

from the uses of domestically produced goods and services (domestic use table) shown in the

middle of the left-hand side of Figure 9.1.

9.22. The domestic use table at basic prices is deflated using appropriate price deflators (or use

of volume indicators) applied across each product in both the supply table and use table, allowing

for the separation of domestically consumed products and exported products, deflated using export

price indices (EPIs). This assumes that the sale price charged by the seller is the same as the

purchase price paid by the purchaser. Implicitly, this may not, for example, allow for bulk

purchases made at a discounted price. If, however, the deflation is carried out at a very detailed

level, then the impact will be insignificant.

9.23. In the ideal case, producer price indices (PPIs) are a correctly weighted average of domestic

sales and exports. In practice, however, often a weighted average has to be constructed by applying

weights of the previous year. The preferred option is to split produced goods and services between

those consumed domestically and those exported, and then to deflate using PPIs and EPIs,

respectively.

9.24. The same approach holds for the imports of goods and services when applying import price

indices (IPIs).

9.25. Within this step, the use of volume indicators, if appropriate, may form better quality than

deflating current prices with inappropriate deflators. Further areas where a different approach for

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

276

deflation may be considered is handling self-balanced concepts such as FISIM. In current prices,

FISIM is balanced across the production, income and expenditure approaches and can be shown

in the form of SUTs and there is no need for any balancing adjustments. Therefore, if the SUTs in

current prices are balanced at this step, you can remove FISIM as a balanced change leaving the

SUTs in current prices (excluding FISIM) still balanced. FISIM can then be deflated separately

using an alternative, or more appropriate, deflation approach generating a balanced FISIM in

volume terms. Balanced FISIM in volume terms can be added to balanced SUTs in volume terms

(excluding FISIM) to give balanced SUTs in volume terms. This ensures good quality, appropriate,

consistent and coherent deflation of self-balanced concepts. There are several examples that may

be addressed using a self-balanced approach such as insurance, consumption of fixed capital for

non-market units, imputed rental of owner-occupied dwellings, etc.

9.26. The above step results in SUTs at basic prices in previous years’ prices. These data in

volume terms sit in the middle of the right-hand side of Figure 9.1.

9.27. In the case when the SUTs in current prices and at purchasers’ prices are already balanced,

the above step should result in balanced SUTs at basic prices in previous years’ prices, assuming

the requirements on prices mentioned above are fulfilled. In practice, however, some balancing

will be required.

3. Step 3

9.28. Step 3 consists in the deflation of the valuation matrices for taxes, subsidies and margins

(trade and transport) by applying the previous year rates to the volumes at basic prices (or volume

change as appropriate).

9.29. In combination with the SUTs at basic prices, in previous years’ prices, the volume changes

can be calculated, and these form the basis for the volume estimates for all the individual entries

of the valuation matrices. In this compilation step, there are two key issues:

• First, it is important to check the plausibility of the volume estimates, in particular trade and

transport margins for the goods and services which have large changes in quality, as the

quality change forms part of the volume changes in the national accounts. The consequences

for GVA in the trade and transport industries might be unacceptable.

• Second, separate independent PPIs might be available for transport services. Confrontation

with the implicit price indices resulting from the process indicated by Figure 9.1 might lead

to unacceptable differences with the observed price indices. This, in turn, may necessitate a

re-evaluation of earlier estimates.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

277

4. Step 4

9.30. Step 4 consists in the compilation of SUTs at purchasers’ prices in previous years’ prices

– as shown in the top-right corner of Figure 9.1 – by adding the SUTs at basic prices and the

valuation matrices in previous years’ prices obtained in the previous steps.

9.31. At this stage, additional plausibility checks are available, and in some cases, essential.

These may include, as described earlier, the use of household final consumption expenditure in

volume terms using CPIs. Confrontation of the resulting implicit price indices from the SUTs at

purchasers’ prices with observed purchasers’ price indices like the CPI may reveal implausible

results, leading to the need to re-evaluate and adjust earlier estimates.

9.32. The reassessment and allocation of adjustments may ultimately mean, in some cases, that

the current price SUTs at purchasers’ prices may need to change or that one or several of the

intermediate steps may need to be altered.

9.33. In the case of starting with balanced SUTs in current prices and at purchasers’ prices, each

transitional step thereafter creates a balanced matrix. As a consequence, the resulting SUTs in

previous years’ prices both at basic prices and at purchasers’ prices will also be balanced. Any

adjustments resulting from the additional plausibility checks will be incorporated in a balanced

manner and improve plausibility and, in turn, quality.

5. Step 5

9.34. Step 5 consists in the compilation of IOTs in previous years’ prices using the same

assumptions as applied for the compilation of IOTs in current prices. This is shown in the bottom

right-hand side of Figure 9.1.

6. Other points to note

9.35. Options on the starting point, in other words, whether to use the current price SUTs at

purchasers’ prices and at basic prices balanced or unbalanced as a starting point, is discussed in

chapter 11 of this Handbook. This choice does not alter the steps or the processes of deflation but

it will have different impacts on the processes, resources, systems and schedules needed to compile

these tables.

9.36. Figure 9.2 shows the links between the table in Figure 9.1 at the current price estimates

and in volume estimates for two successive years. In particular:

• The link between the SUTs of year t-1 in current prices of t-1 and the SUTs year t in prices

of t-1 are the SUTs with the corresponding volume indices

• The connecting link between the SUTs in current prices of year t and the SUTs of year t in

prices of t-1 are the SUTs with the information on price indices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

278

From the SUTs in current prices of year t and year t-1, the SUTs with value indices can be derived.

9.37. At the end of this estimation process, a complete picture is available as laid out in Figure

9.2, for every column of the SUTs, not only in current prices but also in previous years’ prices for

outputs, intermediate uses, final uses and imports of goods and services. It should be noted the

SUTs may be unbalanced at this stage.

9.38. This set of data makes it possible to check the consistency of the data, in which process,

even if the results in current prices look plausible, the analysis of volume and price data may reveal

significant problems. Such analysis may include, for example, a comparison of changes in the

volume of output by industry with the corresponding changes in the volume of intermediate

consumption and the volume of GVA.

9.39. The analysis in volume terms is far superior, in particular when prices are changing rapidly.

In several cases, these data can be checked with actual data in volumes, for example, the use of

energy products or the volume of sales by product as in agriculture.

9.40. The value-price-volume analysis can lead to amendments on either of the estimated

variables before the balancing process has commenced.

Figure 9.2 Link between SUTs in current prices and in volume terms

9.41. An empirical example of SUTs reflecting the components of Figure 9.2 (including prices

indices, volume indices and value indices) is shown in Table 9.1 and Table 9.2. From these tables,

information on inflation, real growth and nominal growth can be extracted at a detailed level, along

with nominal GDP, real GDP growth and the GDP deflator. Table 9.3 also shows that the growth

rates for GDP can be directly derived from the SUTs by applying the formulae for the production,

income and expenditure approaches.

9.42. It should be noted that the estimates of “direct purchases abroad by residents” and

“purchases on the domestic territory by non-residents” in row (11) of Table 9.1 and rows (11) and

Supply and use tables

(Real growth)

(Price change)

Supply and use tables

of year t-1

of year t

in prices of t-1

in prices of t

Supply and use tables

Value indice s

(Nominal growth)

Volume indices

Price indices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

279

(12) of Table 9.2 are not shown separately in these tables but incorporated in the products of the

table. This is an alternative presentation of direct purchases abroad by residents and domestic

purchases by non-residents: these adjustments form the difference between the national concept

and domestic concept of household final consumption expenditure. The breakdown by product

allows for more appropriate deflation by product, whereas the two aggregates form a

heterogeneous suite of products.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

280

Table 9.1 Supply table in current prices and in volume terms

Netherlands 2011

Agriculture

Manufacturing

Constructi o n

Trade, transport

and

communication

Finance and

business

services

Other

services

Output at

basic prices

Trade and

transport

margins

Taxes less

subsidies on

products

Total

(1)

(2) (3)

(4)

(5) (6)

(7) (8)

(9)

(10) (11)

(12) (13)

Agriculture (1) 25 773

153

20 16

25 962 15 384

41 346

10 903 535

11 438 52 784

Manufacturing

(2) 1 262 316 757

1 757 18 741

5 157

11 821 355 495

336 807

692 302 112 979

39 419 152 398

844 700

Constructi o n

(3) 92 1 104

87 896 296

2 062

2 134 93 584

1 564 95 148

8 260

8 260 103 408

Trade (4) 144 9 919

356

116 902 2 893

672 130 886

3 429

134 315 - 109 321

829 - 108 492 25 823

Transport (5) 385 17

63 645 1 348

977 66 441

10 671

77 112 - 14 864

- 996 - 15 860 61 252

Communication (6) 4 722 41 679

2 895

585 49 881

6 277

56 158 198 3 160 3 358 59 516

Finance and business services (7) 489

6 992 1 053

13 015

195 536 73 409

290 494 55 242

345 736 9 124 9 124 354 860

Other services

(8) 304

2 512

2 687

3 540

216 628

225 671

16 638 242 309

105 3 008

3 113

245 422

Total

(9) 28 449 342 176

91 131

256 965 213 451

306 242

1 238 414 446 012 1 684 426 63 339 63 339 1 747 765

CIF/FOB adjustments on imports (10) - 3 569 - 3 569 - 3 569

Direct purchases abroad by

residents

(11)

Total (12) 28 449

342 176

91 131 256 965

213 451 306 242

1 238 414

442 443 1 680 857

63 339 63 339 1 744 196

Agriculture (1) 102.0

104.1

100.0

102.0 106.9 103.8 97.5 107.6 97.9 102.4

Manufacturing

(2) 107.5

108.0

101.3

99.9

100.5

101.3

107.2

108.0

107.6 100.2

101.9 100.7

106.2

Constructi o n (3) 103.4 100.8

100.5 102.1

100.9

99.8 100.5 101.6 100.5 100.1 100.1 100.5

Trade (4) 98.0 100.1

99.2

99.7 99.9

99.9 99.7 100.0 99.7 99.5 102.3 99.5 100.4

Transport (5) 102.4 100.0

101.5

102.0 101.6 101.6 102.0 102.2 102.0 103.5 99.4 103.2 101.7

Communication (6)

97.5

99.6 97.4 98.3 99.2 98.3 99.1 101.5 100.5 100.6 99.2

Finance and business services (7) 102.9

101.0 100.5

100.7 100.2 101.0 100.4 101.0 100.5 93.6 93.6 100.3

Other services (8) 102.4 100.2

102.2 115.9 100.8 101.0

103.4 101.1 99.1 110.2

109.8 101.2

Total

(9) 102.2

107.4 100.5 100.3

100.4

100.8 102.4 106.5 103.4 100.8 100.8 103.3

CIF/FOB adjustments on imports

(10)

108.1 108.1

108.1

Direct purchases abroad by

residents

(11)

Total

(12) 102.2 107.4

100.5 100.3 100.4 100.8 102.4

106.5 103.4

100.8 100.8 103.3

Agriculture (1) 25 274 147 20 16

25 457 14 386 39 843 11 183 497 11 680 51 523

Manufacturing (2) 1 174

293 265 1 734 18 755

5 129

11 673 331 730 311 934

643 664 112 711 38 690

151 401 795 065

Constructi o n

(3) 89 1 095

87 486 290 2 043

2 138 93 141 1 540

94 681 8 253

8 253 102 934

Trade

(4) 147

9 913 359 117 309

2 897 673

131 298 3 428 134 726 - 109 828

810 - 109 018

25 708

Transport (5)

376 17 68 62 377

1 327 962 65 127

10 444 75 571 - 14 367

- 1 002 - 15 369 60 202

Communication (6) 4 845 41 862

2 973 595 50 275

6 385 56 660 195 3 144

3 339 59 999

Finance and business services

(7) 475 6 925 1 048 12 924

195 168 72 709 289 249

54 690 343 939 9 745

9 745 353 684

Other services (8)

297 2 507 2 628 3 055

214 997 223 484 16 094

239 578 106 2 729 2 835 242 413

Total (9)

27 832 318 714 90 695 256 145 212 612 303 763 1 209 761 418 901 1 628 662 62 866 62 866 1 691 528

CIF/FOB adjustments on imports

(10) - 3 303 - 3 303

- 3 303

Direct purchases abroad by

residents

(11)

Total (12)

27 832 318 714 90 695

256 145 212 612 303 763

1 209 761 415 598 1 625 359

62 866 62 866

1 688 225

Agriculture

(1) 99.9 96.1

87.0 88.9

99.9 103.5 101.1 103.6 111.9 103.9

101.8

Manufacturing (2)

95.4 104.5 89.4 98.5 94.4

103.1 103.8 103.3 103.6 103.6 98.7 102.3

103.3

Constructi o n

(3) 127.1

129.4 104.1 117.9

89.3 99.9

103.9

101.2 103.8

98.2

98.2 103.4

Trade

(4) 79.5 94.4 104.1 104.8 99.0 90.5

103.6 101.4 103.6 103.4 98.8 103.5 104.1

Transport (5) 98.9 100.0 82.9 103.2 129.2 101.9 103.6 100.3 103.1 104.9 106.5

105.0 102.7

Communication (6) 109.4

101.6 109.6 97.2 102.7 97.8 102.1 97.0 97.5 97.5 101.9

Finance and business services

(7) 99.6 103.0

97.1 103.3 102.7 100.4

102.1 107.4 102.9

98.2

98.2 102.8

Other services (8)

95.2 99.1

99.4 100.3 100.5

100.5 98.3 100.3

94.6 101.9 101.6 100.4

Total (9) 99.6

104.2 103.6 103.3 102.4

100.6 102.6 103.5

102.8 98.6 98.6 102.7

CIF/FOB adjustments on imports

(10)

100.9 100.9

100.9

Direct purchases abroad by

residents

(11)

Total (12)

99.6 104.2

103.6 103.3

102.4 100.6

102.6 103.5 102.8

98.6

98.6 102.7

Agriculture

(1) 25 299 153 23

18 25 493 13 900

39 393 10 799 444

11 243 50 636

Manufacturing

(2) 1 230 280 614 1 939

19 032 5 431 11 317

319 563 301 843 621 406

108 774 39 201 147 975 769 381

Constructi o n (3) 70

846 84 076 246 2 289

2 141 89 668 1 521

91 189 8 404 8 404 99 593

Trade (4) 185

10 503 345 111 978 2 927 744

126 682 3 381 130 063 - 106 185 820 - 105 365

24 698

Transport (5) 380

17 82 60 429 1 027 944

62 879 10 408 73 287 - 13 701 - 941 - 14 642

58 645

Communication

(6) 4 428

41 199 2 713

612

48 952 6 530 55 482 201

3 224 3 425

58 907

Finance and business services

(7) 477 6 724 1 079 12 509 190 079 72 409 283 277 50 908 334 185 9 925

9 925 344 110

Other services (8) 312 2 531

2 644 3 045 213 878 222 410

16 366 238 776 112 2 678

2 790 241 566

Total (9) 27 953

305 816 87 521 248 037 207 534 302 063 1 178 924 404 857 1 583 781 63 755 63 755 1 647 536

CIF/FOB adjustments on imports (10) - 3 272 - 3 272

- 3 272

Direct purchases abroad by

residents

(11)

Total (12) 27 953 305 816 87 521

248 037 207 534 302 063 1 178 924 401 585 1 580 509 63 755 63 755 1 644 264

Agriculture (1)

101.9 100.0 87.0 88.9 101.8 110.7 105.0 101.0

120.5 101.7 104.2

Manufacturing (2)

102.6 112.9 90.6 98.5 95.0 104.5 111.2 111.6 111.4 103.9 100.6 103.0 109.8

Constructi o n (3) 131.4 130.5 104.5 120.3 90.1 99.7 104.4 102.8 104.3 98.3 98.3 103.8

Trade

(4) 77.8 94.4 103.2 104.4 98.8 90.3

103.3 101.4 103.3 103.0 101.1 103.0 104.6

Transport (5) 101.3 100.0 84.1 105.3

131.3 103.5 105.7 102.5 105.2 108.5 105.8 108.3 104.4

Communication (6) 106.6 101.2 106.7 95.6 101.9

96.1 101.2 98.5 98.0 98.0 101.0

Finance and business services (7) 102.5 104.0 97.6 104.0 102.9 101.4

102.5 108.5 103.5 91.9 91.9 103.1

Other services (8) 97.4 99.2 101.6 116.3 101.3 101.5 101.7 101.5

93.8 112.3 111.6 101.6

Total (9) 101.8 111.9 104.1 103.6 102.9

101.4 105.0 110.2 106.4 99.3 99.3 106.1

CIF/FOB adjustments on imports (10) 109.1 109.1 109.1

Direct purchases abroad by

residents

(11)

Total (12) 101.8 111.9

104.1 103.6 102.9 101.4 105.0 110.2 106.3 99.3 99.3 106.1

OUTPUT OF INDUSTRIES

Imports

VALUATION

Total supply

purchasers’

prices

Total supply

at basic

prices

Supply table at current prices in year

PRODUCTS

Value index (t -1 = 100)

PRODUCTS

Price index (t

-1 = 100)

PRODUCTS

Volume index (t -1 = 100)

Supply table of year t

at previous years' prices

Supply table of t -1 at current prices

Adjustme

PRODUCTS

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

281

Table 9.2 Use table in current prices and in volume terms

Households NPISH

General

government

(1) (2) (3) (4) (5) (6) (7) (8) (9)

(10)

(11) (12)

(13)

(14) (15) (16)

Agriculture

(1) 5 514 17 236

114 205 248 1 091

24 408 5 937

143

- 91 22 387 28 376 52 784

Manufacturi ng

(2) 10 133 175 862 26 285 26 880 7 573 30 984 277 717 122 814 9 119

51 545

1 409 382 096 566 983 844 700

Constructio n

(3) 298 1 718 24 504 1 608 2 212 15 934 46 274 453 566

54 015

2 100 57 134 103 408

Trade

(4) 107

2 975

336

7 421 2 115

576 13 530

4 750

7 543 12 293 25 823

Transport

(5) 291

3 230 162 23 566 2 738 1 903

31 890 5 679

492

23 191 29 362 61 252

Communication

(6) 170

3 027 1 068 13 842 7 911 6 089 32 107 10 794 868 6 717

2 9 028 27 409 59 516

Finance and business services

(7) 2 067 36 341 7 131 46 730 61 900 52 851 207 020 77 968 5 3 585 16 820 49 462 147 840 354 860

Other services

(8) 172

2 306 1 236

4 214

4 569 13 381

25 878 55 061

4 610

153 396 1 162

206 5 109 219 544 245 422

Total

(9) 18 752 242 695

60 836 124 466 89 266

122 809 658 824

283 456 5 483

167 158

130 402 1 526

500 916 1 088 941 1 747 765

CIF/FOB adjustments on exports

(10)

- 3 569 - 3 569 - 3 569

Direct purchases abroad by

residents

(11)

Purchases in the domestic

territory by non-residents

(12)

Total

(13) 18 752 242 695

60 836 124 466 89 266 122 809 658 824 283 456 5 483 167 158 130 402 1 526 497 347 1 085 372 1 744 196

Compensation of employees

(14) 2 677 42 141

18 786 71 831 70 666 111 939 318 040

318 040

Other taxes less subsidies on

production

(15)

- 722

176 5 18

- 930 1 190

- 263 - 263

Consumption of fixed capital

(16) 3 683 18 241 2 234 18 024 13 259 51 627 107 068 107 068

Net operating surplus

(17) 4 059 38 923 9 270 42 626 41 190 18 677 154 745 154 745

GVA

(18) 9 697 99 481

30 295 132 499 124 185 183 433 579 590

579 590

Total input at basic prices

(19) 28 449 342 176 91 131 256 965 213 451 306 242 1 238 414 283 456 5 483 167 158 130 402 1 526 497 347 1 085 372

Agriculture

(1)

105.8

109.9 102.7

108.5

102.5 96.2

108.2 100.4

94.7

568.8

97.7

98.0 102.4

Manufacturi ng

(2) 114.1 111.5

103.3 107.0 103.9 103.5

109.2

102.1

99.5 99.6

85.7 106.8 104.9 106.2

Constructio n

(3) 101.4 103.1 100.3 102.4 101.1 101.8 101.0 97.4

97.1 100.0

102.0 100.0 100.5

Trade

(4) 100.0 101.2

101.2 101.7 102.4 101.8 101.7 102.2

97.3 99.1 100.4

Transport

(5)

101.7

101.4 101.9 102.4

100.4 102.1

102.1

101.8

103.1

101.3 101.4 101.7

Communication

(6) 100.0

98.3 97.9 98.7

99.6 99.2

99.0

102.4 100.9

96.7

100.0 98.1 99.5 99.2

Finance and business services

(7) 101.4 100.8 100.4 100.7 101.7 98.7 100.5 101.7

100.0 101.8

95.7

99.0 100.1 100.3

Other services

(8) 102.4 101.4 102.1 102.5 102.4 105.6 103.9 102.7 103.8 100.2

101.7

99.5 101.8

100.9 101.2

Total

(9) 109.3

109.0 101.6 102.2

101.7 101.1

104.5

102.1 103.3

100.2

99.1

83.0 104.9

102.6 103.3

CIF/FOB adjustments on exports

(10)

108.1

108.1 108.1

Direct purchases abroad by

residents

(11)

Purchases in the domestic

territory by non-residents

(12)

Total

(13) 109.3

109.0 101.6 102.2

101.7 101.1

104.5 102.1 103.3 100.2

99.1 83.0

104.8 102.6 103.3

Compensation of employees

(14) 102.7 102.1 101.3 101.8 101.5

101.4 101.6

101.6

Other taxes less subsidies on

production

(15)

98.0

193.4

- 500.0 - 1 800.0 98.8

113.9 48.3

48.3

Consumption of fixed capital

(16) 99.1 100.7 99.4 99.9 98.8 97.7 98.8

98.8

Net operating surplus

(17) 79.8 106.1 92.4 93.1 96.3 103.8 97.7

97.7

GVA

(18) 90.8 103.5 98.3 98.6 99.5 100.6 100.1

100.1

Total input at basic prices

(19) 102.2 107.4 100.5 100.3 100.4

100.8 102.4 102.1 103.3

100.2 99.1

83.0 104.8 102.6

Agriculture

(1) 5 212 15 677 111 189 242 1 134 22 565 5 912 151

- 16

22 911 28 958 51 523

Manufacturi ng

(2)

8 882 157 696 25 450

25 123 7 291 29 932

254 374 120 255 9 167

51 739 1 645 357 885

540 691 795 065

Constructio n

(3) 294

1 667 24 422

1 570 2 188

15 654

45 795 465

583 54 033 2 058

57 139

102 934

Trade

(4) 107 2 940

332 7 299 2 065 566 13 309 4 646 7 753

12 399 25 708

Transport

(5) 286

3 186 159

23 022 2 728

1 863

31 244 5 579

477 22 902

28 958

60 202

Communication

(6)

170 3 079 1 091 14 028 7 939

6 139 32 446 10 538 860 6 947 2

9 206 27 553 59 999

Finance and business services

(7)

2 039 36 056

7 102 46 387

60 856

53 536 205 976

76 632 5

3 523 17 570

49 978 147 708

353 684

Other services

(8) 168 2 274 1 210 4 112 4 461

12 675 24 900 53 632 4 443 153 068 1 143

207 5 020 217 513

242 413

Total

(9) 17 158

222 575

59 877 121 730 87 770

121 499

630 609 277 659

5 308

166 818 131 583 1 838 477 713

1 060 919 1 691 528

CIF/FOB adjustments on exports

(10) - 3 303 - 3 303 - 3 303

Direct purchases abroad by

residents

(11)

Purchases in the domestic

territory by non-residents

(12)

Total

(13) 17 158

222 575

59 877 121 730

87 770 121 499

630 609

277 659 5 308

166 818 131 583

1 838 474 410 1 057 616

1 688 225

Compensation of employees

(14) 2 606 41 261 18 537 70 563 69 589 110 413 312 969 312 969

Other taxes less subsidies on

production

(15)

- 737 91

- 1 - 1 - 941 1 045 - 544

- 544

Consumption of fixed capital

(16) 3 718 18 116

2 247 18 045 13 424 52 821 108 371 108 371

Net operating surplus

(17) 5 087 36 671 10 035 45 808 42 770 17 985 158 356 158 356

GVA

(18) 10 674 96 139 30 818 134 415 124 842 182 264 579 152 579 152

Total input at basic prices

(19)

27 832 318 714 90 695

256 145 212 612

303 763 1 209 761 277 659 5 308

166 818 131 583 1 838 474 410

1 057 616

Agriculture

(1)

102.0

103.3 119.4 71.3 103.4 101.1 102.6 101.5

84.8 - 9.1 102.0 101.1 101.8

Manufacturi ng

(2)

99.5 106.3 101.6 99.3 99.9

97.5 103.6 99.4 102.8

112.5 43.1 104.0

103.2 103.3

Constructio n

(3) 100.7 126.0 109.8 102.3 101.2

99.8 105.9 103.3 96.0 101.9 91.6

101.4 103.4

Trade

(4) 97.3 99.0 108.1 99.3 99.2 95.0 99.2 99.2 117.4 109.9 104.1

Transport

(5) 100.0 97.7 100.0 101.7 108.4 96.4 101.5 100.8

97.3 104.9 103.9

102.7

Communication

(6) 98.8 102.5 100.6 104.7 101.3 95.3 101.6 97.5 100.0 102.9 7.4 107.9 102.2 101.9

Finance and business services

(7) 98.9 104.9 104.0 104.8 104.7

99.1 103.2 100.5 100.0

96.2 100.2 106.2 102.2 102.8

Other services

(8) 99.4 103.6 98.5 107.8 102.7 99.1 101.5 101.8 99.8 99.7 98.2 72.6 103.1 100.2 100.4

Total

(9) 100.2

105.7 105.1 102.7 103.8 98.6 103.2

100.2 99.8 99.8 105.6 42.7 104.4

102.4 102.7

CIF/FOB adjustments on exports

(10)

100.9 100.9 100.9

Direct purchases abroad by

residents

(11)

Purchases in the domestic

territory by non-residents

(12)

Total

(13) 100.2 105.7 105.1 102.7 103.8 98.6 103.2 100.2 99.8 99.8 105.6 42.7 104.4 102.4 102.7

Compensation of employees

(14) 100.1 100.5

98.8 101.4 101.6 100.4 100.8

100.8

Other taxes less subsidies on

production

(15)

137.2

- 104.6 5.3 .9 94.4 136.4 55.5

55.5

Consumption of fixed capital

(16) 101.6 100.8

101.5 100.3 100.4 102.0 101.3

101.3

Net operating surplus

(17) 99.7 101.2

104.8 109.0 101.5 110.7 104.7

104.7

GVA

(18) 98.6 101.0

100.9 103.8 101.5 101.9 102.0

102.0

Total input at basic prices

(19) 99.6 104.2

103.6 103.3 102.4

100.6 102.6 100.2 99.8 99.8

105.6 42.7 104.4

102.4

Use table at previous years' prices

PRODUCTS

Adjustmen

GVA

Volume index (t -1 = 100)

PRODUCTS

Adjustmen

GVA

PRODUCTS

Adjustmen

GVA

PRODUCTS

Adjustmen

GVA

Total

Agriculture

Manufacturi ng

Constructio n

Trade, transport

and

communication

Finance and

business

services

Other

services

Total

Final consumption expenditure

Exports

Chang es i n

inventories

Changes

valuables

Gross fixed

capital

formation

Total use at

purchasers’

prices

INPUT OF INDUSTRIES

FINAL USE

Use table at current in year

Price index (t -1 = 100)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

282

Netherlands 2011

Households

NPIS H

General

government

(1) (2)

(3) (4)

(5) (6)

(7) (8) (9) (10) (11) (12) (13) (14) (15) (16)

Agriculture

(1) 5 108 15 176

93 265 234 1 122 21 998 5 823 178 176 22 461 28 638 50 636

Manuf acturing

(2) 8 926 148 356

25 043 25 305

7 297 30 707

245 634 120 960

8 919

45 974 3 820

344 074 523 747

769 381

Construc tion

(3) 292

1 323 22 247

1 534 2 161

15 686 43 243

450

607 53 047

2 246 56 350

99 593

Trade

(4) 110 2 970 307 7 347 2 082 596 13 412 4 683 6 603

11 286 24 698

Transport

(5)

286 3 260

159 22 630 2 517 1 933 30 785 5 535

490 21 835

27 860 58 645

Communication

(6)

172 3 004 1 084 13 401 7 835 6 439 31 935 10 806 860 6 748

27 8 531

26 972

58 907

Finance and business s ervices

(7) 2 062 34 384 6 828 44 243 58 107 53 998 199 622 76 231 5 3 664

17 538 47 050

144 488

344 110

Other services

(8)

169 2 194 1 229

3 815

4 345 12 786

24 538 52 706 4 451 153 552

1 164

285 4 870

217 028 241 566

Total

(9)

17 125 210 667 56 990

118 540 84 578 123 267 611 167 277 194 5 316

167 232

124 649 4 308

457 670 1 036 369

1 647 536

CIF/FOB adjustments on

exports

(10)

- 3 272 - 3 272

- 3 272

Direct purchases abroad by

residents

(11)

Purchases in the domestic

territory by non-residents

(12)

Total

(13)

17 125

210 667

56 990 118 540 84 578 123 267 611 167 277 194 5 316

167 232 124 649

4 308 454 398

1 033 097

1 644 264

Compensation of employees

(14) 2 603 41 042 18 765 69 595 68 466 110 000 310 471 310 471

Other taxes less subsidies on

production

(15)

- 537 - 87 - 19 - 106 - 997 766 - 980

- 980

Consumption of f ixed capital

(16)

3 660

17 973 2 213 17 989

13 366 51 781 106 982

106 982

Net operating surplus

(17)

5 102 36 221 9 572 42 019 42 121 16 249 151 284

151 284

GVA

(18) 10 828 95 149 30 531 129 497 122 956 178 796 567 757 567 757

Total input at basic prices

(19)

27 953 305 816 87 521 248 037 207 534 302 063 1 178 924 277 194 5 316

167 232

124 649 4 308

454 398 1 033 097

Agriculture

(1) 107.9 113.6

122.6 77.4 106.0 97.2 111.0 102.0

80.3

- 51.7

99.7

99.1 104.2

Manuf acturing

(2)

