ARETROSPECTIVE
ASSESSMENTOFCLEAN
ENERGYINVESTMENTSIN
THERECOVERYACT
February2016
**Draft**
1
Contents
ExecutiveSummary.......................................................................................................................................2
HighlightsoftheARRACleanEnergyRelatedInvestments..........................................................................4
I.Introduction...............................................................................................................................................7
II.EconomicRationaleforCleanEnergyRelatedInvestment......................................................................9
III.OverviewofCleanEnergyFundingUndertheRecoveryAct.................................................................12
IV.DramaticTrendsinCleanEnergyMarkets............................................................................................16
Growth
inRenewableEnergyGeneration..............................................................................................16
GrowthinRenewableEnergyCapacity...................................................................................................16
LoweredCleanEnergyCosts...................................................................................................................18
MajorJobGrowthintheRenewableEnergySector...............................................................................19
V.HighlightsoftheARRACleanEnergyInvestments.................................................................................21
RenewableGeneration...........................................................................................................................21
EnergyEfficiency.....................................................................................................................................22
CleanandEnergy
EfficientTransportation.............................................................................................22
CleanEnergyManufacturing...................................................................................................................22
OtherInitiatives......................................................................................................................................23
VI.Conclusions............................................................................................................................................24
References..................................................................................................................................................25
AppendixI:DetailsoftheARRACleanEnergyRelatedInvestments.........................................................28
RenewableEnergyInvestments..............................................................................................................28
InvestmentsinEnergyEfficiency............................................................................................................35
TransitInvestments.................................................................................................................................41
Grid
ModernizationInvestments............................................................................................................42
InvestmentsinAdvancedVehiclesandFuels.........................................................................................44
InvestmentsinCarbonCaptureandStorage..........................................................................................47
GreenInnovationandJobTraining.........................................................................................................48
CleanEnergyEquipmentManufacturing................................................................................................49
AppendixII:CEACalculatio n ofARRACleanEnergySpending...................................................................50

2
ExecutiveSummary
President Obama took office in the middle of the worst economic crisis since the Great
Depression.Inthepreviousyear,privateemployersshed3.8millionjobs.Trillionsofdollarsof
household wealth had beenwiped out, and the economy’s total output, as measured by real
grossdomesticproduct(GDP),wasinthemidstofitsmostseveredownturnofthepostwarera.
Inthefaceofthiscrisis,thePresidenttookimmediate,bold,andeffectiveaction.
OnFebruary17,2009,lessthanamonthintohisfirstterm,PresidentObamasignedintolawthe
American Recovery and Reinvestment Act of 2009, also known as the Recovery Act, or ARRA.
ARRAwasnotonlyahistoricactiontohelpbringaboutamacroeconomicrecovery,butitwas
also a dramatic investment in the future of the U.S. economy. In 2009, there was an initial
allocation of $90 billion dollars of ARRA funds towards clean energyrelated investments—an
unprecedentedinvestmentincleanenergyandtowardsasustainable21
st
centuryeconomy.
Theseinvestmentscontributedtotherecovery—GDPpercapitastartedexpandinginthethird
quarter of 2009 and reached its precrisis level in nearly four years. The Council of Economic
Advisers(CEA)(2014)estimatedthatfromlate2009throughmid2011,ARRAliftedGDP2to3
percentabovewhere itwouldhave been, andover6 millionjobyears(afull time job forone
year)weresupportedbyARRAfrom2009to2012.Thecleanenergyrelatedfundingmadeup
roughly one eighth of the total, representing a substantial direct boost. In this report, CEA
estimates that ARRA clean energyrelated programs supported roughly 900,000 jobyears in
innovativecleanenergyfieldsfrom2009to2015.
TheshortruneffectsofARRAareonlyhalfthestory.Theseinvestmentslaidthegroundworkfor
theremarkablegrowthincleanenergyintheUnitedStatesthathasoccurredoverthepastseven
years.Solarelectricitygeneration has increased over 30fold since 2008.Wind generation has
increasedoverthreefoldsince2008. Throughavarietyofmechanisms,ARRAfundingreached
nearly every aspect of the value chain for numerous key clean energy technologies, including
advanced vehicles, batteries, carbon capture and sequestration, and technologies to enhance
energyefficiency.These investments are a down payment towards an innovative 21
st
century
cleaneconomyandpromisetoyieldbenefitsformanyyearsintothefuture.
Some of the highlights of the ARRA funding include the Advanced Research Projects Agency
Energyprogramforearlystageinnovations,severalofwhicharealreadyontheirwaytomarket;
electric vehicle battery manufacturing facilities in many states; over 180 awards to advanced
clean energy manufacturing; deployment of smart meters in states across the nation; and
weatherizing more than 800,000 homes. There is an economic rationale for using funding
designedtostimulatetheeconomytoinvestin projectsoflongtermvalue.The cleanenergy
related projects
funded by ARRA help address a variety of market failures, including
environmentalandinnovationmarketfailures.
3
SinceARRA,therehasbeencontinualgrowthincleanenergyjobsthathaswithstoodsignificant
headwinds. While there is more work to do to keep us on track to a sustainable 21
st
century
economy,thecleanenergyARRAinvestmentswereanimportantstepintherightdirection.
4
HighlightsoftheARRACleanEnergyRelatedInvestments
TheboldinvestmentsincleanenergythroughARRAhadasubstantialmacroeconomicimpact,
supportingtheU.S.economyatacriticalmomentbyraisingoutputandsupportingjobs.
Over $90 billion of ARRA funding was invested in clean energy and related technologies.
These investments covered renewable energy generation, clean transportation, energy
efficiency, grid modernization, advanced vehicles and fuels, carbon capture and storage,
greeninnovationandjobtraining,andcleanenergymanufacturing.
TheARRAcleanenergyrelatedprogramssupportedroughly900,000jobyears(fulltimejobs
over one year), from 2009 to 2015. Many of these jobs provided employment during a
significant labor market downturn. Previous CEA analysis estimated roughly 720,000 job
yearsweresupportedfrom2009to2012.
CleanenergyinvestmentsmadeupoveroneeighthoftotalARRAspendingandprovideda
meaningfulboosttoeconomicoutput.CEAestimatesthatARRAraisedthelevelofGDPby
between2and3percentfromlate2009throughmid2011.
ARRA’s clean energy policies laid the foundation for a longterm transition to a cleaner
economy by improving clean energy markets, unlocking private capital, helping drive down
cleanenergytechnologycosts,andexpandingresearchanddevelopmentofnewtechnologies.
TheARRAfundingnot onlyprovidedastimulus,butalsohelpedaddressmarketfailuresin
clean energy markets. The diverse set of funding mechanisms helped to address market
failuressuchasenvironmentalexternalities,innovationmarketfailures,andcapitalmarket
failuresbyinvestinginmeasuresacrossthecleanenergyvaluechain.
Renewable
Generation
29%
Other
0%
Clean
EnergyMfg
2%
Advanced
Vehicles
andFuels…
Grid
Modern
ization
12%
CarbonCapture 4%
GreenInnovation
andJobTraining
4%
Energy
Efficiency
22%
Transit
20%
Figure1:Distribution ofInitialCleanEnergy
AppropriationsbyCategory
Source:CEA(2010).
5
Solar electricity generation has increased over 30 times from 2008 levels, and wind
generationhasincreasedoverthreetimes.TheinvestmentsmadethroughARRAhelpedto
catalyze dramatic trends in clean energy over the past seven years. Along with the lift in
renewable energy generation is a commensurate notable decline in technology costs for
thesetechnologies,makingthemallthemorecompetitiveagainstfossilfuelgeneration.
ARRAinvestmentsinthedeploymentofcleanenergytechnologiesalsohelpedcontributeto
dramaticcostreductionsforthosesametechnologiesaspartofavirtuouscycle.Forexample,
theovernightcapitalcostofutilityscalephotovoltaic(PV)systemsfellfrom$4.1/watt(W)in
2008 to $2.0/W in 2014—a decrease of 50 percent. Cost reductions for this and other
technologiesresulted from anumberof factors—including economiesofscale,technology
learning, and new business practices—that were assisted by the widespread
deployment
madepossiblebyARRA.
ARRAhassupportedover$27billionininvestmentincleanenergygenerationandcapacity.
Thesefundswerecomplementedbyinvestmentstomodernizethegridsuchthatthesenew
energysourcescouldplayaroleinprovidingcleanenergytohouseholdsandfirms.
SubstantialARRAcleanenergyfundingwasinvestedinlongrunresearchanddevelopment,
helping to bring transformative new technologies to fruition. The benefits from this
investmentarebeginningtopayoff,withnewtransformativetechnologiessuchasa1MW
(megawatt) silicon carbide transistor, engineered microbes that use hydrogen and carbon
dioxidetomakeliquidtransportationfuel,andmajoradvancesinlithiumionbatteries.
ARRA helped catalyze substantial shifts toward cleaner energy sources, raising energy
efficiency,supportingjobs,andbuildingtheinfrastructureneededforlargescaleuseofclean
energy.
ARRA support helped improve residential energy efficiency by weatherizing over 800,000
homes,leadingtooveronemillionhomesbeingweatherizedbetween2009and2012with
federalsupport.TheWeatherizationAssistanceProgramscaledupquicklyandreachedlow
incomehomesacrossthecountry.
ARRAenergyefficiencyinvestmentsareprojectedtosaveover400millionMMBtu(million
Britishthermalunits)ofenergyoverthenextfourdecades,orthe equivalentofannualenergy
consumptionfor10,000homes.OakRidgeNationalLaboratory(ONRL2015c)estimatesthat
all ARRA investments will save 400 million MMBtu of energy from 2009 to 2050. This is
equivalent to the entire annual energy consumption of 10,000 homes, if the savings are
distributedequallyovertheyears.
Amajordemonstrationcarboncaptureandstoragefacility in the UnitedStates was made
possiblethrough ARRAfunding.AsofMay15,2015,theAirProductsandChemicalshydrogen
6
facility in Port Arthur, Texas had successfully captured its second millionth metric ton of
carbondioxide.
ARRAhelpedtosupportthetrainingofover30,000studentsforsolarcareersandotherclean
energycareers.Over 25 Energy Training Partnership grants provided support for electrical
apprenticeship training, programs targeting unemployed dislocated workers, women,
minoritiesandveterans,andcommunitygroupsservingunemployedanddislocatedworkers.
ARRA helped lay the groundwork for a transition to a cleaner and more energy efficient
transportationsystemwithover$18billioninitiallyallocatedtowardscleanertransportation.
This investment supported public transit with purchase of 12,000 buses, vans, and rail
vehicles and the constructionor renovation of over 850 transit facilities. It also supported
highspeed rail through $8 billion allocated to 49 projects, 98 percent of which are either
completeorwithconstructionunderway.
The Clean Energy Manufacturing ARRA funding, along with the more stable wind turbine
market facilitated by ARRA support, helped support a dramatic increase in the share of
domesticallyproducedwind turbinesusedintheUnitedStatesfrom25percentin20062007
to72 percentin2012. The 30percentclean energy manufacturingtaxcreditwasawidely
oversubscribedsuccess,withnumerousinnovativeprojectsacrossthecleanenergyspace.
The Smart Grid Investment Program helpedto support the installation of 16 million smart
metersby2016.Smartmeterprojectsaroundthecountrycanhelpconsumersunderstand
theirenergyuseandpricing,helpingtofacilitatethetransitiontoamoreresilientelectricgrid
infrastructure.
ARRA funds provided a matching grant to enable the deployment of over 4,600 electric
vehiclechargingstations.ChargePointreceiveda$15milliongranttoleverageprivatefunds
toinstallchargingstationsinstatesacrossthenationtohelpsupporttheemergingelectric
vehiclemarket.

7
I. Introduction
The American Recovery and Reinvestment Act (ARRA), also known as the Recovery Act, was
signed by President Obama on February 17, 2009, at a time when the U.S. economy was
contractingataratenotseensincetheGreatDepression.Therewasaseverefinancialcrisis,and
asteepdeclineinconsumerandbusinessconfidence,householdwealth,andaccesstocredit.In
thelastquarterof2008,employmentwasfallingbymorethan700,000jobspermonthandU.S.
realgrossdomesticproduct(GDP)contractedatan8.9percentannualizedrate.ARRAwaspart
ofacomprehensiveandboldcountercyclicalfiscalpolicyresponsetotheeconomic turmoilthat
grippedtheUnitedStatesandtheworldeconomy.
ARRA’simmediategoalwasstabilizetheeconomy,preservingandrestoringjobs,andassisting
deeplysufferingindustries.Inacontextofweakaggregate demand, already aggressive use of
monetarypolicytoolsbringinginterestratestonearzerolevels,highlyconstrainedcredit,and
expectations of protracted contraction, there is a strong economic case for a significant fiscal
stimulustoincreaseneartermeconomicoutput.Asdocumentedinnumerousreports,ARRAis
estimatedtohaveincreasedoutputandemploymentsubstantiallyrelativetoabaselinewithout
thefiscalstimulus.Forexample,theCouncilofEconomicAdvisers(CEA)estimatesthattheentire
ARRApackageincreasedGDPbetween2and3percentfromlate2009tomid2011.
1
ARRA also provided a major opportunity for laying the groundwork for sustainable longrun
growthandwellbeing.Indeed,thereisareasonwhytheword“Reinvestment”wasincludedin
the title of the Act. A second goal of ARRA was invest in the foundation for a robust and
sustainable21
st
centuryeconomy.AsizablefractionoftheARRAfundswereinvestedinprojects
thatimprovedlongrunproductivity,suchastransportationinfrastructureimprovements,aswell
as investments in innovative technologies, including clean energy technologies and related
innovations.
Thisreportfocusesontheseinvestmentsincleanenergyandasustainable21
st
centuryeconomy.
ARRAappropriated$787billionatthetimeofpassage,andthiswaslaterrevisedto$831billion
over the 2009 to 2019 period.
2
Of the initial allocations, $90 billion was allocated towards
investing in a cleaner, more sustainable energy future.
3
These investments can be seen as a
“downpayment” on the transition to a sustainable 21
st
century economy, and each has an
economic rationale based on addressing multiple market failures, such as environmental
externalitiesandinnovationmarketfailures.