113.5

118.5 105.0 106.2

103.8 100.9

113.1 101.5

102.2 112.1

36.9

111.1

108.3 109.8

Construc tion

(3) 102.1

129.9 110.1

104.8 102.4 101.6 107.0 100.7 93.2

101.8

93.5 101.4 103.8

Trade

(4) 97.3

100.2

109.4 101.0 101.6 96.6 100.9 101.4

114.2

108.9

104.6

Transport

(5) 101.7 99.1

101.9

104.1 108.8

98.4

103.6 102.6

100.4

106.2 105.4

104.4

Communication

(6) 98.8 100.8 98.5 103.3 101.0 94.6 100.5 99.9 100.9

99.5

7.4

105.8 101.6

101.0

Finance and business s ervices

(7) 100.2 105.7

104.4 105.6 106.5 97.9 103.7 102.3 100.0

97.8 95.9

105.1

102.3 103.1

Other services

(8) 101.8 105.1

100.6 110.5 105.2 104.7 105.5 104.5 103.6

99.9

99.8

72.3 104.9

101.2

101.6

Total

(9)

109.5

115.2 106.7

105.0 105.5

99.6 107.8 102.3

103.1 100.0

104.6

35.4

109.4 105.1

106.1

CIF/FOB adjustments on

exports

(10)

109.1 109.1 109.1

Direct purchases abroad by

residents

(11)

Purchases in the domestic

territory by non-residents

(12)

Total

(13)

109.5 115.2

106.7 105.0

105.5 99.6 107.8

102.3 103.1

100.0 104.6

35.4

109.5 105.1

106.1

Compensation of employees

(14) 102.8 102.7 100.1

103.2 103.2 101.8 102.4

102.4

Other taxes less subsidies on

production

(15)

134.5

- 202.3 - 26.3

- 17.0 93.3 155.4 26.8

26.8

Consumption of f ixed capital

(16)

100.6 101.5

100.9

100.2 99.2 99.7

100.1

Net operating surplus

(17)

79.6 107.5 96.8

101.4 97.8

114.9 102.3

102.3

GVA

(18)

89.6 104.6

99.2 102.3

101.0 102.6

102.1

Total input at basic prices

(19)

101.8 111.9

104.1 103.6

102.9 101.4

105.0 102.3 103.1

100.0

104.6

35.4 109.5

105.1

Exports

Total

FINAL USE

INPUT OF INDUSTRIES

Total use at

purchasers’

prices

Agriculture

Manuf acturing

Construc tion

Trade, transport

and

communication

Finance and

business

services

Other

services

Total

Final cons umption expenditure

Gross f ixed

capital

formation

Changes

valuables

Changes in

inventories

Adjustments

GVA

Va lue index (T-1 = 100)

Use Table of T-1 at current prices

PRODUCTS

Adjustments

GVA

PRODUCTS

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

283

Table 9.3 Gross domestic product in current prices and in volume terms

Netherlands 2011

Production approach Income approach Expenditure approach

t at current prices

Total output at basic prices

1 238 414 Compensation of employees 318 040 Household final consumption expenditure 283 456

- Intermediate consumption

- 658 824 + Other net taxes on production - 263 + NPISH final consumption expenditure

5 483

+ Capital consumption 107 068 + Government consumption expenditure 167 158

+ Net operating surplus 154 745

+ Gross fixed capital formation

130 402

+ Acquisitions less disposals of valuables

= Value added at basic prices

579 590 = Value added at basic prices 579 590 + Changes in inventories 1 526

+ Exports of goods and services

497 347

+ Taxes less subsidies on products 63 339 + Taxes less subsidies on products 63 339 - Imports of goods and services - 442 443

= Gross domestic product

642 929 = Gross domestic product 642 929 = Gross domestic product

642 929

Annual change of prices in percent

Total output at basic prices

2.4

Compensation of employees 1.6

Household final consumption expenditure

2.1

- Intermediate consumption

4.5 + Other net taxes on production -51.7 + NPISH final consumption expenditure 3.3

+ Capital consumption -1.2

+ Government consumption expenditure

0.2

+ Net operating surplus -2.3 + Gross fixed capital formation -0.9

+ Acquisitions less disposals of valuables 0.0

= Value added at basic prices

0.1 = Value added at basic prices 0.1

+ Changes in inventories -17.0

+ Exports of goods and services

4.8

+ Taxes less subsidies on products 0.8 + Taxes less subsidies on products 0.8 - Imports of goods and services 6.5

= Gross domestic product

0.1 = Gross domestic product

0.1

t at prices of previous year

Total output at basic prices 1 209 761 Compensation of employees 312 969

Household final consumption expenditure 277 659

- Intermediate consumption - 630 609

+ Other net taxes on production - 544 + NPISH final consumption expenditure 5 308

+ Capital consumption

108 371 + Government consumption expenditure

166 818

+ Net operating surplus 158 356 + Gross fixed capital formation 131 583

+ Acquisitions less disposals of valuables

= Value added at basic prices 579 152

= Value added at basic prices 579 152 + Changes in inventories 1 838

+ Exports of goods and services 474 410

+ Taxes less subsidies on products 62 866

+ Taxes less subsidies on products 62 866 - Imports of goods and services

- 415 598

= Gross domestic product 642 018 = Gross domestic product

642 018 = Gross domestic product 642 018

Annual real growth rates in percent

Total output at basic prices 2.6 Compensation of employees 0.8

Household final consumption expenditure 0.2

- Intermediate consumption 3.2 + Other net taxes on production -44.5 + NPISH final consumption expenditure -0.2

+ Capital consumption 1.3 + Government consumption expenditure

-0.2

+ Net operating surplus 4.7 + Gross fixed capital formation 5.6

+ Acquisitions less disposals of valuables 0.0

= Value added at basic prices 2.0

= Value added at basic prices 2.0 + Changes in inventories

-57.3

+ Exports of goods and services

4.4

+ Taxes less subsidies on products -1.4

- Imports of goods and services

3.5

= Gross domestic product

1.7 = Gross domestic product 1.7 = Gross domestic product 1.7

t-1 at current prices

Total output at basic prices 1 178 924 Compensation of employees 310 471

Household final consumption expenditure 277 194

- Intermediate consumption

- 611 167 + Other net taxes on production - 980 + NPISH final consumption expenditure 5 316

+ Capital consumption

106 982 + Government consumption expenditure 167 232

+ Net operating surplus 151 284 + Gross fixed capital formation 124 649

+ Acquisitions less disposals of valuables

= Value added at basic prices 567 757 = Value added at basic prices 567 757 + Changes in inventories 4 308

+ Exports of goods and services 454 398

+ Taxes less subsidies on products 63 755 + Taxes less subsidies on products 63 755 - Imports of goods and services

- 401 585

= Gross domestic product 631 512 = Gross domestic product

631 512 = Gross domestic product 631 512

Annual nominal growth rates in percent

Total output at basic prices 5.0

Compensation of employees 2.4 Household final consumption expenditure 2.3

- Intermediate consumption 7.8 + Other net taxes on production -73.2 + NPISH final consumption expenditure

3.1

+ Capital consumption 0.1 + Government consumption expenditure 0.0

+ Net operating surplus

2.3 + Gross fixed capital formation 4.6

+ Acquisitions less disposals of valuables 0.0

= Value added at basic prices 2.1

= Value added at basic prices 2.1 + Changes in inventories -64.6

+ Exports of goods and services 9.5

+ Taxes less subsidies on products -0.7 + Taxes less subsidies on products -0.7 - Imports of goods and services 10.2

= Gross domestic product 1.8 = Gross domestic product 1.8 = Gross domestic product 1.8

NOMINAL GROWTH

GROSS DOMESTIC PRODUCT

INFLATION

REAL GROWTH

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

284

D. Price and volume indicators in theory

9.43. The price and volume changes for aggregates are derived from price and volume changes

of separate products, preferably at a low level of aggregation. Ideally, price changes on the product

level should refer to specific products; for example, every year the prices of exactly the same

product can change (in accordance with brand type, weight, quality, and other properties) and

should be observed. Generally speaking, in terms of quality, the greater the number of products in

the SUTs, the better the required matching of appropriate values and prices.

9.44. The price and volume indicators must meet a number of requirements in order to be

appropriate for the estimation of price and volume indices within the SUTs framework. These

requirements are discussed in relation to the concept of output. They also apply, however, to all

other transactions in goods and services. Examples of the requirements include:

• The prices and quantities should relate directly to output. This means that they should refer

to complete end-products and not to contributory activities or to contributory intermediate

or primary inputs. In the case of prices, they must also refer to the right valuation, for

example, output at basic prices.

• The prices and quantities should have sufficient stratification, meaning that different prices

and quantities should be available for all different product groups making up the output.

• The product classification of prices and quantities should have sufficient and detailed

matching. This requirement will be fully met, for example, if there is only one product in a

product group. If there is more than one product within a product group, an additional

requirement is that the composition of the product group does not change over time, which

is a weak assumption and not very likely.

• The prices and quantities should be representative for the product group. Usually, prices and

quantities available do not cover all products of the product group or are based on a sample

survey. Changes in the prices and quantities that are observed should be representative of

changes in the prices and quantities that are not observed.

• If prices differ among users for the same products, then separate price indices should be

collected and used; this is very important when, for example, distinguishing price changes

between domestic users and for export.

• The changes in values resulting from changes of quality should be excluded from the price

index and included in the volume index.

9.45. The requirement of matching for volume estimation means that the compilation of SUTs

in volume terms will require much more detail in terms of products and prices than is necessary

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

285

for the compilation of SUTs in current prices. It is preferable, however, that the classification of

the SUTs in volume terms should be similar to the level of detail of the SUTs in current prices.

9.46. Considerations relating to the compilation of SUTs in volume terms constitute one of the

many criteria to be borne in mind when determining the size of the SUTs – this covered in more

detail in chapter 4. The balance between, on the one hand, the assumptions to be made in order to

obtain sufficient detail to ensure homogeneity of price and values and, on the other, the gain in

quality must be taken into account when determining the classifications in the SUTs.

9.47. For the most part, comprehensive information is not available, either on prices or on

quantities. Accordingly, estimates will be based on limited information. It is recognized that

limited price data and limited quantity data do not provide the same possibilities. Price information

from a sample with a certain size may well be more representative than quantity information from

a sample of the same size. This observation is based on the consideration that, if there is a

competitive market for a specific product grouping, there will be a tendency to use one price for

the total supply of that product. In that case, a relatively small sample will be sufficient for

observing the price and price changes of the total supply of that product.

9.48. For their part, however, changes in quantities are less liable to such equalizing tendencies.

Thus it is true that, in an expanding market, all producers will try to increase their supply but the

realization will depend on restrictive factors, such as production capacity and financing facilities.

Alongside fast-growing producers, there will be slow-growing producers and possibly even

shrinking producers. This means that, in order to obtain reliable estimates for quantities, the

samples will have to be larger – even much larger. As a consequence, it is common practice to

derive price indices from price samples and, afterwards, to compile data in volume terms by

combining current price data and price indices. In many cases, this approach is efficient and cost-

saving.

9.49. Although the standard price method can be applied for many goods and services, there are

still a number of transactions for which observation of prices has not been realized or is even not

possible. The latter condition applies to certain cases in which, owing to the nature of the definition

and measurement of output in current prices, the direct observation of appropriate prices is not

possible, such as with non-market services, FISIM and insurance services.

E. Price and volume indicators in practice

9.50. This section briefly covers a range of price and volume indicators that may be available as

the building blocks for the preparation of SUTs in volume terms and are not meant to be

exhaustive. These indicators should be considered applicable, as appropriate, to the various parts

of the H-Approach, for example, in ensuring consistency as far as possible between the supply

table components and the use tables components. The section covers:

• Supply table at basic prices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

286

• Use table at basic prices

• GVA by industry

• Valuation matrices

9.51. More information on the choice of index formulae may be found in chapter 15 of the 2008

SNA and the Handbook on Price and Volume Measures in National Accounts (Eurostat, 2016).

1. Supply table at basic prices

9.52. Under this heading, the domestic production of goods and services and imports of goods

and services are separately considered.

(a) Domestic production: producers’ price indices

9.53. The PPIs usually fulfil the general requirements for price indicators, such as valuation,

adjustment for quality change and level of detail, and separate prices should preferably be available

for domestic sales and for exports. For total output, a weighted average should be used. For this

reason, PPIs are the best indicators for the deflation of output of goods and services (either by

product as required or by industry).

9.54. Most PPIs are collected as producers’ prices, and these will suffice for many industries.

Industries paying large amounts of excise duties, however, such as oil, alcohol and tobacco, will

require adjustments when, for example, rates have changed. One disadvantage of PPIs is that they

are mostly Laspeyres-type indices and might use fixed weighting schemes, generally updated only

once every five years. This militates in favour of applying PPIs at the lowest possible level of

detail when deflating the domestic supply of products for domestic consumption.

9.55. It is not always possible to observe prices directly because the products of concern are not

the same over time, as is the case, for example, with unique goods and services and goods that

change rapidly in quality, such as computers, mobile phones, tablets, and so forth.

9.56. If the products in the domestic output part of the supply table which are sold for domestic

consumption are deflated using PPIs (or adjusted to form basic price indices), then the same indices

need to be used for the corresponding products in the domestic use table on the right-hand side of

the H-Approach, thereby ensuring consistency together with balance. The domestic output part of

the supply table that is sold for export should be deflated using export price indices for consistency.

Export price indices are covered later in this chapter.

9.57. For some products, price indicators other than direct observation may need to be

considered. These include tariff indices, model pricing, hedonic price indices, unit value indices,

consumer price indices, extrapolation by quantity indicators, input methods, and non-market

production.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

287

(i) Tariff indices

9.58. Certain types of services (for example, commercial services and services of general

medical practitioners) are paid for on a tariff basis, for example, with a fee per time unit. One

problem arising with indices of these types is that, in this approach, no account is taken of changes

in the quality of the services provided and of changes in the productivity per time unit. Thus, tariff-

based price indices are only appropriate deflators if adjustments can be made for changes of quality

and productivity or when it is known for sure that such changes are within acceptable limits.

(ii) Model pricing

9.59. In the model pricing approach, the producers are asked to provide price estimates for

typical products. Model price indices are candidate approximate deflators when there are

significant changes in product specification from one year to the next and, in particular, in areas

where products are unique. An important advantage of these indices is that, since the same product

or project is priced, the quality is unchanged. They do have some disadvantages, however, in such

areas as rapid product change and the degree of representation of the observed price change for

total supply is questionable.

(iii) Hedonic price indices

9.60. Hedonic price indices are candidate deflators when product specification and quality

change significantly. This method is based on an assessment of certain measurable characteristics

of products. For example, in personal computers, memory and processing speed are two such

characteristics. The main advantage is that quality changes are explicitly captured, enabling

productivity changes also to be taken into account. A serious drawback is the complexity of the

method. Furthermore, the resulting quality adjustment factor seems to be highly dependent on the

choice of the characteristics and the choice of the regression model.

(iv) Unit value indices

9.61. Unit value indices of a product may be derived when, for both the current year and the base

year, information on value and quantity is available for domestic supply (for example, from

production surveys). Dividing the values by the corresponding quantities gives the so-called “unit

values”. Under certain conditions, a unit value index can be applied as a price index, for example,

when a PPI is not available. Unit value indices often cover a heterogeneous set of products, and,

therefore, their usefulness is generally limited when they are applied to homogeneous products

where the quality does not change rapidly over time.

9.62. Unit values are quantity weighted as opposed to the regular price indices, which will be

time weighted (such as the annual average of monthly indices). In some cases, where quantities do

not move smoothly over the year and prices vary considerably, unit values may be the only

approach to get the correct relationship between current price value and volume. Thus we may

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

288

often choose to rely on unit values for energy products and some agricultural products, with

changes in inventories in current prices adjusted accordingly.

(v) Consumer price indices

9.63. For some products, CPIs may be used as approximate deflators for domestic supply. This

is only possible, however, in cases where private households buy a considerable part, or all, of the

supply of a product, and trade and transport margins and taxes and subsidies play a very small part

in the value at purchasers’ prices. Special attention must be given to changes in tax rates, in

particular such taxes as VAT. As only non-deductible VAT should be taken into account, any

modification in the legal entitlements to VAT deduction must be treated in the same way as a

modification of the rates of invoiced VAT and therefore as a variation of the price of the tax. This

effect would not be fully detected if CPI were used to deflate taxes because only the products of

final consumption are considered in CPI, and not, for example, those of intermediate consumption

of exempt industries.

9.64. The CPIs are serious candidate deflators for service products mainly provided to private

households. An advantage of CPIs is that they take due account of changes in quality. On the other

hand, most CPIs are of the Laspeyres type and might use fixed weighting schemes, generally

updated only once every five years, and this is a disadvantage. This argues in favour of applying

CPIs at the lowest possible level of detail when deflating the domestic supply of products for

domestic consumption.

(vi) Extrapolation by quantity indicators

9.65. Although the standard price method can be applied for many goods and services, there are

still some transactions for which observation of prices has not been realized or is even not possible.

The latter category includes those cases where, owing to the nature of the definition and

measurement of output in current prices, direct observation of appropriate prices is not possible:

for example, non-market services, FISIM and insurance companies. If price observation is not

possible, then the use of quantity information is an alternative.

9.66. As mentioned above, for some products, it is impossible to collect price data and, in order

to decompose a value change in a price change and a volume change, quantity indicators must be

used. For some industries, mostly the object of government involvement (for example, public

transport, medical services and cultural services) or government supervision (for example, banking

and insurance), a considerable volume of detailed quantity data are already collected by the

national statistical offices or government agencies. Examples may be found in the medical sector

(for example, number of inpatients of short-stay hospitals classified by diagnosis-related groups),

the cultural sector (for example, the number of visitors attending theatrical performances), and the

banking sector (for example, the number of saving accounts, number of credits granted to

commercial and private customers, number of payments on bank accounts). Of course, quantity

indicators must fulfil the general requirements concerning quality adjustment.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

289

(vii) Input methods

9.67. Input methods use the weighted price or volume changes of intermediate and primary

inputs as a proxy for the price or volume change of the output of an industry. The advantage of

input methods for deflation within a SUTs framework is that the necessary data are readily

available, as all inputs exist in the SUTs in current prices and can be deflated on the use table side

in a manner consistent with the supply table side. It should be noted that the input method is

recommended for capitalized research and development, although it is not the total production of

an industry. On the other hand, a considerable disadvantage is that the price and volume indicators

are not directly related to output. As a result, the change in GVA in volume terms, and also

productivity changes of an industry, cannot be properly calculated. For that reason, input methods

must be avoided as far as possible. Another disadvantage is that input methods can only be applied

for the total production of an industry and a separate deflation of the different products of an

industry is impossible. It should be noted that, although the output method is recommended by the

SNA, it should only be used when the method and the results have been tested carefully for an

appropriate number of years.

(viii) Non-market production

9.68. Special attention must be paid to non-market production by general government and

NPISHs. By definition, the output of non-market producers in current prices equals the sum of the

costs of inputs. As the SUTs accounting rules are valid both in current prices and in volume terms,

it can be argued that the output of non-market producers in volume terms equals the sum of inputs

in volume terms. This means that, in fact, an input method is applied. This approach, however,

places a considerable restraint on the estimation of volume and price indices for the non-market

services. Independent estimates of GVA in volume terms and productivity changes are not possible

if the input method is used. If non-market services contribute a significant amount in an economy,

estimates of the volume growth of such macroeconomic variables as GDP are liable to be biased

if input methods are applied. Applying quantity methods can improve the estimates of output and

GVA in volume terms. The quantity indicators should preferably be adjusted for changes in

quality.

(b) Imports of goods and services

(i) Import price indices

9.69. Import price indices usually meet the general requirements for price indicators, such as

valuation, adjustment for quality change and detail. Accordingly, import prices are the best

indicators for the deflation of import of goods and services. One disadvantage of most import price

indices are that they tend to be Laspeyres-type indices and may use fixed weighting schemes,

generally updated only once every five years. This militates in favour of the application of import

price indices at the lowest possible level of detail when deflating the imports of goods and services.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

290

One problem here is that import price data covering services are of limited availability,

necessitating the search for alternatives.

(ii) Unit value indices

9.70. Foreign trade statistics often provide the value of imports along with the corresponding

quantities at a detailed level. Using this information, unit value indices can be derived. One

problem arising with unit value indices when used for deflation purposes is that they often cover

a heterogeneous set of product groups. Sometimes the unit of measurement is the kilogram or the

unit is simply the number of items. This means that, in many cases, unit value indices have

heterogeneity problems, limiting the scope for their use as deflators. If, however, no appropriate

information from price statistics is available, and the unit values refer to similar mass products for

which the quality does not change rapidly over time, they may be applied as useful proxy deflators.

(iii) Other proxies

9.71. Generally, directly observed deflators for services are limited in terms of availability,

which means that the deflation of services within SUTs may need to use proxies based on rough

assumptions – nonetheless, deflation through SUTs ensures consistency. A good assumption may

be that, for every product, the price generating conditions at the domestic market tend to bring

about one price, which may hold for both domestic supply and imported services. In terms of the

strength of this assumption, the price index of the domestic supply of a service is an acceptable

proxy for the price index of the imports of that service. The validity of the assumption depends on

whether imports are a large part of the domestic market and on whether imported services are of

same quality as domestically produced services.

2. Use table at basic prices

9.72. In the ideal scenario, the use table at basic prices can be derived as the sum of the domestic

use table at basic prices and the imports use table at basic prices both in volume terms.

(a) Domestic use table at basic prices

9.73. Assuming a competitive economy, for the deflation of the use table with goods and services

from domestic production, PPIs are appropriate price indices, as (in the main, except in such areas

as duty-related industries) they are valued at basic prices. In cases where volume indicators are

used for the compilation of the volume estimates in the supply table, the residually derived price

indices (in other words, the implied price index derived from the difference between the value

index and the volume index) can be used. If available, it is better to apply dedicated volume or

price indices for specific transactions.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

291

(b) Imports use table at basic prices

9.74. Assuming a competitive economy, for the deflation of the imports use table, import prices

are appropriate price indices. In cases where volume indicators are used for the compilation of the

volume estimates in the supply table, the residually derived price indices can be used. If available,

again, dedicated volume or price indices should preferably be applied for specific transactions.

3. GVA by industry

9.75. Although GVA is not deflated directly, this section covers how GVA is established in

volume terms and its constituents.

(a) Double deflation approach

9.76. Total GVA in current prices by industry is compiled as the difference between output and

the intermediate consumption of goods and services. For the estimates in volume terms, the same

method is applied. As a result, the following condition holds:

GVA in volume terms equals deflated output

less deflated intermediate consumption

9.77. The corresponding price and volume indices are derived afterwards. This approach is also

known as the “double deflation” approach.

9.78. From a theoretical perspective, this approach is superior to the so-called “single deflation”

methods, since it takes into account changes in both the composition of outputs and composition

of inputs in order to derive GVA as a residual. With a double deflation approach, the volume index

of GVA is the result of independent estimates of the volume indices of output and intermediate

consumption, and the results are pre-eminently appropriate for productivity analysis. It is

important to note that additional quality assurance checks are often needed when establishing

plausible volume growth rates, in particular, where intermediate consumption forms a large

proportion of output. There are three variations of the double deflation approach.

9.79. Double deflation: As described, “double deflation” covers the deflation of current price

estimates for output and of intermediate consumption separately using appropriate price indices.

The volume estimate of GVA is derived by subtracting the volume of intermediate consumption

from the volume of output.

9.80. Double extrapolation: In double extrapolation, the previous year values of output and

intermediate consumption are extrapolated using appropriate volume indices, and then the volume

estimate of GVA is derived by subtracting the volume of intermediate consumption from the

volume of output.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

292

9.81. Extrapolation and deflation: This is a combination of the extrapolation of output of the

previous year by a volume index and the deflation of intermediate consumption of the current year

by a price index; then the volume estimate of GVA is derived by subtracting the volume of

intermediate consumption from the volume of output.

(b) Compensation of employees

9.82. The compensation of employees is part of total GVA and it is useful to estimate it in volume

terms as it increases the range of options for economic analysis using SUTs: for example, the

results can be used in the analysis of labour productivity. Another application is in price analysis,

where, for example, the price change of the output of an industry is linked to, and explained by,

the price changes of the inputs, including compensation of employees.

9.83. The compensation of employees consists of two parts, wages and salaries, both in cash and

in kind, and employers’ social contributions. The deflation of both parts should be closely

connected, since both relate to the same labour input. Thus, both volume indices must be the same,

rendering it unnecessary to estimate price and volume indices for both parts separately.

9.84. Since employers’ social contributions are liable to complex legislation, it is difficult to

observe their price index. This means that, in practice, wages and salaries will be deflated, and the

resulting volume index will also be applied in the calculation employers’ social contributions in

volume terms.

9.85. An important question relates to the appropriate unit of the volume of labour. Many

candidate units suffer from heterogeneity; for example, the numbers of employed persons do not

give an indication of the number of hours worked per person. Even full-time equivalent jobs are

not really an adequate unit since they do not carry information about reductions in working hours

and differences in the education level, skill and other attributes of the employees. For this reason,

for the purpose of measuring the volume of the input of labour in an industry, the most appropriate

quantity unit may be the actual number of hours worked, classified by education level, skill, and

other properties. The corresponding price is the value of this unit.

9.86. The payment of other taxes on production is related to the use of certain inputs in the

production process or to socially unwelcome results of production processes. Examples of the

former include taxes on real estate property, taxes on motor vehicles owned by producers. Levies

on pollution caused by production processes represent an example of the latter.

9.87. Taxes can be based on values (for example, the value of a building) or quantities (for

example, tons of pollutants), which means that the deflation of other taxes on production is in

principle comparable with the deflation of taxes on products, and that the same formulae are

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

293

applicable. In practice, however, the deflation of other taxes on production is more difficult to

achieve because of a serious lack of appropriate indicators for price and volume.

9.88. In principle, price or quantity indicators can be used to derive other taxes on production in

volume terms. Because of the complexity of the tariff structure of most taxes and the lack of

appropriate data, quantity methods will prevail. The use of quantity indicators requires a direct

link between them and the tax. For instance, the indicator for the tax on real estate property needs

a direct relation to the amount of real estate property owned by producers. A candidate proxy

indicator is the volume index of the total stock real estate property. The index of the total tons of

emitted pollutants per kind of pollution tax could serve as an appropriate indicator for taxes on

pollution. The price indices are derived afterwards from the combination of the value index and

the volume index, and they can be applied for the deflation of the tax payments by industry.

9.89. The practical elaboration of the volume estimation of taxes on production presented above

can be similarly applied to subsidies on production.

(d) Gross operating surplus

9.90. Gross operating surplus in volume terms is a residual item calculated as GVA minus

compensation of employees and minus other taxes on production plus subsidies on production.

Direct deflation of gross operating surplus is not possible because no appropriate price or volume

indices are available. Furthermore, the economic interpretation of gross operating surplus in

volume terms is questionable, and many view it as a meaningless concept.

4. Valuation matrices

(a) Trade margins

9.91. Trade margins are the remuneration for the services mainly provided by the trade industry

to producers, consumers and exports in the distribution of goods. Trade margins can also be

generated by industries other than the trade industry. As with other services, the appropriate

deflation of trade services requires price or volume indicators directly related to the service

provided. Alongside the difficulty of defining the services provided by the trade industry precisely,

numerous aspects influence the quality of the services of the trade industry, such as the amount of

information given to the customers, after-sales services, delivery time, assortment, competence of

shop assistants and availability of parking lots. As a result, methods to observe or derive price and

volume indices based on direct price and quantity indicators are not available.

9.92. As trade margins in current prices are defined as the difference between the value of goods

sold and the value of the same goods purchased for resale by trade industry, double deflation would

offer an alternative, and theoretically sound solution. This would comprise the independent

deflation of sales and purchases for resale, and the subsequent calculation of trade margins in

volume terms as the difference. This approach requires high quality price indices for both

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

294

purchases for resale and sales of products by the trade industry (and other industries as

appropriate).

9.93. A third option is to apply a proxy for the estimation of the volume index of the trade margin

on a product, based on the assumption that the volume change of trade margins equals the volume

change of the underlying product flow. An alternative means of formulating this proxy is to take

as the percentage of trade margins in volume terms to be applied on the product flow in volume

terms, the percentage of the current prices of period t-1. The percentages of trade margins are

defined as the ratio of trade margins and the relevant product flow valued at basic prices. In this

option, the price change is a residual item derived from the current price trade margins and the

trade margins in volume terms. This method provides better quality results when applied at a

detailed product and industry level. Application of the margin rate in the previous year assumes

no change in quality on the margin. Some countries do this by taking the mid-point rate between

the two years, but other approaches can also be applied.

9.94. For every entry of the use table, if applicable, the trade margins in volume terms can be

estimated as:



/

= 

/

× 



where



/

= trade margins of  in prices of 1



/

= trade margins of 1 in prices of 1





= volume change of the underlying product

9.95. The underlying assumption is more valid, if the degree to which trade involved in the

concerning transactions does not change from one year to another. The position of trade in a

market, however, is reflected in what might be termed the “involvement rate”, which can be

defined as the ratio between turnover of trade and the relevant product flow which can differ from

year to year. These changes influence the estimates in volume terms.

9.96. The relation of the flow and the turnover of trade can be written as:





= × 



where





= volume index of turnover trade

F = rate of trade in the product flow

Trade margins in volume terms can be written as:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

295



/

= 

/

× 



9.97. If = 1, then the involvement rate of trade in the product flow has not been changed from

period 1 to period , and the volume index of the turnover of trade equals the product flow.

9.98. If 1, then the product flow assumption is not valid. In order to refine the estimates,

data must be collected on involvement rates by product (and preferably industry).

9.99. A further major improvement can be achieved by collecting a detailed breakdown of trade

margins, by type of product, and by type of outlet, assuming that different outlets provide different

qualities of services. In this way, the quality changes due to turnover shifts between outlets can be

addressed.

(b) Transport margins

9.100. For transport margins, there is more than one way of estimating the volume estimates.

9.101. The first approach is similar to the method for compiling trade margins in volume terms –

that is using the rates of the previous year. This implicitly assumes that transport costs are

proportional to the value of the product, something which may not be universally true.

9.102. An alternative option for the deflation of transport margins is the use of price indices for

the output of transport industries. A necessary condition is the existence of a matrix of transport

margins, by type of transport (column) and by type of product (row). By column, the price index

of the relevant type of transport can be applied. The resulting volume change of the transport

margins can be checked for plausibility with the volume changes resulting from the so-called

“margin method”. Generally, it is expected that these two volume changes should be similar.

9.103. A further approach is first to derive volume estimates, by applying the volume changes of

the transported products on the previous years’ results, and, second, to inflate these with the

appropriate price indices in order to arrive at current price estimates. Consequently, the initial

current price estimates will then be overruled.

9.104. Taxes on products are taxes that are payable per unit of a certain good or service purchased.

The tax may be a specific amount of money per unit of quantity of a good or service, or it may be

calculated as a specified percentage of the price per unit or value of the goods and services

purchased.

9.105. Taxes on products affect the price of a product and not the volume. This means that, for

deflation, for a specific product, it is a requirement that the volume index including any taxes on

products equals the volume index excluding any taxes on products. As a result, the volume index

of the tax must also equal the volume index at basic prices of the product on which the tax is

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

296

applied. It should be noted, however, that the volume index of GVA for the whole economy will

not necessarily move in line with the volume index of GDP as there is no direct link between the

volume of taxes on products (or subsidies on products) and the volume of GVA. The taxes on

products are directly linked to the sales of goods and services and therefore relate to the volume

of output and not to GVA. The volume change of GVA is not necessarily the same as the volume

change of output, because the volume change of intermediate consumption might – or will – be

different as a consequence of more efficient production, outsourcing and other effects.

9.106. In the case of taxes on products on “quantities”, for every entry of the use table, if

applicable, taxes on products in volume terms can be estimated as:



/

= 

/

× 



where



/

=tax on products  in prices of 1



/

=tax on products 1 in prices of 1

9.107. Examples of this application cover excise duties on tobacco, alcoholic drinks and fuel.

9.108. In the case of taxes on products on “values”, for every entry of the use table, if applicable,

taxes on products in volume terms can be estimated as:



/

= 

/

× 



where



/

=tax on products  in prices of 1



/

=tax on products 1 in prices of 1

9.109. An example of taxes levied on prices is VAT.

9.110. In the case of taxes on products on “value”, the price index of the tax is usually different

for different transactions. The reason is that they depend on the price index of the value at basic

prices of the transactions. Furthermore, different tariffs exist for different products.

9.111. This approach to the calculation of taxes on products can lead to odd-looking results when

a new tax appears or an existing tax disappears, as shown in Box 9.1.

(d) Subsidies on products

9.112. The practical elaboration of the estimation in volume terms of the taxes on products

presented above also applies in the same way to subsidies on products – thus the equations in that

section also apply to subsidies on products (with the replacement of T by S).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

297

9.113. It must be recognized that the assumption that the volume change of trade (and transport)

margins and taxes and subsidies on products equals the volume change of the transactions at

purchasers’ prices can lead to unacceptable results – the focus is the volume change at basic prices.

For products with a rapid increase of quality (for example, computers, mobile phones, and so forth)

and also the volume changes of the relevant valuation layers, this change in quality should be

included. This may lead to unacceptable growth rates of GVA and labour productivity for specific

branches in wholesale and retail, necessitating ad hoc adjustments.

9.114. Similar to taxes, new subsidies can appear or an existing subsidy disappears, and this is

covered in Box 9.1.

Box 9.1 Treatment of newly introduced and disappearing taxes and subsidies

Newly introduced and disappearing taxes and subsidies

When using Laspeyres volume indices and Paasche price indices, taxes on products and subsidies on products

affect the price of a product and not the volume, implying that the volume index of the value including tax (or

subsidy) of a product equals the volume index of the value excluding tax (or subsidy).

Therefore the volume index of the value including tax (or subsidy) also equals the volume index of the tax (or

subsidy) value. In the case of newly introduced or disappearing taxes (or subsidies), these conditions give rise

to remarkable results. In the example used to demonstrate the impact, trade and transport margins are omitted

for convenience. However, these results are in conformity with the registration of changes in taxes on products

as a price change.

Newly introduced taxes on products

Applying the guidelines, the volume change at purchasers’ prices equals the volume change at basic prices,

which means that taxes on products in volume terms equal zero, while the current price amount is not zero, as

shown in the table below.

As expected, the introduction of a tax on products results in an increase in the purchasers’ prices.

Disappearing taxes on products

Applying the guidelines, the volume change at purchasers’ prices equals the volume change at basic prices,

which means that taxes on products in volume terms are not zero, while in current prices the amount equals zero,

as shown in the table below.

As expected, the disappearance of a tax on products results in a decrease in the purchasers’ prices.

Year Pr ic e Year t

Volume Year t-1

Current prices index Volume terms index Current prices

Output at basic prices 1 000 100 1 000 100 1 000

Taxes on products 100 0 0 0 0

Output at purchasers' prices 1 100 110 1 000 100 1 000

Year Pr ic e Year t Volume Year t-1

Current prices index Volume terms index Current prices

Output at basic prices 1 000 100 1 000 100 1 000

Taxes on products 0 0 100 100 100

Output at purchasers' prices 1 000 91 1 100 100 1 100

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

298

5. Use table at purchasers’ prices

9.115. The use table at purchasers’ prices can be derived from the use table at basic prices and the

valuation matrices. In order to ensure consistency in the system, the bridge columns between the

supply table at basic prices and the use table at purchasers’ prices are derived from the valuation

matrices (row totals).

9.116. For the use table at purchasers’ prices, alternative options for price and volume estimation

are available using indicators appropriate for this valuation. This option entails additional

plausibility checks on volumes and prices, in particular for the valuation matrices.

(a) Intermediate consumption by industries

9.117. Intermediate consumption price indices (ICPIs) usually meet the general requirements such

as valuation, adjustment for quality change and detail. For that reason, ICPIs are the best indicators

for the deflation of intermediate consumption of goods and services. A key problem, however, is

that ICPIs are very rarely collected by national statistics offices and, if available, do not cover

intermediate consumption of services. Thus, in a sense, the H-Approach removes this problem by

deflating at basic prices (or producers’ prices) and at a very disaggregated level by product, in

which process a single price can be used for both output and intermediate consumption by product,

thereby matching the price paid by the purchaser with the price received by the seller.

9.118. In some cases, where ICPIs are not available, CPIs can be used as proxy deflators for

intermediate consumption of products. An important requirement is that market conditions for

intermediate use and such areas as household final consumption expenditure are comparable. This

means, for example, that the share of wholesale and retail margins in the purchasers’ price should

be the same. One example of goods where intermediate use and household final consumption

expenditure often show comparable price changes is fuel for motor-vehicles.

9.119. In a number of cases, the error in the estimation of total GDP due to the use of less

appropriate price indices will be limited. When intermediate consumption accounts for the bulk of

the turnover of a domestically produced product, the under-estimation of intermediate

consumption and, thus, over-estimation of GVA in one industry will be counter-balanced by an

under-estimation of output (product, trade or transport margins), thus an under-estimation of GVA

in another industry.

(b) Exports of goods and services

9.120. Export price indices usually fulfil the general requirements such as valuation, adjustment

for quality change and detail. For that reason, export price indices are the best indicators for the

deflation of exports of goods and services. One problem here is that the availability of export price

statistics covering services tends to be limited. In addition, they often have the added disadvantage

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

299

of being Laspeyres-type indices and use fixed weighting schemes generally updated only once

every five years. This militates in favour of their application at the lowest possible level of detail.

9.121. Where deflation by unit value indices is concerned, foreign trade statistics often provide

the value of exports along with the corresponding quantities at a detailed level. From this

information, unit value indices can be derived. A problem with unit value indices when used for

deflation purposes is that they often refer to heterogeneous product groups. The unit of

measurement can be kilos or simply the number of products. That means that, in many cases, unit

value indices suffer from heterogeneity issues, limiting the possibilities for their use as deflators.

If, however, no appropriate information from producer’s price statistics is available and the unit

values refer to similar mass products where the quality does not change rapidly over time, then

they can be applied as useful proxies of deflators.

9.122. Currently, deflation using exports price data and unit value indices is only possible for the

export of goods. Direct deflators for services are limited in availability. A general problem consists

in the exact observation of the exports by product group according to the classification in the SUTs.

The second, and for deflation most important, problem is that the price observation of exported

services is not well developed in many countries. For that reason, in the national accounts, there is

a tendency for the deflation of the exports of services to resort to proxies based on rough

assumptions.

9.123. A simple but rough assumption is that, for every product, the price index for exports equals

the price index of domestic production. Another possibility would be to collect information on the

price changes of that service in the customer countries (see imports of services).

9.124. CPIs usually fulfil the general requirements such as valuation, adjustment for quality

change and detail. For this reason, CPIs are the best indicators for the deflation of household final

consumption expenditure (for both goods and services). Most CPIs are Laspeyres-type, which

militates in favour of the application of CPIs at the lowest possible level of detail. Balancing the

SUTs at basic prices is complicated when CPIs are used, owing to the differences in valuation.

Thus household final consumption expenditure deflated using CPIs should be used to validate the

household final consumption expenditure deflated at basic prices and transformed into household

final consumption expenditure at purchasers’ prices, which is the right-hand side of the H-

Approach.

(d) Government consumption

9.125. Collective and individual government consumption equals government production minus

sale of market production by government and own account fixed capital formation. Estimates in

volume terms can be derived following the same approach. For social benefits in kind, similar

indicators may be used as for consumption of households.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

300

(e) Gross fixed capital formation

9.126. Specific price indices for fixed capital goods usually fulfil the general requirements such

as valuation, adjustment for quality change and detail. For that reason, directly collected specific

price indices for capital goods form the best indicators for deflation of gross fixed capital formation

in goods. A major problem is that price indices for capital goods are hardly ever collected as part

of the collection of prices by national statistics offices, and proxy producer price-type indices are

used instead. Again, the disadvantages are that price indices for capital goods are often Laspeyres-

type and that they use fixed weighting schemes generally updated only once every five years. This

militates in favour of the application of price indices for capital goods at the lowest possible level

of detail when deflating the domestic supply of products. For gross fixed capital formation, more

weight will be given to deflation through the basic price valuation to purchasers’ price valuation

on the right-hand side of the H-Approach, when compared with the results using proxy price

indices for capital goods.

(f) Changes in inventories

9.127. The calculations of changes in inventories in current prices and in volume terms are often

closely interlinked. If high quality current price estimates can be made because reliable and

appropriate data are available, then it is often possible to make high quality estimates in volume

terms as well, since the same data are used.

9.128. In the ideal case, information is available on the exact times and quantities of additions to

and withdrawals from the inventory and the price of the product at those times. It is then in

principle straightforward to calculate the changes in inventories in current prices and in volume

terms. Additions and withdrawals have to be valued at the prices prevailing at the times at which

they take place. The changes in inventories in volume terms can be calculated by valuing the

quantities of additions and withdrawals at the average prices of the previous year.

9.129. In practice, the data available for the calculation of changes in inventories are not sufficient

for a perfect estimation. Assumptions and approximations have to be made. The estimation

methodology for changes in inventories (both in current prices and in volume terms) is highly

dependent on the kind of information on inventories that is available. In general, enterprises will

not provide data on quantities but only on the value of their inventories at the beginning and end

of the year according to their own bookkeeping system. This makes it difficult to calculate the

current price value from the volume change, as the adjustment for holding gains and losses is

incorrect or missing.

9.130. These bookkeeping systems also do not generally value inventories according to SNA rules

but follow other systems, such as historic cost systems, the LIFO system, and others. In

consequence, these values cannot be used directly in the national accounts. In order to calculate

correctly the change in volume of inventories, information is needed on the bookkeeping system

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

301

used in the enterprise. The first step is to estimate the change in volume, and the result should then

be multiplied with an appropriate price index to arrive at changes in inventories in current prices.

As a result, the quality of this process, and the quality of the subsequent estimates, provides some

scope for adjusting the current price changes in inventories in the balancing process.

F. Input-output tables in volume terms

9.131. As with SUTs, IOTs can be compiled in volume terms. The reference year can in particular

be a problem for IOTs, however, because they are often compiled at irregular, non-annual intervals,

such as once every five years. This does not line up with other national accounts data which are

required by international standards to make use of the previous year as base year in the calculation

of volume measures and the chaining of annual data at prices of a fixed reference year.

9.132. The SUTs in volume terms will generally be compiled at prices of the previous year. This

makes possible the calculation of growth rates by comparing the volume measures with the current

price values of the previous year. Results of the chaining of complete SUTs will not be additive:

that is to say, in the resulting SUTs the elements of a given row will not add up to the row total,

and the same will be true for the columns. The resulting tables can only be used to analyse the time

path of one particular element at a time. They cannot be used to any real effect for such purposes

as the analysis of the time path of the total input structure of an industry or the market shares for a

product.

9.133. As for IOTs, if they are only compiled once every five years, it is of course possible to use

the previous years’ prices as well but the results cannot be used to calculate growth rates, and this

clearly limits the usefulness of such tables.

9.134. The alternative would be to compile IOTs in prices of the year five years prior to the current

year, for example, the year 2010 in prices of 2005. This could be done by performing the same

transformation process as for the current price IOTs, but it would require the availability of

coherent SUTs in the same valuation, which is a problem when chaining is used, as already

mentioned.

9.135. Another possibility for the derivation of such IOTs is to deflate directly the IOTs in current

prices by finding appropriate price or volume indices for the products. These should be indices of

the price or volume change in the five years between the base year and the current year. This

procedure will entail the use of a different base year than for the SUTs in volume terms, introducing

the risk of inconsistencies.

9.136. The recommended approach would be to compile IOTs annually, and to apply the same

price and volume methodology as used for SUTs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

303

Chapter 10. Linking the supply and use tables to the institutional sector

accounts

A. Introduction

10.1. When compiling SUTs, it is important to ensure that they are linked to, and consistent with,

the institutional sector accounts of the SNA. This is fundamental to ensuring a full integration of

the accounts and also a full integration of the SUTs with the regular annual compilation of the

national accounts. In this way data in the SUTs, such as GVA and GDP, are made to be consistent

and coherent with the institutional sector accounts, and vice versa. This is achieved through the

compilation and balancing process of SUTs incorporating a table that cross-classifies data by

industry, by type of factor incomes and by institutional sector.

10.2. Linking SUTs to the institutional sector accounts extends the role of SUTs, enabling them

to increase the quality, consistency and coherence of the national accounts, where the SUTs have

specific links bringing together parts of the national accounts. Figure 10.1 shows the links between

the industry accounts (for example, the SUTs system) and the institutional sector accounts as part

of the balancing framework within the national accounts. Figure 10.1 provides a different

perspective from that illustrated in figure 1.1 but contains a more detailed diagram linking the two

parts of the national accounts framework.

10.3. This chapter describes the links between the SUTs and the institutional sector accounts.

Section B provides a description of the institutional sectors in the national accounts and the

differences in perspective between the institutional accounts and SUTs. Section C provides a

description of how the SUTs are linked to the institutional sector accounts and gives the layout of

the table linking these accounts. Section C also provides a numerical example of the linking table

and how the SUTs are linked to the goods and services accounts, production accounts and

generation of income accounts. Section D presents various approaches to establishing the link

between the SUTs and the institutional accounts and describes some issues that may arise in the

compilation of the linking table.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

304

Figure 10.1 Links between the industry accounts and the institutional sector accounts

B. Institutional sectors and subsectors

10.4. The institutional sectors of the 2008 SNA group together institutional units on the basis of

their principal functions, behaviour and objectives. The following institutional sectors are

distinguished in the 2008 SNA:

• Non-financial corporations: these are institutional units that are principally engaged in the

production of market goods and non-financial services (2008 SNA, para. 4.94).

NPISH

Balancing

Input-output approach

Institutional approach

Domestic

output at basic

prices

Use table at

purchasers'

prices

Imports use

table at basic

prices (CIF)

Valuation

matrices

Production

approach

Income

approach

Expenditure

approach

Supply table at

basic prices

with transforma-

tion into

purchasers'

prices

Use table at

basic prices

Total economy

Non-financial corporations

Financial corporations

General government

Households

Institutional sector accounts

Supply and use tables system

GDP

Domestic

output at basic

prices

unbalanced

Use table at

purchasers'

prices

unbalanced

Imports use

table at basic

prices (CIF)

unbalanced

Valuation

matrices

unbalanced

Production

approach

unbalanced

Income

approach

unbalanced

Expenditure

approach

unbalanced

Rest of the

world

Goods and

services

account

Production

account

Distribution and

use of income

accounts

Accumu-

lation

accounts

Table linking institutional sector

accounts to the SUTs

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

305

• Financial corporations: these consist of all resident corporations that are principally

engaged in providing financial services, including insurance and pension funding services,

to other institutional units (2008 SNA, para. 4.98).

• General government: this consists of the following groups of resident institutional units:

first, all units of central, state or local government; second, all non-market non-profit

institutions that are controlled by government units (2008 SNA, para. 4.127). It consists of

institutional units that, in addition to fulfilling their political responsibilities and their role

of economic regulation, produce services (and possibly goods) for individual or collective

consumption mainly on a non-market basis and redistribute income and wealth.

• Households: these consist of all resident households (2008 SNA, para. 4.158). Households

are institutional units consisting of one individual or a group of individuals. All physical

persons in the economy must belong to one and only one household. The principal

functions of households are to supply labour, to undertake final consumption and, as

entrepreneurs, to produce market goods and non-financial (and possibly financial) services.

The entrepreneurial activities of a household consist of unincorporated enterprises that

remain within the household except under certain specific conditions.

• NPISHs: these consist of all resident non-profit institutions, except those controlled by

government, that provide non-market goods or services to households or to the community

at large (2008 SNA, para. 4.31).

10.5. Table 10.1 provides a summary of institutional sectors and subsectors with a link to the

concepts of market and non-market producers. Often there is a misconception that the distinction

between market and non-market producers corresponds to the distinction between the private and

public sectors. This is not the case in the national accounts. The public sector includes all resident

institutional units controlled directly or indirectly by resident government units. In other words,

the public sector consists of all units of the general government sector plus all resident public

corporations (2008 SNA, para. 22.164). Accordingly, the public sector includes both market and

non-market producers so long as they are controlled directly or indirectly by resident government

units.

10.6. It should be noted that the SUTs and the institutional sector accounts reflect different ways

of looking at and measuring the economy. In the SUTs, the analysis by products and industries

emphasizes the production processes, the flows of goods and services, and the use of primary

inputs (for example capital, labour, and others). In this way the units are chosen to reflect technical-

economic relations, for example, in units of production such as establishments. As a result,

economic activities are studied from the viewpoint of the specific units that carry out the

production. The balance between supply (resources) and uses of products constitutes the central

element of this type of functional analysis.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

306

10.7. In the institutional approach, the analysis focuses on the generation and distribution of

income, and the investment and financing of capital by institutional sectors. In this case, the units

that are chosen reflect the general economic behaviour of so-called “institutional units” according

to their economic objectives, functions and behaviour.

Table 10.1 Summary of institutional sectors and subsectors

Social security funds may also be grouped at the various government levels.

10.8. Table 10.2 provides a summary of the key features of the SUTs approach and institutional

approach. The two types of approaches are linked and should be integrated via the linking table

10.9. Ideally, if a single common unit could meet the needs of both SUTs and the institutional

sector accounts, this would further improve the coherence, consistency and compilation of the two

areas of the national accounts framework (see also section D on the compilation methods). In the

2008 SNA, however, the recommended unit for the SUTs is the establishment, while for the

institutional sector account it is the institutional unit. Information on the legal status and ownership

of the establishments is important for the cross-classification of establishments into institutional

sectors.

Market / non-

market

producers

Institutional sectors

Sub-sectors (summary)

Central Bank

Deposit-taking corporations except the Central Bank

Money market funds (MMF)

Non-MMF investment funds

Other financial intermediaries except insurance corporations and pension funds

Financial auxiliaries

Captive financial institutions and money lenders

Insurance corporations

Pension funds

Central government

State government

Local government

Social security funds

Market Households

Non-market

Non-profit institutions serving

households

n/a Rest of the world

Non-market

General government

Market

Non-financial corporations

Market

Financial corporations

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

307

Table 10.2 Main features of the SUT approach and the institutional sector approach

10.10. As an establishment always belongs to an institutional unit, it is possible to link the

production activities of industries and institutional sectors. The output of an institutional unit is

equal to the sum of the outputs of the individual establishments of which the institutional unit is

composed, thus including deliveries between establishments within the institutional unit.

10.11. To clarify relationships and contents of industries and institutional sectors, the 2008 SNA

proposes the cross-classification of GVA and its components (and if possible, also for output and

intermediate consumption) by both industry and by institutional sector. This is essentially the GVA

part of the use table broken down also by sectors, in addition, to become the table linking SUTs to

institutional sectors.

10.12. In order to implement the table linking SUTs and institutional sector accounts, it would be

a great advantage to have clear links between units and institutional sectors on the business register,

and as a feature of business survey results. The split by institutional sector of units classified by

industry would meet the compilation requirements. This would also facilitate similar cross-

classifications for output, intermediate consumption and variables such as gross fixed capital

formation and the compensation of employees.

C. Table linking SUTs and institutional sector accounts

10.13. Figure 10.2 shows how the SUTs are linked to the sequence of accounts by institutional

sector through a linking table. The rows of the linking table contain information by institutional

sector on the following:

• Transaction of production accounts: total output and intermediate consumption

• Transactions on the generation of income account: GVA, compensation of employees,

other taxes less subsidies on production and imports

• Transaction of the accumulation accounts: gross fixed capital formation

10.14. The linking table thus records complete data of three specific sector accounts of the system

as a whole: the production account, the generation of income account and the accumulation

SUT approach

Institutional sector approach

Production relationship

Goods and services flows

(equilibrium of resources and uses)

Goods and services account

Production account

Generation of income account

(integrated in the SUT framework)

Types of units:

Elementary Production units (establishments, etc.)

Institutional units (households, corporations, etc.)

Aggregates Industries (type of economic activity)

Institutional sectors

Objectives

To record the economic data of institutional units

grouped in terms of their economic objectives,

functions and behaviour

Accounts

A complete system of accounts:

goods and services account; production account;

distribution and use of income account;

accumulation accounts, balance sheets

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

308

accounts, broken down simultaneously by industries (by column) and by institutional sector (by

row). In this way, the interrelations between the systems become clear and their coherence is

ensured in both the institutional sector accounts and the SUTs.

10.15. The starting point for linking the SUTs to the institutional sector accounts is the supply

table at basic prices, including a transformation at purchasers’ prices and the use table at

purchasers’ prices. Table 10.3 shows a numerical example of the linking table linking the SUTs in

table 5.2 of chapter 5 and table 6.1 of chapter 6 to the institutional sector accounts.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

309

Figure 10.2 Link between the SUTs and institutional sector accounts

Domestic supply

Supply

at basic prices

Trade and transport

margins

Taxes less subsidies

on products

1 2 n

Non-financial corporations

Total output

Market output

Output for own fi nal use

Non-market output

Intermediate consumption

GVA at basic prices

Compensation of employees

Other net taxes on production and imports

Consumption of fixed capital

Operating surplus, net

Gross fixed capital formation

Financial corporations

General government

Households

NPISHs

Total Economy

Production

accounts

Income

accounts

GVA at basic

prices

Imports CIF

Products

Domestic supply

Taxes and

subsidies

Imports

Table linking the institutional sector accounts and the supply and use tables

Industries

Supply table

Industries

Use table

Use at purchasers’

prices

Final use

Supply purchasers’

prices

Intermediate consumption at

purchasers' prices

Final use at purchasers'

prices

Output by product at basic

prices

GVA at basic prices

Components

Supply at basic

prices

3 …………………...………………...………

Total

Supply and use tables framework

Total

Output at basic prices

Total

Products

Margins

Supply

purchasers’

prices

Output at basic prices

Use at

purchasers’

prices

Final uses at purchasers’

prices

Uses Resources

Financial corporations

…

Institutional Sector Accounts

Households

Uses Resources

Total economy

…

Non-financial

corporations

NPISHs

Uses Resources

…

----

---

----

…

General

government

Uses Resources

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

310

Table 10.3 Numerical example showing the table linking

the SUTs and institutional sector accounts

Millions of euros

Table based on 2011 figures from Austria

Agriculture Manufacturing Construction

Trade, transport and

communication

Finance and

business services

Other services

(1) (2)

(3) (4) (5) (6)

1 837 196 180 37 517 112 204 53 204 8 809 409 751

Market output

1 828

192 193 37 244 110 468 52 633 8 685 403 050

Output for own final use

9 3 988 273 1 736 571 124 6 701

Other non-market output

972 136 868 24 907 52 394 24 513 3 129 242 784

864 59 313 12 611

59 810 28 691 5 679 166 967

Compensation of employees 349 29 839

9 247 32 573 13 008 4 912 89 928

Other net taxes on production and imports

- 133 1 017 479 1 360 955 - 340 3 337

Consumption of fixed capital 165 12 533 1 370 8 598 9 208 526 32 401

Operating surplus, net 482 15 923 1 515 17 279 5 519 582 41 300

277 14 376 1 058

10 419

14 605 931 41 665

25 058 25 058

Market output

24 802 24 802

Output for own final use 256 256

Other non-market output

12 351

12 706

Compensation of employees 8 125 8 125

Other net taxes on production and imports 926 926

Consumption of fixed capital 1 810 1 810

Operating surplus, net 1 846

1 846

1 923 1 923

225

5 753 3 045

53 382 62 423

Market output 18

221 151 1 489 578 2 457

Output for own final use 300 314 2 223 2 837

Other non-market output 4 5 302

1 241

50 581 57 129

13 149 2 555 1 366 15 609 19 692

5 77 3 198

1 678

37 773 42 731

Compensation of employees 3

12 1 494 782 30 725 33 017

Other net taxes on production and imports 1 1

127 34

1 260 1 423

Consumption of fixed capital 1 20

1 582 625 5 890 8 118

Operating surplus, net

0 44

- 5 237 - 103 173

2 11

2 318 1 251 5 661 9 243

8 012

3 544 7 413

16 880 28 141 9 096 73 086

Market output 7 918 3 503

4 218 16 782 9 391 8 847 50 659

Output for own final use 95 41 3 195

98 18 749

249 22 428

Other non-market output

4 455 1 975 2 559

6 270 8 307

2 746 26 311

3 558 1 569 4 854 10 610

19 834

6 350 46 775

Compensation of employees 198 828 992 3 839

1 081 1 390 8 329

Other net taxes on production and imports - 1 495 60 67 267 89 - 70 - 1 081

Consumption of fixed capital 1 622 249 183 777 7 264 580 10 675

Operating surplus, net 3 233 432

3 612 5 726 11 399 4 450 28 852

2 036 129 101 586 12 114 887 15 853

8 029 8 029

Market output 7 7

Output for own final use 74 74

Other non-market output 7 948 7 948

2 356 2 356

5 672 5 672

Compensation of employees 4 944 4 944

Other net taxes on production and imports 253 253

Consumption of fixed capital 475 475

Operating surplus, net 0 0

734

9 867 199 950 44 931 134 837 109 447 79 315 578 347

Market output 9 763 195 916

41 462 127 401 88 315 18 116 480 975

Output for own final use 104 4 029 3 468 2 134 19 890 2 670

32 295

Other non-market output 4 5 302 1 241

58 529 65 077

5 440 138 991 27 466 61 219 46 538 23 841 303 495

4 428 60 958 17 465 73 618 62 909 55 475 274 852

Compensation of employees 551 30 679 10 239 37 906 22 997 41 971 144 343

Other net taxes on production and imports - 1 627 1 077 546 1 755 2 004 1 103 4 858

Consumption of fixed capital 1 788 12 803 1 553 10 958 18 908 7 472 53 480

Operating surplus, net 3 715 16 400 5 128 22 999 19 001 4 929 72 171

Gross fixed capital formation 2 314 14 516 1 160 13 323 29 892 8 212 69 418

Gross fixed capital formation

6. Total

Total output

Intermediate consumption

Gross value added at basic prices

Intermediate consumption

INSTITUTIONAL SECTORS

INDUSTRIES

Total

1. Non-financial corporations

Total output

Gross fixed capital formation

Gross value added at basic prices

Gross fixed capital formation

2. Financial corporations

Total output

Intermediate consumption

Gross value added at basic prices

Gross fixed capital formation

3. General government

Total output

Intermediate consumption

Gross value added at basic prices

Total output

Intermediate consumption

Gross value added at basic prices

4. Households

Total output

Intermediate consumption

Gross value added at basic prices

Gross fixed capital formation

5. Non-profit institutions serving households

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

311

10.16. Among the institutional sector accounts as a whole, it is the goods and services accounts,

the production accounts and the generation of income accounts that are important for balancing

the SUTs and the institutional sector accounts.

10.17. If the goods and services accounts, production accounts, and generation of income accounts

are compiled and balanced in an integrated manner as part of the SUTs compilation and balancing

process, then all the components of the first three accounts of the national accounts framework

become available from the balanced SUTs. This ensures a high degree of consistency and

coherence between the SUTs and the institutional sector accounts. In addition, there is a powerful

data quality feedback loop from the institutional sector accounts affecting SUTs, and vice versa.

10.18. Furthermore, when the data for gross fixed capital formation is compiled by industry, by

product and by institutional sector and as an integrated input to the use table, this also provides a

key link between the SUTs and part of the accumulation accounts. Although these could be

compiled as a satellite system, they should be integrated as an input to the SUTs process and are

available on a consistent basis after the SUTs are balanced.

10.19. Although this type of approach is recommended, in many countries the SUTs are compiled

separately from the institutional sector accounts. See also section D on compilation methods.

1. Goods and services accounts

10.20. The goods and services accounts show the total supply of a product for the whole economy,

together with how it has been used. The main components for the whole economy balance are:

Output + imports + taxes on products – subsidies on products (Total resources)

equals

Intermediate consumption + final consumption + gross capital formation + exports (Total uses)

10.21. The goods and services are traced through the economy from their original producers

(either resident producers or producers abroad) to their users (either resident users or users abroad).

With output being valued at basic prices and uses at purchasers’ prices, then taxes on products less

subsidies on products must be included in the resources part to ensure that a purchasers’ prices

balance can be struck.

10.22. It is important to note that the goods and services account is by definition in balance and

therefore has no balancing item, and all the components are available from the SUTs. In essence,

all these are totals of variables available in the SUTs. Table 10.4 shows a numerical example

covering goods and services.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

312

Table 10.4 Goods and services for the whole economy

Millions of euros

Table based on 2011 figures from Austria

2. Production account

10.23. The production account shows the transactions relating to the production process and is

drawn up for institutional sectors and for industries. For the whole economy, and for each

institutional sector, the resources include output and the uses include intermediate consumption.

10.24. The production account generates one of the most important balancing items in the system,

namely, GVA, the value generated by any unit engaged in production activity and in turn, the link

to the major aggregate, GDP. GVA is economically significant for both the institutional sectors

and the industries.

10.25. As with balancing items for all the accounts, value added may be calculated before or after

allowance is made for consumption of fixed capital, and is therefore available on a gross or net

basis. Given that output is valued at basic prices and intermediate consumption at purchasers’

prices, GVA will not include taxes on products and will include subsidies on products.

10.26. The production account at the whole economy level includes among resources, in addition

to the output of goods and services, the taxes on products less subsidies on products. This enables

GDP at market prices to be obtained as a balancing item.

10.27. Again, all the components are available from the SUTs and are totals for the variables

available in the SUTs. Table 10.5 shows a numerical example covering the production account for

the whole economy. The same variables underpin the whole economy by the institutional sectors,

except that GVA is shown instead of GDP as the balancing item on the uses side.