1
SeeCEA(2014)forestimatesoftheaggregateeconomiceffectsoftheARRAandtheliteratureontheuseoffiscal
stimulusinarecession.
2
Thosefigures,though,include$69billionallocatedtoaroutinesetofpatchesfortheAlternativeMinimumTax
(AMT).ThispartoftheAct,acontinuationofalongstandingpractice,isbestthoughtofasongoingfiscalpolicy,not
asa temporaryfiscalimpulsedesignedspecificallytocounterthe
effectsofaneconomicrecession. Excluding the
AMTpatch,theRecoveryActprovidedatotalfiscalimpulseof$763billion.
3
Asdescribedbelow,certainprogramswereextendedorhadgreatertakeupthananticipated. Assuchthetotal
allocationofARRArelatedcleanenergyprogramswillbeover$90billion.CEAcalculationsindicatethatjustunder
$90billionofARRAcleanenergyrelatedfundinghadbeenspentbytheend
of2015.
8
Thisretrospectivereviewofthecleanenergyandrelatedinvestmentsprovidesseveralinsights.
Ithighlights dramatictrendsof increasedactivity andinnovationincleanenergymarketsover
thepastsevenyears.Forexample,solarelectricitygeneration increasedover30fold,andwind
generation increased over threefold since 2008. It reviews
the direct contributions of the
RecoveryActforjobsandeconomicactivity.TheARRAcleanenergyprogramsareestimatedto
havesupportedroughly900,000jobyears,whereajobyearisafulltimejobforoneyear.Italso
connectsthedotsbetweeninvestmentsmadeintheRecoveryAct
andmajorinnovativeactivity,
suchasthe400potentiallytransformativeAdvancedResearchProjectsAgencyEnergy(ARPAE)
projects.Whileformallyprovingcausalityforanypolicy,includingARRA,ishighlychallengingand
requiresareliablecounterfactual,thisreviewprovidesevidencethatARRAcontributedtooneof
thegreatestincreasesincleanenergyactivityinthehistoryoftheUnitedStates,andsupported
hundredsofthousandsofjobsthatweremuchneededinthewakeofthefinancialcrisis.
This report is a followon from a series of quarterly reports by the CEA on the effects of the
RecoveryActonoveralleconomicactivityandemploymentthatranthroughthefourthquarter
of2013.Thereportaddsnewdataonthedramatictrendsoccurringincleanenergymarketsand
providesinsightintowheretheARRAcleanenergyfundingwasspentandthewaysinwhichit
contributedtocleanenergydevelopment.Assuch,thereportprovidesanassessmentoftherole
of ARRA in leading the transition of the energy economy in the United States and a more
completepictureoftheAct’slongertermimpactsontheU.S.economy.
Theremainderofthisreportbeginswithadiscussionoftheeconomicrationaleforcleanenergy
investmentaspartoftheARRApackage.Nextisanoverviewofthecleanenergyfundingunder
ARRA,which isfollowedbythenotable trendsin cleanenergyduringandfollowingARRAand
thenabriefdiscussionofsomeofthehighlightsoftheimpactsofthekeycleanenergyrelated
components of ARRA. Finally the report concludes with a discussion of why the ARRA clean
energyrelated investments were successful. Appendix I provides much greater detail on the
majorcategoriesofprograms,includingmanymoreexamplesofprogramimpacts.AppendixII
providesanoverviewofCEA’s
calculationoftheARRAcleanenergyspendingovertimeusedfor
thejobsestimate.
9
II. EconomicRationaleforCleanEnergyRelatedInvestment
Economictheoryprovidesusefulguidanceforthedevelopmentofpoliciesandprogramsinthe
clean energy space. In particular, there are an array of market failures that can lead to
underinvestment in clean energy, energy efficiency, and research and development in these
areas relative to the socially optimal levels of investment. These market failures will come up
throughoutthediscussioninthisreportofthedifferentcleanenergyrelatedinvestmentsmade
underARRA.
Thefirstcategoryofmarketfailurethatafflictsmarketsforcleanenergy andenergyefficiency
consists of environmental externalities from the burning of fossil fuels. These environmental
externalities
includedamages from greenhousegasemissionsandotherlocal air pollutants.If
theseenvironmentalexternalitiesareinternalizedintothemarketprice forenergyservices,the
priceoffossilfuelbasedenergywouldincreaseandtherelativefinancialattractivenessofclean
energy would improve, resulting in more clean energy investment. Thus,
without some
interventiontointernalizethesenegativeexternalitiesorotherwiseaddressthismarketfailure,
themarketwillunderprovidecleanenergy.
Asecond categoryofmarket failureistheresultof thepositiveexternalityofknowledge spill
overs.These marketfailures,calledinnovationmarketfailures, lead to an underinvestment in
innovative activity on new technologies by private actors because some of the returns from
investmentspillovertootherfirms,implyingthatthesocialrateofreturnoninnovati veactivity
is much higher than the private rate of return. Innovation market failures are particularly
important for fledgling technologies that are just being developed, for foundational advances
may be difficult to patent and are more likely to spillover to other firms than incremental
advancesinmaturetechnologies.
The most common example of an innovation market failure is the research and development
(R&D) market failure from imperfect appropriability of innovations (i.e., those who
innovate
cannot appropriate or capture the returns from the innovation). R&D market failures lead to
underinvestmentinR&Dactivity.Anotherexampleofaninnovationmarketfailureisthemarket
failurefromcostdeclines duetolearningbydoing(i.e.,improvingprocessesandloweringcosts
withadditionalexperience)inthetechnologythatspillovertootherfirmsinamarket.Learning
bydoing spillovers would dampen the incentive to quickly ramp up production in a new
technology to gain experience. Clean energy technologies are often susceptible to both R&D
marketfailuresand learningbydoing marketfailures becausetheyarenewtechnologieswith
significant gains still to be made (and spillover to others in society) from both research and
learning.
4
Again, this implies that the private market left to its own devices will underprovide
cleanenergyinnovationinvestment.

4
SeeJaffeandStavins(1994)orGillinghamandSweeney(2012)foramoredetaileddiscussionofinnovationmarket
failuresinthecontextofcleanenergy.
10
Athirdmarketfailureisthepossibilityofenergysecurityexternalitiesfromthe useofoil,which
many clean energy technologies, particularly in the transportation sector, can help offset. As
discussedinworkbyBrownandHuntington(2013)andNordhaus(2009),whentheU.S.economy
isexposedtooilpriceshocks,
therecanhavesignificantmacroeconomiceffects.Theseoilprice
shocks may occur from crises on the other side of the world, such as wars, revolutions, and
decisionsmadebyforeigncountries.Thusoilisunderpricedwhenoneconsidersthecosttothe
economyfromrelyingheavily onoil, leading to
underinvestmentincleanenergytechnologies
thathelpreduceoilconsumption.
A fourth category of market failures that may lead to underinvestment in clean energy and
energy efficiency consists of information market failures. These may be due to inadequate or
poor information about new clean energy or energy efficient consumer technologies. In the
contextofenergyefficiency,theseinformationmarketfailuresmayeveninteractwith
behavioralfailures,suchasconsumermyopia(i.e.,consumersonlyfocusingontheveryshort
runattheexpenseofthelongrunforaparticulargoodorserviceinawaythatisdifferentthan
otherchoicesconsumersmake)leadingto anoverweightingoftheupfrontcostsforenergyusing
durablegoods relativetoother decisionsconsumers make.Thereissome evidencethatthese
issuescouldleadtounderinvestmentinconsumerenergyefficiencydecisions.
5
Itisalsopossible
thattheycouldinfluencecleanenergypurchases.
Othermarketfailuresmayalsoaffectcleanenergyinvestmentsincertaincircumstances.During
times with extremely tight credit markets, such as just after the 2008 financial crisis, there is
oftenlimitedavailabilityofcapitalforinvestmentsinpromisingnewtechnologies.Thesecapital
constraints very likely afflicted clean energy investments just as the ARRA funding became
available. Such capital constraints at times of crisis call for greater availability of financing for
promisingnewtechnologies.
In some cases, there may also be network externalities that influence the adoption of new
technologies
thatrelyonnetworkeffects.Networkeffectsimplythatthevalueofaproductis
greater when there is a larger network of users of that product. For instance, electric vehicle
charging stations are more valuable if there is a large network of charging stations and many
electricvehicledrivers.Thesameistrueforbiofuelrefuelingstations.Sometimesthesenetwork
effectsareinternalizedalreadybythe produceroftheproduct. However,inthe caseofmany
cleanenergytechnologies,eitherthebenefitsdonotallaccruetotheproducerorthescaleof
investmentrequiredtocreatethenetworkissolarge(especiallyimportantduringtimesoftight
credit markets), that these network effects are not internalized, and thus there is an
underinvestmentinthecleanenergytechnologies.Governmentsupportcannotonlyprovidethe
needed funds, but it can also act as a signal to the market that a particular technology has
support,influencingexpectationsoffuturegrowthandinvestmentdecisions.

5
E.g.,seeGillinghamandPalmer(2014)andAllcottandGreenstone(2012)forreviewsoftheacademicliterature
onmarketandbehavioralfailuresinenergyefficiencypolicy.
11
Such a diverse array of market failures affecting clean energy markets provides an economic
rationalefortheuseofavarietyofpolicytoolsforgovernmentintervention.Italsounderscores
that when a sizable stimulus package is needed for macroeconomic purposes, there is an
economic rationale for investing this funding in
clean energyrelated technologies that will
providelongtermbenefits.NotonlydidtheARRAfundingalleviatemacroeconomicconcerns,
but the investments help to address several important market failures. The package of clean
energyrelatedinvestmentinARRAhelpstospurcleanenergymarketsallalongthevaluechain,
aswillbeseeninthefollowingsections.
12
III. OverviewofCleanEnergyFundingUndertheRecoveryAct
The over $90 billion ARRA investment in clean energyrelated sectors represents one of the
largestinvestmentsinthesesectorsinU.S.history.Forexample,itwasanorderofmagnitude
largerthanthefiveyear$6.3billionClimateTechnologyInitiativeproposedduringtheClinton
Administration.Further,theARRAcleanenergyrelatedinvestmentsleveragedontheorderof
$150billion inadditionalprivateandnonfederalcapitalininvestmentstoward advancing the
deploymentof energyefficiency,wind, solar,geothermal,biomass,lowcarbon fossil fuel,and
othertechnologies.
6
DirectinvestmentsfromARRA fundedavarietyofdifferentprojectswithlongrunimplications.
These projects were broadly targeted to help address market failures, such as environmental
externalitiesandinnovationmarketfailures.Forexample,grantsfordeploymentofrenewable
energyprojects,suchassolarphotovoltaic(PV)projects,helpedtoreducegreenhousegasand
otherpollutantemissions.Similarly,thesegrantsmayhavefosteredlearningbydoingspillovers
through a growing and publicizing of the market. Many of the projects involved research and
developmentatcriticalpointsinthevaluechainwhenthe technologyspilloverstoothersare
likelytobelarge,andthustheprivateinvestmentespeciallylow.
The ARRA clean energyrelated investments can be divided into two categories based on the
fundingmechanism.First,therewere45investmentprovisionswithaninitiallyestimatedtotal
allocation of $60.7 billion. These provisions were primarily focused on areas of highvalu e
investmenttoprimeasustainable21
st
centuryeconomy.Second,therewere11taxincentives,
whichtheU.S.Treasuryestimatedwillinvest$29.5billionthroughfiscalyear2019.
7
Thesetax
incentiveswere primarily focused on fostering newtechnologiesin the renewableenergyand
advancedvehicletechnologyspace,butalsoprovidestrongsupportforenergyefficiency.Such
investments accelerated the growth of key new technologies, which will help reduce
environmentalexternalitiesandU.S.relianceonoilforyearstocome.
AnotherwaytocategorizetheARRAcleanenergyrelatedinvestmentsisbythebroadcategory
of project. Figure 1 illustrates the distribution of the initially allocated $90 billion investment
acrosseightbroadcategoriesofcleanenergyprojects:renewablegeneration,energyefficiency,
grid modernization, advanced vehicles and fuels, transit, carbon capture
and storage, green
innovationandjobtraining,cleanenergymanufacturing.Figure2illustratesthebreakdownof
initial allocations by spending category, and shows that 67 percent of the allocations went
towardsspendingand33percentwenttowardstaxcredits.

6
CEA(2010)calculatesthat$46billionofthe$90billioninitialallocationhasacoinvestmentcomponentandthat
this$46billionleveragesover$150billioninprivateandnonfederalcapitalinvestment.SeeAldy(2013)foramore
detaileddiscussionofthecoinvestmentcomponentoftheARRAclean
energyinvestments.
7
Thisestimateonlyconsideredtaxcreditsinplacein2010,anddoesnotincludeinvestmentsfromlaterextensions
ofthetaxcredits(CEA2010).
13

Atnearly30percent,thelargestcategoryofallocationsconsistedofinvestmentsinrenewable
energy generation, which both induced innovation and helped address environmental
externalities.At22percent,thenextlargestcategoryofallocationswenttowardsinvestments
inenergyefficiency,alsoreducingenvironmentalexternalities.Transitinvestmentsmadeup20
percentofallocations,andservedtobuildinfrastructuretobothimproveeconomicproductivity
andreducetransportationemissions.At12percentofallocations,gridmodernizationalsoserved
to improve infrastructure, increase electricity system reliability, and demonstrate new
technologiesforamoreresilientandsustainableelectricitygrid.Theremaininginvestmentsin
advancedvehiclesandfuels,greeninnovationandjobtraining,carboncaptureandstorage,and
cleanenergymanufacturingallservedtofosterinnovationinnew technologies,preparing the
UnitedStatesforalongertermshifttoaclean21
st
centuryeconomy.
A key element in all of the ARRA clean energyrelated investments is that while they were
designedtoprovidelongtermbenefits,theallocationsfocusedasmuchaspossibleonprojects
thatwere“shovelready”andcouldbedeployedrelativelyquickly,inordertotakeadvantageof
resources in the economy that were underutilized due to the Great Recession. In short, the
allocations aimed to put people back to work and contributed to both the recovery and
reinvestment goals of the legislation. More broadly, the choice of allocation across the many
possibleprojectswasbasedprimarily
onaseveralcriteria:theabilitytodeployresourcesquickly,
thepotentialforfederalsupporttostimulateprivatefinancing,existenceofadministrativeand
authoritativecapacityforpolicyimplementation,CO
2
reductionpotential, andimpactperdollar
in employment, economic activity, and changes to the energy system.
8
Federal capacity to
administer programs and funds were also considered to help assess feasibility. All of these
considerations led to a wide variety of funding mechanisms and policy approaches, as will be
seenlaterinthisreport.