Uses Total Resources Total

Intermediate consumption 303 492 Output 578 360

Final consumption expenditure 226 258 Imports of goods and services 157 871

Final consumption by households

159 792 Taxes less subsidies on products 33 778

Final consumption by non-profit organisations

5 416 Taxes on products 34 416

Final consumption by government

61 050 Subsidies on products (-) - 638

GCF 74 612

GFCF 69 418

Changes in inventories 2 335

Acquisition less disposal of valuables 2 859

Exports of goods and services 165 648

Total 770 009 Total 770 009

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

313

Table 10.5 Production account for the whole economy

Millions of euros

Table based on 2011 figures from Austria

10.28. All the estimates for the production account for the whole economy in Table 10.5 can be

derived from the SUTs in order to derive GDP. Alongside this, GVA by industry can also be linked

and similarly derived from the same SUTs, as shown in Table 10.6.

Table 10.6 Link between GDP and industry GVA

Millions of euros

Table based on 2011 figures from Austria

3. Generation of income account

10.29. The generation of income account analyses the extent to which GVA can cover

compensation of employees and other taxes less subsidies on production. It measures the gross

operating surplus, which is the surplus (or deficit) on production activities before account has been

taken of the interest, rents or charges which the production unit must pay on financial assets or on

tangible non-produced assets which it has borrowed or rented and must receive on financial assets

or on tangible non-produced assets of which it is the owner.

10.30. The gross operating surplus corresponds to the income which the units obtain from their

own use of their production facilities. Although the institutional sector accounts have balancing

items in each of the accounts, gross operating surplus is the last balancing item in the national

accounts framework that can be calculated linking industries, institutional sectors and subsectors.

10.31. In the case of unincorporated enterprises in the household sector, the balancing item of the

generation of income account implicitly includes an element corresponding to remuneration for

Uses Total

Resources Total

Intermediate consumption

303 492 Output

578 360

Market output 480 989

Output for own final use

32 295

Non-market output 65 075

GDP 308 647 Taxes less subsidies on products

33 778

Consumption of fixed capital

53 469 Taxes on products

34 416

NDP 255 177 Subsidies on products (-)

- 638

Total 612 138 Total

612 138

Industries

Output

Intermediate

consumption

Gross value

added at

basic prices

Taxes on

products

Subsidies on

products

GDP at

market prices

Agriculture 9 867 5 440 4 427

Manufacturing 199 950 138 991 60 959

Construction 44 931 27 466 17 465

Trade, transport and communication 134 837 61 219 73 618

Finance and business services 109 461 46 538 62 923

Other services 79 314 23 839 55 475

Total 578 360 303 492 274 868 34 416 638 308 647

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

314

work carried out by the owners or members of their family which cannot be distinguished from

their profits as entrepreneur. This is referred to as “mixed income”.

10.32. In the case of own account production of accommodation services by owner-occupier

households, the balancing item of the generation of income account is an operating surplus, and

not mixed income.

10.33. The generation of income account can also be presented by industries and is usually

published with the main national accounts releases. These can be shown as the industry columns

of the use table and presented as sectors, subsectors and industries which are the source, rather

than the destination, of primary income.

10.34. It is essential that these industries (and underlying units) are the same as those applied in

the SUTs, the industry-by-industry IOTs and the institutional sector accounts. Where this is not

possible at best a clear bridge should be provided, to explain the differences.

10.35. All the components could be available from the SUTs and would consist of totals of

variables available in the GVA part of the use table if the SUTs incorporated these components as

part of the SUTs compilation and balancing process.

10.36. Table 10.7 shows a numerical example of the generation of income account for the whole

economy. It is underpinned by a similar breakdown by institutional sector (and by industry), except

that GVA is shown instead of GDP as the starting point, in line with the function of GVA as the

balancing item of the production account for each institutional sector.

Table 10.7 Generation of income account for the whole economy

Millions of euros

Table based on 2011 figures from Austria

10.37. The income approach to measuring GDP is obtained by summing together:

• Gross operating surplus

• Compensation of employees

Uses

Total

Resources

Total

Compensation of employees 144 343 GDP 308 647

Wages and salaries

Employers' social contributions

Taxes on production and imports

Taxes on products 34 416

VAT type taxes

Taxes and duties on imports excuding VAT

Taxes on products except VAT and import taxes

Other taxes on production 4 858

Subsidies - 638

Subdsidies on products

Other subsidies on production

Gross operating surplus 125 667

Gross mixed income

Total 308 647 Total 308 647

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

315

• Taxes on production and imports less any subsidies on production

• Taxes on products and imports less any subsidies on products

D. Compilation methods

10.38. Using the tables linking the SUTs to the institutional sector accounts, it is possible to make

a direct comparison with information from the SUTs and the institutional sector accounts for each

period. This at least guarantees that, after the balancing process, consistency is ensured between

the SUTs and the sector accounts. Even after independently compiling the SUTs and institutional

sector accounts, the linking table may be established to check the consistency of results.

10.39. The compilation procedure is designed in such a way that, in the first stage, the SUTs on

the one hand and the institutional sector accounts on the other are independently compiled. In the

second stage, the comparison between the two types of information is made in the linking table.

In the event of data incompatibilities, a revision process will start on both approaches until a new

assessment can be reached.

10.40. There are, however, other possible compilation methods (see Eurostat, 2008) as shown in

Box 10.1, used to link the SUTs to the institutional sector accounts.

Box 10.1 Compilation methods used to link SUTs to the institutional sector accounts

General structure of the national accounts

compilation procedure

Role of the linking matrix in the compilation

process

Method A

Independent compilation of SUTs and institutional sector

account

Ex post reconciliation of both approaches

Method B

Compilation based on instruments and statistical sources

of SUTs as a core element with a secondary role assigned

to institutional sector accounts

The linking matrix as the first stage in the compilation

of institutional sectors accounts

Method C

Compilation based on instruments and statistical sources

of institutional sectors accounts with a secondary role

assigned to SUTs

The linking matrix as the first stage in the compilation

of SUTs

Method D

Simultaneous compilation of SUTs and institutional

sector accounts

The linking matrix as a central instrument in the

compilation of the system of national accounts

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

316

10.41. In method A, the two approaches are elaborated in an independent way. The particular role

of the linking matrix is to contrast them and to facilitate the reconciliation process. Method B and

method C represent opposite alternatives in the compilation of national accounts. Method B builds

on the production SUTs methods and corresponding sources of information. Method C starts from

the institutional sectors from which SUTs elements will be derived. In both alternatives the role of

the linking matrix is similar: it represents the missing link between the two approaches. When the

SUTs approach is the starting point, the linking matrix gives data for the first two accounts of

institutional sectors – the production and generation of income accounts. In the event that

institutional sector accounts are the main and initial stage, the linking matrix helps to distribute

data over the different industries as the first stage in the compilation of SUTs.

10.42. The recommended approach would be to start simultaneously from an institutional and

SUTs standpoint, as indicated in method D. The advantage of this method is that the two different

standpoints are wholly compatible from the very beginning of the national accounts compilation

process. In terms of the relevance of the linking table, such a method would mean that it would

form the core of the whole compilation process.

10.43. The statistical requirements for this method are significant but, by that token, so are the

benefits. The main aspect is that the databases should be structured according to the institutional

sector with which the units are associated. As indicated in the linking table, five basic types of

information are required (broken down by institutional sectors) to prepare SUTs:

• Production data broken down into matrices by products and industries and valued at basic

prices, or at least the total for each industry.

• Intermediate consumption data, broken down by products and industries and valued at

purchasers’ prices, or at least the total for each industry.

• Data on the cost of primary inputs, particularly wage earners’ compensation (with a

breakdown of wages and salaries and employers’ social contributions), and consumption

of fixed capital: these data should be disaggregated by industries.

• Data on the gross fixed capital formation and stock variations broken down by types of

products and industries. In the case of gross fixed capital formation, the data are valued at

purchasers’ prices, while in the case of changes in inventories (stock variations) the data

are shown at basic prices.

• Data on labour input broken down by the employers’ industries and by employment

category (wage and salary earners, self-employed) – also defined by the numbers of people

employed and hours worked.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

317

10.44. The availability of such a cross-classified database makes possible the simultaneous

compilation of SUTs and institutional sector accounts and increases the overall improvement of

all requirements to compile the national accounts as follows:

• One of the main advantages of the linking table is that it allows for the possibility of stating

and analysing the different types of production (market, non-market, for own final use)

which depend on the institutional approach. Definition of the concepts of market output,

output for own final use and other non-market output can only be understood by looking,

in addition, at features of the institutional unit and the establishment that produce that

output. The distinctions are defined in a top-down way: that is to say, the distinction is first

defined for institutional units, then for establishment and then for their output.

• The simultaneous compilation of aspects of institutional sectors and industries is a

prerequisite for the estimation of taxes of the value-added type. If some details of

intermediate consumption and gross fixed capital formation are available in the suggested

approach, then it is possible to achieve a more accurate compilation of VAT.

• The linking table allows for a clear identification of non-market household production

activities: the (imputed) production of rental services of owner-occupied dwellings; output

of household services produced by employing paid staff; own-account construction, and so

forth.

10.45. When the compilation of the linking table is at the core of the compilation of national

accounts (as in method D in Box 10.1), it is important to cross-classify the original data by industry

and sector. In this regard, it is important to keep the link between national accounting, on the one

side, and business accounting and public finance, on the other side, as close as possible.

10.46. From the perspective of compiling the linking table, major problems arise from vertically

integrated enterprises. A vertically integrated enterprise is one in which different stages of

production, which are usually carried out by different enterprises, are carried out in succession by

different parts of the same enterprise (2008 SNA, para. 5.23). Business accounting data will be

consolidated, without specific detail on the stages and intra-enterprise transactions involved among

the different units. This causes difficulties in distinguishing intermediate consumption and other

current costs as output of one stage which is, for example, intermediate consumption of another

stage. Moreover, gross operating surplus may not be differentiated among the different parts of the

enterprise, thus appropriate adjustments would be required; more details of the type of redefinition

required may be found in chapter 5.

10.47. The 2008 SNA recommends (see 2008 SNA, para. 5.26) that, when a vertically integrated

enterprise spans two or more sections of ISIC, at least one establishment must be distinguished

within each section. With such a treatment, the activities of units engaged in vertically integrated

activities will not cross section boundaries of ISIC.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

319

Chapter 11. Balancing the supply and use tables

A. Introduction

11.1 The balancing of SUTs is a fundamental step in the compilation process of SUTs. The

usefulness of the SUTs is underpinned by the set of identities between elements of the tables which

enable the consistent integration of the components of the three approaches to measuring GDP.

11.2 In fact, since the SUTs are populated in the first stage with data derived from many sources,

each of which has its own sample and reliability margins, definitions and peculiarities, the basic

identities of the SUTs are not satisfied when the tables are first put together (as described in the

previous chapters) and the resulting estimates of GDP emerging from the three approaches are

likely to be very different, and to differ from year to year. In order to achieve a single, coherent

and consistent estimate, all the identities and plausibility relations in the SUTs have to be satisfied,

and the initial unbalanced SUTs need therefore to be balanced, preferably using a time series

perspective.

11.3 The ideal scenario, linked to the H-Approach to the compilation of SUTs (as shown in

figure 9.1), is for the full set of SUTs to be balanced simultaneously at basic prices and at

purchasers’ prices, as well as in current prices and volume terms. In addition, if the balancing also

takes into account the institutional sector accounts, IOTs, PSUTs and EE-IOTs, balanced as a

single package or sequentially, the integration and reliability of the system are greatly enhanced.

11.4 This approach, however, is demanding in terms of data, resources and computer systems.

In practice, balancing will often be less extensive and sequential procedures will be applied. The

sequential theme may, for example, concern volume estimates, valuation matrices or the import

matrix. The choice of a variant for the balancing process in practice depends upon criteria such as

the availability of data. In the estimation of volume data, the application of appropriate price

indices is a key factor to consider.

11.5 Whatever choice is made with regard to the set-up of the balancing, it is important to

recognize that any version of the SUTs or a particular stage of the process is not finished until all

the subsequent estimates are made and checked for plausibility. The balancing phase is then an

iterative process and feedback loops to earlier stages in the process will improve the quality of the

final result and can also indicate future improvements to source data.

11.6 The main objective of this chapter is to provide an overview of the balancing of SUTs.

Section B provides an overview of the basic identities that need to be satisfied in the SUTs system.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

320

Section C describes different methods of balancing – sequential and simultaneous methods – and

a general approach to the investigation and resolution of inconsistencies. Section D describes a

step-by-step procedure for simultaneous balancing in current prices and volume terms. Section E

describes the benefits of extending the balancing to include the institutional sector accounts, IOTs,

PSUTs and EE-IOTs. Lastly, section F provides a list of practical considerations for balancing.

These include, for example, the use of automated and manual balancing procedures, the role of

balancing in benchmark years and the importance of documenting adjustments to the data. Annex

A to this chapter provides a numerical example of how the unbalanced initial SUTs are balanced

through a simultaneous balancing process.

B. Overview of the system and basic identities

11.7 The balancing starts with a set of tables which consists of the following (in current prices

and previous years’ prices):

• SUTs at purchasers’ prices

• Valuation matrices

• SUTs at basic prices

• Use table at basic prices with a split between the domestic use table and the imports use

table

11.8 The full system of SUTs, as presented in figure 2.2 in chapter 2, therefore consists of SUTs

both at purchasers’ prices and basic prices and a set of valuation matrices bridging the valuation

gap between the supply table and the use table, together with the corresponding dimension

covering previous years’ prices. The use table at basic prices is also split between a table showing

uses of domestically produced goods and services (domestic use table) and a table showing the

imports of goods and services (imports use table). Although not shown here, the domestic output

table at basic prices needs to be split between that for domestic consumption and that for export

for the purposes of deflation, as covered in chapter 9. Figure 11.1 shows the full set of tables and

matrices irrespective of the price basis. In addition, the IOTs at basic prices, IOTs of domestic

output at basic prices and the input tables of imports also play a key feedback role in terms of

quality, coherence and consistency (whether through sequential or simultaneous balancing).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

321

Figure 11.1 Simplified SUTs system

11.9 In the ideal case, all tables and matrices presented in Figure 11.1 are balanced

simultaneously both in current prices (top-down in the scheme and the top left-hand side of the H-

Approach) and in volume terms (bottom-up in the scheme and the top right-hand side of the H-

Approach). By so doing, it is possible to avoid implausible results such as those showing imports

Supply table at purchasers' prices

Use table at purchasers' prices

Final use at

purchasers' prices

Domestic output at

basic prices

Total

supply at

purchas-

ers'

prices

Imports

CIF

Total

supply at

basic

prices

Taxes less

subsidies on

products

Trade

and

transport

margins

Valuation

Intermediate

consumption at

purchasers' prices

Total use

at purchas-

ers' prices

Valuation matrices

GVA at basic prices

Total output at basic

prices

Total output

at basic prices

Supply table of domestic output

Domestic use table

Final use of domestic

output at basic prices

Use of imported

products cif

Supply table at basic prices

Total output

at basic prices

Total

supply at

basic

prices

Taxes less subsidies

on products

Use table at basic prices

Domestic output at

basic prices

Imports

CIF

Final use at basic

prices

Total output at basic

prices

Total use

domestic

output at

basic

prices

Total

GVA at basic prices

Total output at basic

prices

Domestic output at

basic prices

Domes-

tic output

at basic

prices

Total output

at basic prices

Imports use table

Final use of imported

products at basic

prices

Total

Total use

imported

products

at basic

prices

Total

Intermediate

consumption of

imported products at

basic prices

Total

Retail trade margins

Wholesale trade margins

Transport margins

VAT

Other taxes on products

Subsidies on products

Total use

at basic

prices

Total

Intermediate

consumption at basic

prices

GVA at basic prices

Taxes less subsidies

on products

Total

Intermediate

consumption of

domestic output at

basic prices

Use of imported

products cif

Taxes less subsidies

on products

Taxes less subsidies

on products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

322

for a product smaller than re-exports (apart from cases when the re-exports have come from

inventories) and negative values at basic prices. In addition, the volume and price changes can be

judged on plausibility with possible implications for the current price estimates.

11.10 Being an accounting framework, the SUTs have basic identities which are directly linked

to the three approaches to measuring GDP. With the inclusion of taxes and subsidies in the SUTs,

differences will exist in the identities at the macro (total economy) and the meso (product or

industry) level.

1. Basic identities of SUTs

(a) Supply = use

11.11 The “total supply equals total use” identity must be satisfied for the whole economy (macro

level) and also for each product (product level). In the first case, at the macro level, this identity

has to be satisfied at purchasers’ prices. Total supply at purchasers’ prices consists of domestically

produced, and imported, goods and services plus taxes on products less subsidies on products. At

the macro level, trade and transport margins do not appear separately in this identity because they

are part of the output of goods and services at basic prices. Total use consists of intermediate

consumption, final consumption of households and government, gross capital formation and

exports, which are all valued at purchasers’ prices.

11.12 At the product level, the “total supply equals total use” identity is defined both at

purchasers’ prices and at basic prices. In the first case, the total supply consists of domestically

produced and imported goods and services, trade and transport margins plus taxes on products less

subsidies on products. The total use consists of intermediate consumption, final consumption,

gross capital formation and exports, which are all valued at purchasers’ prices.

11.13 In the case of basic prices, the total supply consists only of domestically produced and

imported goods and services. The total use consists of intermediate consumption, final

consumption, gross capital formation and exports, which are all valued at basis prices. In the basic

price case, the trade and transport margins are treated as ordinary services.

(b) Output = input

11.14 The “total output equals total input” identity is also defined at different levels: for the whole

economy (macro level) and by industry (industry level). For the whole economy, the output is at

basic prices and the input consists of intermediate consumption at purchasers’ prices and GVA at

basic prices.

11.15 At the industry level, this identity is also defined at basic prices. The output is defined at

basic prices and the input consists of intermediate consumption at purchasers’ prices and GVA at

basic prices. The column totals of the valuation matrices appear as separate rows in the SUTs

system at basic prices.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

323

imported

11.16 In the supply table at purchasers’ prices, the trade and transport margins appear in separate

columns. These columns provide the total constraint for the relevant valuation matrices. This

means that, for each product, the trade and transport margins in the supply table have to be equal

to the sum by columns of the relevant valuation matrices (namely, retail trade, wholesale trade and

transport margins).

(d) Value change = volume change x price change

11.17 When the above identities are not satisfied, it is not always easy to discover the causes and,

for that reason, it is always helpful to have additional information. For example, incorporating

price and volume information helps considerably with identifying and analysing inconsistencies

within the SUTs. Preferably, the basic identities referred to above should apply both in current

prices and in volume terms. This requirement depends upon the choice of the index formulae. In

this case, the combination of the Laspeyres volume index and Paasche price index formula ensures

that the identities in this section also hold for the volume terms.

11.18 With the inclusion of volume estimates, the SUTs identities in current prices and in volume

terms have to be satisfied, and the less strict relations between variables based on price and volume

changes can be judged on plausibility.

11.19 Looking at the industries, it can be seen that the volume change of production is very

similar to the volume change of intermediate consumption. This relation is stronger for the output

goods and input of raw materials than for services. Nevertheless, when there is a large difference

between the two volume changes, this indicates that there may be something wrong in the data and

further investigation is advisable.

11.20 When combined with labour data, the volume changes of GVA can be used to calculate

changes in labour productivity. It is important to note that the labour data should be calculated on

the same basis (for example, using the same statistical unit) as the economic data. This being so,

labour productivity is expected to rise gradually every year (except for periods such as the start of

a recession). A decrease or a high growth of productivity can also indicate possible mistakes in the

data.

11.21 When looking at products in a competitive economy, it is expected that price changes

should be more or less the same for all economic agents (except for areas like foreign trade). If the

price change of a certain user deviates significantly from the average, this may indicate that

something is wrong and further investigation is advisable.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

324

11.22 For household final consumption expenditure, the plausibility of volume changes of

products for general use, such as food, can be evaluated comparing them with other indicators such

as population growth.

11.23 One optional check would be to view the time series of variables. Sudden breaks in time

series can indicate a signal of implausible data in the SUTs. Again, further investigation would be

necessary before concluding that adjustments are necessary. For example, the impact of

globalization and rapidly changing ways of organizing production processes by enterprises, can

lead to justified breaks in time series.

C. Balancing

11.24 Balancing of the SUTs refers to the iterative process of reconciling differences between the

different parts of the SUTs. For balancing, there is no general theory or mathematical programme

available enabling the entire process to be automated. There is a clear, controlled role for

automated balancing techniques but after, and only after, all the significant imbalances have been

resolved manually. In balancing, however, it is very important to follow a systematic approach to

problem solving involving basic identities, checks on plausibility and credibility, and the

investigation of possible causes of inconsistencies. This section reviews the two main approaches

to balancing (sequential and simultaneous balancing) and provides a general guide on how to

investigate sources of inconsistency.

1. Simultaneous or sequential balancing

11.25 The compilation of SUTs in current prices and in volume terms can be organized in two

ways: through a sequential approach whereby the SUTs are balanced first in current prices,

subsequently deflated and finally balanced in volume terms; or through a simultaneous approach

whereby the SUTs in current prices and in volume terms are balanced at the same time. At the end

of the balancing process, the tables in current prices and in volume terms are available and

balanced. There are advantages and disadvantages to each approach but, in general, the

simultaneous balancing approach is recommended.

11.26 The main advantage of sequential balancing is that it is, in general, less complicated

because it is only necessary to deal with values in current prices during balancing and also because

there may be a lack of reliable price data at a sufficiently detailed level. The major disadvantage

of the sequential approach, however, is that problems encountered while compiling SUTs in

volume terms sometimes make it necessary to amend the current price tables that have already

been finished, and perhaps even published.

11.27 In general, in the sequential approach, it is preferable to employ an iterative procedure with

feedback loops to the SUTs in current prices. Moreover, the SUTs in current prices should not be

considered as final until all tables of the SUTs system (including SUTs in volume terms) are

checked for coherence and plausibility.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

325

11.28 The main advantage of the simultaneous approach is that it makes it possible to analyse

value, price and volume indices in relation to one another. The outcome of the analysis may affect

data in volume terms and also current price data. In other words, all three indices must give a

plausible picture. This clearly improves the quality of the outcome of the balancing process. It

should also be noted that the simultaneous approach can be useful not only in the balancing phase

but also in the phase in which basic data are prepared for national accounts purposes. This approach

affords an opportunity to check the data by comparing price and volume indices before they are

entered in the SUTs system. Simultaneous balancing in current prices and in volume terms may

result in a different allocation of adjustments than balancing in current prices only.

11.29 The simultaneous approach requires every transaction of the SUTs to be available,

including current prices, deflation detail and prices of the previous year. In order to calculate

indices, the system also requires values in current prices of the previous year. For every entry in

the SUTs, three values must be available:

• Value for year t in prices of t-1

• Value in current prices for year t-1

• Value in current prices for year t

11.30 Figure 11.2 illustrates the above configuration, presented in the form of what might be

termed a “six-pack”.

Figure 11.2 Six-pack

11.31 The six-pack table can be used by compilers of national accounts and SUTs to cross-check

consistency of data – analytical tools should ensure that such analyses are readily available to aid

validation and balancing. Although the results in current prices may at first sight look plausible,

an analysis of the volume and price data can show implausible results and lead to adjustments in

the current price data. It is important to check the comparison between changes in the volume of

output by industry, its intermediate consumption and GVA. When prices are changing rapidly it is

particularly evident that analysis in volume terms is to be preferred, for example, in the oil and

chemical industries.

11.32 One major advantage of the simultaneous approach is that it provides an opportunity to

analyse value, price and volume indices in relation to one another, and the impact of any

Description Data Description Data

at current prices 525 Price index 102.9

t in prices of

t-1 510 Volume index 102.0

t-1

at current prices 500 Value index 105.0

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

326

adjustments on all items of the six-pack immediately, in terms of plausibility, including the impact

on macro and meso economic aggregates, such as GDP and GVA by industry.

2. Balancing: investigative dimension

11.33 In general, an effective way to approach the balancing process is to investigate the

inconsistencies in the SUTs in a systematic manner. The first step would be to select the major

inconsistencies. The second step would be to carry out a critical search for results of data processed

in compiling the national accounts. With regard to the use table in particular, the main items are

the result of partitioning source data into product groups. The allocation may be changed without

altering the original aggregates. In practice, it is apparent that not all problems can be solved in

this way. The third step would be to consult the expert knowledge of the statistician who is

compiling the source statistics. If major inconsistencies still remain, the fourth step would be to

contact the reporting company and have a critical discussion regarding the data that it has provided.

11.34 The balancing is driven by two linked underlying themes: the reconciliation of estimates

of industry GVA between the income-based and production-based approaches; and the

reconciliation of supply and use for each product, essentially through the matching of production

and expenditure. As all the components of production, income and expenditure are integrated

within a single framework, when the identities are reconciled, the estimates based on the three

approaches will be equal.

11.35 It should be noted that these reconciliations must also ensure that consistency and

coherence over time is achieved. For example, consistency over time of individual series, both

within the SUTs and in suppliers’ own detailed series; consistency over time of aggregated series;

consistency of estimates in current prices, estimates in volume terms and the implied deflators,

both at the aggregate and component level; and consistency in terms of growth rates and levels.

11.36 When assessing these aspects, the impact of revisions to earlier years and the quality of the

relative data sources should also be taken into account.

11.37 It should be noted that, during the balancing, the basic identities of SUTs in current prices

and, if applicable, in volume terms, must be satisfied and that the values in the SUTs are consistent

and plausible, providing a coherent set of price and volume changes. In a set of balanced SUTs,

the identities of the framework are satisfied as well as less strict plausibility relationships, such as

the volume change of output of goods resembling the volume change of intermediate consumption.

Through the process of balancing, the detection of inconsistencies and implausibilities, on the one

hand, and the identification of their causes, on the other, forms the most important part of the

exercise. With this knowledge, the resolution of any inconsistencies is much more straightforward.

11.38 Any difference between the total supply and total use of any product points to an

inconsistency in the system, and forms the start for a balancing procedure in the cause should be

sought by:

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

327

• Analysing the transformation of source data and the validity of assumptions made (see the

various compilation chapters)

• Analysing the underlying source data, if necessary, at the unit level

• Discussing the data with experts in the respective areas or even survey respondents

• Analysing the data in the form of time series

• Carrying out a number of credibility checks, for example, of the following:

o GVA to total output ratios, while recognizing that activities such as processing require

careful consideration

o Changes in the composition of GVA weights

o Taxes on products, trade and transport margins as a proportion of the supply and use

of products

o Search for outliers in price and volume ratios (if applicable)

• Comparing data with other data sources (which are not from the statistical office or central

bank), for example qualitative and quantitative sources covering specific industries or

products such as company reports, regulatory reports, trade association analysis, and others

• Comparing and reconciling inconsistencies between different survey data sources providing

different estimates for the same or similar variables (for example, turnover from monthly

sources compared with audited annual sources)

• Using other proxy indicators to facilitate the identification of plausible SUTs variables, for

example, VAT-based indicators to compare with GVA and turnover

• Analysing related volume ratios for variables such as output and intermediate consumption

11.39 Working with statistical data based on sample surveys and questionnaires, and influenced

by, among other things, non-response type issues, involves working with reliability margins (for

each cell), so there will inevitably be inconsistencies. The cause would then be a statistical

measurement issue. In such a case, balancing could be automated using the reliability margins of

the statistics concerned as weights. Some of the methods of automated balancing described later

in this chapter are based on this principle.

11.40 Statistics, however, are never ideal and inconsistencies are not only caused by sampling

and other such activities, but may be caused by issues that are of a non-statistical nature. It is the

sources of these inconsistencies that make manual balancing essential, as a preliminary step

towards any form of automated balancing.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

328

3. Examples of causes of data inconsistencies

11.41 There could be several reasons for data inconsistencies and they can arise at various stages

of the collection and processing of data. Some of the inconsistencies that are frequently

encountered in the compilation of SUTs are presented below.

(a) Inconsistencies in data at the unit level

11.42 For the collection of data on sales and purchases, most statistical units such as enterprises,

establishments or kind-of-activity units are defined. These consist of sets of legal units. In the

simplest case, the statistical unit is the same as the legal unit but often the statistical unit consists

of more than one legal unit. Having a well-defined statistical unit does not necessarily mean that

it corresponds, for example, to tax units used by the company concerned for its tax declaration or

to the level of consolidation in the bookkeeping. Where the reporting unit uses their bookkeeping

or tax records, the reporting unit is not likely to be the same as the statistical unit. This can lead to

data missing from certain legal units or even double-counting. This risk increases when data are

collected by different agencies, for example, the national statistical offices, national central banks

and the tax authorities.

11.43 Another ever-increasing and widespread cause of inconsistencies is the impact of

globalization, reflecting issues such as production abroad and the trade flows associated with

intellectual property products, together with the impact of any change – or lack of change – of

economic ownership. When the unit in a country is the economic owner of all goods and services

purchased and sold, it will report its worldwide activity in business statistics, even when the goods

concerned never enter the country of residence of the unit. On the other hand, foreign trade

statistics on goods are based on goods crossing borders, so data on goods that never enter the

country of residence of the unit will be missing. In this case, there is an inconsistency between

business statistics and foreign trade statistics, both of which serve as a source for the SUTs system.

The Guide to Measuring Global Production (UNECE, 2015) provides extensive detail on how to

handle these types of issues.

11.44 Examples of other causes of inconsistencies at the unit level are mismatches and mistakes.

One example of a mismatch is the difference between the calendar year and the bookkeeping year,

where for a significant number of units the bookkeeping year differs from the calendar year used

in the national accounts (and other annual statistics). Entering the bookkeeping data in the

questionnaire causes inconsistencies in the SUTs when these data are confronted with other

statistics.

11.45 The survey questionnaires for business statistics are designed in such a way that data

covering different branches can be compared and added together. The needs of users like national

accounts require specific definitions of variables in the survey questionnaires, which cannot always

be derived directly from bookkeeping records. When respondents use their own definitions of

variables, this may also cause inconsistencies in the SUTs.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

329

11.46 When detailed information on such variables as output and intermediate consumption is

sought via survey questionnaires, it is possible that respondents may allocate products to the wrong

CPC product code, with the result that the contents of product codes in the SUTs are no longer

comparable.

11.47 Last but not least, a business can provide incomplete data. If, for example, data on changes

in inventories are lacking, the transformation from either sales to output or purchases to

intermediate consumption cannot be made. This will in turn also affect GVA and GDP.

(b) Inconsistencies in processing survey data

11.48 The processing of collected micro data into subject matter statistics can create

inconsistencies. Although procedures for grossing up are routine, the target population is less

straightforward. An important issue in this regard is linked to the updating of the business register

and the consequence of correctly identifying active or non-active units during the reporting period.

A further related issue is the detection and treatment of outliers.

11.49 Small enterprises are often given less detailed survey questionnaires, making it necessary

to break down the aggregated variables to the level of detail required of large enterprises. The

assumptions made for this calculation may be incorrect. The same holds for the breakdown of

variables from business statistics to the product classification used in the SUTs. For the

compilation of valuation matrices, the trade and transport margins and taxes and subsidies on

production must be allocated to the various users (industries and final consumption categories). If

scant detailed information is available, then a range of assumptions is applied, and these may also

lead to inconsistencies, in particular in the SUTs at basic prices.

11.50 Another potential cause of inconsistencies is the coverage of the hidden and informal

economy. When no estimates for the hidden and informal economy are included in the SUTs, or

where those estimates are insufficient, then inconsistencies will arise. When, for example,

consumers buy a beer in a bar, they usually do not know whether it is, economically speaking, an

“illegal” (for example, smuggled) or a “legal” beer, meaning that it is reported in household

consumption, while the “illegal” beer will not be recorded in business statistics.

11.51 Deflating SUTs data can itself generate inconsistencies in the SUTs in volume terms. As

most price index numbers based on observation are Laspeyres indices, inconsistencies result when

work is carried out at a level of aggregation above the observation of the price data. The observed

price data often do not keep account of discounts, bulk purchases and negotiated prices (in

particular in business-to-business sales or business-to-households sales), with the result that they

do not always match the actual value of the transactions. The impact is due less to the different

prices paid than to the changing weights on price changes. In addition, if the discount bulk

purchases and other such items are stable and form a constant share over time, there is little impact.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

330

In the process of balancing the full SUTs system, the implicit price indices resulting from the SUTs

system must be reconciled with the observed indices, such as the CPIs and PPIs by specific

products.

4. Reliability of data in the unbalanced SUTs

11.52 One important and very useful step to be taken before starting the balancing process is the

assessment of the reliability and quality of the data in the unbalanced SUTs. In general, less reliable

data will, and should, be adjusted to a relatively higher degree. It should, however, be borne in

mind that even weaker estimates cannot endlessly absorb inconsistencies; for example, positive

changes in inventories for a product for a large number of consecutive years are implausible or

generate implausible ratios or movements in ratios, for GVA as a proportion of output or trade

margin as a proportion of domestic output at basic prices.

11.53 The quality of the estimates will influence the role that the variable will play in the

balancing process of the SUTs. Some variables are predetermined when entered directly into the

system and kept at their original value throughout the entire balancing process; for example, data

on taxes and subsidies which are directly derived from government administrative data sources

and data derived from exhaustive sources (such as regulatory sources).

11.54 A perfect sample with a 100 per cent response rate can still generate inconsistencies.

Although such source statistics can be judged as very reliable, they may still be adjusted in the

balancing process. Estimates using models, for example fixed input structures based on the

previous period, expert guesses, use of data for the previous period, and others, are likely to be

adjusted earlier in the balancing process.

11.55 A ranking of the reliability of estimates for entries and aggregates in the SUTs should

always be borne in mind, in particular in the manual balancing phase. This ranking information is

an essential input for any automated balancing procedures and is covered later in this chapter.

11.56 Box 11.1 and Box 11.2 provide two examples illustrating the simultaneous balancing

process. With each example it is very important to have details on the reliability of the data before

starting to look for a solution or implementing any adjustments.

11.57 Box 11.1 illustrates a situation where the discrepancies are balanced in current prices and

in volume terms. The value, price and volume analysis can lead to adjustment of any of the

estimated variables. Sometimes the results can be checked using observed quantity data, which are

also available for the supply and use of energy products. Another possible check in the

simultaneous approach is the ratio of the volume of GVA to the input of labour.

11.58 The example in Box 11.2 shows that the comparison of volume indices of the main supplier

and the main user offers one solution for a balancing problem.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

331

Box 11.1 Example of discrepancies balanced in current prices and in volume terms

The price and volume changes of domestic production and exports can be compared in the simplified example

below (there is need to accept possible inconsistencies between the price indices from the supply and use sides

in the use of a simplified example). This example, for demonstration purposes only, excludes margins, taxes,

subsidies and imports. The supply minus use shows the discrepancies between supply (domestic production)

and use (exports and by other users) in current prices and in volume terms.

In this example, there is a discrepancy both in current prices and in volume terms. The first step is to get an

idea about the reliability of the data. In this case, data on both domestic production and exports in current prices

are considered to be very reliable. Thus, a sensible solution would be to adjust “other uses”. If the price index

(102.9) is considered to be correct, the adjustment should be made both in current prices and in volume terms.

This results with the following situation.

The discrepancy in current prices has been eliminated but, in volume terms, a discrepancy remains. Assuming

that the figures for the price of domestic production are reliable, and that the difference between the volume

index of domestic production and exports should not be too large, then the balancing results in an adjustment

of the price of the export and a minor adjustment of other uses.

Box 11.2 Example of simultaneous balancing comparing volume indices

Large discrepancies between volume changes of the main user of important raw materials and volume changes of

the main supplier (for instance imports) are an indication for inconsistent data.

In this example, no discrepancy between supply and use in current prices is assumed. The value indices of imports

and the main user are both plausible: 104.0 and 103.9 respectively. Analysis reveals, however, that volume indices

of imports and the main user differ: 100.0 versus 103.9, which is not plausible. Further analysis is necessary to find

the solution for this balancing problem. It is not inconceivable that the value in current prices also needs to be

adjusted.

Supply

minus use

Domestic

production

Exports Other uses

Domestic

production

Exports Other uses

Value t in current prices -10 525 420 115 Price index 102.9 100.0 103.6

Value t in prices of t-1 -21 510 420 111 Volume index 102.0 105.0 111.0

Value t-1 in current prices 0 500 400 100 Value index 105.0 105.0 115.0

Supply

minus use

Domestic

production

Exports Other uses

Domestic

production

Exports Other uses

Value t

in current prices 0 525

420 105

Price index 102.9

100.0 102.9

Value t in prices of

t-1 -12 510

420 102 Volume index 102.0

105.0 102.0

Value t-1

in current prices 0

500 400

100 Value index 105.0

105.0 105.0

Supply

minus use

Domestic

production

Exports Other uses

Domestic

production

Exports Other uses

Value t in current prices

0 525 420 105

Price index 102.9 102.7 104.0

Value t in prices of

t-1 0

510 409 101

Volume index 102.0

102.3 101.0

Value t-1 in current prices 0 500

400 100 Value index 105.0

105.0 105.0

Supply

minus use

Domestic

production

Imports Main user Other uses

Domestic

production

Imports Main user Other uses

Value t in current prices 0

468 426 92 Price index 100.0 104.0 100.0 100.0

Value t in prices of t-1 -18 50 450 426 92 Volume index 100.0 100.0 103.9 102.2

Value t-1 in current prices 0 50 450 410 90 Value index 100.0 104.0 103.9 102.2

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

332

5. Documentation

11.59 Many decisions leading to corrections, adjustments and subjective estimates are entered by

the balancers, and these may provoke a struggle when referred to other statistical and available

sources or when common sense considerations are taken into account. Thus, it is important that

the considerations and rationale behind the solutions implemented are visible to other balancers,

and the solutions are sustainable and can be reproduced if the same problems are encountered in

subsequent years. Such corrections should be recorded in a systematic way.

11.60 It is also important to record separately the steps and links between the source data through

to the balanced data so that they can be reviewed in subsequent balancing exercises to investigate

source data incoherence, bias and other such factors (Mahajan and Penneck, 1999). For example:

National accounts source data (covering business survey data, household survey data,

census data, administrative based data, extrapolations and models (for example, PIM,

FISIM), company accounts based data, etc.)

plus coverage (including exhaustiveness) adjustments

plus conceptual adjustments

plus quality (data validation) adjustments

plus balancing and coherence adjustments

equals national accounts final estimates on 2008 SNA basis

11.61 Balancing adjustments can, and should, be part of a process table describing the steps taken

from the source statistics to the final estimates in the balanced SUTs. If the balancing adjustments

are recorded in a systematic manner, they can point to flaws in source statistics or even a bias in

the balancing process itself. Again, the feedback loop can be powerful in that suppliers of source

data can improve survey questionnaires, data collection, data processing, and other processes, thus

cumulatively improving the quality of the national accounts estimates.

D. Step-by-step procedure for simultaneous balancing

11.62 This section sets out a step-by-step process for the simultaneous balancing of SUTs in

current prices and in volume terms. The process presented below relies on a sequence of tables

which starts from the SUTs at purchasers’ prices, valuation matrices, SUTs at basic prices, and the

domestic use table and imports use table at basic prices. An alternative sequence could be to split

the SUTs at purchasers’ prices into the domestic use table and imports use table at purchasers’

prices. Nevertheless the latter is not a commonly used sequencing of tables and the first is the

recommended approach to the compilation of SUTs.

11.63 Figure 11.3 provides a flowchart showing how to carry out the balancing of SUTs

indicating what types of balancing are performed at each step. Note that balancing is an iterative

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

333

process, so the figures shows a number of feedback loops that need to be carried out in order to

arrive at a final set of balanced SUTs in current prices and volume terms.

11.64 At the start of the balancing process, an estimate is available for all entries in the full SUTs

system both in current prices and in volume terms. In combination with balanced and fixed data

of the previous year, volume changes can then be compiled. The balancing effort starts by checking

all the inconsistencies and implausible estimates in the system. This is summarized in the sequence

of steps below.

Figure 11.3 Overview of the SUTs balancing framework for simultaneous balancing

(a) Differences between supply and use of products at purchasers’ prices in current

prices

11.65 These types of checks are represented in current prices in part A of Figure 11.3. Differences

between the supply and use of products at purchasers’ prices in current prices point towards

inconsistencies, possibly caused by data processing with national accounts (for example,

transformation to national accounts definitions and requirements) or by inconsistencies in observed

data (for example, as a result of the impact of globalization).

(b) Unwanted negative entries in the SUTs at basic prices: parts B and C in current prices

11.66 These types of checks are represented in current prices in parts B and C of Figure 11.3. The

unwanted negative entries in the SUTs at basic prices can be caused by mistakes in the calculation

of the valuation matrices, which would lead to a recalculation of these matrices. This is a process

that should be continued until all unwanted negatives are eliminated and the valuation matrices

look plausible. There are cases, however, where negative entries are plausible in areas such as

Current prices

DUT

at basic

prices

IUT

at basic

prices

SUTs

at purchasers’ prices

Valuation matrices

SUTs

at basic prices

T-1

DUT

at basic

prices

IUT

at basic

prices

SUTs

at purchasers’ prices

Valuation matrices

SUTs

at basic prices

Previous Years’ Prices

SUTs

at purchasers’ prices

SUTs

at basic prices

DUT

at basic

prices

IUT

at basic

prices

Valuation matrices

Deflation

Volume changes

Price changes

2,3,4

Key:

Feedback loops Iterative process as in the text Process flow

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

334

changes in inventories or exports of goods (for example, merchanting). This step forms the first

key iterative process of balancing the SUTs system (heptagon 1 in the figure).

11.67 These checks are represented in current prices in part C of Figure 11.3. These differences

will point towards inconsistencies of the type covered in subsection (a) above, due to issues such

as data processing inconsistencies in observed data.

(d) Inconsistencies between domestic supply and use of products from domestic origin

and use of imported products in current prices

11.68 These checks are represented in current prices in part D of Figure 11.3. The use table is

initially split into the domestic use table and imports use table independently of the supply table.

Thus there may be inconsistencies at the product level such as, for example, the value of exports

being larger than the value of domestic supply or the value of re-exports larger than the value of

imports. In these cases, values need to be adjusted. The breakdown of the supply of domestic and

imports in the supply table can also be used to inform the split between the domestic use table and

imports use table.

(e) Differences between the supply and use of products at basic prices in previous years’

prices

11.69 These checks are represented in previous years’ prices in parts C and D of Figure 11.3.

When starting with balanced SUTs in current prices, the differences between supply and use of

products in volume terms point towards inconsistencies in the applied price indices; for example,

failure to deflate domestic output and exports separately or failure to use an appropriate weighted

average. Furthermore, there may be weaknesses in the PPIs regarding details on discounts, bulk

purchases and negotiated prices which can also cause inconsistencies. In addition, the use of a

Laspeyres-type index can also play a role in generating inconsistencies.

11.70 These checks can also reveal errors in the SUTs in current prices. In this case, the SUTs in

current prices need to be rebalanced. This forms the second key iterative process in balancing the

full SUTs system (heptagon 2 in the figure).

(f) Plausibility of volume changes of output and intermediate consumption

11.71 These checks are represented in current prices in part C of Figure 11.3, comparing the

SUTs in previous years’ prices and the SUTs for period t-1. When combined with previous years

data, the deflated SUTs at basic prices in previous years’ prices provide a framework for judging

the volume changes of output, intermediate consumption and GVA at the industry level.

Implausible results will need adjustment of the estimates in volume terms, and if necessary, the

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

335

current price estimates in the SUTs. This forms the third iterative process in balancing the SUTs

system (heptagon 3 in the figure).

(g) Plausibility of changes in labour productivity

11.72 These checks are represented in part C comparing the SUTs in previous years’ prices and

the SUTs for period t-1 in current prices in Figure 11.3. As with the changes in volume of output

and intermediate consumption, the changes in labour productivity can be used to assess the

plausibility of the resulting GVA in volume terms at both the macro level and the industry level.

Implausible results will require adjustment of the estimates in volume terms, and if necessary, the

estimates in current prices in the SUTs. This forms the fourth iterative process in balancing the

SUTs system (heptagon 4 in the figure).

(h) Confrontation of implicit price indices of valuation matrices and observed PPIs and

changes in tariffs

11.73 These checks are represented in part (B) in Figure 11.3, confronting the valuation matrices

for period t in current and previous years’ prices. The volumes for the valuation layers are

calculated by applying the rates of the previous year to the estimates in volume terms. Therefore,

for all entries of the valuation matrices, implicit prices can be compiled. If available, observed

producer prices indices can be compared with these implicit prices indices. It is likely that there

will be possibilities for a data confrontation for specific areas, for example, transport services. For

taxes and subsidies linked to the value of the transaction concerned (for example, VAT), the

changes in tariffs can be used to assess the plausibility of the implicit prices. Implausible results

will need adjustment of the estimates in volume terms, and if necessary, the current price estimates

in the SUTs. This forms the fifth iterative process in balancing the SUTs system (heptagon 5 in

the figure).

(i) Confrontation of implicit purchasers’ price indices resulting from calculation and

observed purchasers’ price indices such as the CPIs

11.74 These checks are represented in in current prices and in previous years’ prices in part A of

Figure 11.3, confronting the SUTs for period t. The SUTs in volume terms at purchasers’ prices

are compiled as the sum of the SUTs at basic prices and the valuation matrices (all in volume

terms). When the SUTs at basic prices are balanced, the SUTs at purchasers’ prices are by

definition also balanced. At this point in the balancing process, a confrontation of observed

purchasers’ price-type indices such as the CPI and the calculated implicit purchasers’ price indices

may show that the latter may be implausible, while accepting that they should not be the same. If

there are significant differences, then the estimates in volume terms of all underlying component

tables (the valuation matrices, and the SUTs at basic prices) may need to be reconsidered. This

forms the sixth iterative process in balancing the SUTs system (heptagon 6 in the figure).

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

336

(j) Overall assessment of the second order effects of balancing steps (a)–(i)

11.75 Through the balancing procedure, the trade and transport margins are very likely to be

adjusted as a result of manual and automated corrections. As a consequence, the total use of trade

and transport margins will probably not equal the total supply, even if they were in balance in the

initial version of the system, and will need to be constrained.

11.76 Similarly, VAT should be recalculated based on the adjusted results in the use table. The

total of non-deductible VAT which is the result of the balancing procedure cannot be expected to

be an exact match for the VAT receipts based on government accounts. If official rates and tax

legislation alone are used in the calculations, the computed VAT total will normally exceed the

target; this is closer to the concept of theoretical VAT as opposed to cash-collected VAT (on an

accrued basis). To be realistic, however, the model used to estimate VAT should take into account

the expected patterns of tax evasion by keeping account of various issues like the hidden economy.

Nevertheless, the total estimated VAT will not necessarily equal the government data, so final

corrections will be needed. It may be preferable to adjust VAT proportionally in specific columns,

where the exact share of VAT liable is uncertain. A final proportional adjustment of VAT on many

products, most likely to be household final consumption expenditure, can be used to eliminate the

remaining difference.

11.77 One important final check is to ensure that the resulting effective (and implied) tax rates

do not exceed the legal rates, for example, the standard rate of VAT.

E. Alternative balancing methods

11.78 The ideal balancing scenario covered in the previous section based on the H-Approach

consists in simultaneously balancing SUTs at basic prices and at purchasers’ prices both in current

prices and in volume terms. This balancing is data demanding and the choice between

simultaneous balancing and any other variation of balancing methods depends heavily upon the

availability of data, human resources and information technology systems. If the ideal scenario is

not possible, then alternatives can be considered such as, for example, balancing the SUTs at

purchasers’ prices and balancing at basic prices or prioritizing them or launching an iterative

process with feedback loops.

11.79 The choice made among the alternative scenarios will have various consequences, in

particular for the use of price indicators. In the ideal scenario, the price indicators optimally match

the SUTs being deflated in terms of underlying flows and valuations. Diversion from the ideal

scenario will require additional compilation, assumptions and approximations in the use of price

indicators.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

337

1. Balancing SUTs at basic prices

11.80 Assuming that the balancing process is not ended before all components of the full SUTs

are checked for plausibility, the balancing at basic prices only provides a close approximation to

the ideal scenario.

11.81 Balancing at basic prices requires stripping out the trade margins, transport margins, taxes

on products and subsidies on products from the initial current price use table at purchasers’ prices.

The deflation of the SUTs then takes place at basic prices, with the application of PPIs and import

prices for the supply table and a weighted average of those indicators for the use table. Weights

could be derived from the domestic use table and imports use table of the previous year. When the

SUTs are balanced both in current prices and in volume terms, the volume changes of the valuation

matrices can be compiled, applying the volume changes of the corresponding entries of the use

table.

11.82 Subsequently, the SUTs at purchasers’ prices including non-deductible VAT, can be

derived both in current prices and in volume terms. The resulting price indices can be checked for

plausibility with observed indices on household consumption (for example, the CPIs) and export

price indices as in the ideal scenario. The price indices resulting from the sequentially compiled

domestic use table and imports use table can also be checked for plausibility with the observed

PPIs and import price indices.

2. Balancing SUTs at purchasers’ prices

11.83 Balancing at purchasers’ prices requires a very different approach for a number of entries

in the SUTs, and in general, more approximations and assumptions because of the lack of

appropriate price indices, in particular for those cases where trade and transport margins play a

substantial role. As a first step, non-deductible VAT may have to be stripped out from the initial

use table, which includes VAT (this step may often be carried out in the preprocessing of source

data for the SUTs).

11.84 As the supply table is valued at basic prices, the deflation for this part will be similar to the

ideal scenario applying PPIs, import price indices or other appropriate indicators.

11.85 The compilation of the volumes of trade margins, transport margins, taxes on products and

subsidies on products using this balancing approach will be performed at an aggregate level. If

applicable, at this stage, only the total trade margins, transport margins, taxes on products and

subsidies on products for each product are included in the system as part of the bridge columns

between the supply table and the use table. In order to compile the volume changes, the volume

change of the underlying aggregated flows must be determined. For an accurate estimate of the

volume index of the valuation layers, it is therefore very important to determine which part of the

supply or use of a product is liable to this valuation layer. For example, the retail trade margins are

mainly linked to household final consumption expenditure.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

338

11.86 For deflation of the use table at purchasers’ prices (excluding the deflation of VAT), price

indicators are required other than those applied in the ideal scenario in particular for intermediate

consumption of goods and services and gross fixed capital formation, where the ideal price indices

are often not available and must be replaced by proxies. More details may be found in chapter 9.

F. Extending the balancing of SUTs to include institutional sector accounts, IOTs,

PSUTs and EE-IOTs

11.87 The previous section describes a process for a simultaneous balancing of SUTs at basic

prices and purchasers’ prices both in current prices and in volume terms. It therefore focuses on

balancing within a SUTs system, even though the compilation of SUTs is not seen here as a

separate and isolated exercice from the compilation of the national accounts and from the

compilation of IOTs, PSUTs, EE-IOTs or other satellite accounts that may also be compiled. . This

implies that the balancing process has to be extended to ensure a coherent and consistent

integration of SUTs with the national accounts (namely, intitutional sector accounts) and related

products (for example, IOTs, PSUTs and EE-IOTs).

11.88 The balancing of SUTs can be extended to include additional accounts either in a

simultaneous or sequential manner. There are clear benefits in this extended balancing due to the

additional feedback loops which would eventually lead to further improvements in the quality of

the SUTs, and also the other products in terms of consistency, coherence and integration. Thus, in

general, it is recommended that the balancing be extended to include the features outlined in the

following subsections.

1. Institutional sector accounts

11.89 Together with the SUTs, the institutional sector accounts lie at the core of the national

accounts. The sector accounts provide an overview of the various economic activities covering

production, consumption, generation of income and distribution of income, accumulation of

wealth and relations with the rest of the world. The SUTs and institutional sector accounts thus

have several variables in common such as output, intermediate consumption, GVA and its

components linked by industry and by institutional sector. Analysing and balancing SUTs and the

institutional sector accounts can point to implausible data in SUTs, suggesting the need for a

rebalancing of the SUTs, for example highlighting classification issues or showing where cells

should have zero or non-zero entries.

11.90 Like the SUTs system, the institutional sector accounts constitute a balancing framework

consisting of a set of well-defined variables and a number of basic identities. For the total economy,

the production account and the generation of income account are in fact an aggregate of the

domestic production part of the SUTs without the dimensions products and industries. One-to-one

links exist for production, intermediate consumption and GVA. In addition, compensation of

employees is directly linked to the SUTs system. Other macroeconomic variables with a strong

link to SUTs and institutional sector accounts include household and government consumption

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

339

(use of disposable income account), and fixed capital formation (capital account). Lastly, taxes

and subsidies on products and other taxes on production appear in both systems.

11.91 From a conceptual point of view, the links between SUTs and institutional sector accounts

are strong. In statistical practice, however, it is not always easy to transform industry data on

production into institutional sector data and vice versa. For that purpose, a set of tables is

constructed with a dual classification. In this table the transactions are classified by industry

(SUTs) and by institutional sector (sector accounts) (see the linking table in chapter 10).

11.92 The SUTs are the most developed and detailed framework use for the estimation of GDP

and other macroeconomic variables in the scope of production, consumption, gross capital

formation, exports, import, and income. The three approaches to measuring GDP are combined in

one system based on a great variety of source data which are confronted and compared with one

another in order to find possible causes of inconsistencies. The high reliability, strengths, and

quality of SUTs estimates ensure that they have a strong influence on the sector accounts. It could

be said that, in the main, there is a one-way traffic between SUTs and institutional sector accounts,

but the use of dual classification provides possibilities for feedback in both directions. For the time

being, feedback is limited because the level of aggregation in the institutional sector accounts is

very high, making it difficult to trace back inconsistencies and implausible results on a sector level

to specific industries in the SUTs system.

11.93 When GVA by industry from the production approach is available, it should be balanced

against the GVA from the income approach for the corresponding industry, linking the factor

incomes and the institutional sectors. This link is extremely important between the industry

accounts and the institutional sector accounts, as illustrated below:

(a) For each industry, using the production approach:

Total output at basic prices

less total intermediate consumption at purchasers’ prices

equals GVA at basic prices (production approach)

(b) For the corresponding industry, using the income approach reflecting the different factor

incomes:

Self-employment income (mixed income and quasi-corporations)

plus gross trading profits of private financial corporations

plus gross trading profits of private non-financial corporations

plus gross trading surplus of public corporations (financial and non-financial)

plus rental income

plus non-market consumption of fixed capital

less holding gains and losses

less intermediate consumption of FISIM

plus other taxes on production and imports

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

340

less other subsidies on production

equals GVA at basic prices (income approach)

11.94 Note that, for each of the factor income components shown in (b), there is an institutional

sector breakdown.

11.95 The approaches shown in (a) and (b) bring together within the SUTs framework a range of

data from sources such as administrative data and business surveys, and this, when balanced,

ensures a better quality estimate first of GVA by industry, then for the whole economy GVA, and

hence GDP and GNI.

11.96 The full breakdown in (b) in terms of data may not be available by detailed industry (and

by institutional sector) for a range of reasons. The minimum that should be incorporated in the

SUTs compilation and balancing process is covered in (c) below:

plus compensation of employees

plus other taxes on production and imports

less other subsidies on production

equals GVA at basic prices (income approach)

11.97 It should be noted again that, for each component shown in (c), there is an institutional

sector breakdown.

2. IOTs

11.98 The links between the SUTs and IOTs in the bottom left-hand side and right-hand side of

the H-Approach have been covered through the links to the separation of imports of goods and

services and valuation matrices needed to produce the SUTs at basic prices and the SUTs in volume

terms, and then in turn, the IOTs.

11.99 Integrating the production of IOTs with the production and balancing of SUTs makes

possible the effective and timely use of powerful feedback loops in indicating data problems within

the SUTs or with the steps transforming SUTs to IOTs. For example, addressing negative cell

entries in the IOTs can improve the quality of the SUTs. It is recommended that IOTs should be

produced (simultaneously or sequentially) alongside SUTs rather than at a much later stage or even

less frequently than the SUTs.

3. PSUTs and EE-IOTs

11.100 The links with the PSUTs and EE-IOTs are of a different nature because the transactions

in the physical SUTs are expressed in other units (for example, kilograms, terajoules, and others)

or are more detailed in terms of industries and products. The balancing of PSUTs and EE-IOTs in

combination with SUTs and IOTs is described in chapter 13.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

341

4. Key feedback loops generated by the balancing process

11.101 In the ideal scenario, all balancing is performed simultaneously, implying that all feedback

loops are part of an integrated process. This includes the feedback loops within the SUTs process

together with the loops going back to earlier steps in the full statistical process chain.

11.102 Figure 11.4 illustrates key feedback loops that can be generated from within the balancing

process as well as examples of the sources of the feedback loops.

11.103 When the balancing is extended to cover IOTs, the links to the institutional sector accounts,

PSUTs, EE-IOTs, and other satellite accounts, there will be more feedback loops that become

available than are shown in Figure 11.4, and, in turn, each adds a further data quality improvement

dimension.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

342

Figure 11.4 Sources of feedback loops emanating from the balancing process

G. Practical aspects of balancing

11.104 The balancing of SUTs is not a simple task, as it requires priorities to be set because of

time and resources constraints. Below are some practical considerations to be considered in the

balancing process.

1. Automated and manual balancing

11.105 The balancing process covering the full SUTs system incorporates automated and manual

balancing. The first step is to separate the inconsistencies between those needing further research

and those which can be resolved using automated procedures. In general, large inconsistencies

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

343

require more attention than smaller ones but such indicators as time series, revision analyses, input-

output ratios and labour productivity can also point to serious problems in the data. For the smaller

inconsistencies, automated procedures may be used, although they will still require assessment in

terms of the quality and plausibility of the results.

11.106 For the inconsistencies selected for manual balancing, a more or less reversed version of

the initial full statistical process from surveys to national accounts can be applied. It would start

with a critical investigation of the national accounts process, transforming source data into data

for use in SUTs. In the use table in particular there are many entries which are the result of splitting

aggregated source data to the product level as required in the SUTs. The allocation may be changed

without altering the initial aggregate total, and thus GVA. In practice, many of the problems will

not be resolved through this approach. The next step is to have recourse to expert knowledge on

the specific subject matter and, if major inconsistencies persist, it may prove necessary, as a final

step, to contact the survey respondent or source supplier to instigate a critical discussion of the

data that they have provided.

11.107 The adjustments to entries in the SUTs will obviously affect other flows and ratios. To

systematize the balancing process in order, for example, to take into account inter-industrial

relationships, it is helpful to distinguish separable blocks of industries in which the main producers

and users of products are represented and to assign to these blocks separate groups of input-output

specialist statisticians responsible for balancing. For example, cement is definitely required by the

construction industry and cement supplies can therefore be used to cross-check the plausibility of

construction estimates. The SUTs can be divided into various groupings of related industries; one

section of groupings, for example, could include the following, although this is not an exhaustive

list:

• Agriculture, and fishing (ISIC Rev. 4, divisions 01 and 03), manufacture of food products

and beverages (ISIC Rev. 4, divisions 10–12), hotels and restaurants (ISIC Rev. 4, divisions

55 and 56)

• Manufacture of metals and metal products (ISIC Rev. 4, divisions 24 and 25) and

manufacture of machinery and means of transport industries (ISIC Rev. 4, divisions 26, 29

and 30)

• Forestry (ISIC Rev. 4, division 02) and industries producing wood and wood products (ISIC

Rev. 4, divisions 16 and 31)

• Quarrying and non-metallic mineral products (ISIC Rev. 4, divisions 08 and 23) and

construction (ISIC Rev. 4, divisions 41 and 43)

• Manufacture of textile (ISIC Rev. 4, division 13) and textile products, footwear (ISIC Rev.

4, divisions 14 and 15)

• Chemical industries, including plastic products (ISIC Rev. 4, divisions 19–22)

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

344

• Energy sectors (ISIC Rev. 4, divisions 05, 06 and 35)

11.108 Automation is essential for the preparation and management of the SUTs system. The

SUTs information technology system (including the supporting modules and analytical tools and

function) will play various roles in the compilation process so it will need to be designed to a high

standard and with an eye to the whole range of functions that it has to carry out.

11.109 Many of the calculations in the preparation stage are carried out by automated procedures.

At every stage of the process, the information technology system can provide quick and clear

overviews of the data in every chosen configuration. The information technology system produces

the first parts of the SUTs, which are essential for the detection of major integration problems and,

down the line, it enables investigation of the finest details, so that the causes of balancing problems

can be detected in the most efficient manner possible. Lastly, the information technology system

can help to develop appropriate solutions along with professional and orderly documentation of

all the adjustments made during the compilation and balancing processes. The information

technology system is a powerful instrument which is indispensable for all operations from the

source data through to the final set of balanced SUTs.

11.110 Modules can be designed to eliminate small discrepancies between supply and use at the

product level in an adequate and efficient way. This may imply one-dimensional proportional

distribution of the discrepancies over a selected set of users.

11.111 Balancing the whole SUTs system requires a multidimensional approach to the

reconciliation of inconsistencies and a range of human inputs which cannot be automated. The

experiences to date of balancing SUTs in an automated manner have shown that full automated

balancing does not yet work; ,the quality generated is poor, many issues require further

investigation and, overall, the savings gained are tenuous at best. .Most experiences have shown

that a combination of automated and manual statistical techniques and procedures is the best

workable solution to the underpinnng of a SUTs system.

11.112 Working with statistical data based on sample surveys, survey questionnaires and

influenced by such issues as non-response, means working with sample margins of error. Even

when samples are perfect and there is a100 per cent response rate there will be inconsistencies.