8
Summers(2008)suggeststhatanypackageintendedtoprovidestimulusshouldbe,atleastinitially,drivenbythe
“3T’s”;theyshouldbe“timely,targeted,andtemporary.”Theseprinciplesechothestrongpreferenceforshovel
readyprojectsandusingexistingprogramstodistributethefundsinastimelyamanner
aspossiblewhiletheneed
isthegreatest.
Renewable
Generation
29%
Other
0%
Clean
EnergyMfg
2%
Advanced
Vehicles
andFuels…
Grid
Modern
ization
12%
CarbonCapture 4%
GreenInnovation
andJobTraining
4%
Energy
Efficiency
22%
Transit
20%
Figure1:Distribution ofInitialCleanEnergy
AppropriationsbyCategory
Source:CEA(2010).
Spending
67%
TaxIncentive
33%
Figure2:DistributionofInitialCleanEnergy
AllocationsbyType
Source: CEACalculations.
14
AllARRAspendingwaschanneledthroughfederalagenciesbeforebeingdisbursedtostateand
localgovernmentsorotherrecipients.Muchoftheenergyrelatedfunding;includingforenergy
efficiency,carbon capture and storage, and grid modernization; was administered by the U.S.
DepartmentofEnergy(DOE).
9
TheU.S.TreasuryDepartmenthandledmuchofthesupportfor
renewable energy through tax incentives and cash grant program; the Federal Transportation
Administration(FTA)andFederalRailroadAdministration(FRA)handledtransportationfunding.
Table1shows the breakdownofARRAspending across thecategoriesofcleanenergyrelated
investment.Thesecond
columnshows theinitial allocationoffundingacrosscategoriesbased
onthe2009allocations.Thesesumuptothetotalof$90billion,whichisthestandardnumber
usedtodescribetheARRAcleanenergyinvestments.However,someoftheprogramsweretax
creditprograms,ratherthanallocationsoffunding.Forexample,the1603CashGrantprogram
forrenewablesandthe30percentcleanenergymanufacturingtaxcreditwerebothtaxcredit
programs. Both of these programs were in high demand, and in fact the clean energy
manufacturingtaxcreditwassignificantlyoversubscribed.Otherprograms,suchas taxcredits
foradvancedvehicles,continuepast2015andmaynothavespentalloftheallocatedfundingby
2015.CEAestimatesthatthefinalallocationoutto2019willaddroughly$15billiontothetotal
sum, bringing it to roughly $105 billion. Of that total, CEA estimates roughly $88.5 billion has
beenspentsofar.NonARRArelatedextensionsofanyoftheprograms,includingtheproduction
taxcreditandinvestmenttaxcredit,areexcludedfromthesetotals.
BasedontheARRAspending,itispossibletoestimatetheimpactonemployment.PreviousCEA
analysisestimatedthatARRAcleanenergyrelatedinvestmentshelpedsupportroughly720,000
jobyears(whereajobye ar isoneyearoffulltimeequivalentemployment)overtheperiodfrom

9
CarleyandHyman(2013).
15
2009to2012.
10
Thiswasbasedontheextenttowhichinitiallyallocatedfundinghadbeenspent
outoverthatperiod.Includingfurtherspendingoverthelastthreeyears,CEAestimatesroughly
900,000jobyearsweresupportedbythe$88.5billionARRAcleanenergyrelatedinvestments
from 2009 to 2015.
11
When all of the ARRA funds have been spent out by 2019, CEA analysis
suggests that over one million jobyears will have been supported from 2009 to 2019. Many
programshavebeencontinuedlongbeyondtheARRAtimeframeandcontinuetoprovideboth
employmentandlongtermbenefits
totheeconomy.Further,manyprogramsalsoleveragedand
made possible significant private investment, which supported additional jobs. This report
focusesontheinvestmentsdirectlyfundedororiginatedunderARRA.

10
TheseCEAestimatesincludebothdirect,indirect,andinducedjobs.Directjobsincludepeoplehireddirectlyby
thegovernmentoritsgranteesorcontractorstodotheworkpaidforbythestimulus.Indirectjobsareadditional
people hired by suppliers to and subcontractors of the original recipients to provide
the inputs needed for the
projects.Inducedjobsarepeoplewhoarehiredbecauseoftheextraspendingbyworkershiredindirectandindirect
jobs.Roughlytwothirdsofthejobyearsaredirectandindirectcleanenergyjobs;theremainderareinducedjobs.
11
Theestimateof900,000jobyearsiscalculatedbasedonamacroeconomicmultipliersimilartopreviousCEAARRA
analysisadjustedforchangesinnominalGDPandthelaborforce.Otherapproachesyieldanevenlargerestimate;
forexample,usingafullmacroeconomicmodelyieldsanestimateofoveronemillion.
Theestimateincludesdirect,
indirect,andinducedjobs,asdefinedinfootnote10.
16
IV. DramaticTrendsinCleanEnergyMarkets
Cleanenergyinthe UnitedStateshasmade enormousstridesover the pastsevenyearssince
2009. Electricity generation from renewable energy sources has skyrocketed, and costs for
technologieslike wind turbines and solar panels haveplummeted.The clean energyeconomy
has also supported high quality jobs and robust employment growth. While the vast progress
thathasbeenmadeacrossthecleanenergysectorcannotbeentirelyattributedtoARRA,ARRA
supporthasplayedaroleinthedramatictrendsthatcanbeobservedincleanenergymarkets.
GrowthinRenewableEnergyGeneration
Tobegin,theshareofelectricitygeneratedbynonhydrorenewableshasincreasedfromroughly
2 percent in 2005 to over 8 percent in 2015 (Figure 3). Growth in windpowered and solar
poweredelectricitygenerationhasbeenparticularlynotable.DatafromtheEnergyInformation
Administration (EIA) show that since 2008, electricity generation from wind has more than
tripled,andelectricityfromsolarhasincreased30fold(Figure4).
Manyfactorshavecontributedtothisdramaticgrowth.Therehavebeenimprovementsinboth
windandsolarPVtechnologies.Manystateshaveimplementedrenewableportfoliostandards.
And many states and municipalities have other policies to encourage renewable energy
generation. All of these other factors complement the major investments in wind and solar
technologydevelopmentanddeploymentinARRA.
GrowthinRenewableEnergyCapacity
Alongwith therapidgrowth in newelectricitygeneration from renewablesourceshas beena
commensurate surge in renewable energy capacity. Electric generation capacity refers to the
maximumoutputthatageneratorcanproduce,whileelectricitygenerationreferstotheactual
electricity produced. As illustrated in Figure 5, nonhydro renewable energy capacity in the
UnitedStatesmorethantripledbetween2008and2015,fromatotaloflessthan30thousand
0
2
4
6
8
10
2000 2003 2006 2009 2012 2015
Figure3: MonthlyShareofNonHydroRenewablesin
NetElectric PowerGeneration
Percentof TotalNetGeneration
Note:Dotted lineisasmoothedtrend,showntodampenthestrong
seasonalpatterns(theshareofnonhydrorenewablesdropsduringthe
winterandsummerbothseasonsofhighpowergenerationdemand).
Source:EnergyInformationAdministration.
0
15
30
45
60
75
90
0
100
200
300
400
500
600
2008 2009 2010 2011 2012 2013 2014 2015
Figure4:ElectricityGenerationfromWindandSolar
ThousandMegawatthoursperDay
ThousandMegawatthoursperDay
Source:EnergyInformationAdministration.
Wind
(leftaxis)
Solar
(rightaxis)
17
megawattsto almost 100 thousand megawatts.Mostoftheincreasewas driven by growth in
windandsolarcapacity.
Thegrowthinrenewableenergycapacityhasnotonlybeenimpressive,butitalsofarexceeded
expectations prior to the enactment of ARRA. The dashed lines in Figures 6 and 7 show EIA’s
projections for installed wind and solar capacity made in an early release of the 2009 Annual
EnergyOutlook(AEO)thatexcludedanyforecastedimpacts from ARRA. The solid line in each
figureshowsactualinstalledcapacity,whichiswellabovethepreARRAprojections.
Just as
described above, the impressive performance of the clean energy sector since 2008
cannotbesolely attributeddirectlytoARRApolicies.Otherfactorsverylikelycomplementedthe
ARRAcleanenergyinvestmentsandalsocontributedtorenewablecapacitygrowth.That said,
ARRAfundscontributedsignificantlytoinvestmentsincleanenergyprojectsinseveralwaysas
willbefurtherdiscussedbelow,andareamajor contributortotheremarkableandunexpected
growthinthemarketobservedoverthepastseveralyears.
0
10
20
30
40
50
60
70
80
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Source:EnergyInformationAdministration.
Figure6:ElectricPowerSectorInstalledWindCapacity
Gigawatts(GW)
Actual
2009PreARRARelease
0
2
4
6
8
10
12
14
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Note:Includes solarPVandsolarthermal.
Source:EnergyInformationAdministration.
Figure7:ElectricPowerSectorInstalledSolarCapacity
Gigawatts(GW)
Actual
2009PreARRARelease
18
LoweredCleanEnergyCosts
Along with the considerable growth in generation by renewables, there have been sizable
declines in the cost of renewable and othe r clean energy technologies since the ARRA
investments. For example, LED lighting has seen a nearly 90 percent decrease in cost per kilo
lumensince2008.Batterycostsforelectricvehicleshavefallenfromalmost$1,000/kWhin2008
tounder$300/kWhin2014.
12
Butinparticular,thecostsofwindandsolarPVtechnologyhave
notablydeclinedinrecentyears,andarerapidlybecomingcostcompetitivewithtraditionalfossil
fuelsources.
A common metric for comparing cost competitiveness between renewable and conventional
technologiesisthe“levelizedcostofenergy”(LCOE).TheLCOErepresentstheperkilowatthour
cost(inrealdollars)ofbuilding andoperatingageneratingplantoveranassumedfinanciallife
anddutycycle.SeveralkeyinputsaretakenintoaccountwhencalculatingLCOE,includingcapital
costs,fuelc osts, fixed andvariable operations and maintenancecosts,financingcosts, and an
assumedutilizationrateforeachplanttype.
13
Importantly,theavailabilityofvariousincentives
includingstateorfederaltaxcreditsarealsofactoredin.Becausesolarand windte chnologies
havenofuelcosts,theirLCOEsarehighlydependentontheestimatedcapitalcostsofgeneration
capacityandavailableincentives,andcanvarysubstantiallybyregion.
BasedonbothLCOEandovernightcapitalcosts(thecosttobuildapowerplantifnointerestwas
incurredduringconstruction),windandsolarcostshavefallensubstantiallyoverthepastdecade.
Figure8indicatesthepatternofcostsovertimesince2005,withthe shadedareaindicatingthe
time period in which the ARRA programs were in the greatest effect (note that several ARRA
programshavecontinuedpasttheendof2013).AsillustratedinFigure9,theLCOEforonshore
windtechnologieswasdown32percentin2014fromits2005level,andthecostofsolarPVwas
down60percent.Overnightcapitalcostshavealsodeclined,byalmost10percentforwindand
60percentforsolar.

12
DOE(2015).
13
EIA(2015).
19
Thedeclineinthelevelizedcostof windtechnologyoverthelast decade has been somewhat
slowerthanthedeclinesseenforsolartechnology.Infact,averagewindturbinepricesactually
morethandoubledbetween2002and2008beforeresumingadownwardtrend.Theincreasein
turbine prices during the pre
ARRA period was influenced by several factors, including higher
prices for materials, energy, and labor inputs and an increase in turbine manufacturer
profitabilitydueinparttorobustdemandgrowthandturbineandcomponentsupplyshortages.
Inaddition,overthepast15yearstherehasbeenasubstantialincreasein
turbinesize,including
hub height and rotor diameter. According to the U.S. Department of Energy, the average
nameplatecapacityofnewlyinstalledwindturbinesin2014wasup172percentsince19981999,
andaveragehubheightsandrotordiametersincreased48and108percent,respectively,over
the same time period.
14
So, while the levelized costs were declining modestly,the technology
wasalsoimproving,allowingforwindinstallationsatsitesthatwerepreviouslyuneconomicand
possiblyimplyingevenlargeractualcostsavingsoverthepastdecadethanifthetechnologywas
heldfixed.
Currently,thereisawiderangeofcostsfornewwindandsolargeneration.Therearealready
some places in the United States where new wind—without any subsidies or accounting for
externalities—cancompeteheadtoheadwithnewfossilfuelgenerationonpurelyatechnology
costbasis.Moreover,inmanylocationsintheUnitedStates,thecostfornewwindandsolaris
atorbelow thecostofnewfossilfuelgenerationwhenincludingsubsidiesandtaxcreditsused
to promote renewable generation. Adjusting for environmental externalities would tip the
balanceevenfurther.Withouttakingintoaccount subsidiesorexternalities,however,windand
solar on average remain more expensive forms of new generation than coal or natural gas,
underscoring the continued need for government support to catalyze greater investment in
renewablegeneration.Promisingly,forecastsforwindandsolarPVcostsfromthe EIAandthe
InternationalEnergyAgency(IEA)suggestthattheunsubsidizedtechnologycostofnewwindand
solarwithout subsidieswillbe on parorbelow thatof newcoalby2020(Figure 9).TheARRA
investmentsinwindandsolarunquestionablyplayedaroleishelpingthese technologiesgetto
scaleandbecomemorecompetitive.
MajorJobGrowthintheRenewableEnergySector
EmploymentintherenewablesectorspansseveralcategoriesinFederaldatacollectionsystems,
which complicates direct estimation of job growth and output in the sector. However, trade
association data suggest that, along with the rapid expansion in wind and solar electricity
generation,therehasalsobeenasharpriseinemployment.From2010to2015,employmentin
thesolarenergyindustrymorethandoubled,andisexpectedtoincreasebyanother15percent
in2016(seeFigure10).