Statistics therefore constitute a major cause of differences and it could be argued that balancing

could be carried out in an automated manner using the sample margins of the statistics concerned

as weights. Statistics, however, are never ideal and inconsistencies are not only caused by sampling

but also have various causes of a non-statistical nature. It is these causes of inconsistencies that

make manual balancing necessary as a preliminary step to be taken prior to using automated

balancing.

11.113 The decision as to what can be balanced using automated procedures and what should be

done manually depends not only on the nature of the inconsistencies but also on the context of

where the balancing takes place.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

345

11.114 If the automated balancing of SUTs is carried out independently for consecutive years, and

is applicable to current prices and volume estimates, then the initial discrepancies must be small.

Experience has shown that even small differences in initial data can lead to totally different results

in RAS (covered in box 11.3) and optimization procedures, which indicates that the balanced SUTs

are not comparable over time. As long as the discrepancies are small, the chances that the results

will no longer be comparable should also be small.

11.115 One major step forward is to include linkages (for example, growth rate expectations)

between consecutive years, preferably including price and volume indicators. In setting such a

scenario, the basic identities of the SUTs can be used as a restrictive measure in the automated

balancing process and also in the criteria relating to plausibility referred to above. Including price

and volume ratios and defined restrictions attributable to them would help to assure the plausibility

of growth rates and comparability over time. This also allows for the option of leaving major

discrepancies to the automated system.

11.116 Box 11.3 provides a short description of automated methods that are often used to remove

some of the inconsistencies in the SUTs.

Box 11.3 Methods used for automated balancing SUTs

RAS method

The RAS method (the abbreviation is derived from the phrase “raking and scaling”) is a well-known and widely

used method for data reconciliation. Its aim is to achieve consistency between the entries of some non-negative

matrix and pre-specified column totals and row totals. It is very easy to apply and to understand. RAS has a

narrow range of applicability, however; for example, it can only be applied to non-negative matrices.

It is used to revise the internal entries in a matrix so that they agree with the margin totals. RAS is used when

the margin totals – total supply and use of products, or total output by kind of activity, for example – are believed

to be correct but the breakdown inside the matrix is not consistent with the margin totals. Over the years, many

extensions, variations and improvements of the RAS method have been developed. Some examples include:

• GRAS (generalized RAS) allows for matrices in which some of the elements are predefined, in addition

to the row totals and the column totals. The GRAS method allows for matrices with negative entries

(see Lenzen and others, 2007).

• TRAS (three-stage RAS) extends RAS by including constraints on arbitrary subsets of the matrix

elements, instead of only fixing row totals and column totals (see Cole, 1992, and Gilchrist and St.

Louis, 1999).

• KRAS, as developed by Lenzen and others (2009), includes the aforementioned features of GRAS and

TRAS and further generalises RAS for the case of conflicting source data. The simplest case is when

two data sources prescribe two different values for the same matrix entry. In order to converge, the

original RAS method can use only conflicting values, whereas the KRAS method will use both values

and allow for different reliabilities of the data sources.

More details on different updating methods may be found in chapter 18 of this Handbook and in United Nations

(1999).

Stone method

The Stone method is another method of data reconciliation whereby data are adjusted in order to satisfy a set of

linear constraints. In adjusting the initial data, the Stone method uses information on the relative reliabilities of

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

346

the initial data given in a covariance matrix. Data that are considered to be the most reliable are modified least,

and vice versa. The Stone method yields a set of fully reconciled data, with minimum variance. It translates the

reconciliation problem into a mathematical (weighted quadratic) optimization problem under linear constraints.

In practical applications of the method, a covariance matrix of the initial data is often unavailable. Accordingly,

applications generally use estimates of relative variances. There are several ad hoc methods for estimating

relative variances. One method is to have a specialist estimate of 95 per cent confidence intervals and to use the

interval sizes as an approximation for variances. Another method may be to distinguish several categories such

as “relatively unreliable”, “normally reliable” and “relatively reliable”, and all variables within the same group

are assigned the same variance.

It is often desirable in practice for reconciliation to affect large values more than small values in an absolute

sense. If this is the intention, then the following variances may be chosen:

Var

(



)

= 









where, 



is a parameter that depends on the reliability, or reliability category, of x



In practice, determining the correct ratios between the various variances is a process of trial and error, which

means that one particular ratio is chosen based on a degree of prior knowledge and simple assumptions (for

example, that variances are equal in the absence of prior knowledge), and then judging whether the results are

acceptable. If not, the variances are then modified.

Convex quadratic constrained optimization

An option for balancing SUTs both in current prices and in volume terms simultaneously is the application of

loss function, which includes current price and volume estimates and also linear price and volume ratios. The

loss function must be minimized under a set of linear constraints, and the loss is defined as the difference

between the initial data and the balanced data. For the price and volume ratios, linear constraints are applied.

The constraints are either strong or weak.

Strong constraints include, for example, identities of the SUTs and upper and lower boundaries (subsidies should

be less than or equal to zero). Weaker constraints include, for example: volume change of output related to

volume change of input; ratio of taxes and subsidies to the basic price estimate (in the form of a percentage of

the variable-like tax); ratio of margins to the basic price estimate (in the form of a percentage of the margins);

and re-exports are smaller than the corresponding imports (weak because of differences in valuation).

More specific constraints at a product level can be specified in this optimization problem. It is also possible to

extend the optimization problem and to include, for example, the transformation to industry-by-industry IOTs,

including the estimation of valuation matrices.

2. Balancing benchmark and consecutive years

11.117 In the ideal situation, the annual SUTs and IOTs present the current state of the art when it

comes to balancing of the basic statistics. The data from the annual SUTs and IOTs are further

improved, however, with every new benchmarking exercise, leading to an entire new time series.

Benchmarking is a regular process in economic statistics, whereby data sources for the same target

variable with different frequencies are reconciled and the inconsistencies between the different

estimates are corrected. Benchmarking leads to revisions of earlier estimates of the target variables.

This section deals with two types of benchmark revisions for SUTs and IOTs – periodic

benchmarking and annual benchmarking – and considerations about balancing of SUTs and IOTs

in these two cases.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

347

11.118 Periodic benchmarking refers to significant revisions, for example, conceptual changes,

new or changed basic data sources that originate from incorporating data from periodic benchmark

censuses (which are carried out every five to ten years), revised international guidelines like the

2008 SNA and BPM 6 and other changes that cannot be incorporated on a continuous basis because

of resource contraints. Backward revisions to the time series affecting several previous years are

also carried out based on the benchmark excersise.

11.119 In general, planning for period benchmark revisions should seek to coordinate all major

changes to be synchronized for a common year for implementation, thereby achieving, at a

minimum of about once every five years, a maximum degree of consistency within national

accounts, balance of payments and other related domains and the statistics concerned are based on

the best possible data.

11.120 In this context annual benchmarking mainly refers to the regular revisions to annual

accounts made possible by the availability of new and more complete data sources. Annual

benchmarking, however, also includes revisions due to the alignment of short-term survey results

(say, turnover variable) based on small survey samples with much larger sample-based annual

structural surveys. Following the reconciliation of short-term survey sources before the

benchmarking process with more complete and detailed annual data sources such as structural

statistics, these sources are then fed through the SUTs framework. Combining annual

benchmarking and annual chain-linking also ensures that improved accuracy of the levels and

growth rates of the economy is reached more quickly, and again, achieved through the SUTs

framework.

11.121 Consistency is one of the key elements of mational accounts. Theoretically, the whole time

series and every level of detail should be consistent. In practice, it may not be possible to publish

all results at the same time, although the SUTs and the main aggregates such as GDP should be

fully consistent. More and more years of SUTs are being produced and this poses a growing

challenge as to how to maintain their consistency as a long-run dataset. Statistics Denmark, for

example, as part of its implementation of the 2008 SNA and BPM 6, ensured that the Danish IOTs

were retained on a consistent basis going back to 1966 – a significant exercise in its own right.

11.122 Annual benchmark revisions, as carried out in Ireland and the United Kingdom, entail less

of a planning burden in synchronizing a common year, as the changes form an integral process to

compiling annual SUTs.

11.123 In theory, the balancing procedure for period benchmark SUTs is the same as for SUTs

compiled annually and also for investigations into the causes of inconsistencies and the search for

solutions. For non-benchmark years, however, much information can be derived from looking at

the previous years. When balanced product-flow systems already exist for the previous year, it can

safely be assumed that the general structure of the system will be more or less similar to the

preceding year, unless of course specific information is available showing that there are major

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

348

structural changes in some industries. Together with price and volume estimates, the T-1 data

provide a valuable source in detecting inconsistencies in the SUTs.

11.124 For periodic benchmark SUTs, all inconsistencies must be investigated thoroughly and

exhaustively. This is even more important as the level estimates of the subsequent years are based

on this benchmark. Consequently, the balancing of benchmark SUTs should be carried out

manually to a great degree, so that a good quality base can be achieved and the use of automated

procedures should be limited to small discrepancies.

3. Organization of the balancing function

11.125 The organization of the balancing function can be set up in different ways across teams.

The following are possible examples:

• Centralized balancing team – a single person or a very small team of people form the central

balancing team whose designated role is to take all the validated and investigated data from

the compilation teams and balance the SUTs using the tools (manual and automated) at their

disposal. They will lead and coordinate the implementation of balancing adjustments across

the components and record the adjustments as “pure” balancing adjustments. They may

generate the areas for investigation in liaison with the compilers. The team may also be

organized in such a way that it does not simply produce a balanced SUTs dataset but also

provides feedback to the compilers on the balancing adjustments as appropriate, for example,

household final consumption expenditure or gross fixed capital formation, to enble the

compilers to generate publications consistent with the final dataset. Alternatively, the

compilers could use the final dataset to generate the publications with no additional data

flows in the system.

• Decentralized balancing – here the balancing is devolved to industry, product or topic

specialists (for example, energy, household final consumption expenditure or the compilers).

They will undertake a row and column balancing process related to their allocated row,

column and topic. After so many iterations, the central team may simply use automated tools

to achieve a final balance. The role of the central team is different in that it focuses solely

on the automated part, whereas the manual balancing is left to the specialists and is also

coordinated across the balancers.

11.126 There are pros and cons to either approach:

• In centralized balancing, the control and order assessment and also the overall quality

control sit with the central team and form an efficient way of achieving a coherent and

high-quality balance. The specialist experts’ knowledge can be used for quality

assurance of the inputs and the balanced picture.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

349

• In decentralized balancing, the specialist experts’ knowledge is more extensively used

but requires much more communication and coordination to be effective, and the

balancing adjustments and quality adjustments may be less clear.

11.127 In both cases, such issues as documentation, communication and coordination are crucial,

and it is also essential to ensure that staff have appropriate skills and knowledge. It is preferable,

however, to have a centralized rather than a decentralized balancing arrangement.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

350

Annex A to chapter 11. Balancing supply and use tables

A11.1 This annex presents a numerical example of the balancing of SUTs in current prices and in

previous years’ prices in line with the SUTs balancing framework shown in figure 11.3 (fully

consistenct with the H-Approach). This provide an example of the type of thinking and issue

resolution that goes to achieve a balanced SUTs system. The numbers in the tables in this annex

have been divided by 1,000 for readability and presentation purposes; in reality the differences are

larger than the small numbers shown. This annex consists of three sections.

A11.2 Section 1 of the annex shows the following (unbalanced) tables:

• Table A11.1: Supply and use tables 2011 in current price (SUTs at purchasers’ prices, Taxes

less subsidies on products, trade and transport margins, SUTs at basic prices, imports use

table and domestic use table).

• Table A11.2: Price indices for supply and use tables 2011 (SUTs at purchasers’ prices, taxes

less subsidies on products, trade and transport margins, SUTs at basic prices, imports use

table and domestic use table). It should be noted that some prices may be based on actual

source data or derived from independent volume estimates (using the valuation matrices) or

implicit prices, but a complete set of prices is shown for the purposes of this balancing

example.

• Table A11.3: Supply and use tables 2011 in previous years’ prices (SUTs at purchasers’

prices, taxes less subsidies on products, trade and transport margins, SUTs at basic prices,

imports use table and domestic use table).

• Table A11.4: Volume indices tables for supply and use tables 2011 (SUTs at purchasers’

prices, taxes less subsidies on products, trade and transport margins, SUTs at basic prices,

imports use table and domestic use table).

• Table A11.5: Supply and use tables 2010 in current price (SUTs at purchasers’ prices, taxes

less subsidies on products, trade and transport margins, SUTs at basic prices, imports use

table and domestic use table).

A11.3 Section 2 of the annex follows the same sequence of tables but shows the corresponding

balanced SUTs system for both price bases. The cells highlighted in yellow are those changed in

order to achieve balanced SUTs.

A11.4 Section 3 of the annex provides an overview of the numerical adjustments required to

achieve the balanced system.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

351

A11.5 The inconsistencies in the system are revealed by differences between the supply and use

at the product level and discrepancies between trade and transport margins in the SUTs at

purchasers’ prices and the row totals of the respective valuation matrices. The same holds for taxes

less subsidies on products. In general, the net operating surplus is used as a balancing item in

achieving the identity that the total output equals the total input.

A11.6 If the inconsistencies are relatively small, then automated procedures can be applied to

balance the SUTs system simultaneously in the different valuations. In the case of large

discrepancies or implausible input-output ratios in volume terms or implausible movements in the

price changes on a row, then further research would have to be undertaken before any automated

balancing could be applied.

A11.7 In the numerical example, the estimates for agriculture products in current prices show a

discrepancy both at purchasers’ prices and at basic prices. Conversely, the estimates in previous

years’ prices are balanced. The latter does not mean that the previous years’ price-based data are

plausible although, having examined the data on volumes and prices at a more detailed level than

presented in the example, it was concluded that the prices were too high and the volumes looked

plausible. As a consequence, the estimates in current prices for the supply of agriculture products

by the agriculture industry needed to be adjusted.

A11.8 The estimates for manufacturing products show a discrepancy in current and previous

years’ prices both at purchasers’ prices and at basic prices. The difference in the previous years’

prices based data at purchasers’ prices is somewhat larger than in current prices because the initial

estimate of taxes less subsidies on products and trade and transport margins in previous years’

prices seems to be very low. The separate use table based on domestic output (namely, domestic

use table) and imports (namely, imports use table) shows a big discrepancy for this product group

in the domestic use table, which is counterbalanced in the imports use table. In addition to the

difference between supply and use of products in current prices and in previous years’ prices, there

is an implausible price index for exports. In this case, the estimates in previous years’ prices can

be adjusted in order to get plausible price indices. A second reason for the inconsistency is the

delineation of exports and re-exports. At the point when products are imported, it is not always

known whether or not the products will be re-exported in the same form. As a consequence, the

data available for re-exports from source statistics should be a minimum and the actual estimate in

the SUTs will be much higher, as shown in this example. The delineation problem caused a small

inconsistency which can be solved by adjusting total exports.

A11.9 The balancing adjustments and approaches applied to the valuation matrices differ between

estimates in current prices and estimates in previous years’ prices.

A11.10 The current price estimates of trade and transport margins (TTM) do not show an

inconsistency in the purchasers’ prices table; the TTM column is consistent with the output of

trade. In the basic price table, however, there is an inconsistency caused by the difference between

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

352

the TTM matrix and the TTM column in the supply table at purchasers’ prices. The gap is 680,

which is found on the manufacturing product row.

A11.11 Similarly, for taxes less subsidies on products (TLS), there were some differences between

the TLS column of the supply table and the total column of the TLS matrix. These inconsistencies

appear on the product rows for maufacturing (MAN) and fnance and business services (FBS).

A11.12 In both cases, the data in the current price valuation matrices need to be adjusted, as in

most cases the current price TLS on products is derived from government data and therefore fixed.

The same approach is less strong for TTM, in which the output control totals generally prevail

over the estimates of the valuation matrices.

A11.13 For the previous years’ prices-based estimates of TLS and TTM, the opposite holds

because, while the estimates for TTM and TLS on products are compiled using the volume change

of the relevant transactions, the estimates in the valuation matrices determine TTM and TLS in

previous years’ prices, so the TLS and TTM columns of the SUTs at purchasers’ prices have to be

adjusted.

A11.14 Both in current prices and in previous years’ prices, the transport products show a

discrepancy. In this case the inconsistency was not caused by the valuation matrices but, on looking

at the imports use table and through additional research, it was identified that the estimate of import

of transport services was too high.

A11.15 The discrepancy in communication is shown in both valuation matrices in current prices

and in previous years’ prices and can also be seen in the domestic use table. On examining the

data, it was decided that consumption of households had to be adjusted.

A11.16 A similar procedure was followed for financial and business services and other services, in

which cases it was decided that the intermediate consumption of TIC and consumption of

households should be adjusted respectively. It should be noted that the discrepancy for the FBS

product in the basic price table is somewhat larger than in the purchasers’ price table because of

the wrong estimate in the TLS matrix.

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

353

1. Unbalanced SUTs system (table A11.1–table A11.5)

Table A11.1: Supply and use tables 2011 in current prices

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 26.0 0.2 0.0 26.2 15.4 41.6 10.9 0.5 53.0 AGR 5.5 17.4 1.5 24.4 5.9 0.1 -0.1 22.4 28.4 52.8 0.2

MAN 1.3 318.5 35.7 355.5 336.8 692.3 113.0 39.4 844.7 MAN 10.1 202.1 65.4 277.7 122.8 9.1 51.5 1.4 383.1 568.0 845.7 -1.0

CON 0.1 88.0 4.5 92.6 1.6 94.1 8.3 102.4 CON 0.3 26.2 19.8 46.3 0.5 0.6 53.7 2.1 56.8 103.1 -0.7

TTC 0.5 15.1 231.6 247.2 73.4 320.6 -124.0 3.0 199.6 TTC 0.6 10.8 66.2 77.5 21.4 0.9 0.5 6.7 0.0 92.3 121.8 199.3 0.3

FBS 0.5 8.0 282.0 290.5 55.2 345.7 9.1 354.9 FBS 2.1 43.5 161.8 207.3 78.0 0.0 3.6 16.8 49.5 147.8 355.2 -0.3

OSE 0.3 2.5 222.9 225.7 16.6 242.3 0.1 3.0 245.4 OSE 0.2 3.5 22.2 25.9 54.7 4.6 153.4 1.2 0.2 5.1 219.1 245.0 0.4

TOT 28.7 432.3 776.7 1237.7 499.1 1736.7 0.0 63.3 1800.1 TIC 18.8 303.5 336.8 659.1 283.3 5.5 167.2 130.1 1.5 554.5 1142.0 1801.1 -1.0

OTLS -0.7 0.2 0.3 -0.3 -0.3

COE 2.7 60.9 254.4 318.0 318.0

CFC 3.7 20.5 82.9 107.1 107.1

NOS 4.3 47.2 102.2 153.7 153.7

GVA 9.9 128.8 439.8 578.5 578.5

TOT 28.7 432.3 776.7 1237.7 283.3 5.5 167.2 130.1 1.5 554.5 1142.0

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 0.5 0.5 AGR 0.0 0.0 0.0 0.1 0.4 0.1 0.5 0.5

MAN 39.4 39.4 MAN 0.2 1.7 6.4 8.3 26.8 0.4 2.2 0.0 1.7 31.1 39.4 0.0

CON 8.3 8.3 CON 0.0 2.0 2.0 0.0 0.1 6.1 0.0 6.2 8.3

TTC 3.0 3.0 TTC 0.0 -0.1 1.1 1.0 1.7 -0.1 0.4 -0.1 2.0 3.0

FBS 9.1 9.1 FBS 0.0 0.0 4.4 4.4 1.6 0.1 3.0 0.0 4.7 9.1 0.0

OSE 3.0 3.0 OSE 0.0 0.1 0.1 2.3 -0.4 0.7 0.3 2.9 3.0

TOT 63.3 63.3 TOT 0.2 1.6 14.0 15.8 32.9 0.1 12.5 0.0 2.0 47.5 63.3 0.0

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 10.9 10.9 AGR 0.4 1.6 0.4 2.5 3.2 0.0 0.0 5.2 8.4 10.9

MAN 113.0 113.0 MAN 0.9 20.5 9.2 30.6 39.1 3.1 5.1 -0.1 34.5 81.7 112.3 0.7

CON CON

TTC -124.0 -124.0 TTC -1.3 -22.2 -9.6 -33.1 -42.3 -3.1 -5.1 0.1 -39.7 -90.2 -123.3 -0.7

FBS FBS

OSE 0.1 0.1 OSE 0.1 0.0 0.1 0.1

TOT 0.0 0.0 TOT 0.0 0.0 0.0 0.0 0.0

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 26.0 0.2 0.0 26.2 15.4 41.6 AGR 5.1 15.7 1.1 21.9 2.4 0.1 -0.1 17.1 19.5 41.3 0.3

MAN 1.3 318.5 35.7 355.5 336.8 692.3 MAN 9.0 179.9 49.8 238.8 56.9 5.6 44.3 1.5 346.9 455.2 694.0 -1.7

CON 0.1 88.0 4.5 92.6 1.6 94.1 CON 0.3 26.2 17.7 44.3 0.4 0.5 47.6 2.1 50.6 94.8 -0.7

TTC 0.5 15.1 231.6 247.2 73.4 320.6 TTC 1.8 33.0 74.8 109.6 62.0 0.9 3.7 11.4 -0.1 132.1 210.0 319.7 1.0

FBS 0.5 8.0 282.0 290.5 55.2 345.7 FBS 2.1 43.5 157.4 202.9 76.4 0.0 3.5 13.8 49.5 143.1 346.1 -0.3

OSE 0.3 2.5 222.9 225.7 16.6 242.3 OSE 0.2 3.5 22.1 25.8 52.3 4.6 153.8 0.4 0.2 4.8 216.1 241.9 0.4

TOT 28.7 432.3 776.7 1237.7 499.1 1736.7 TLS 0.2 1.6 14.0 15.8 32.9 0.1 12.5 0.0 2.0 47.5 63.3

TIC 18.8 303.5 336.8 659.1 283.3 5.5 167.2 130.1 1.5 554.5 1142.0 1801.1 -1.1

OTLS -0.7 0.2 0.3 -0.3 -0.3

COE 2.7 60.9 254.4 318.0 318.0

CFC 3.7 20.5 82.9 107.1 107.1

NOS 4.3 47.2 102.2 153.7 153.7

GVA 9.9 128.8 439.8 578.5 578.5

TOT 28.7 432.3 776.7 1237.7 283.3 5.5 167.2 130.1 1.5 554.5 1142.0

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 15.4 AGR 0.8 6.4 0.5 7.7 1.2 0.0 -0.1 6.6 7.7 15.4

MAN 336.8 MAN 1.3 98.0 20.9 120.2 24.5 1.6 19.2 -0.2 121.3 166.5 286.6 50.2

CON 1.6 CON 0.0 0.5 0.4 0.9 0.7 0.7 1.6

TTC 73.4 TTC 0.0 2.7 14.0 16.8 0.7 2.7 0.0 52.7 56.1 72.9 0.5

FBS 55.2 FBS 0.2 14.0 24.5 38.8 0.2 2.2 14.1 16.5 55.2

OSE 16.6 OSE 0.0 0.5 3.9 4.5 12.0 0.0 0.1 0.0 12.2 16.6

TOT 499.1 TLS

TIC 2.3 122.1 64.2 188.7 38.7 1.6 24.8 -0.1 194.7 259.7 448.4 50.7

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 26.0 0.2 0.0 26.2 AGR 4.3 9.3 0.6 14.2

1.2 0.1 0.0 10.5 11.7 26.0 0.2

MAN 1.3 318.5 35.7 355.5 MAN 7.8 81.9 28.9 118.6 32.4 4.0 25.1 1.6 225.6 288.7 407.4 -51.9

CON 0.1 88.0 4.5 92.6 CON 0.3 25.7 17.4 43.4 0.4 0.5 46.9 2.1 49.9 93.3 -0.7

TTC 0.5 15.1 231.6 247.2 TTC 1.8 30.3 60.7 92.8 61.3 0.9 3.7 8.6 -0.1 79.4 153.9 246.7 0.5

FBS 0.5 8.0 282.0 290.5 FBS 1.8 29.5 132.9 164.2 76.2 0.0 3.5 11.6 35.4 126.6 290.8 -0.3

OSE 0.3 2.5 222.9 225.7 OSE 0.2 3.0 18.1 21.3 40.2 4.6 153.8 0.4 0.1 4.8 203.9 225.3 0.4

TOT 28.7 432.3 776.7 1237.7 TLS 0.2 1.6 14.0 15.8 32.9 0.1 12.5 0.0 2.0 47.5 63.3

TIC 16.4 181.4 272.6 470.5 244.6 5.5 165.5 105.3 1.6 359.7 882.3 1352.8 -51.8

IMP 2.3 122.1 64.2 188.7 38.7 1.6 24.8 -0.1 194.7 259.7 448.4

TOT 18.8 303.5 336.8 659.1 283.3 5.5 167.2 130.1 1.5 554.5 1142.0 1801.1 -51.8

OTLS -0.7 0.2 0.3 -0.3 -0.3

COE 2.7 60.9 254.4 318.0 318.0

CFC 3.7 20.5 82.9 107.1 107.1

NOS 4.3 47.2 102.2 153.7 153.7

GVA 9.9 128.8 439.8 578.5 578.5

TOT 28.7 432.3 776.7 1237.7 283.3 5.5 167.2 130.1 1.5 554.5 1142.0

Check

Use Table at basic prices

Supply Table at basic prices

Imports Use Table

Imports C IF

Domestic Use Table at basic prices

Supply Table for domestic output at basic prices

Taxes less subsidies on products

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Use Table for taxes less subsidies on products

Use Table for trade and transport margins

Trade and transport margins

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

354

Table A11.2: Price indices for supply and use tables 2011

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 102.0 104.1 102.0 106.9 103.8 97.5 107.6 102.5 AGR 105.8 109.9 98.7 108.2 100.4 94.7 568.8 97.7 98.0 102.4 0.0

MAN 107.5 108.0 100.5 107.2 108.0 107.6 101.1 103.2 106.4 MAN 114.1 110.4 105.0 109.2 102.1 99.5 99.6 85.7 106.7 104.9 106.2 0.2

CON 103.4 100.5 100.5 100.5 101.6 100.5 100.1 100.4 CON 101.4 100.5 101.8 101.0 97.4 97.1 100.0 102.0 100.0 100.5 0.0

TTC 101.1 99.2 100.3 100.2 110.7 102.4 100.8 101.4 103.4 TTC 100.9 100.1 100.8 100.7 102.2 108.0 106.2 104.0 -0.5

FBS 102.9 100.9 100.4 100.4 101.0 100.5 93.6 100.3 FBS 101.4 100.7 100.4 100.5 101.7 101.8 99.0 100.1 100.3 0.0

OSE 100.2 101.0 103.4 101.1 99.1 110.2 101.2 OSE 102.4 101.7 104.3 103.9 102.7 103.8 101.7 99.5 101.8 100.9 101.2 0.0

TOT 102.2 105.8 100.5 102.4 107.3 103.7 101.6 103.7 TIC 109.3 107.5 101.7 104.5 102.1 103.3 100.2 99.1 83.0 105.7 103.2 103.6 -0.3

OTLS 98.0 201.1 269.9 48.3 48.3

COE 102.7 101.9 101.5 101.6 101.6

CFC 99.1 100.6 98.4 98.8 98.8

NOS 80.9 103.2 96.2 97.7 97.7

GVA 91.1 102.2 99.7 100.1 100.1

TOT 102.2 105.8 100.5 102.4 102.1 103.3 100.2 99.1 83.0 105.7 103.2

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 107.6 AGR 92.3 612.5 91.4 230.0 100.8 96.2 99.8 107.6

MAN 103.2 MAN 103.7 107.3 103.9 104.6 101.7 98.7 93.8 -100.0 104.4 101.3 101.9 1.3

CON 100.1 CON 102.2 102.1 90.7 95.1 99.6 100.0 99.4 100.1

TTC 101.4 TTC 99.4 103.3 102.4 101.4

FBS 93.6 FBS 123.1 100.0 102.0 102.1 105.7 103.8 44.4 86.4 93.3 0.3

OSE 110.2 OSE 241.7 109.2 101.1 108.5 110.2

TOT 101.6 TOT 105.1 110.1 103.1 103.8 102.5 102.3 92.5 -100.0 105.0 99.8 100.7 0.8

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 97.5 AGR 90.4 96.7 100.7 96.3 104.9 85.7 100.0 94.0 97.9 97.5

MAN 101.1 MAN 113.7 102.6 102.3 102.8 97.8 97.9 97.5 93.5 99.5 98.5 99.6 1.5

CON CON

TTC 100.8 TTC 105.0 102.1 102.2 102.3 98.3 97.9 97.5 93.8 98.8 98.4 99.4 1.4

FBS FBS

OSE 99.1 OSE 97.6 104.2 99.1 99.1

TOT TOT

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 102.0 104.1 102.0 106.9 103.8 AGR 107.2 111.2 98.1 109.5 94.9 95.6 725.0 98.9 98.0 103.8 0.0

MAN 107.5 108.0 100.5 107.2 108.0 107.6 MAN 114.4 111.4 105.6 110.2 105.5 100.5 100.2 85.7 107.5 106.3 107.6 -0.1

CON 103.4 100.5 100.5 100.5 101.6 100.5 CON 101.4 100.5 101.7 101.0 98.3 97.4 100.0 102.0 100.1 100.5 0.0

TTC 101.1 99.2 100.3 100.2 110.7 102.4 TTC 103.8 101.5 101.0 101.2 99.5 98.5 97.0 93.7 105.0 102.8 102.2 0.2

FBS 102.9 100.9 100.4 100.4 101.0 100.5 FBS 101.2 100.7 100.4 100.5 101.7 101.7 100.6 99.0 100.6 100.5 0.0

OSE 100.2 101.0 103.4 101.1 OSE 102.4 101.8 104.1 103.7 102.4 103.8 102.5 99.5 101.5 100.8 101.1 0.0

TOT 102.2 105.8 100.5 102.4 107.3 103.7 TLS 105.1 110.1 103.1 103.8 102.5 102.3 92.5 -100.0 105.0 99.8 100.7

TIC 109.3 107.5 101.7 104.5 102.1 103.3 100.2 99.1 83.0 105.7 103.2 103.6 0.1

OTLS 98.0 201.1 269.9 48.3 48.3

COE 102.7 101.9 101.5 101.6 101.6

CFC 99.1 100.6 98.4 98.8 98.8

NOS 80.9 103.2 96.2 97.7 97.7

GVA 91.1 102.2 99.7 100.1 100.1

TOT 102.2 105.8 100.5 102.4 102.1 103.3 100.2 99.1 83.0 105.7 103.2

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 106.9 AGR 113.0 114.5 98.0 113.2 100.1 100.0 207.1 102.1 101.4 106.9

MAN 108.0 MAN 112.6 114.4 107.2 113.1 102.0 99.6 99.5 -311.5 89.4 92.1 99.9 8.1

CON 101.6 CON 100.0 101.0 102.0 101.4 101.8 101.8 101.6

TTC 110.7 TTC 100.0 100.5 101.2 115.2 113.9 110.7 -0.1

FBS 101.0 FBS 97.0 101.4 100.4 100.8 106.7 101.6 101.0

OSE 103.4 OSE 100.0 100.8 101.4 101.4 104.2 99.0 97.1 104.1 103.4

TOT 107.3 TLS

TIC 110.7 112.3 102.8 108.9 102.6 99.6 99.4 -100.8 96.5 97.6 102.0 5.3

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 102.0 104.1 102.0 AGR 106.2 109.0 98.2 107.6 90.0 94.6 -181.3 97.0 95.9 102.0 0.0

MAN 107.5 108.0 100.5 107.2 MAN 114.7 107.9 104.5 107.5 108.4 100.8 100.7 98.0 120.7 116.7 113.9 -6.7

CON 103.4 100.5 100.5 100.5 CON 101.4 100.5 101.7 101.0 98.3 97.4 100.0 102.0 100.0 100.5 0.0

TTC 101.1

99.2 100.3 100.2 TTC 103.9 101.5 100.9 101.2 99.5 98.5 97.2 93.5 99.2 99.2 99.9 0.3

FBS 102.9 100.9 100.4 100.4 FBS 101.8 100.4 100.4 100.4 101.7 101.7 100.5 98.0 100.5 100.4 0.0

OSE 100.2 101.0 OSE 102.5 101.9 104.7 104.2 101.9 103.8 102.5 100.0 101.5 100.7 101.0 0.0

TOT 102.2 105.8 100.5 102.4 TLS 105.1 110.1 103.1 103.8 102.5 102.3 92.5 -100.0 105.0 99.8 100.7

TIC 109.1 104.4 101.4 102.8 102.0 103.3 100.2 99.0 96.1 111.5 104.9 104.2 -6.4

IMP 110.7 112.3 102.8 108.9 102.6 99.6 99.4 -100.8 96.5 97.6 102.0

TOT 109.3 107.5 101.7 104.5 102.1 103.3 100.2 99.1 83.0 105.7 103.2 103.6 -6.4

OTLS 98.0 201.1 269.9 48.3 48.3

COE 102.7 101.9 101.5 101.6 101.6

CFC 99.1 100.6 98.4 98.8 98.8

NOS 80.9 103.2 96.2 97.7 97.7

GVA 91.1 102.2 99.7 100.1 100.1

TOT 102.2 105.8 100.5 102.4 102.1 103.3 100.2 99.1 83.0 105.7 103.2

Domestic Use Table at basic prices

Supply Table at basic prices, transf. to purchasers' prices

Taxes less subsidies on products

Trade and transport margins

Supply Table at basic prices

Imports cif

Supply Table for domestic output at basic prices

Use Table at purchasers' prices

Use Table for taxes less subsidies on products

Use Table for trade and transport margins

Use Table at basic prices

Imports Use Table

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

355

Table A11.3: Supply and use tables 2011 at previous years' prices

AGR MMC

SER

DP IMP

SUPbp

TTM

TLS

SUPpp

AGR MMC

SER TIC

FCH

FCN FCG

GFCF

INV E XP TOTfi n TOTpp Check

AGR 25.5 0.1 0.0 25.7 14.4 40.1 11.2 0.5 51.8 AGR 5.2 15.8 1.6 22.6 5.9 0.2 0.0 22.9 29.0 51.5 0.2

MAN 1.2 295.0 35.6 331.7 311.9 643.7 111.7 38.2 793.6 MAN 8.9 183.1 62.3 254.4 120.3 9.2 51.7 1.6 358.9 541.7 796.1 -2.5

CON 0.1 87.6 4.5 92.2 1.5 93.7 8.3 102.0 CON 0.3 26.1 19.4 45.8 0.5 0.6 53.7 2.1 56.8 102.6 -0.7

TTC 0.5 15.2 231.0 246.7 66.4 313.1 -123.0 3.0 193.0 TTC 0.6 10.8 65.6 77.0 21.0 0.9 0.5 6.9 0.0 85.5 114.7 191.7 1.3

FBS 0.5 8.0 280.8 289.2 54.7 343.9 9.7 353.7 FBS 2.0 43.2 161.1 206.3 76.6 0.0 3.5 17.6 50.0 147.7 354.0 -0.3

OSE 0.3 2.5 220.7 223.5 16.1 239.6 0.1 2.7 242.4 OSE 0.2 3.5 21.2 24.9 53.2 4.4 153.1 1.1 0.2 5.0 217.1 242.0 0.4

TOT 28.1 408.4 772.5 1209.0 465.0 1674.0 0.0 62.4 1736.4 TIC 17.2 282.5 331.3 630.9 277.5 5.3 166.8 131.3 1.8 524.3 1107.0 1737.9 -1.5

OTLS -0.7 0.1 0.1 -0.5 -0.5

COE 2.6 59.8 250.6 313.0 313.0

CFC 3.7 20.4 84.3 108.4 108.4

NOS 5.3 45.7 106.3 157.3 157.3

GVA 10.9 126.0 441.2 578.1 578.1

TOT 28.1 408.4 772.5 1209.0 277.5 5.3 166.8 131.3 1.8 524.3 1107.0

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 0.5 AGR 0.0 0.0 0.0 0.0 0.4 0.1 0.5 0.5

MAN 38.2 MAN 0.2 1.6 6.2 8.0 26.4 0.4 2.3 0.0 1.6 30.7 38.7 -0.5

CON 8.3 CON 0.0 2.0 2.0 0.1 0.1 6.1 0.0 6.3 8.3

TTC 3.0 TTC 0.0 -0.1 1.1 1.0 1.7 -0.1 0.5 -0.1 2.0 3.0

FBS 9.7 FBS 0.0 0.0 4.3 4.3 1.5 0.1 3.9 0.0 5.5 9.7

OSE 2.7 OSE 0.0 0.0 0.0 2.1 -0.4 0.7 0.3 2.7 2.7

TOT 62.4 TOT 0.2 1.5 13.6 15.3 32.1 0.1 13.5 0.0 1.9 47.6 62.9 -0.5

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 11.2 AGR 0.4 1.7 0.4 2.6 3.1 0.0 0.0 5.6 8.6 11.2

MAN 111.7 MAN 0.8 20.0 9.0 29.8 39.9 3.2 5.2 -0.1 34.7 82.9 112.7 -1.0

CON CON

TTC -123.0 TTC -1.2 -21.7 -9.4 -32.3 -43.1 -3.2 -5.2

0.1 -40.2 -91.7 -124.0 1.0

FBS FBS

OSE 0.1 OSE 0.1 0.0 0.1 0.1

TOT 0.0 TOT 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.5 0.1 0.0 25.7 14.4 40.1 AGR 4.8 14.1 1.1 20.0 2.5 0.1 0.0 17.2 19.9 39.8 0.2

MAN 1.2 295.0 35.6 331.7 311.9 643.7 MAN 7.9 161.6 47.1 216.6 53.9 5.6 44.2 1.7 322.6 428.0 644.7 -1.0

CON 0.1 87.6 4.5 92.2 1.5 93.7 CON 0.3 26.1 17.4 43.8 0.4 0.5 47.6 2.1 50.6 94.4 -0.7

TTC 0.5 15.2 231.0 246.7 66.4 313.1 TTC 1.8 32.5 74.0 108.3 62.4 0.9 3.8 11.7 -0.1 125.8 204.4 312.8 0.3

FBS 0.5 8.0 280.8 289.2 54.7 343.9 FBS 2.0 43.2 156.8 202.0 75.1 0.0 3.4 13.7 50.0 142.2 344.2 -0.3

OSE 0.3 2.5 220.7 223.5 16.1 239.6 OSE 0.2 3.5 21.2 24.9 51.0 4.4 153.5 0.4 0.2 4.7 214.3 239.2 0.4

TOT 28.1 408.4 772.5 1209.0 465.0 1674.0 TLS 0.2 1.5 13.6 15.3 32.1 0.1 13.5 0.0 1.9 47.6 62.9

TIC 17.2 282.5 331.3 630.9 277.5 5.3 166.8 131.3 1.8 524.3 1107.0 1737.9 -1.0

OTLS -0.7 0.1 0.1 -0.5 -0.5

COE 2.6 59.8 250.6 313.0 313.0

CFC 3.7 20.4 84.3 108.4 108.4

NOS 5.3 45.7 106.3 157.3 157.3

GVA 10.9 126.0 441.2 578.1 578.1

TOT 28.1 408.4 772.5 1209.0 277.5 5.3 166.8 131.3 1.8 524.3 1107.0

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 14.4 AGR 0.7 5.6 0.5 6.8 1.2 0.0 0.0 6.4 7.6 14.4

MAN 311.9 MAN 1.1 85.6 19.5 106.2 24.0 1.6 19.3 0.1 135.7 180.7 286.9 25.0

CON 1.5 CON 0.0 0.5 0.3 0.9 0.7 0.7 1.5

TTC 66.4 TTC 0.0 2.7 13.9 16.6 0.7 2.8 0.0 45.8 49.3 65.9 0.5

FBS 54.7 FBS 0.2 13.8 24.4 38.5 0.2 2.1 13.9 16.2 54.7

OSE 16.1 OSE 0.0 0.5 3.9 4.4 11.6 0.0 0.1 0.0 11.7 16.1

TOT 465.0 TLS

TIC 2.1 108.7 62.5 173.3 37.7 1.6 24.9 0.1 201.8 266.2 439.5 25.5

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.5 0.1 0.0 25.7 AGR 4.1 8.5 0.6 13.2 1.3 0.1 0.0 10.8 12.2 25.5 0.2

MAN 1.2

295.0 35.6 331.7 MAN 6.8 75.9 27.7 110.4 29.9 3.9 24.9 1.7 186.9 247.3 357.7 -26.0

CON 0.1 87.6 4.5 92.2 CON 0.3 25.6 17.1 43.0 0.4 0.5 46.9 2.1 49.9 92.8 -0.7

TRA 0.5 15.2 231.0 246.7 TTC 1.7 29.9 60.2 91.8 61.7 0.9 3.8 8.9 -0.1 80.0 155.1 246.9 -0.2

TRN 0.5 8.0 280.8 289.2 FBS 1.8 29.4 132.4 163.5 74.9 0.0 3.4 11.6 36.1 126.0 289.5 -0.3

COM 0.3 2.5 220.7 223.5 OSE 0.2 3.0 17.3 20.5 39.5 4.4 153.5 0.4 0.1 4.7 202.6 223.1 0.4

TOT 28.1 408.4 772.5 1209.0 TLS 0.2 1.5 13.6 15.3 32.1 0.1 13.5 0.0 1.9 47.6 62.9

TIC 15.1 173.7 268.8 457.6 239.8 5.3 165.2 106.4 1.7 322.5 840.8 1298.4 -26.5

IMP 2.1 108.7 62.5 173.3 37.7 1.6 24.9 0.1 201.8 266.2 439.5

TOT 17.2 282.5 331.3 630.9 277.5 5.3 166.8 131.3 1.8 524.3 1107.0 1737.9 -26.5

OTLS -0.7 0.1 0.1 -0.5 -0.5

COE 2.6 59.8 250.6 313.0 313.0

CFC 3.7 20.4 84.3 108.4 108.4

NOS 5.3 45.7 106.3 157.3 157.3

GVA 10.9 126.0 441.2 578.1 578.1

TOT 28.1 408.4 772.5 1209.0 277.5 5.3 166.8 131.3 1.8 524.3 1107.0

Check

Supply Table at basic prices

Use Table at basic prices

Imp o rts C IF

Imports Use Table

Supply Table for domestic output at basic prices

Domestic Use Table at basic prices

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Taxes less subsidies on products

Use Table for taxes less subsidies on products

Trade and transport margins

Use Table for trade and transport margins

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

356

Table A11.4: Volume indices for supply and use tables 2011

AGR

MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 100.8 96.1 87.8 100.8 103.5 101.8 103.6 111.9 102.2 AGR 102.0 103.4 96.5 102.6 101.5 84.8 -9.1 102.0 101.1 101.8 0.5

MAN 95.4 104.4 99.4 103.8 103.3 103.6 102.7 97.4 103.1 MAN 99.5 105.6 98.5 103.6 99.4 102.8 112.5 43.1 104.3 103.4 103.5 -0.3

CON 127.1 103.2 95.6 102.8 101.2 102.8 98.2 102.4 CON 100.7 110.7 100.2 105.9 103.3 96.0 101.3 91.6 100.9 103.1 -0.7

TTC 92.6 98.9 103.8 103.4 101.9 103.1 102.8 95.1 103.2 TTC 99.1 100.0 101.3 101.1 99.7 104.5 103.4 102.5 0.7

FBS 99.6 102.2 102.1 102.1 107.4 102.9 98.2 102.8 FBS 98.9 104.7 103.0 103.3 100.5 96.2 106.2 102.2 102.9 -0.1

OSE 95.2 99.1 100.5 100.5 98.3 100.3 94.6 101.9 100.4 OSE 99.4 101.8 101.4 101.5 101.0 99.8 98.2 72.6 103.1 100.0 100.2 0.2

TOT 100.4 103.8 102.0 102.6 103.4 102.8 104.2 97.8 102.6 TIC 100.2 105.5 101.5 103.2 100.1 99.8 99.8 105.3 42.7 104.3 102.4 102.7 0.2

OTLS 137.2 -84.9 -30.6 55.5 55.5

COE 100.1 100.0 101.0 100.8 100.8

CFC 101.6 100.9 101.4 101.3 101.3

NOS 104.4 99.9 105.9 104.0 104.0

GVA 100.8 100.2 102.3 101.8 101.8

TOT 100.4 103.8 102.0 102.6 100.1 99.8 99.8 105.3 42.7 104.3 102.4

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 111.9 AGR 92.9 -40.0 134.6 -375.0 101.4 110.5 103.3 111.9

MAN 97.4 MAN 94.3 97.2 97.6 97.5 98.2 100.8 105.1 -100.0 104.0 99.0 98.7 -1.3

CON 98.2 CON 98.1 98.1 105.9 96.5 98.2 100.0 98.2 98.2

TTC 95.1 TTC 92.2 98.2 96.7 95.1

FBS 98.2 FBS 130.0 -14.3 100.4 100.7 102.9 89.0 225.0 96.3 98.2

OSE 101.9 OSE 240.0 101.0 101.1 101.1 101.9

TOT 97.8 TOT 96.4 99.7 98.3 98.4 98.7 79.8 98.2 -100.0 103.5 98.7 98.6 -0.8

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 103.6 AGR 101.4 104.6 95.9 102.5 101.7 82.4 -50.0 105.4 103.9 103.6

MAN 102.7 MAN 97.9 107.5 97.8 104.1 100.7 101.5 118.6 -183.3 105.2 103.4 103.6 -0.9

CON CON

TTC 102.8 TTC 99.2 107.3 97.7 104.0 100.8 101.5 118.5 -162.0 105.2 103.5 103.6 -0.8

FBS FBS

OSE 94.6 OSE 95.3 92.3 94.6 94.6

TOT TOT

AGR MMC SER

DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 100.8 96.1 100.8 103.5 101.8 AGR 102.1 103.1 95.9 102.4 101.3 85.1 -7.1 100.9 99.9 101.1 0.6

MAN 95.4 104.4 99.4 103.8 103.3 103.6 MAN 99.8 105.5 98.7 103.7 99.0 103.7 112.3 45.7 104.2 103.8 103.7 -0.2

CON 127.1 103.2 95.6 102.8 101.2 102.8 CON 100.7 110.7 100.4 106.3 103.0 96.0 101.7 91.6 101.2 103.5 -0.7

TTC 92.6 98.9 103.8 103.4 101.9 103.1 TTC 99.1 104.7 101.0 102.1 100.5 101.0 109.7 -102.6 104.7 103.5 103.0 0.1

FBS 99.6 102.2 102.1 102.1 107.4 102.9 FBS 98.7 104.7 103.1 103.4 100.5 96.4 102.0 106.2 102.5 103.0 -0.1

OSE 99.1 100.5 98.3 100.3 OSE 99.4 101.6 101.4 101.4 101.0 99.8 93.7 72.6 103.2 100.0 100.2 0.2

TOT 100.4 103.8 102.0 102.6 103.4 102.8 TLS 96.4 99.7 98.3 98.4 98.7 79.8 98.2 -100.0 103.5 98.7 98.6

TIC 100.2 105.5 101.5 103.2 100.1 99.8 99.8 105.3 42.7 104.3 102.4 102.7 -0.1

OTLS 137.2 -84.9 -30.6 55.5 55.5

COE 100.1 100.0 101.0 100.8 100.8

CFC 101.6 100.9 101.4 101.3 101.3

NOS 104.4 99.9 105.9 104.0 104.0

GVA 100.8 100.2 102.3 101.8 101.8

TOT 100.4 103.8 102.0 102.6 100.1 99.8 99.8 105.3 42.7 104.3 102.4

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 103.5 AGR 100.9 106.3 88.2 104.2 105.0 78.8 -32.2 104.5 102.9 103.5

MAN 103.3 MAN 104.1 106.6 99.7 105.2 100.4 92.0 117.4 3.1 86.4 90.0 95.1 8.3

CON 101.2 CON 100.0 110.6 96.6 104.5 97.4 97.4 101.2

TTC 101.9 TTC 102.3 99.0 100.2 101.7 101.4 101.1 0.8

FBS 107.4 FBS 98.8 103.5 107.7 106.1 92.1 110.8 107.4

OSE 98.3 OSE 85.7 104.5 101.8 102.1 97.6 57.8 97.2 97.0 98.3

TOT 103.4 TLS

TIC 102.3 105.9 102.9 104.8 99.7 92.2 114.7 6.3 91.3 93.6 97.7 5.7

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 100.8 96.1 100.8 AGR 102.3 101.0 103.0 101.5 98.0 86.7 19.8 98.9 98.1 99.9 0.9

MAN 95.4 104.4 99.4 103.8 MAN 99.1 104.3 98.0 102.3 98.0 109.4 108.6 79.6 122.5 116.8 111.9 -8.1

CON 127.1 103.2 95.6 102.8 CON 100.7 110.7 100.5 106.3 103.0 96.0 101.8 91.6 101.3 103.5 -0.8

TTC 92.6 98.9 103.8 103.4 TTC 99.0 105.3 101.1 102.4

100.5 101.0 114.6 -110.0 106.6 104.2 103.5 -0.1

FBS 99.6 102.2 102.1 102.1 FBS 98.7 105.3 102.3 102.8 100.5 96.4 98.3 105.3 101.5 102.2 -0.1

OSE 99.1 100.5 OSE 100.0 101.2 101.3 101.2 102.1 99.8 93.7 95.5 103.2 100.2 100.3 0.2

TOT 100.4 103.8 102.0 102.6 TLS 96.4 99.7 98.3 98.4 98.7 79.8 98.2 -100.0 103.5 98.7 98.6

TIC 99.9 105.3 101.2 102.7 100.2 99.8 99.8 103.3 72.5 114.6 105.5 104.5 -8.0

IMP 102.3 105.9 102.9 104.8 99.7 92.2 114.7 6.3 91.3 93.6 97.7

TOT 100.2 105.5 101.5 103.2 100.1 99.8 99.8 105.3 42.7 104.3 102.4 102.7 -8.0

OTLS 137.2 -84.9 -30.6 55.5 55.5

COE 100.1 100.0 101.0 100.8 100.8

CFC 101.6 100.9 101.4 101.3 101.3

NOS 104.4 99.9 105.9 104.0 104.0

GVA 100.8 100.2 102.3 101.8 101.8

TOT 100.4 103.8 102.0 102.6 100.1 99.8 99.8 105.3 42.7 104.3 102.4

Supply Table for domestic output at basic prices

Domestic Use Table at basic prices

Trade and transport margins

Use Table for trade and transport margins

Supply Table at basic prices

Use Table at basic prices

Imp o rts C IF

Imports Use Table

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Taxes less subsidies on products

Use Table for taxes less subsidies on products

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

357

Table A11.5: Supply and use tables 2010 in current prices

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.3 0.2 0.0 25.5 13.9 39.4 10.8 0.4 50.6 AGR 5.1 15.3 1.6 22.0 5.8 0.2 0.2 22.5 28.6 50.6

MAN 1.2 282.6 35.8 319.6 301.8 621.4 108.8 39.2 769.4 MAN 8.9 173.4 63.3 245.6 121.0 8.9 46.0 3.8 344.1 523.7 769.4

CON 0.1 84.9 4.7 89.7 1.5 91.2 8.4 99.6 CON 0.3 23.6 19.4 43.2 0.5 0.6 53.0 2.2 56.4 99.6

TTC 0.6 15.4 222.6 238.5 65.1 303.6 -119.7 3.1 187.1 TTC 0.6 10.8 64.8 76.1 21.0 0.9 0.5 6.7 0.0 81.8 110.9 187.1

FBS 0.5 7.8 275.0 283.3 50.9 334.2 9.9 344.1 FBS 2.1 41.2 156.3 199.6 76.2 0.0 3.7 17.5 47.1 144.5 344.1

OSE 0.3 2.5 219.6 222.4 16.4 238.8 0.1 2.7 241.6 OSE 0.2 3.4 20.9 24.5 52.7 4.5 153.6 1.2 0.3 4.9 217.0 241.6

TOT 28.0 393.3 757.6 1178.9 449.7 1628.6 0.0 63.8 1692.3 TIC 17.1 267.7 326.4 611.2 277.2 5.3 167.2 124.6 4.3 502.5 1081.2 1692.3

OTLS -0.5 -0.1 -0.3 -1.0 -1.0

COE 2.6 59.8 248.1 310.5 310.5

CFC 3.7 20.2 83.1 107.0 107.0

NOS 5.1 45.8 100.4 151.3 151.3

GVA 10.8 125.7 431.2 567.8 567.8

TOT 28.0 393.3 757.6 1178.9 277.2 5.3 167.2 124.6 4.3 502.5 1081.2

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 0.4 AGR 0.0 0.0 0.0 0.0 0.4 0.1 0.5 0.4

MAN 39.2 MAN 0.2 1.6 6.3 8.2 26.9 0.4 2.2 0.0 1.6 31.0 39.2

CON 8.4 CON 0.0 2.0 2.0 0.1 0.1 6.3 0.0 6.4 8.4

TTC 3.1 TTC 0.0 -0.1 1.1 1.1 1.7 -0.1 0.5 -0.1 2.0 3.1

FBS 9.9 FBS 0.0 0.0 4.3 4.3 1.5 0.1 4.1 0.0 5.7 9.9

OSE 2.7 OSE 0.0 0.0 0.0 2.1 -0.4 0.7 0.3 2.7 2.7

TOT 63.8 TOT 0.2 1.5 13.8 15.5 32.5 0.1 13.8 0.0 1.8 48.2 63.8

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 10.8 AGR 0.4 1.6 0.4 2.5 3.0 0.0 0.0 5.3 8.3 10.8

MAN 108.8 MAN 0.8 18.6 9.2 28.6 39.6 3.2 4.4 0.0 32.9 80.2 108.8

CON CON

TTC -119.7 TTC -1.2 -20.2 -9.7 -31.1 -42.7 -3.2 -4.4 -0.1 -38.2 -88.6 -119.7

FBS FBS

OSE 0.1 OSE

0.1 0.0 0.1 0.1

TOT 0.0 TOT 0.0 0.0 0.0 0.0 0.0 0.0

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.3 0.2 0.0 25.5 13.9 39.4 AGR 4.7 13.7 1.2 19.5 2.5 0.2 0.2 17.1 19.9 39.4

MAN 1.2 282.6 35.8 319.6 301.8 621.4 MAN 7.9 153.2 47.8 208.9 54.5 5.4 39.4 3.8 309.6 412.5 621.4

CON 0.1 84.9 4.7 89.7 1.5 91.2 CON 0.3 23.6 17.4 41.2 0.4 0.5 46.8 2.2 50.0 91.2

TTC 0.6 15.4 222.6 238.5 65.1 303.6 TTC 1.8 31.1 73.3 106.2 62.1 0.9 3.7 10.7 0.1 120.1 197.5 303.6

FBS 0.5 7.8 275.0 283.3 50.9 334.2 FBS 2.1 41.2 152.1 195.4 74.8 0.0 3.5 13.4 47.1 138.8 334.2

OSE 0.3 2.5 219.6 222.4 16.4 238.8 OSE 0.2 3.4 20.9 24.5 50.5 4.5 154.0 0.5 0.3 4.6 214.3 238.8

TOT 28.0 393.3 757.6 1178.9 449.7 1628.6 TLS 0.2 1.5 13.8 15.5 32.5 0.1 13.8 0.0 1.8 48.2 63.8

TIC 17.1 267.7 326.4 611.2 277.2 5.3 167.2 124.6 4.3 502.5 1081.2 1692.3

OTLS -0.5 -0.1 -0.3 -1.0 -1.0

COE 2.6 59.8 248.1 310.5 310.5

CFC 3.7 20.2 83.1 107.0 107.0

NOS 5.1 45.8 100.4 151.3 151.3

GVA 10.8 125.7 431.2 567.8 567.8

TOT 28.0 393.3 757.6 1178.9 277.2 5.3 167.2 124.6 4.3 502.5 1081.2

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 13.9 AGR 0.7 5.3 0.6 6.5 1.2 0.0 0.1 6.1 7.4 13.9

MAN 301.8 MAN 1.1 80.4 19.5 101.0 23.9 1.8 16.4 1.7 157.1 200.9 301.8

CON 1.5 CON 0.0 0.5 0.4 0.8 0.7 0.7 1.5

TTC 65.1 TTC 0.0 2.7 13.8 16.6 0.7 2.9 0.0 45.0 48.6 65.1

FBS 50.9 FBS 0.2 13.3 22.7 36.3 0.2 1.7 12.8 14.6 50.9

OSE 16.4 OSE 0.0 0.5 3.8 4.3 11.8 0.2 0.0 12.1 16.4

TOT 449.7 TLS

TIC 2.0 102.6 60.8 165.4 37.8 1.8 21.7 1.9 221.0 284.2 449.7

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.3 0.2 0.0 25.5 AGR 4.0 8.4 0.6 13.0 1.3 0.1 0.1 10.9 12.5 25.5

MAN 1.2 282.6 35.8 319.6 MAN 6.8 72.8 28.2 107.9 30.6 3.6 22.9 2.1 152.5 211.7 319.6

CON 0.1 84.9 4.7 89.7 CON 0.3 23.1 17.0 40.4

0.4 0.5 46.1 2.2 49.3 89.7

TTC 0.6 15.4 222.6 238.5 TTC 1.7 28.4 59.5 89.6 61.4 0.9 3.7 7.8 0.1 75.1 148.9 238.5

FBS 0.5 7.8 275.0 283.3 FBS 1.8 27.9 129.4 159.1 74.5 0.0 3.5 11.8 34.3 124.2 283.3

OSE 0.3 2.5 219.6 222.4 OSE 0.2 2.9 17.1 20.2 38.7 4.5 154.0 0.5 0.1 4.5 202.2 222.4

TOT 28.0 393.3 757.6 1178.9 TLS 0.2 1.5 13.8 15.5 32.5 0.1 13.8 0.0 1.8 48.2 63.8

TIC 15.1 165.0 265.6 445.7 239.4 5.3 165.5 102.9 2.4 281.5 796.9 1242.7

IMP 2.0 102.6 60.8 165.4 37.8 1.8 21.7 1.9 221.0 284.2 449.7

TOT 17.1 267.7 326.4 611.2 277.2 5.3 167.2 124.6 4.3 502.5 1081.2 1692.3

OTLS -0.5 -0.1 -0.3 -1.0 -1.0

COE 2.6 59.8 248.1 310.5 310.5

CFC 3.7 20.2 83.1 107.0 107.0

NOS 5.1 45.8 100.4 151.3 151.3

GVA 10.8 125.7 431.2 567.8 567.8

TOT 28.0 393.3 757.6 1178.9 277.2 5.3 167.2 124.6 4.3 502.5 1081.2

Check

Supply Table at basic prices

Use Table at basic prices

Imp o rts C IF

Imports Use Table

Supply Table for domestic output at basic prices

Domestic Use Table at basic prices

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Taxes less subsidies on products

Use Table for taxes less subsidies on products

Trade and transport margins

Use Table for trade and transport margins

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

358

2. Balanced supply and use tables (tables A11.6–A11.10)

Table A11.6: Supply and use tables 2011 in current prices

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.8 0.2 0.0 26.0 15.4 41.3 10.9 0.5 52.8 AGR 5.5 17.4 1.5 24.4 5.9 0.1 -0.1 22.4 28.4 52.8

MAN 1.3 318.5 35.7 355.5 336.8 692.3 113.0 39.4 844.7 MAN 10.1 202.1 65.4 277.7 122.8 9.1 51.5 1.4 382.1 567.0 844.7

CON 0.1 89.0 4.5 93.6 1.6 95.1 8.3 103.4 CON 0.3 26.2 19.8 46.3 0.5 0.6 54.0 2.1 57.1 103.4

TTC 0.5 15.1 231.6 247.2 72.9 320.1 -124.0 3.0 199.1 TTC 0.6 10.8 66.2 77.5 21.2 0.9 0.5 6.7 0.0 92.3 121.6 199.1

FBS 0.5 8.0 282.0 290.5 55.2 345.7 9.1 354.9 FBS 2.1 43.5 161.5 207.0 78.0 0.0 3.6 16.8 49.5 147.8 354.9

OSE 0.3 2.5 222.9 225.7 16.6 242.3 0.1 3.0 245.4 OSE 0.2 3.5 22.2 25.9 55.1 4.6 153.4 1.2 0.2 5.1 219.5 245.4

TOT 28.4 433.3 776.7 1238.4 498.6 1737.0 0.0 63.3 1800.3 TIC 18.8 303.5 336.5 658.8 283.5 5.5 167.2 130.4 1.5 553.5 1141.5 1800.3

OTLS -0.7 0.2 0.3 -0.3 -0.3

COE 2.7 60.9 254.4 318.0 318.0

CFC 3.7 20.5 82.9 107.1 107.1

NOS 4.1 48.2 102.5 154.7 154.7

GVA 9.7 129.8 440.1 579.6 579.6

TOT 28.4 433.3 776.7 1238.4 283.5 5.5 167.2 130.4 1.5 553.5 1141.5

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 0.5 0.5 AGR 0.0 0.0 0.0 0.1 0.4 0.1 0.5 0.5