14
EERE(2015).
20
Between2010and2015,thesolarindustryhasaddedworkersatapace12timesfasterthanthe
overalleconomy.Infact,solarjobsaccountedfor1.2percentofallnetjobsaddedintheUnited
Statesin2015.Employmentinthewindindustryhasalsobeenquiterobust.In2012,whenjobs
wereneededthemost,employmentreached80,000.Itsincedroppedsomefromthislevel,but
increasedagainfrom2013to2014,reachingatotalwindindustryemploymentofroughly73,000
workersin2014,upfrom50,000intheprioryear.
15
Thesearehighqualityjobstoo—Brookings
(2011)reportsthatthemedianwageforatypicalcleaneconomyjobapproaches$44,000,which
isabovethenationalmedianwage.

15
AWEA(2015).
0
50
100
150
200
250
300
2010 2011 2012 2013 2014 2015 2016
Source: TheSolarFoundation,NationalSolarJobsCensus2015.
Figure10:SolarRelated Employment
ThousandJobs
Projected
21
V. HighlightsoftheARRACleanEnergyInvestments
AcrosstheeightmajorcategoriesofcleanenergyinvestmentslistedinFigure1,ARRAfunding
laidthegroundworkforthetransitiontoacleanenergyeconomy.Thissectionprovidesabrief
overviewofsomeofthehighlightsfromtheARRAcleanenergyrelatedinvestments.Thereview
coversthecategoriesinFigure1,inorderfromlargesttosmallestintermsoffundingallocation.
Appendix I goes into much greater detail on each of the different categories of investment,
includingmuchmorediscussionoftheprogramimpacts.
RenewableGeneration
As noted, clean energy investment may be underprovided by the market due to unpriced
environmentalexternalitiesfromfossilfueluse, innovationrelatedmarketfailures,andcapital
marketconstraintsduetothechallengesofafinancialcrisis.ARRAsupportedrenewableenergy
generation by both expanding existing incentives and adding programs. The two primary
programsthatwere expanded were the production tax credit (PTC) and investment taxcredit
(ITC).ARRAextendedtheproductiontaxcreditforrenewableenergygenerationbythreeyears,
expandedeligibilityforthe30percentinvestmenttaxcreditforrenewableprojects,andremoved
acaponinvestmenttaxcredits,whichallowedforsmallwindprojects.
Acknowledgingthechallengesthatcertainrenewabledevelopershadintakingadvantageofthe
taxincentivesavailableatthetime,ARRAalsocreatedtwonewprogramstosupportrenewable
energygeneration.First,byprovidingloanguaranteesforrenewableenergyprojects,the1705
LoanGuaranteeProgramaddressedthedifficultyofsecuringfinancingrenewableprojectsinthe
thenprevailingmarketconditions.The1705programsupportedtheconstructionofthefirstfive
solarPVprojectsover100MWintheUnitedStatesforatotalof1,502MWinrenewableenergy
capacity.
16
Further,ARRAcreatedthe1603CashGrantprogram,providingrenewableenergyprojectswith
acashgrantequalto30percentofprojectcostsasanalternativetotakingtheinvestmenttax
credit.Astaxequitymarketstightenedduringthefinancialcrisis,the1603CashGrantprogram
allowedforthedevelopment
ofrenewableenergyprojectsbyentitiesthatlackedsufficienttax
liability to take advantage of other tax incentives. Through 2015, the 1603 Cash Grant has
provided $25 billion in awards to support 9,915 businesses in a range in technology types,
supportingatotalinstalledcapacity33.3GW,andanestimated
annualelectricitygenerationof
88,700GWh.
17
Foratypicalhouseholdusing11,000kWhperyear,thiswouldimplythatthetotal
electricitygenerationwouldpowerover8millionhomesperyear.

16
LPO(2015).
17
UST(2016a).
22
EnergyEfficiency
Inadditiontoinvestinginacleanerenergysupply,theRecoveryActalsohelpedreduceenergy
demand through a suite of investments designed to increase energy efficiency in homes,
commercial buildings, and factories. All of these programs are designed to help address
environmental externalities by reducing energy use, most of which is provided by fossil fuels.
Those programs thatreducedtheuseoffueloilalsocouldhelpreducereliance on foreign oil
productionandthusmayhelpreduceenergysecurityexternalities.
Oak Ridge National Laboratory (ORNL 2015c) projected that total ARRA energy efficiency
investmentswillsaveover400millionMMBtusofenergyoverthenextfourdecades.Assuming
thesavingshappenuniformlyoverthatperiod,thisisnearly10millionMMBtuperyear,orthe
equivalentoftheannualenergyconsumptionfor10,000homes.Inparticular,theexpansionof
the Weatherization Assistance Program (WAP) with nearly $5 billion in additional funds
supported the weatherization of over 800,000 sites over the 2009 to 2013 period, with an
estimated97millionMMBtuenergysavedandover5millionmetrictonsofcarbonsaved.
18
One
ofthemostvaluablefeaturesofallocatingfundingtowardsWAPwastheabilitytoutilizeexisting
administrativecapacityandrampupquickly.Infact,mostoftheARRAWAPfundswerespentby
theendof2013.Thissupportedjobsatatimetheyweremuchneeded;WAPsupported28,000
directandindirectjobsin2010alone.
19
CleanandEnergyEfficientTransportation
ARRA helped lay the groundwork for a transition to a cleaner and more energy efficient
transportation system with over $18 billion initially allocated towards cleaner transportation.
Thisinvestmentsupportedpublictransitwithpurchaseof12,000buses,vans,andrailvehicles
andtheconstructionorrenovationofover850transitfacilities.Italsosupportedhighspeedrail
through $8 billion allocated to 49 projects, 98 percent of which are either complete or with
constructionunderway.Further,theARRAtransportationinvestmentsenabledthedeployment
ofover4,600electricvehiclechargingstationsaroundthecountry.
CleanEnergyManufacturing
ARRAsupportedcleanenergymanufacturingthroughinvestmenttaxcreditsforawiderangeof
clean energy products such as equipment for renewable energy generating facilities, energy
storage,energyconservation,carboncaptureandsequestration,fuelcells,andtherefining and
blending of renewable fuels. The credit was awarded for 30 percent of
project costs to clean
energymanufacturerson acompetitive bases.Interest intheprogramgreatlyexceededinitial
expectations,withover500applicationsseeking$8billionincredits.Sincethiswasmuchabove
the allocated funding level, funding was increased,allowingforawards in the amount of$2.3
billionto183
domesticcleanenergymanufacturers,promotingthedevelopmentofadomestic
supplychaintosupportagrowingcleanenergyindustry.