MAN 39.4 39.4 MAN 0.2 1.6 6.4 8.3 26.8 0.4 2.2 0.0 1.7 31.1 39.4

CON 8.3 8.3 CON 0.0 2.0 2.0 0.0 0.1 6.1 0.0 6.2 8.3

TTC 3.0 3.0 TTC 0.0 -0.1 1.1 1.0 1.7 -0.1 0.4 -0.1 2.0 3.0

FBS 9.1 9.1 FBS 0.0 0.0 4.4 4.4 1.6 0.1 3.0 0.0 4.7 9.1

OSE 3.0 3.0 OSE 0.0 0.1 0.1 2.3 -0.4 0.7 0.3 2.9 3.0

TOT 63.3 63.3 TOT 0.2 1.6 14.0 15.9 32.9 0.1 12.5 0.0 2.0 47.5 63.3

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 10.9 10.9 AGR 0.4 1.6 0.4 2.5 3.2 0.0 0.0 5.2 8.4 10.9

MAN 113.0 113.0 MAN 0.9 20.4 9.2 30.5 39.9 3.1 5.1 -0.1 34.5 82.5 113.0

CON CON

TTC -124.0 -124.0 TTC -1.3 -22.0 -9.6 -33.0 -43.1 -3.1 -5.1

0.1 -39.7 -91.0 -124.0

FBS FBS

OSE 0.1 0.1 OSE 0.1 0.0 0.1 0.1

TOT 0.0 0.0 TOT 0.0 0.0 0.0 0.0 0.0 0.0

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.8 0.2 0.0 26.0 15.4 41.3 AGR 5.1 15.7 1.1 21.9 2.4 0.1 -0.1 17.1 19.5 41.3

MAN 1.3 318.5 35.7 355.5 336.8 692.3 MAN 9.0 180.1 49.8 238.9 56.1 5.6 44.3 1.5 345.9 453.4 692.3

CON 0.1 89.0 4.5 93.6 1.6 95.1 CON 0.3 26.2 17.7 44.3 0.4 0.5 47.9 2.1 50.9 95.1

TTC 0.5 15.1 231.6 247.2 72.9 320.1 TTC 1.8 32.9 74.8 109.5 62.6 0.9 3.7 11.4 -0.1 132.1 210.6 320.1

FBS 0.5 8.0 282.0 290.5 55.2 345.7 FBS 2.1 43.5 157.1 202.6 76.4 0.0 3.5 13.8 49.5 143.1 345.7

OSE 0.3 2.5 222.9 225.7 16.6 242.3 OSE 0.2 3.5 22.1 25.8 52.7 4.6 153.8 0.4 0.2 4.8 216.5 242.3

TOT 28.4 433.3 776.7 1238.4 498.6 1737.0 TLS 0.2 1.6 14.0 15.9 32.9 0.1 12.5 0.0 2.0 47.5 63.3

TIC 18.8 303.5 336.5 658.8 283.5 5.5 167.2 130.4 1.5 553.5 1141.5 1800.3

OTLS -0.7 0.2 0.3 -0.3 -0.3

COE 2.7 60.9 254.4 318.0 318.0

CFC 3.7 20.5 82.9 107.1 107.1

NOS 4.1 48.2 102.5 154.7 154.7

GVA 9.7 129.8 440.1 579.6 579.6

TOT 28.4 433.3 776.7 1238.4 283.5 5.5 167.2 130.4 1.5 553.5 1141.5

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 15.4 AGR 0.8 6.4 0.5 7.7 1.2 0.0 -0.1 6.6 7.7 15.4

MAN 336.8 MAN 1.3 98.0 20.9 120.2 24.7 1.6 19.2 -0.2 171.3 216.7 336.8

CON 1.6 CON 0.0 0.5 0.4 0.9 0.7 0.7 1.6

TTC 72.9 TTC 0.0 2.7 14.0 16.8 0.7 2.7 0.0 52.7 56.1 72.9

FBS 55.2 FBS 0.2 14.0 24.5 38.8 0.2 2.2 14.1 16.5 55.2

OSE 16.6 OSE 0.0 0.5 3.9 4.5 12.0 0.0 0.1 0.0 12.2 16.6

TOT 498.6 TLS

TIC 2.3 122.1 64.2 188.7 38.9 1.6 24.8 -0.1 244.7 309.9 498.6

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.8 0.2 0.0 26.0 AGR 4.3 9.3 0.6 14.2 1.2 0.1 0.0 10.5 11.7 26.0

MAN 1.3 318.5 35.7 355.5 MAN 7.8 82.1 28.9 118.8 31.4 4.0 25.1 1.6 174.6

236.7 355.5

CON 0.1 89.0 4.5 93.6 CON 0.3 25.7 17.4 43.4 0.4 0.5 47.2 2.1 50.2 93.6

TTC 0.5 15.1 231.6 247.2 TTC 1.8 30.2 60.7 92.7 61.9 0.9 3.7 8.6 -0.1 79.4 154.5 247.2

FBS 0.5 8.0 282.0 290.5 FBS 1.8 29.5 132.6 163.9 76.2 0.0 3.5 11.6 35.4 126.6 290.5

OSE 0.3 2.5 222.9 225.7 OSE 0.2 3.0 18.1 21.3 40.6 4.6 153.8 0.4 0.1 4.8 204.3 225.7

TOT 28.4 433.3 776.7 1238.4 TLS 0.2 1.6 14.0 15.9 32.9 0.1 12.5 0.0 2.0 47.5 63.3

TIC 16.4 181.4 272.3 470.2 244.6 5.5 165.5 105.6 1.6 308.7 831.6 1301.8

IMP 2.3 122.1 64.2 188.7 38.9 1.6 24.8 -0.1 244.7 309.9 498.6

TOT 18.8 303.5 336.5 658.8 283.5 5.5 167.2 130.4 1.5 553.5 1141.5 1800.3

OTLS -0.7 0.2 0.3 -0.3 -0.3

COE 2.7 60.9 254.4 318.0 318.0

CFC 3.7 20.5 82.9 107.1 107.1

NOS 4.1 48.2 102.5 154.7 154.7

GVA 9.7 129.8 440.1 579.6 579.6

TOT 28.4 433.3 776.7 1238.4 283.5 5.5 167.2 130.4 1.5 553.5 1141.5

Check

Supply Table for domestic output at basic prices

Domestic Use Table at basic prices

Impo rts C IF

Imports Use Table

Supply Table at basic prices

Use Table at basic prices

Trade and transport margins

Use Table for trade and transport margins

Taxes less subsidies on products

Use Table for taxes less subsidies on products

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

359

Table A11.7: Price indices for supply and use tables 2011

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfi n TOTpp Check

AGR 102.0 104.1 102.0 106.9 103.8 97.5 107.6 102.4 AGR 105.8 109.9 98.7 108.2 100.4 94.7 568.8 97.7 98.0 102.4

MAN 107.5 108.0 100.5 107.2 108.0 107.6 100.2 101.9 106.2 MAN 114.1 110.4 105.0 109.2 102.1 99.5 99.6 85.7 106.8 104.9 106.2

CON 103.4 100.5 100.5 100.5 101.6 100.5 100.1 100.5 CON 101.4 100.5 101.8 101.0 97.4 97.1 100.0 102.0 100.0 100.5

TTC 101.1 99.2 100.3 100.2 110.7 102.4 100.0 101.4 104.0 TTC 100.9 100.1 100.8 100.7 102.2 108.0 106.2 104.0

FBS 102.9 100.9 100.4 100.4 101.0 100.5 93.6 100.3 FBS 101.4 100.7 100.4 100.5 101.7 101.8 99.0 100.1 100.3

OSE 100.2 101.0 103.4 101.1 99.1 110.2 101.2 OSE 102.4 101.7 104.3 103.9 102.7 103.8 101.7 99.5 101.8 100.9 101.2

TOT 102.2 105.8 100.5 102.4 107.3 103.7 100.8 103.6 TIC 109.3 107.5 101.7 104.5 102.1 103.3 100.2 99.1 83.0 105.8 103.2 103.6

OTLS 98.0 201.1 269.9 48.3 48.3

COE 102.7 101.9 101.5 101.6 101.6

CFC 99.1 100.6 98.4 98.8 98.8

NOS 79.8 103.2 96.2 97.7 97.7

GVA 90.8 102.2 99.7 100.1 100.1

TOT 102.2 105.8 100.5 102.4 102.1 103.3 100.2 99.1 83.0 105.8 103.2

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfi n TOTpp Check

AGR 107.6 AGR 92.3 612.5 91.4 230.0 100.8 96.2 99.8 107.6

MAN 101.9 MAN 103.7 106.1 103.9 104.3 101.7 98.7 93.8 -100.0 104.4 101.3 101.9

CON 100.1 CON 102.2 102.1 90.7 95.1 99.6 100.0 99.4 100.1

TTC 101.4 TTC 99.4 103.3 102.4 101.4

FBS 93.6 FBS 123.1 100.0 102.7 102.8 105.7 103.8 44.4 86.4 93.6

OSE 110.2 OSE 241.7 109.2 101.1 108.5 110.2

TOT 100.8 TOT 105.1 108.7 103.3 103.9 102.5 102.3 92.5 -100.0 105.0 99.8 100.8

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfi n TOTpp Check

AGR 97.5 AGR 90.4 96.7 100.7 96.3 104.9 85.7 100.0 94.0 97.9 97.5

MAN 100.2 MAN 113.7 102.0 102.3 102.4 99.8 97.9 97.5 93.5 99.5 99.5 100.2

CON CON

TTC 100.0 TTC 105.0 101.6 102.2 101.9 100.2 97.9 97.5 93.8 98.8 99.3 100.0

FBS FBS

OSE 99.1 OSE 97.6 104.2 99.1 99.1

TOT TOT

AGR MMC SER DP IMP SUPbp TTM

TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfi n TOTpp Check

AGR 102.0 104.1 102.0 106.9 103.8 103.8 AGR 107.2 111.2 98.1 109.5 94.9 95.6 725.0 98.9 98.0 103.8

MAN 107.5 108.0 100.5 107.2 108.0 107.6 107.6 MAN 114.4 111.5 105.6 110.3 104.0 100.5 100.2 85.7 107.6 106.2 107.6

CON 103.4 100.5 100.5 100.5 101.6 100.5 100.5 CON 101.4 100.5 101.7 101.0 98.3 97.4 100.0 102.0 100.1 100.5

TTC 101.1 99.2 100.3 100.2 110.7 102.4 102.4 TTC 103.8 101.1 101.0 101.1 100.8 98.5 97.0 93.7 105.0 103.1 102.4

FBS 102.9 100.9 100.4 100.4 101.0 100.5 100.5 FBS 101.2 100.7 100.4 100.5 101.7 101.7 100.6 99.0 100.6 100.5

OSE 100.2 101.0 103.4 101.1 101.1 OSE 102.4 101.8 104.1 103.7 102.4 103.8 102.5 99.5 101.5 100.8 101.1

TOT 102.2 105.8 100.5 102.4 107.3 103.7 103.7 TLS 105.1 108.7 103.3 103.9 102.5 102.3 92.5 -100.0 105.0 99.8 100.8

TIC 109.3 107.5 101.7 104.5 102.1 103.3 100.2 99.1 83.0 105.8 103.2 103.6

OTLS 98.0 201.1 269.9 48.3 48.3

COE 102.7 101.9 101.5 101.6 101.6

CFC 99.1 100.6 98.4 98.8 98.8

NOS 79.8 103.2 96.2 97.7 97.7

GVA 90.8 102.2 99.7 100.1 100.1

TOT 102.2 105.8 100.5 102.4 102.1 103.3 100.2 99.1 83.0 105.8 103.2

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfi n TOTpp Check

AGR 106.9 AGR 113.0 114.5 98.0 113.2 100.1 100.0 207.1 102.1 101.4 106.9

MAN 108.0 MAN 112.6 114.4 107.2 113.1 102.8 99.6 99.5 -311.5 106.6 105.3 108.0

CON 101.6 CON 100.0 101.0 102.0 101.4 101.8 101.8 101.6

TTC 110.7 TTC 100.0 100.5 101.2 115.2 113.9 110.7

FBS 101.0 FBS 97.0 101.4 100.4 100.8 106.7 101.6 101.0

OSE 103.4 OSE 100.0 100.8 101.4 101.4 104.2 99.0 97.1 104.1 103.4

TOT 107.3 TLS

TIC 110.7 112.3 102.8 108.9 103.2 99.6 99.4 -100.8 107.9 106.4 107.3

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfi n TOTpp Check

AGR 102.0 104.1 102.0 AGR 106.2 109.0 98.2 107.6 90.0 94.6 -181.3 97.0 95.9 102.0

MAN 107.5 108.0 100.5 107.2 MAN 114.7 108.1 104.5 107.6 105.0 100.8 100.7 98.0 108.5 106.9 107.2

CON 103.4 100.5 100.5 100.5 CON 101.4 100.5 101.7 101.0 98.3 97.4 100.0 102.0 100.0 100.5

TTC 101.1 99.2 100.3 100.2 TTC 103.9 101.1 100.9 101.0 100.8 98.5 97.2 93.5 99.2 99.7 100.2

FBS 102.9 100.9

100.4 100.4 FBS 101.8 100.4 100.4 100.4 101.7 101.7 100.5 98.0 100.5 100.4

OSE 100.2 101.0 OSE 102.5 101.9 104.7 104.2 101.9 103.8 102.5 100.0 101.5 100.7 101.0

TOT 102.2 105.8 100.5 102.4 TLS 105.1 108.7 103.3 103.9 102.5 102.3 92.5 -100.0 105.0 99.8 100.8

TIC 109.1 104.4 101.4 102.8 101.9 103.3 100.2 99.0 96.1 104.1 102.0 102.3

IMP 110.7 112.3 102.8 108.9 103.2 99.6 99.4 -100.8 107.9 106.4 107.3

TOT 109.3 107.5 101.7 104.5 102.1 103.3 100.2 99.1 83.0 105.8 103.2 103.6

OTLS 98.0 201.1 269.9 48.3 48.3

COE 102.7 101.9 101.5 101.6 101.6

CFC 99.1 100.6 98.4 98.8 98.8

NOS 79.8 103.2 96.2 97.7 97.7

GVA 90.8 102.2 99.7 100.1 100.1

TOT 102.2 105.8 100.5 102.4 102.1 103.3 100.2 99.1 83.0 105.8 103.2

Supply Table for domestic output at basic prices

Domestic Use Table at basic prices

Imp o rts C IF

Imports Use Table

Supply Table at basic prices

Use Table at basic prices

Trade and transport margins

Use Table for trade and transport margins

Taxes less subsidies on products

Use Table for taxes less subsidies on products

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

360

Table A11.8: Supply and use tables 2011 at previous years' prices

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.3 0.1 0.0 25.5 14.4 39.8 11.2 0.5 51.5 AGR 5.2 15.8 1.6 22.6 5.9 0.2 0.0 22.9 29.0 51.5

MAN 1.2 295.0 35.6 331.7 311.9 643.7 112.7 38.7 795.1 MAN 8.9 183.1 62.3 254.4 120.3 9.2 51.7 1.6 357.9 540.7 795.1

CON 0.1 88.6 4.5 93.1 1.5 94.7 8.3 102.9 CON 0.3 26.1 19.4 45.8 0.5 0.6 54.0 2.1 57.1 102.9

TTC 0.5 15.2 231.0 246.7 65.9 312.6 -124.0 3.0 191.5 TTC 0.6 10.8 65.6 77.0 20.8 0.9 0.5 6.9 0.0 85.5 114.5 191.5

FBS 0.5 8.0 280.8 289.2 54.7 343.9 9.7 353.7 FBS 2.0 43.2 160.8 206.0 76.6 0.0 3.5 17.6 50.0 147.7 353.7

OSE 0.3 2.5 220.7 223.5 16.1 239.6 0.1 2.7 242.4 OSE 0.2 3.5 21.2 24.9 53.6 4.4 153.1 1.1 0.2 5.0 217.5 242.4

TOT 27.8 409.4 772.5 1209.8 464.5 1674.3 0.0 62.9 1737.1 TIC 17.2 282.5 331.0 630.6 277.7 5.3 166.8 131.6 1.8 523.3 1106.5 1737.1

OTLS -0.7 0.1 0.1 -0.5 -0.5

COE 2.6 59.8 250.6 313.0 313.0

CFC 3.7 20.4 84.3 108.4 108.4

NOS 5.1 46.7 106.6 158.4 158.4

GVA 10.7 127.0 441.5 579.2 579.2

TOT 27.8 409.4 772.5 1209.8 277.7 5.3 166.8 131.6 1.8 523.3 1106.5

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 0.5 AGR 0.0 0.0 0.0 0.0 0.4 0.1 0.5 0.5

MAN 38.7 MAN 0.2 1.6 6.2 8.0 26.4 0.4 2.3 0.0 1.6 30.7 38.7

CON 8.3 CON 0.0 2.0 2.0 0.1 0.1 6.1 0.0 6.3 8.3

TTC 3.0 TTC 0.0 -0.1 1.1 1.0 1.7 -0.1 0.5 -0.1 2.0 3.0

FBS 9.7 FBS 0.0 0.0 4.3 4.3 1.5 0.1 3.9 0.0 5.5 9.7

OSE 2.7 OSE 0.0 0.0 0.0 2.1 -0.4 0.7 0.3 2.7 2.7

TOT 62.9 TOT 0.2 1.5 13.6 15.3 32.1 0.1 13.5 0.0 1.9 47.6 62.9

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 11.2 AGR 0.4 1.7 0.4 2.6 3.1 0.0 0.0 5.6 8.6 11.2

MAN 112.7 MAN 0.8 20.0 9.0 29.8 39.9 3.2 5.2 -0.1 34.7 82.9 112.7

CON CON

TTC -124.0 TTC -1.2 -21.7 -9.4 -32.3 -43.1 -3.2 -5.2 0.1 -40.2 -91.7 -124.0

FBS FBS

OSE 0.1 OSE 0.1

0.0 0.1 0.1

TOT 0.0 TOT 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.3 0.1 0.0 25.5 14.4 39.8 AGR 4.8 14.1 1.1 20.0 2.5 0.1 0.0 17.2 19.9 39.8

MAN 1.2 295.0 35.6 331.7 311.9 643.7 MAN 7.9 161.6 47.1 216.6 53.9 5.6 44.2 1.7 321.6 427.0 643.7

CON 0.1 88.6 4.5 93.1 1.5 94.7 CON 0.3 26.1 17.4 43.8 0.4 0.5 47.9 2.1 50.9 94.7

TTC 0.5 15.2 231.0 246.7 65.9 312.6 TTC 1.8 32.5 74.0 108.3 62.2 0.9 3.8 11.7 -0.1 125.8 204.2 312.6

FBS 0.5 8.0 280.8 289.2 54.7 343.9 FBS 2.0 43.2 156.5 201.7 75.1 0.0 3.4 13.7 50.0 142.2 343.9

OSE 0.3 2.5 220.7 223.5 16.1 239.6 OSE 0.2 3.5 21.2 24.9 51.4 4.4 153.5 0.4 0.2 4.7 214.7 239.6

TOT 27.8 409.4 772.5 1209.8 464.5 1674.3 TLS 0.2 1.5 13.6 15.3 32.1 0.1 13.5 0.0 1.9 47.6 62.9

TIC 17.2 282.5 331.0 630.6 277.7 5.3 166.8 131.6 1.8 523.3 1106.5 1737.1

OTLS -0.7 0.1 0.1 -0.5 -0.5

COE 2.6 59.8 250.6 313.0 313.0

CFC 3.7 20.4 84.3 108.4 108.4

NOS 5.1 46.7 106.6 158.4 158.4

GVA 10.7 127.0 441.5 579.2 579.2

TOT 27.8 409.4 772.5 1209.8 277.7 5.3 166.8 131.6 1.8 523.3 1106.5

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 14.4 AGR 0.7 5.6 0.5 6.8 1.2 0.0 0.0 6.4 7.6 14.4

MAN 311.9 MAN 1.1 85.6 19.5 106.2 24.0 1.6 19.3 0.1 160.7 205.7 311.9

CON 1.5 CON 0.0 0.5 0.3 0.9 0.7 0.7 1.5

TTC 65.9 TTC 0.0 2.7 13.9 16.6 0.7 2.8 0.0 45.8 49.3 65.9

FBS 54.7 FBS 0.2 13.8 24.4 38.5 0.2 2.1 13.9 16.2 54.7

OSE 16.1 OSE 0.0 0.5 3.9 4.4 11.6 0.0 0.1 0.0 11.7 16.1

TOT 464.5 TLS

TIC 2.1 108.7 62.5 173.3 37.7 1.6 24.9 0.1 226.8 291.2 464.5

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 25.3 0.1 0.0 25.5 AGR 4.1 8.5 0.6 13.2 1.3 0.1 0.0 10.8 12.2 25.5

MAN 1.2 295.0 35.6 331.7 MAN 6.8 75.9 27.7 110.4 29.9 3.9 24.9 1.7 160.9 221.3 331.7

CON 0.1 88.6 4.5 93.1 CON 0.3

25.6 17.1 43.0 0.4 0.5 47.2 2.1 50.2 93.1

TRA 0.5 15.2 231.0 246.7 TTC 1.7 29.9 60.2 91.8 61.5 0.9 3.8 8.9 -0.1 80.0 154.9 246.7

TRN 0.5 8.0 280.8 289.2 FBS 1.8 29.4 132.1 163.2 74.9 0.0 3.4 11.6 36.1 126.0 289.2

COM 0.3 2.5 220.7 223.5 OSE 0.2 3.0 17.3 20.5 39.9 4.4 153.5 0.4 0.1 4.7 203.0 223.5

TOT 27.8 409.4 772.5 1209.8 TLS 0.2 1.5 13.6 15.3 32.1 0.1 13.5 0.0 1.9 47.6 62.9

TIC 15.1 173.7 268.5 457.3 240.0 5.3 165.2 106.7 1.7 296.5 815.3 1272.6

IMP 2.1 108.7 62.5 173.3 37.7 1.6 24.9 0.1 226.8 291.2 464.5

TOT 17.2 282.5 331.0 630.6 277.7 5.3 166.8 131.6 1.8 523.3 1106.5 1737.1

OTLS -0.7 0.1 0.1 -0.5 -0.5

COE 2.6 59.8 250.6 313.0 313.0

CFC 3.7 20.4 84.3 108.4 108.4

NOS 5.1 46.7 106.6 158.4 158.4

GVA 10.7 127.0 441.5 579.2 579.2

TOT 27.8 409.4 772.5 1209.8 277.7 5.3 166.8 131.6 1.8 523.3 1106.5

Check

Supply Table for domestic output at basic prices

Domestic Use Table at basic prices

Imports Use Table

Use Table at basic prices

Imp o rts C IF

Supply Table at basic prices

Trade and transport margins

Use Table for trade and transport margins

Taxes less subsidies on products

Use Table for taxes less subsidies on products

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

361

Table A11.9: Volume indices for supply and use tables 2011

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 99.9 96.1 87.8 99.9 103.5 101.1 103.6 111.9 101.8 AGR 102.0 103.4 96.5 102.6 101.5 84.8 -9.1 102.0 101.1 101.8

MAN 95.4 104.4 99.4 103.8 103.3 103.6 103.6 98.7 103.3 MAN 99.5 105.6 98.5 103.6 99.4 102.8 112.5 43.1 104.0 103.2 103.3

CON 127.1 104.3 95.6 103.9 101.2 103.8 98.2 103.4 CON 100.7 110.7 100.2 105.9 103.3 96.0 101.9 91.6 101.4 103.4

TTC 92.6 98.9 103.8 103.4 101.1 102.9 103.6 95.1 102.4 TTC 99.1 100.0 101.3 101.1 98.8 104.5 103.2 102.4

FBS 99.6 102.2 102.1 102.1 107.4 102.9 98.2 102.8 FBS 98.9 104.7 102.8 103.2 100.5 96.2 106.2 102.2 102.8

OSE 95.2 99.1 100.5 100.5 98.3 100.3 94.6 101.9 100.4 OSE 99.4 101.8 101.4 101.5 101.8 99.8 98.2 72.6 103.1 100.2 100.4

TOT 99.6 104.1 102.0 102.6 103.3 102.8 104.2 98.6 102.6 TIC 100.2 105.5 101.4 103.2 100.2 99.8 99.8 105.6 42.7 104.1 102.3 102.6

OTLS 137.2 -84.9 -30.6 55.5 55.5

COE 100.1 100.0 101.0 100.8 100.8

CFC 101.6 100.9 101.4 101.3 101.3

NOS 99.7 102.0 106.2 104.7 104.7

GVA 98.6 101.0 102.4 102.0 102.0

TOT 99.6 104.1 102.0 102.6 100.2 99.8 99.8 105.6 42.7 104.1 102.3

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 111.9 AGR 92.9 -40.0 134.6 -375.0 101.4 110.5 103.3 111.9

MAN 98.7 MAN 94.3 97.2 97.6 97.5 98.2 100.8 105.1 -100.0 104.0 99.0 98.7

CON 98.2 CON 98.1 98.1 105.9 96.5 98.2 100.0 98.2 98.2

TTC 95.1 TTC 92.2 98.2 96.7 95.1

FBS 98.2 FBS 130.0 -14.3 100.4 100.7 102.9 89.0 225.0 96.3 98.2

OSE 101.9 OSE 240.0 101.0 101.1 101.1 101.9

TOT 98.6 TOT 96.4 99.7 98.3 98.4 98.7 79.8 98.2 -100.0 103.5 98.7 98.6

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 103.6 AGR 101.4 104.6 95.9 102.5 101.7 82.4 -50.0 105.4 103.9 103.6

MAN 103.6 MAN 97.9 107.5 97.8 104.1 100.7 101.5 118.6 -183.3 105.2 103.4 103.6

CON CON

TTC 103.6 TTC 99.2 107.3 97.7 104.0 100.8 101.5 118.5 -162.0 105.2 103.5 103.6

FBS FBS

OSE 94.6 OSE 95.3 92.3 94.6 94.6

TOT TOT

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH

FCN FCG GFCF INV E XP TOTfi n TOTpp Check

AGR 99.9 96.1 99.9 103.5 101.1 AGR 102.1 103.1 95.9 102.4 101.3 85.1 -7.1 100.9 99.9 101.1

MAN 95.4 104.4 99.4 103.8 103.3 103.6 MAN 99.8 105.5 98.7 103.7 99.0 103.7 112.3 45.7 103.9 103.5 103.6

CON 127.1 104.3 95.6 103.9 101.2 103.8 CON 100.7 110.7 100.4 106.3 103.0 96.0 102.3 91.6 101.8 103.8

TTC 92.6 98.9 103.8 103.4 101.1 102.9 TTC 99.1 104.7 101.0 102.1 100.2 101.0 109.7 -102.6 104.7 103.4 102.9

FBS 99.6 102.2 102.1 102.1 107.4 102.9 FBS 98.7 104.7 102.9 103.2 100.5 96.4 102.0 106.2 102.5 102.9

OSE 99.1 100.5 98.3 100.3 OSE 99.4 101.6 101.4 101.4 101.8 99.8 93.7 72.6 103.2 100.2 100.3

TOT 99.6 104.1 102.0 102.6 103.3 102.8 TLS 96.4 99.7 98.3 98.4 98.7 79.8 98.2 -100.0 103.5 98.7 98.6

TIC 100.2 105.5 101.4 103.2 100.2 99.8 99.8 105.6 42.7 104.1 102.3 102.6

OTLS 137.2 -84.9 -30.6 55.5 55.5

COE 100.1 100.0 101.0 100.8 100.8

CFC 101.6 100.9 101.4 101.3 101.3

NOS 99.7 102.0 106.2 104.7 104.7

GVA 98.6 101.0 102.4 102.0 102.0

TOT 99.6 104.1 102.0 102.6 100.2 99.8 99.8 105.6 42.7 104.1 102.3

Check

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 103.5 AGR 100.9 106.3 88.2 104.2 105.0 78.8 -32.2 104.5 102.9 103.5

MAN 103.3 MAN 104.1 106.6 99.7 105.2 100.4 92.0 117.4 3.1 102.3 102.4 103.3

CON 101.2 CON 100.0 110.6 96.6 104.5 97.4 97.4 101.2

TTC 101.1 TTC 102.3 99.0 100.2 101.7 101.4 101.1

FBS 107.4 FBS 98.8 103.5 107.7 106.1 92.1 110.8 107.4

OSE 98.3 OSE 85.7 104.5 101.8 102.1 97.6 57.8 97.2 97.0 98.3

TOT 103.3 TLS

TIC 102.3 105.9 102.9 104.8 99.7 92.2 114.7 6.3 102.6 102.4 103.3

AGR MMC SER DP IMP SUPbp TTM TLS SUPpp AGR MMC SER TIC FCH FCN FCG GFCF INV E XP TOTfin TOTpp Check

AGR 99.9 96.1 99.9 AGR 102.3 101.0 103.0 101.5 98.0 86.7 19.8 98.9 98.1 99.9

MAN 95.4 104.4 99.4 103.8 MAN 99.1 104.3 98.0 102.3 98.0 109.4 108.6 79.6 105.5 104.6 103.8

CON 127.1 104.3 95.6 103.9 CON 100.7 110.7 100.5 106.3 103.0 96.0 102.4 91.6 101.9 103.9

TTC 92.6 98.9 103.8 103.4 TTC 99.0 105.3 101.1 102.4 100.1 101.0 114.6 -110.0 106.6 104.0 103.4

FBS 99.6 102.2 102.1 102.1 FBS 98.7 105.3 102.1 102.6 100.5 96.4 98.3 105.3 101.5 102.1

OSE

99.1 100.5 OSE 100.0 101.2 101.3 101.2 103.1 99.8 93.7 95.5 103.2 100.4 100.5

TOT 99.6 104.1 102.0 102.6 TLS 96.4 99.7 98.3 98.4 98.7 79.8 98.2 -100.0 103.5 98.7 98.6

TIC 99.9 105.3 101.1 102.6 100.2 99.8 99.8 103.6 72.5 105.3 102.3 102.4

IMP 102.3 105.9 102.9 104.8 99.7 92.2 114.7 6.3 102.6 102.4 103.3

TOT 100.2 105.5 101.4 103.2 100.2 99.8 99.8 105.6 42.7 104.1 102.3 102.6

OTLS 137.2 -84.9 -30.6 55.5 55.5

COE 100.1 100.0 101.0 100.8 100.8

CFC 101.6 100.9 101.4 101.3 101.3

NOS 99.7 102.0 106.2 104.7 104.7

GVA 98.6 101.0 102.4 102.0 102.0

TOT 99.6 104.1 102.0 102.6 100.2 99.8 99.8 105.6 42.7 104.1 102.3

Supply Table at basic prices

Use Table at basic prices

Supply Table for domestic output at basic prices

Domestic Use Table at basic prices

Imp o rts C IF

Imports Use Table

Taxes less subsidies on products

Use Table for taxes less subsidies on products

Trade and transport margins

Use Table for trade and transport margins

Supply Table at basic prices, transf. to purchasers' prices

Use Table at purchasers' prices

Handbook on Supply and Use Tables and Input-Output Tables with Extensions and Applications

362

Table A11.10: Supply and use tables 2010 in current prices