18
EnergysavingscalculationbasedonsiteestimatesfromORNL,furtherdescribedinAppendixI.
19
ORNL(2015a).
23
OtherInitiatives
Initiativesundertheotherfivecategoriesofspendingalsomadeimportantcontributionstohelp
lay the groundwork for a longterm shift toward a clean energy economy. The Smart Grid
InvestmentProgramhelpedtosupporttheinstallationof16millionsmartmetersby2016,and
a major demonstration carbon capture and storage facility in the United States was made
possiblethroughARRAfunding.Finally,substantialARRAcleanenergyfundingwasinvestedin
longrun research and development—an investment that will continue to pay off far into the
futurebyhelpingtobringaboutmajoradvancesintransformativetechnologies.
24
VI. Conclusions
The American Recovery and Reinvestment Act represented an unprecedented clean energy
investment in United States history. The investment had a dual purpose: jumpstarting the
economytogetAmericansbacktoworkand atthesametimemakingalongruninvestmentor
down payment towards the transition to a sustainable 21
st
century economy. Thus, it was
designed to leverage existing administrative and technical capacities so that the investments
couldgetoffthegroundquickly.Thismeantbuildingonexistingtaxcredits,providingfundingon
a competitive basis to grantees with experience, leveraging private capital, and expanding
currently existing successful programs in addition to creating new ones. These clean energy
investments through ARRA provided a strong economic boost, and by the end of 2015 have
supportedroughly900,000jobyears.
Whilemostoftheseinvestments occurredquickly—roughly70percentoftheinitialallocation
wasspentbymid2013,providingaboostatthetimeitwasmostneeded—theinvestmentscan
beexpectedtoprovidebenefitslongintothefuture.Moreover,thereisastrongeconomiccase
fortheinvestmentsbasedonhelpingtoaddressmarketfailures.Nearlyalloftheinvestments
willhelpaddressenvironmentalexternalitieseitherintheshortorlongrun.Butmanyalsohelp
address innovation market failures that slow the innovation so critical to longrun economic
growth.Still others help address further market failures, such as energy securityexternalities,
information market failures, and severe capital constraints that occurred as a result of the
financialcrisis.Justastherearemultiplemarketfailures,theARRAinvestmentsinvolvedmultiple
policymechanisms.Further,suchamixofpolicymechanismsfocusingonmultipletechnologies
andmultipleinnovationopportunitiesnotonlyaddressesbarriersallalongtheinnovationvalue
chain,butalsoenablesthebesttechnologiestosucceed.
While many of the investments are in earlystage technologies that may take time to pay
dividends to society, there are already remarkable trends in clean energy markets. The
concentratedfocusonencouragingcleanenergyinnovationanddeploymentsentastrongand
consistent signal to markets, contributing to the greatest growth in renewable electricity
generation in history.
This growth occurred alongside, and helped facilitate, dramatically
decreasingcostsforcleanenergytechnologies.Infact,therearemanyplacesintheUnitedStates
whererenewablesarelessexpensive thantraditionalfossil fuel generationonalevelized cost
basis.ThesedramatictrendsinrenewableenergymarketswerenotprojectedpriortoARRA.
ThelongterminvestmentsmadeunderARRAsupportedjobsatacriticaltime,helpedaddress
marketfailures,andaresettingthestageforthetransitiontoacleanenergybased,sustainable
21
st
centuryeconomy.
25
References
Aldy,JosephE.2013.“APreliminaryAssessment of the AmericanRecovery andReinvestment
Act’sCleanEnergyPackage.”ReviewofEnvironmentalEconomicsandPolicy7(1):136155.
Allcott,Hunt,andMichael Greenstone.2012.“IsThereandEnergyEfficiencyGap?”TheJournal
ofEconomicPerspectives26(1):328.
American Wind Energy Association (AWEA). 2015. “Wind Energy Facts at a Glance.”
http://www.awea.org/Resources/Content.aspx?ItemNumber=5059
Beresteanu,Arie,andShanjunLi.2011."GasolinePrices,GovernmentSupport,andtheDemand
forHybridVehiclesintheUnitedStates."InternationalEconomicReview52(1):161182.
Brookings Institution. 2011. “Sizing the Clean Economy: A National and Regional Green Jobs
Assessment.”
Brown,StephenP.A.,andHillardG.Huntington.2013."AssessingtheUSoilsecuritypremium."
EnergyEconomics38:118127.
Carley, Sanya, and Martin Robert Hyman. 2013. "The 'Grand Experiment': An Early Review of
EnergyRelatedRecoveryActEfforts."AvailableatSSRN2316827.
ChargePoint. 2016. “ChargePoint Announces the Successful Completion of its ARRAFunding
ChargePointAmericaProgram.”
Congressional Research Service (CRS). 2013. Battery Manufacturing for Hybrid and Electric
Vehicles:PolicyIssues.”
______.2016.“ResidentialEnergyTaxCredits:OverviewandAnalysis.”
CouncilofEconomic Advisers (CEA).2010. “The Economic Impactofthe AmericanRecovery&
ReinvestmentActof2009,ThirdQuarterlyReport.”
______.2014.“The
EconomicImpactoftheAmericanRecoveryandReinvestmentActFiveYears
Later.”
Diamond, David. 2009. "The Impact of Government Incentives for HybridElectric Vehicles:
EvidencefromU.S.States."EnergyPolicy37(3):972983.
Energy Information Administration (EIA). 2009a. Annual Energy Outlook. U.S. Department of
Energy.
______. 2009b. “Residential Energy Consumption Survey.”
http://www.eia.gov/consumption/residential/data/2009/
______.2015.“LevelizedCostandLevelizedAvoidedCostofNewGenerationResourcesinthe
AnnualEnergyOutlook2015.”U.S.DepartmentofEnergy.
Environmental Protection Agency (EPA). 2016. “Global Greenhouse Gas Emissions Data.”
http://www3.epa.gov/climatechange/ghgemissions/global.html.
Federal Railroad Administration (FRA 2016a). “High Speed Rail Overview.”
https://www.fra.dot.gov/Page/P0060.
______. (FRA 2016b). “High Speed Intercity
Passenger Rail Funding by Region.”
https://www.fra.dot.gov/Page/P0554.
FederalTransitAdministration(FTA).2010.“HearingonFTAFY2011BudgetRequests.”
Fowlie, Meredith, Michael Greenstone, and Catherine Wolfram. 2015. “Do Energy Efficiency
InvestmentsDeliver? EvidencefromtheWeatherizationAssistanceProgram."WorkingPaper
no.w21331.NationalBureauofEconomicResearch.
26
Gallagher,KellySims,andErichMuehlegger.2011."GivingGreentoGetGreen?Incentivesand
ConsumerAdoptionofHybridVehicleTechnology."JournalofEnvironmentalEconomicsand
Management61(1):115.
Gillingham, Kenneth, and James Sweeney. 2012. “Barriers to Implementing Low Carbon
Technologies.”ClimateChangeEconomics3(4):125.
Gillingham,Kenneth,andKarenPalmer.2014.“BridgingtheEnergyEfficiencyGap:PolicyInsights
from Economic Theory and Empirical Evidence.” Review of Environmental Economics and
Policy8(1):1838.
Intergovernmental Panel on Climate Change (IPCC). 2005. “IPCC Special Report on Carbon
DioxideCaptureandStorage.”IPCC,CambridgeUniversityPress.
Jacobsen, Grant D., and Matthew J. Kotchen. 2013. "Are Building Codes Effective at Saving
Energy?EvidencefromResidentialBillingDatainFlorida."ReviewofEconomicsandStatistics
95.1:3449.
Jaffe, Adam B., and Robert N. Stavins. 1994. "The Energy Paradox and the Diffusion of
ConservationTechnology."ResourceandEnergyEconomics16(2):91122.
Kotchen, Matthew J. 2015. “Do Building Energy Codes Have a Lasting Effect on Energy
Consumption? New Evidence from Residential Billing Data in Florida.” Working Paper no.
w21398.NationalBureauofEconomicResearch.
LawrenceBerkeleyNationalLaboratory(LBNL).2010."PreliminaryEvaluationoftheSection1603
TreasuryGrantProgramforRenewablePowerProjectsintheUnitedStates."EnergyPolicy
38(11):68046819.
Lazard. 2014. Lazard’s Levelized Cost of Energy Analysis—Version 8.0.
https://www.lazard.com/media/1777/levelized_cost_of_energy__version_80.pdf.
Levinson,Arik.2014.“HowmuchEnergydoBuildingEnergy CodesReallySave? Evidencefrom
California.”WorkingPaperno.w20797.NationalBureauofEconomicResearch.
LoanProgramsOffice(LPO).2015.“PoweringNewMarkets:UtilityScalePhotovoltaicSolar.”U.S.
DepartmentofEnergy.
Mundaca, Luis, and Jessika Luth Richter. 2015. "Assessing ‘Green Energy Economy Stimulus
Packages: Evidence from the U.S. Programs Targeting Renewable Energy." Renewable and
SustainableEnergyReviews42:11741186.
NationalRenewableEnergyLaboratory(NREL).2009.“FinancingPublicSectorProjectswithClean
Renewable
EnergyBonds(CREBs).
______. 2012. "Preliminary Analysis of the Jobs and Economic Impacts of Renewable Energy
ProjectsSupportedbythe§1603TreasuryGrantProgram."TechnicalReport303:2753000.
Nordhaus, William D. 2009. "The Economics of an Integrated World Oil Market.” Keynote
Address,InternationalEnergyWorkshop.Venice,Italy.
Oak Ridge National Laboratory (ORNL). 1994. “Weatherization Works: Final Report of the
NationalWeatherizationEvaluation.”
______. 2015a. “Weatherization Works II Summary of Findings from the ARRA Period
EvaluationoftheU.S.DepartmentofEnergy’sWeatherizationProgram.”
______. 2015b. “National Evaluation of the State Energy Program: An Evaluation of Select
ActivitiesConductedUndertheStateEnergyProgram.”
27
______. 2015c. “National Evaluation of the Energy Efficiency and Conservation Block Grant
Program.”
OfficeofElectricityDeliveryandEnergyReliability(EDER).2012.“CaseStudyFloridaPower&
Light:SmartGridSolutionsStrengthenElectricReliabilityandCustomerServicesinFlorida.”
U.S.DepartmentofEnergy.
______SmartGrid.gov.U.S.DepartmentofEnergy.
OfficeofEnergyEfficiencyandRenewableEnergy(EERE).2013.“2012WindTechnologiesMarket
Report.”U.S.DepartmentofEnergy.
______.2015.“2014WindTechnologiesMarketReport.”U.S.DepartmentofEnergy.
______. 2016. “Advance Metering Infrastructure and Customer Systems.” U.S. Department of
Energy.
Sallee,JamesM.2011."TheSurprisingIncidenceofTaxCreditsfortheToyotaPrius."American
EconomicJournal:EconomicPolicy3(2):189219.
Schey, Stephen, Don Scoffield, and John Smart. 2012. "A First Look at the Impact of Electric
VehicleChargingontheElectricGridintheEVProject."26thElectricVehicleSymposium(EVS
26),LosAngeles.
The Solar Foundation. 2015. National
Solar Jobs Census.
http://www.thesolarfoundation.org/national/.
Summers, Lawrence. 2008. “Why America must have a fiscal stimulus.” Financial Times.
http://www.ft.com/intl/cms/s/0/3b3bd570bc7611dcbcf9
0000779fd2ac.html#axzz4173gMaQX.
U.S.Department of Energy.2015. “Revolution…Now: The FutureArrivesfor Five Clean Energy
Technologies—2015Update.”
______.2013.
“RecoveryActRecipientData.”Energy.gov.
______.2016a.RecoveryActRecipientData.
______.2016b.“ConcentratingSolarPowerProjects.”
______.2016c.“ARPAEHistory.”
U.S. Department of the Treasury (UST). 2016a. “Overview and Status Update of the 1603
Program.”
______. 2016b. “List of Awards: Section 1603‐Payments for Specified Renewable Energy
PropertyinLieuofTaxCredits.”
U.S.InternalRevenueService(IRS).2016.“IRC30DPlugInElectricDriveMotorVehicleCredit
QuarterlySales.”
28
AppendixI:DetailsoftheA RRACleanEnergyRelatedInvestments
This report has illustrated the major shifts in renewable energy markets during and after the
major ARRA clean energyrelated investments. This appendix takes a closer look at these
investments, by drawing upon the best evidence available on the impact of the funding. The
discussionwillcovereachofthemajorcategoriesofprogramsillustratedinFigure1(shownagain
here for convenience), proceeding in order of the size of the allocation to the category. The
discussion demonstrates the great diversity of funding approaches and mechanisms, ranging
fromloanguaranteesandcompetitivemeritbasedcostsharedgrantstosupportforleadership
andcapacitybuilding.
RenewableEnergyInvestments
The Recovery Act supported renewable energy generation working through several programs.
Twoofthemajorprogramsweretheproductiontaxcredit(PTC)andinvestmenttaxcredit(ITC),
both of which were expanded or extended by ARRA. In addition, ARRA included two new
programs,the1603CashGrantprogramandthe1705
LoanGuaranteeProgram.
20
Finally,ARRA
alsoprovidedsupportforrenewableenergythroughincreasedfundingforbondsforrenewable
energy,calledCleanRenewableEnergyBonds(CREBs).
Combined, these programs supported the dramatic increase in renewable energy generation
observed over the past decade, and likely also contributed substantially towards spurring
technologicalinnovation.Forexample,over
thepastsevenyears,windinstallationshaveseen
large increases in hub heights and larger rotors, which allows for viable wind project
development in new regions where previously the wind resources could not be successfully

20
The1705programmodifiedthepreexisting1703programto includeconventionalrenewableenergyprojects,
transmissionprojects,andbiofuelsprojects.
29
accessed. While it is impossible to attribute technological innovation to any single cause,
Mundaca and Richter (2015) examine clean energy patents over theperiod2009 to2012and
provideevidencesuggestingthatcleanenergy innovationwassubstantiallyencouragedbythe
ARRAprograms.
ProductionTaxCreditandInvestmentTaxCredit
Prior to ARRA, federal support the renewable energy project investment and development
occurredprimarilythroughthePTC,mostoftenusedbywinddevelopers,andtheITC,mostoften
usedforsolarenergydevelopers.Bothtypesofpoliciescanhelpaddressenvironmentalmarket
failuresbyencouragingrenewableenergydevelopmentthatoffsetsfossilfuelgeneration.Atthe
same time, they also provide incentives for private investment in innovation in these
technologies by raising the expected rate of return on these technologies. They were
exceptionally useful as part of the ARRA because they were alreadywellestablished and thus
couldquicklybeputtoworktoachieverapidstimuluseffects.
Prior to the Recovery Act, production tax credits offered performancebased incentives for
eligible renewable energy generation technologies (excluding solar) through rebates of
$0.023/kWhforwind,geothermal,andclosedloopbiomassand$0.012/kWhforothereligible
technologies(in 2015dollars).ITCssupportedprojectinvestmentby offering tax credits of 30
percentofprojectcostsforsolar,fuelcells,andsmall windand10percentforgeothermal,micro
turbines,andCHPrenewableenergyprojects.
The Recovery Act expanded the impact of these credits through several mechanisms. First, it
extendedtheexpirationdateofthePTC.Specifically,ARRAextendedthePTCforwindfacilities
toDecember31, 2012 andfor other technologies to December 31, 2013 (theITC was already
givenan8yearextensionin2008,sothekeycontributionofARRAonthesolarmarketwasthe
1603CashGrantprogram).Thisextensionwascriticallyimportantbecausepreviouslytherewas
anannualboomandbustcycleinwindandotherrenewablesdevelopmentduetouncertainty
aboutwhetherthesecreditswouldberenewed.
21
SubsequentextensionsfollowingARRAhave
allowedITC/PTCstocontinuetosupportrenewableenergyprojectinvestmentandgeneration
through2016andbeyond.
22
Second,ARRAincreasedITCsupportforsmallwindprojectsbyremovingapreviouslyimposed
capofa$4,000maximumamountofcreditsclaimedforwindlessthan100kWincapacity.Third,
itallowedfacilitieseligibleforthePTCto electtoclaimtheITCinstead.Thisprovidedadditional
optionalityfor
largewindprojectspreviouslyonlyeligibleforaPTC.

21
MundacaandRichter(2015).
22
Mostrecently,PTC/ITCswereagain extendedinDecember2015.ITCsareavailableforsolartechnologiesexcluding
hybridsolar lighting through 2019at30 percent, ratcheting downto10 percent by 2022. ITCs forlarge windare
available at30 percent through 2016, phasing out by 2020; ITCs for geothermalcontinue
at 10percent; ITCs for
othereligibletechnologiesthrough2016.PTCsforwindwereextendedthrough2016withaphasedownover2017
‐2019;forothereligibletechnologiesPTCsexpireafter2016.
30
1603CashGrantProgram
The1603programofferedcashgrantstorenewableenergyprojectinvestmentsforanamount
ofupto30percentofaproject’scapitalcosts.
23
Projectdevelopershadachoicebetweentaking
thiscashgrantoroneofthetaxcredits,suchastheinvestmenttaxcredit.Eligibleprojectsfor
the cash grant were required to commence construction by December 31, 2011, or incur 5
percentofprojectcostsbytheendof2011.
24
Whiletheinvestmenttaxcreditwasalsoavailable
forupto30 percent of a project’s capital costs, there wereseveraladvantagesto providing a
cashgrant.
Thefirstisthatsomerenewableenergydevelopersmaynothavehadasufficienttaxliabilityto
claimthetaxcredits.Manyrenewableenergy developersaresmallcompanies,withalimitedtax
burden.InordertotaketheITC,suchcompanieswouldhavetoenterintoafinancialagreement
withapartnerthatwouldprovidetaxequityinreturnfortheclaimedtaxcredits.Duringnormal
economictimes,thismaybearelativelyeasytransaction,butduringthefinancialcrisis,itbecame
difficulttofindsuchpartners.
25
A second advantage is that even if it may have been possible to find a partner, in a time of
financialturmoilthereisanincreasedriskthatthepartnerwouldbackout.Withcertaintyofthe
grantfunding,renewableenergyproject developers found it easier toraiseprojectdebt. Aldy
(2013) also points out that financiers were hesitant to support longterm investment projects
when there was uncertainty regarding the future of the PTC and ITC. The 1603 Cash Grant
programprovidedthecashgrantsquickly,thusreducinguncertainty.
Apreliminaryanalysis of1603’simpact by LBNL(2010)findsadditionalbenefits to developers
from the cash grant program rather than a tax credit program. These benefits accrue from
reducingthedependenceonscarceorcostlythirdpartytax equitytomonetizeaproject’sfederal
taxbenefits.Financingislessexpensiveusingprojectleveltermdebt andcarryingdepreciation
deductionsforwardintimeuntiltheycanbeabsorbedbytheproject(i.e.,“selfsheltering”).The
report estimates the value of selfsheltering at around 8 percent of installed project costs for
winddevelopers.LBNLalsoestimatesthatthe‘facevalue’ofthe cashgrantishigherthanafter
taxeconomic valueofthe ITC/PTC,estimatedaround2.2 percentof installedproject costsfor
windpowerprojects.
26
LBNL(2010)furtherexaminesearly1603recipientsanddoesnotfindanyevidenceofinefficient
project design, which would lead to poor performance. Inefficient project design could be a
concern for a cash grant program because by its structure, it does not specifically incentivize

23
UST(2016a).
24
Whenthe 1603 program was first established, the project constructionstart date required was December31,
2010.In 2010this deadlinewas extended toDecember31,2011.1603 projectswith aconstructionstart datein
2011arealsoincludedintheCEAcalculationsforinthisreport.
25
NRELreportsthenumberoftaxequityinvestorswillingtomakeinvestmentsdecreasedfrom20toaround5during
20082009.
26
LBNL(2010).
31
efficient project construction. However, nearly all renewable energy projects sell electricity
through a power purchase agreement (PPA), which involves safeguards to attempt to ensure
efficientprojectconstruction.
TheTreasuryDepartmentreportsthat$25billioninfundswereusedtosupportthedevelopment
of renewable energy facilities through the 1603 Cash Grant program as of January 1, 2016.
27
Includingprivate,regional,andstate,andfederalinvestments,$90billionhasbeeninvestedin
1603projects.
28
Fundshavesupported9,915businessestodevelopprojectsofthetechnology
mixbelow,showingthatthedominantfirmtechnologytypewasforsolarelectricity,makingup
80 percent of the firm awards associated with 76,000 solar projects.
29
The next largest firm
technology,wind,comprised9percentofawardedfirms;solarthermalconstitutes5percentof
awardedfirms,andtheremaining10percentcomefromavarietyofothersources.
30
Figure A1 and A2 show that while solar firms dominate the count of awardees, wind projects
receivedmoreofthefundinginaggregate.Theaverageawardforawindprojectismuchlarger
thantheaverageawardforasolarproject.
FigureA3illustrates thegenerationcapacityand
estimatedgeneration from 1603 CashGrant
fundedprojectsthrough2015.WindandnonresidentialsolarPVarethelargestcategories:1603
grants funded projects that generated 56,200 GWh from wind and 14,300 GWh from non
residentialsolarPV asofDecember31,2015.Bytheendof2015,theprogramfunded
a total
installedcapacity33.3GW, translatingtoanestimatedannualelectricitygeneration of 88,700
GWh.FigureA4belowshowstheestimatedgenerationfromsupportedprojectsbyprojecttype.

27
UST(2016b).
28
UST(2016a).
29
MundacaandRichter(2015).
30
The Treasury Department’s status update shows 104,211 projects as of December 31, 2015; the awards
spreadsheetkeeps track offunding byfirm nameand notproject. Allthegraphs below are based on numberof
firms,andnotnumberofprojects.
Solar,34%
Biomass,4%
SolarThermal,3%
Geothermal,3%
Hydropower,2%
Other,2%
Wind,52%
FigureA1:Distributionof1603AwardedFundingby
TechnologyType
Note:"Other" includestheFuelCells,LandfillGas,Marine,Microturbine,
CombinedHeatandPower,andTrashFacility.
Source:U.S.TreasuryDepartment.
Solar,80%
Hydropower,
1%
Other,2%
SolarThermal,5%
Geothermal,1%
Biomass,2%
Wind,9%
FigureA2:1603Awardees byTechnologyType
Note:"Other" includestheFuelCells,LandfillGas,Marine,Microturbine,
CombinedHeatandPower,andTrashFacility.
Source:U.S.TreasuryDepartment.
32
Thecost ofsupportingthisrenewablegenerationwas$0.23/kWhforwindand$0.59/kWhfor
nonresidentialsolarPV(in2015dollars).
31
FigureA5andFigureA6belowillustratethetiminganddestinationofthe1603project funding.
Aswindwasthemostcompetitive technology,themajorityoftheprojectfundswenttowards
windenergy,followedby solarPV,withnearly allofthe solarPVcomingfromnonresidential
solarPV.
NREL (2012) estimates gross jobs as a result of the 1603 program and finds that through
November10,2011,thesolarPVandlargewindprojectsthatreceived1603fundssupportedan
average of 52,000 to 75,000 direct and indirect jobs over the 2009 to 2011 period from
construction and installationrelated expenditures. These estimates reflect the impact of $9
billion in 1603 funds supporting 13.5 GW in electric generating capacity, which is over one
percentofthe2011totalsummergeneratingcapacity.Further,thereportfindsthe1603projects
led to up to a $26 to $44 billion increase in total economic output over the period analyzed.
Takingtheseestimatesfortheperiod2009to 2011atfacevalue,andscalingtheestimatesto

31
ThiscalculationdividestotalfederalprojectfundingfortherespectivetechnologypertheTreasuryAwardListby
totalenergyforprojectspertheTreasury1603Statusupdate(UST2016a).
0
5
10
15
20
25
Wind NonResid.
Solar
Biomass Other Geothermal Resid.
Solar
FigureA3:Generation Capacityfrom1603ProjectsbyType
ThousandMegawatts
Note:FigurerepresentscapacityasofDecember31,2015.
Source:Department oftheTreasury.
0
10
20
30
40
50
60
Wind NonResid.
Solar
Biomass Other Geothermal Resid.
Solar
FigureA4:EstimatedGeneration from1603
ProjectsbyType
ThousandGWh
Note:EstimatesasofDecember31,2015.
Source:Department oftheTreasury.
0
5
10
15
20
25
30
2009 2010 2011 2012 2013 2014 2015
FigureA5:Cumulative 1603ProjectFunding
BillionDollars
Source: U.S.TreasuryDepartment.
0
2
4
6
8
10
12
14
2009201020112012201320142015
Biomass
SolarThermal
Wind
Geothermal
Solar
Hydropower
CombinedHeat&Power
Other
FigureA6:Cumulative1603FundingbyTechnologyType
BillionDollars
Source:U.S.TreasuryDepartment.
33
thecurrenttotalofgenerationcapacitysupportedbytheprogram(33.3GW),suggestsatleasta
doublingoftheseemploymentandeconomicoutputestimates.
1705LoanGuaranteeProgram
The 1705 program provided loan guarantees for renewable energy generating facilities that
began construction prior to September 30, 2011.
32
While a previously existing loan guarantee
program (1703) offered loans for innovative commercial energy programs, 1705 modified the
1703 program to include conventional renewable energy sources, transmission, and biofuel
projects.Anexpandedeligibilityofloanguaranteessupportedrenewableenergyprojectfinance
during the time of the financial crisis, when there were severe credit constraints. The loan
guaranteesturnedouttobeparticularlyusefulforlargeutilityscalesolarPVprojects,whichwere
having great difficulty obtaining financing due to severe credit constraints from the financial
crisis.Asofyearend2015,anestimated$1.47billioninAR RAfundshavebeendirectedtoward
the1705LoanGuaranteeProgram.
33
Over$4.6billioninloanguaranteesfromthe1705programsupportedtheconstructionofthe
first five solar electricity generation projects over 100 MW in the United States. These five
projects,financedbetween2010and2012,ledtoatotalof1502MWofsolarelectricitycapacity
either operating or under construction. The 1705 program also supported Brookfield
Renewable’s Granite Reliable Wind Farm, a 99 MW wind power generation project located in
New Hampshire, with a $169 million partial loan guarantee. The project was one of the first
onshorewindprojectstousea3MWturbineforitselectricalgenerationtechnology,leadingto
costreductionsonapermegawattbasis.
Inaddition,the1705program hassupportedfiveconcentratingsolarpowerprojectswithatotal
generationcapacityof1,252MWandanexpected3,544GWhofannual generation.Onesuch
projectincludesCrescentDunes,a110MWfacilitythatconcentratessolarenergytoheatmolten
saltanduponcompletionwillbethelargemoltensaltpowertowerintheworld,projectedto
mitigate279,000metrictonsofCO
2
annually.The1705programalsosupportedNextEraEnergy’s
Genesis project with a partial loan guarantee through the Financial Institution Partnership
Program (FIPP) to support a 250 MW facility using parabolic trough technology preventing
322,000metrictonsofannualCO
2
emissions.
34
Noinvestmentportfolioisperfect,buttheRecoveryAct’scleanenergyprogramshavehadsome
prettyremarkableresults.Thestandardruleofthumbinventurecapitalisthatthreeoutoffour
startups fail. The DOE Loan Programs Office has provided $34.2 billion in loans and loan
guarantees, and
companies have defaulted on only $780 million‐‐a loss rate of only
approximately2percent.DOEalsohascollected$810millionin interestpayments,puttingthe
program$30millionintheblack.

32
LPO(2015).
33
DOE(2016a).
34
DOE(2016b).
34
CleanRenewableEnergyBonds
Clean Renewable Energy Bonds (CREBS) provide another financing mechanism for renewable
energyprojects.Stateandlocalgovernments,cooperativeelectriccompanies,cleanrenewable
energybondlenders,andIndiantribalgovernmentscanissuethesebonds,whichareatypeof
taxcreditbond.Ratherthantheissuerpayinginterest,thefederalgovernmentpaystheinterest
intheformoftaxcredits,sotheissuingentitycanborrowatlowereffectiveinterestrates,and
sometimesevenaslowaszerointerestrates.
35
Thus,CREBSsupportsinvestmentinrenewable
energybyloweringthecostofdebt.
CREBSwere first established by the EnergyPolicyAct of 2005andreceivedadditionalfunding
throughtheTaxReliefandHealthCareActof2006.TheEnergyImprovementandExtensionAct
of 2008 fur ther allocated $800 million in funds for CREBS, accompanied by new rules and
requirements for CREBS as compared to the program prior to 2008 (“New CREBs”). ARRA
increasedthefundingavailableforNewCREBSby$1.6billion,foratotalNewCREBallocationof
$2.4billion.
In October of 2009 the U.S. Treasury Department announced $2.2billion in allocation of new
CREBs to public power providers, cooperative electric companies, and government bodies
supporting805projectsacrossthenation.Oftheseallocations,publicprovidersreceived$800
million,electriccooperativesreceived$609million,andgovernmentbodiesincludingcitiesand
schooldistrictsreceivedtheremaining$800million.Theremainingunallocated$190millionin
funds were distributed through a solicitation in September 2010 explicitly for electric
cooperativessuchthatCREBvolumesacrossthethreeentitytypeswouldequitable.Bondsare
intendedtobeusedwithin3yearsofnoticeofallocation,andanyunusedbondsafterthistime
framearerelinquishedandmaybeofferedinfuturesolicitations.
In2009,CREBsallocationssupported22renewableenergyprojectsfrommunicipalities,31from
electric cooperatives, and 736 from governments for 711 solar, 47 wind, 15 biomass, 26
hydropower,and3geothermal renewable projects. Governmentsmadegreateruse of CREBs,
butfor
smallerprojects.Justasmuchfundingwasallocatedtomunicipalutilitiesasgovernments,
butmunicipalutilitiesonlywereawarded22largerprojects,farfewerthanthe736government
projects.
San Diego County in California provides one example of successful application of the CREBS
programtosupportpublicsectordevelopmentof
renewableenergyprojects.AnonprofitinSan
Diego led a collaborative effort engaging university and high school students to develop a
financial model for CREBs, focusing on onsite solar projects and application for smaller
allocations of $1 million or less. $154 million in CREBS funding was allocated for 192 projects

35
The New CREBs included programchangesthatreducedannual taxcredit rateallowed by 70percent.If a risk
profileofagivenprojectledthemarkettorequireahigherratethanthisreducedrate,theissuerofthebondwould
likelycompensatetheinvestorwithsupplementalinterestpayments.
Anadditionalprogramchangeremovedthe
requirement that the issuermake equal annualpayments over the term; in the new program the issuer has the
flexibilitytorepaythefullprincipalatthematuritydate(NREL2009).
35
acrossmunicipalities,schooldistricts,universities,andawaterdistrictintheregion,supporting
20MWinnewsolarpower.
InvestmentsinEnergyEfficiency
Over $27 billion funds have been spent on a hostof projects and grant programs to improve
buildingenergyefficiency,forresearchanddemonstrationofenergy efficienttechnologies,and
for support for development of forwardlooking state energy plans.
36
The larger programs
included the Weatherization Assistance Program (WAP), State Energy Programs (SEP), Energy
Efficiency & Conservation Block Grant Program (EECBGP), and Residential Energy and Energy
EfficiencyPropertyTaxCredit.Othersmallerprogramssupportedenergyefficiencytechnology
research and demonstration, smart appliances, green investments in assisted living, public
housing,andfederalbuildings.
All of these programs are designed to help address environmental externalities by reducing
inefficient use of energy, which also reduces energy generationrelated emissions. Those
programsthatreducedtheuseoffueloilalsocouldhelpreducerelianceonforeignoilproduction
and thus may help reduce energy security externalities. At the same time, some of these
programshaveaspecificemphasisonlowincomehouseholdsandthushavetheadditionalaim
ofkeepingthemostvulnerableAmericanswarmandhealthy.WAPisamongthemostnotablein
thiscategory.
WeatherizationAssistanceProgram
WAP started in 1976 and since then has provided gran ts to states for energy efficiency
improvementstolowincomefamilies.Specifically,fundsforWAParedistributedtothestates
asafunctionoftheclimateinthestateandtheaverageenergybillsoflowincomehouseholds.
Statesthendistributethefundstosubgrantees,typicallycommunityactiongroupsorothernon
profitentities that identifyeligiblehouseholds.Eligibleparticipatinghouseholds receivea free
energy audits and home retrofits that can include insulation, window replacement, furnace
replacement,andairinfiltrationreduction.
While WAP had been providing energy efficiency improvements prior to
the Act, ARRA
dramatically increased WAP funding from$230 million per program year to roughly $5 billion
overthreeyears.
37
OnereasonforthisdramaticrampupinfundingisthatWAPwasrelatively
easier to scale up than some other programs, due to the existing administrative capacity and
industryexperiencewiththeprogram.ThebroadgoalsforthisincreaseinfundingtoWAPwere:
1)jobcreationata
timewhenitwasmuchneeded,2)energyconservationandsecurity,and3)
reliefforhouseholdburdenedwithhighenergycosts.

36
Thisamountexceedthe2010estimateof$19.9billiondirectedtowardenergyefficiency.Thedifferencestems
fromhigheroutlaysforresidentialenergyefficiencycreditsthanforecastedin2010.
37
Along with the funding increase, the WAP income eligibility requirement was adjusted upwards to within 200
percent of the poverty line from 150 percent, and the average weatherization cost per site was increased from
$2,500to$6,500.
36
TheWAPprogramwasabletospendmoneyrapidlyandincreaseeconomicactivityduringthe
yearsofthegreatestlabormarketslack.TheDOE’srecordofRecoveryActfundingshows$4.9
billionspentbyMay2013.Figure A7showsthedistributionofARRAWAPexpenditurestracked
bytheDOEPAGEdatabase
overtime,underscoringhowthisfundingnotonlywasaninvestment
inthefuture,butalso wasverywelltargetedto provide stimulus andjobs in the wake of the
financialcrisis.
38
Figure A8 shows the number of ARRA units and nonARRA units (i.e., formula units funded
throughthestandardprocess)weatherizedfromQ12008toQ12014.Thespendingseeninthe
previousfigureleddirectlytohundredsofthousandsofweatherizedunitsfromQ32009toQ3
2012, exactly the time when a stimulus was most needed. In fact, when combining the ARRA
fundsandtheformulafunds,overonemillionunitswereweatherizedbetweenApril2009and
September2012.

38
TheWAPexpendituresintheDOEPAGEdatabasedonotsumuptothetotalARRAexpendituresforWAPasthe
databaseonlytracksasubsetoftotalprogramspending.However,nearlyallARRAWAPspendingwasbeforemid
2013,justasintheDOEPAGEdatabase.
0
50
100
150
200
250
300
2009 2010 2011 2012 2013 2014
FigureA7:ARRAWAPExpenditures
MillionDollars
Source: DepartmentofEnergy,PAGEDatabase.
37
OakRidgeNationalLaboratory(2015a)usesageneralequilibriummodeltoestimatesthejobs
impactoftherapidexpansionofWAPthatwassupportedbyARRA.Forexample,theyestimate
thatabout28,000jobsweresupportedinprogramyear2010duetotheARRAWAPfunds.Figure
A9 plots the recorded number of jobs per quarter supported by the site weatherization that
occurredinthatquarter.Thejobsdepictedreflectdirectemploymentperquarter,anddonot
include indirect impact on jobs through expenditures on each site on energy efficient
technologies.
ORNL estimates that for Program Year 2010 (PY 2010), weatherized sites observe an average
savingsfromheatingandelectricityusedasrecordedbyutilitybillsof26.6MMBtuinfirstyear
0
10
20
30
40
50
60
70
80
90
100
2008 2009 2010 2011 2012 2013 2014
Source:OakRidgeNationalLaboratory.
FigureA8:ARRA WeatherizedUnits
ThousandUnits
ARRA Units
FormulaUnits
0
3
6
9
12
15
18
2009 2010 2011 2012 2013
FigureA9:EstimatedJobsSupportedbyWAP
ThousandJobs
Source:OakRidgeNationalLaboratory.
38
forsitebuilthomes,16.4MMBtuformobilehomes,and15.9MMBtuforlargefamilyhomes.
39
Theseestimatesarewithintherangeofthoseinotherstudies.Forexample,Fowlieetal.(2015)
estimates a savings of 17.2 MMBtu per weatherized unit in Michigan and a previous ORNL
evaluationofWAPProgramYear1989findsasavingsof17.6MMBtuperyear.
40

ForProgramYear2010,65percentofhomesweresitebuilt,15percentweremobilefamily,and
20percentwerelargemultifamily.Applyingtheaboveestimatesofsavingsperunittotheover
800,000sitesORNLnotesareweatherizedwithARRAfundsoverthe2010to2013periodimplies
thattherewouldbeanestimated97millionMMBtusavedasaresultoftheARRAfundsinthe
WAP.
41
Taking the average of the ORNL savings estimates across site types gives an average
savingsof 19.63MMBtupersiteperyear.UsingtheEIA (2009b) summary dataon household
energyconsumptionandexpenditurestoestimateaverageexpendituresperMMBtu,theenergy
savingsfromWAPleadtoanaveragesavingsof$444peryearperweatherizedsite.
42
ORNL calculates equivalent emissions reductions from PY 2010 energy savings using state
specific emissions factors based on statespecific energy portfolios. With this approach they
estimateareductionof7,382,000metrictonsofcarbon.IfattributingtheMMBtutoareduction
inanenergysourcewithanemissionsrateofnaturalgas,thenthisenergysavingscorresponds
toareductioninCO2by403,482fornewlyweatherizedsitesinPY2010.Applyingtheseestimates
toallweatherizedsitesovertheperiod2010to2015impliesareductionofover5millionmetric
tonsofcarbon.Thiscalculationislikelyanunderestimatetotheextentthatenergysavingsoffset
energyusefrommorecarbonintensivesources(e.g.,electricityusefromcoalfiredgeneration).
StateEnergyProgram(SEP)
The State Energy Program is managed by the DOE’s Weatherization and Intergovernmental
Programs Office (WIPO) and was created by Congress in 1996 by merging the State Energy
ConservationProgram(SECP) andtheInstitutionalConservationProgram(ICP).Thepurposeof
SEP is to support states’ development of strategic energy plans by providing leadership and
technicalassistance.Suchassistanceoftenincludesoutreachactivities,technologydeployment,
and access to new partnerships and resources. This assistance may help address information
marketfailuresandassiststatesinaddressingenvironmentalexternalities.

39
Energysavingsarecalculatedbycollectedmonthlyutilitybillsfromnaturalgasand/orelectricutilitiesforthe12
months pre and postweatherization. Heating sources vary including natural gas and electricity; energy savings
reflectcombinedheatandelectricitysavings.
40
Fowlieatal.2015;ORNL1994.NoteFowlieetal.showthatWAPhasasimilareffectivenessinsavingenergy,but
differ substantially fromORNL in estimating a lower costeffectiveness and benefitcostratio forWAP. The most
significantdifferencesrelatetowhethercertaincategoriesofbenefitsareincluded
inthecalculation.
41
This calculation estimates total savings from 2010 to 2015. For example, for a site weatherized in 2010, this
estimateincludes6yearsofenergysavingsandassumesthatenergysavingsinfirstyearpersisttofutureyears.It
alsoassumesamixofsinglefamily,mobileunit,andmultifamily
unitin2010persistsinfutureyears.
42
EIA (2009b) reports $2,024 average energy expenditures per household, with average usage per year of 89.6
MMBtuperhousehold,foranaverageof$23/MMBtu.ThisestimateisusedtomapMMBtusavingsfromWAPto
householdsavings.
39
For the Program Year 2008 prior to ARRA, SEP funding was $33 million. ARRA increased SEP
fundingto$3.1billionovertheperiod2009to2011,withnearlyallspentbyMay2013.Onethird
ofSEPfundswereallocatedequallyacrossstatesandterritories,onethirdbasedonpopulation,
and
onethirdbasedonenergyconsumption.
SEPsupportto develop strategic energyplans included buildingcodesandstandards, building
retrofits,loans/grants/incentivesforenergyefficiencyandrenewableprojectsacrossanarrayof
sectors,andthedevelopmentofrenewableenergycapacity.Fundingisalsoavailabletosupport
preparations for emergencies and natural
disasters. SEP encourages sustainable energy
technologiesacrossabroadhost ofsectorsoftheeconomy fromtheenergyindustrytoschools
andhospitals.
ORNL performed an evaluation of SEP for Program Year 2009. The evaluation separates the
impactoftheSEPintofourkeycategories:buildingretrofits;buildingcodesandstanda rds;loans,
grants,andincentives;andrenewableenergymarketdevelopment.Becauseofthemultitudeof
programtypescoveredunderSEP,theORNLmethodologyinvolvedanassessmentbasedona
random sample of program types from fun ding recipients. This approach is inherently
approximate,butitdoesacknowledgedifferentARRAprogramcoverageratesacrossdifferent
activities.TableA1presentstheORNL(2015b)assessmentsoflifetimeofsavingsduetothe2009
SEPfundedprojects.Theresultshighlightsignificantenergysavings,especiallyby residentialand
public institutional projects. They also highlight the substantial industrial sector renewable
energygenerationsupportedbySEP.
EnergyEfficiency&ConservationBlockGrantProgram(EECBGP)
This grant program was developed as a onetime program to jumpstart local investments in
energy efficiency and conservation. ARRA appropriated $2.8 billion for individual program
formula grants to states, U.S. territories, Indian tribes, counties, and cities to support the
development, implementation, and management of energy efficiency programs, almost the
entiretyofwhichhadbeenspentbythe endof2015.
43
Another$400millionwasallocatedfor
distributionthroughcompetitive grants.Thegrantswereawardedforprojectsthatreducedfossil
emissions,reducedenergyuse,improvedenergyefficiencyintransportation,buildingandother

43
DOE(2016a).
40
sectors,andcreatedorretainedjobs.Giventhelimiteddurationofthisgrantprogramandthe
singlefundingcycle,projectswithneartermexecutiontimelinesreceivedfundingpreference.
Through the program, $2.6 billion in grants were allocated to 2,187 cities, counties, states,
territories,andIndiantribesacross14differentgrantcategories.Thelargestcategorybyfunding
isenergyefficiencyretrofits(38.8percentoffunding),followedbyfinancialincentiveprograms
(17.9percent),andbuildingsandfacilities(9.7percent).
AnORNL(2015b)assessmentofthe programestimatesthat409millionMMBtuwillbereduced
over the 2009 to 2050 period as a result of all of the combined programs included in EECBG.
Assuming the savings happen uniformly over that period, this is nearly 10 million MMBtu per
year.Basedontheaveragenaturalgasusageperhouseholdreportedby theEIA,thesesavings
wouldbemorethantheentireannualenergyconsumptionof10,000homes.TableA2showsthe
ORNL estimated lifetime energy savings from the program by category of grant. Financial
incentivesareestimatedtoprovidethelargestenergysavings,atnearly58percentofthetotal
energysavings,sincetheyleverageprivatefundstoamuchgreaterdegreethantheothergrant
categories.
Thereisalsoevidencefromtheacademicliteraturethatisrelevanttounderstandingtheeffects
oftheblockgrantprogram,aswellastheSEP.Forexample,110oftheblockgrantsareforthe
development of building codes. There is evidence that building codes are effective in saving
energy. Jacobsen and Kotchen (2013) study the impact of building codes and estimate that
building codes in Gainesville, FL led to 4 percent decrease in electricity consumption and 6
percentdecreasein naturalgasconsumption.Theseestimatedsavingsareclosetotheexante
predictionsoftheregulation’simpact.
InupdatedanalysisinGainesvilleusing11yearsofbilling
data,Kotchen(2015)findsthatwhileelectricityconsumptionreductionsfromthebuildingcodes
donotpersist,reductionsinnaturalgasconsumptionarepersistentinthelongterm.
44


44
Levinson (2014) finds that building codes in California do not have impacts on electricity consumption when
controllingforbuildingcharacteristicssuchassize,location,vintage,andcharacteristicsoftheresidence.However,
incontrasttoLevinson(2014),Kotchen(2015)continuestofindpersistentenergysavings.
41
ResidentialCleanEnergyandEnergyEfficiencyTaxCredits
Over $10 billion has gone towards ARRA residential tax credits to encourage homeowner
investments in renewable and efficient energy technologyand property. These residential tax
creditsbuiltuponprevioustaxcreditsforenergyefficientinvestmentsalreadyinexistenceatthe
time. ARRA removed or raised some of the previously imposed maximum credit amounts,
allowingforataxcreditequalto30percentofthecostofaqualifiedinvestment.
45

The expansion of these credits through ARRA contributed to the increase in the value of the
credits by over three times over the 2006 to 2010 period. This larger credit value was
accompaniedbyanincreaseinthenumberofclaimsovertheperiod,withaverageannual claims
in2009and2010increasingtooverthreetimesthenumberofclaimsin2008.
With the support of these credits, homeowners have invested in insulation, energy efficient
exteriorwindows,andenergyefficientheatingandairconditioningsystems.TheuseofEnergy
Star criteria to establish eligibility of certain products helped to streamline rebate claims for
qualifying improvements. Tax rebates were also provided for the residential purchase of
renewableenergysuchasgeothermalpumps,solarheating,solarelectricity,fuelcells,andwind
turbines.Overtheperiod2009to2010,over13millionresidentialenergycreditswereclaimed,
withanaveragecreditamountof$863to$868.
46
TransitInvestments
AsdiscussedinsectionII,cleanenergyprogramscanhelpaddressmarketfailuresintheclean
energygeneration,suchastheexistenceofunpricednegativeexternalitiesfromthecombustion
offossilfuels.Cleanertransportationisanotherwaytoreducetheemissionsfromfossilfueluse.
Thefailuretointernalizetheexternalitiesmayalsobiastransportationsystemsawayfrommass
transitandothercleanertransportationoptions.Accordingly,ARRAallocationsincludedfunding
for investments in rail and transit. By laying the groundwork for a shift towards a cleaner
transportationsystem,theseinvestmentscanhelpreduceenvironmentalexternalities,aswell
asothermarketfailures,suchascongestionandaccidentexternalities.
HighSpeedRailandIntercityRailCapitalGrants
ARRAprovidedcloseto$8billioninfundstostatesacrossthecountrytodevelopanationwide
highspeedintercitypassengerrailservice.Thefundshavesupportedplanningandconstruction
projects to develop largescale highspeed rail corridors across the country, targeted through
investmentsinfivekeyregions.Theprogramreceivedanadditional$2billionin2011foratotal
of$10billiontosupporthighspeedrail.
47
Inadditiontoprovidingamoreenergyefficientmeans
oftravel,therailinvestmentisintendedtocreateandsavejobsintracklaying,manufacturing,

45
ForSection1121rebates,ARRA raisedmaximumlimitto$1,500,and forSection1122rebates,ARRA removed
some of the previous caps. A credit of 30 percent of projectcosts up to $1,500is permitted forenergyefficient
installationswereavailable for2009and2010;thecreditrateis10
percentforinstallationsin2011to2016with
$500maximum.Creditsforrenewableenergyinvestmentsremainedat30percent.
46
CRS(2016).
47
FRA(2016a).
42
planningandengineering,andrailmaintenanceandoperation.Toensurethatthisinvestment
leads to domestic jobs, both domestic and foreign rail companies have agreed to establish or
expandtheirbasetotheUnitedStatesifselectedtosupplytheraillines.
TheFederalRailroadAdministration(FRA)reportsthattodate,$5.3billionhasbeenmobilized
towards49constructionprojectsthatareunderwayorcompleted(24completed)in25states
andtheDistrictofColumbia.Additionally,oneconstructionprojectofapproximately$3.4million
isexpectedto be underway by mid2016.Approximately98percentofobligatedconstruction
projectshavebeencompletedorarecurrentlyunderway.Withalongtermvisionofconnecting
80percentofAmericanstohighspeedrailin25years,ARRAfundsaimedtobalancelongterm
objectives of developing the infrastructure needed for a modern, lowcarbon transportation
system,whileprovidingtimelyneartermjobcreationanddemandfordomesticmanufacturing
supply.
Projectssupportedbythesefundsincludea$4milliongranttodevelopanewtransitcenterin
San Francisco, with construction currently underway andexpectedcompletion in 2017. Funds
were also awarded to support California’s high speed trail system with initial construction
underwayandprojectcompletionin2029.ARRAgrantsallowedforthedevelopmentof30miles
ofnewtrackfortheexpansionoftheAmtrakDowneasterservicetoFreeportandBrunswick,ME,
througha$38milliongrantprovidingservicetothesecitiesstartingin November2012.A$15
milliongrantsupportednewtracksforfreighttrainaccesswithoutdisruptingpassengerservice
inthePortofVancouverinWashington,completedinMarch2015.
48
FormulaTransitGrants
The Federal Transit Administration has awarded $8.78 billion through 1,072 grants tosupport
transitcapitalassistanceinurban,rural,andtribalareasaswellasforinvestmentsingreenhouse
gasandenergyreduction,where$6billionofwhichwasdirectedtowardtransit capitalassistance
inurbanareas.InSeptember2010theFTAreportedhavingawardedallofitsARRAfunding,with
16.4percentoffundsnotyetdisbursed.
ARRAfundinghasalsoprovidedover12,000buses,vans,andrail vehicles,andtheconstruction
or renovation of more than 850 transit facilities, and over $620 million in preventative
maintenance.
49

GridModernizationInvestments
Transitioning to a clean energy electric portfolio involves sourcing electricity from increasing
amounts of intermittent generating sources. Key renewable resources such as wind and solar
requireintegratingintermittentresourceprofilesontotheelectricgrid.Inthiscontext,bothon
going maintenance and modernization of U.S. electric grids with cuttingedge technology
becomes important for keeping up with a changing electricity portfolio. For example, modern
electricgridswithadvancedmeteringinfrastructure(AMI),suchassmartmeters,pavetheway

48
FRA(2016b).
49
FTA(2010).
43
forenduserstoparticipateinthemanagingofanincreasinglylowcarbonsystem.Consumers’
thermalloadscanbeleveragedasenergystorageopportunitiesthroughdemandresponseand
demandsidemanagementprograms,thedeploymentofwhichisgreatlyfacilitated,andinsome
casesnecessitates,AMI.
Investmentsinelectricitydeliveryandreliabilityarecoreinvestmentsineconomicproductivity.
Reliableelectricityisessentialforeconomicgrowthanda flexibleresilientelectricitysystemisan
enabler that will allow for higher levels of penetration of renewable energy on the grid, thus
helpingtoreduceenvironmentalexternalities.
TheRecoveryActallocated$10.4billiontowardprogramsandprojectstoenhancethereliability
ofthe nation’selectricgrid,totransformthegridinpreparationofincreasingintermittentenergy
supply,andforresearchanddevelopmentofadvancedgridtechnologies.$4.487billionofthese
funds were appropriated to support states development of modern and reliable electric and
energysystems.Bytheendof2015,$4.4billionofthesefundshadbeenoutlaid.
50
Abreakdown
ofspendingisbelowinTableA3.Ascanbeseen,mostofthefundingwenttowardsinvestments
insmartgrids,whichisdiscussedinmoredetailbelow.
SmartGridI nvestmentGrantProgram(SGIG)
Through the Smart Grid Investment Grant Program (SGIG, EISA 1306) DOE and the electricity
industry jointly invested $8 billion in 112 costshared projects. Over 200 participating electric
utilities and other organizations were involved. These projects focused on strengthening
cybersecurity, improving interoperability, and collecting data on smart grid operations and
benefits.
Since2009,$3.5billionhasbeenawardedandspentthroughthesmartgridprogramas
ofyearend2015.
ThroughtheSmartGridInvestmentGrantprogram,11synchrophasorprojects(i.e.,projectsto
providerealtimemeasurementofimportantmetricsontheelectricgrid)havebeencompleted,

50
DOE(2016a).
44
installing over 800 networked phase measurement units (PMUs). PMUs improve electric grid
reliabilitybyallowinggridoperatorstoidentifyandcorrectgriddisturbancesbeforetheybecome
major grid stability issues. In addition, the 65 SGIG smart meter projects reached the goal of
installing15.5millionsmartmeters,withover16million
nowinstalledandoperational.
51
AnexampleofSGIGatworkistheFloridaPowerandLightCompany’s(FPL)$800milliondollar
projecttomodernizeitselectricitygridsystem,$200millionofwhichweresupportedbyARRA
funds.WithSGIGfunding, FPLwasableto expand itsplanforsmartgrid projectsbyincluding
over
5,000 intelligent monitors, sensors, and controls on the transmission and distribution
system.SGIGfundsalsosupportedFPLsenhanceddiagnosticsystemtocollectandinterpretdata
from substation devices and transmit information to FPL diagnostic centers for problem
detection and outage prevention. For example, FPLs new Transmission Performance and
DiagnosticCenter(TPDC)remotelymonitors500FPLsubstationstomonitorvoltagelevelsand
impedance, allowing for substation problems to be detected and repaired before leading to
outages. Further, the new smart grid capabilities provide customers with access to detailed
informationaboutusageandcoststhroughanonline“EnergyDashboard”displayingdailyenergy
metrics.
52

SmartGridRegionalandEnergyStorageDemonstrationProject
TheSmartGridRegionalandEnergy StorageDemonstrationProject(EISA1304)alsosupported
modernizationof the electricgrid. It directed $684 million, along with a $900 million industry
costshare,towards32RegionalSmartGridDemonstrationandEnergyStorageDemonstration
projects under the Smart Grid Demonstration Program (SGDP).
53
One example of a successful
EISA1304grantincludeda$45millionawardtoConsolidatedEdisonCompanyofNewYork,Inc.,
to demonstrate a scalable Smart Grid prototype that promotes cybersecurity, distributed
resources,electricvehiclechargingandconsumerparticipationinenergymix.
InvestmentsinAdvancedVehiclesandFuels
Thetransportationsectoris the secondlargestcontributorofgreenhousegases inthenation,
comprising 27 percent of national emissions in 2013, and its emissions have increased by 16
percentsince1990.
54
Thecombustionofpetroleumbasedproductsfrompassengervehiclesand
lightdutytrucksmakesupoverhalfoftheemissionsfromthetransportationsector.Developing
newtechnologiesiscriticalforreducingemissionsfromtransportation.
The ARRA investments in advanced vehicles and fuels help to address environmental
externalities, energy security externalities,
innovation market failures, and even network
externalities. The investments can help address environmental externalities in the longrun
throughadecarbonizingofthevehiclefleet.Theycanhelpaddressenergysecurityexternalities
by replacing petroleum products with other fuels in transportation. They can help innovation

51
EDER(2012).
52
EDER(2012).
53
Ofthe$684millionappropriated,almostallhasbeenspentasofyearend2015(DOE2016).
54
EPA(2015).
45
marketfailuresinboththeshortrunandlongrunbyfosteringinnovationwithhighdegreesof
spilloversandhighsocialreturns.Theycanhelpaddressnetworkexternalitiessincethereare
network effects from sufficient refueling or recharging infrastructure. For instance, with few
rechargingstationsfordedicatedelectricvehicles(EVs),
eachrechargingstationwouldhavelittle
valueforfewconsumerswouldbuyelectricvehicles.Butwithmanyrechargingstations,thereis
morelikelytobeanactivemarketforelectricvehicles,andthuseachstationhashighersocial
value.
ARRAdirected$6.1billiontowardprogramstopromoteresearchonand
deploymentofthe next
generationofautomobilebatteries,advancedbiofuels,pluginhybrids,andallelectricvehicles,
andtheinfrastructureneededtosupportoperationalizingthesetechnologies.Programsincluded
fundingforataxcreditforpluginhybridelectricvehiclesanddedicatedelectricvehiclesofupto
$7,500pervehicle,with$2.2billionallocated.
55
Thetaxcreditforeachqualifyingvehiclemodel
phasesoutafterthemanufacturersells200,000qualifiedvehiclesofthatmodel.Inaddition,$2.4
billion in grants were awarded, alongwith costsharing, through a competitive process. These
competitivegrantssupportdomesticmanufacturinganddeploymentofadvancedbatteriesand
electricand
pluginhybridvehiclecomponents.$600millionoftheARRAfundingwasdirected
towardsadvancingbiofuels.Further,$300millionoftheARRAfundingwasdirectedtowardseach
of the following: alternative fueled vehicles pilot grant program, diesel retrofits, and federal
motorvehiclefleetprocurement.
AdvancedVehicleTaxIncentives
Agrowingbodyofeconomicliteraturedemonstratesthattaxincentives,suchastheincentives
inARRA,increasetheadoptionratesandmarketsharesofadvancedvehicles.Mostofthiswork
has focused on hybrid vehicles, due to the lack of data on plugin hybrid electric vehicles and
electric vehicles. As a first example of how tax incentives increase the adoption of advanced
vehicles,GallagherandMuehlegger(2011)studytheeffectofgovernmentincentivesonhybrid
electricvehicleadoptionfrom2000to2006. Theyfindthatbothsalestaxwaiversandincome
taxcreditsincreasehybridsales.Diamond(2009)alsofindsthatgovernmentfinancialincentives
increase the market share of hybrids, although gasoline prices have the strongest impact.
Beresteanu and Li (2011) find a 20 percent increase in hybrid vehicle sales due to federal
incentives using data from 2006. Sallee (2011) examines data on consumer purchases of the
ToyotaPriushybridvehicle
andfindsthatthetaxincentivesarelargelycapturedbyconsumers,
ratherthanproducers.
Whiletheseresultsintheeconomicsliteraturemaynotentirelyapplytopluginhybridelectric
vehicles or electric vehicles, they are indicative of the effects that may be expected from the
ARRAtaxincentivesas
well.Indeed,withthesupportofthetaxincentive,salesofpluginelectric
drivevehicleseligibleforthecreditreachednearly69,000from2012totheendof2015.
56

55
Only$202millionofthisestimatedtotalallocationisforecastedtohavebeenspentbytheendof2015basedon
OSTFY2012forecast.
56
IRS(2016).
46
InvestmentsinBatteryTechnologyandTransportationElectrification
The cost, size, durability, and safety of battery technology remains a critical hurdle to the
widespreaddeploymentofpluginhybridelectricvehicles(PHEVs)anddedicatedelectricvehicles
(EVs).ThelimiteddrivingrangeandcostofmostcommercialEVscreateabarriertoconsumer
adoption,andarelargelyduetothestateofdevelopmentofcurrentbatterytechnologies.The
higherenergyandpowerdensitiesoflithiumionbatteriesareessentialforthefeasibilityofthe
current PHEVs and EVs, but there is still a great need for further improvement in battery
technology.
Developmentsinimprovedbatterytechnologyhaveaveryhighsocialreturn,withbatteriesbeing
used in a wide variety of commercial and military applications. ARRA investments in battery
technology help address innovation market failures in battery technology development,
encouragingthedevelopmentoftechnologythatwillhavespilloverbenefitsacrosstheeconomy.
ARRA funds provided grant funding to support both the use and development in lithiumion
batterytechnologyandtheestablishmentofadomesticbatterymanufacturingsupplychain.
Thebatterysupplychainincludesextractingrawmaterials,developingbatterycellcomponents,
fabricating battery cells, and assembly of the battery pack. ARRA funds have supported
companiesallalongthesupplychain.Specifically,theRecoveryActsupportedthedevelopment
ofthelithiumbatterysupplychainthroughgrantsof$28.4milliontodeveloplithiumsupplies,
$260milliontoproducecellcomponents,$730millionforcellswithdifferentchemistries,$460
millionforpackassemblyfacilities,and$9.5millionforalithium recyclingfacility.
57
One example of a grant awarded to support development of domestic electric vehicle supply
chainwasthe$95.5milliontoSaftAmerica,Inc.,forthe productionoflithiumioncells,modules,
and battery packs for industrial and agricultural vehicles and defense application markets.
Another example is the $161 million grant awarded to Dow Kokam, now XALT energy, for
manganesecathodesandlithiumbatteryproductionatitsplantinHolland,Michigan.Another
wenttoGeneralMotorsfor$105.9milliontoproducelithiumioncellsandpacksfortheVolt.
Thetransportationelectrificationgrantprogramalsodirected$400millioningrantsforprojects
toadvance thedevelopment ofelectricdrive vehicle systemsand infrastructure.Forexample,
ChargePointreceiveda$15millionmatchinggrantthroughtheprogramforthedeploymentof
over4,600home,public,andcommercialelectricchargingpointsatanaveragecostof$3,300of
federalfundsperchargingstation.Completedin2013,theprojectalsoprovidesdatatoIdaho
National Laboratory to provide researchers and planners the information needed to better
understandchargingpatternsandfutureEDVinfrastructureneeds.
58

In addition, ARRA funding supported “The EV Project” with a $115 million matching grant,
59
deploying5,700NissanLeafs,2,600 PHEVChevroletVolts,and14,000EVchargersand300DC

57
CRS(2013).
58
ChargePoint(2013).
59
Matchingfundingleadto$230millionprojectbudget(Scheyetal.2012).
47
chargers.ThroughauniquepartnershipacrossNissanNorthAmerica,GeneralMotors,theIdaho
NationalLaboratory,city,localandstategovernmentsandutilities,theEVProjectimplemented
alargesaleelectricvehiclecharginginfrastructuredemonstration.Thedemonstrationdeployed
EVsandchargerstocollectdatacharacterizingchargingstationusageacrossdifferentlocalities,
evaluateeffectivenessof charging infrastructure,andanalyzechargingimpacts on the electric
grid.Suchdemonstration projects provide valuable data toresearchersseeking tounderstand
chargingpatternsandgridimpacts,pavingthewayforimprovedandinformeddesignofpublic
andprivatechargingnetworks.
ARRA also supported domestic manufacturing in batteries and energy storage though Section
48CgrantsforCleanEnergyManufacturing,asdiscussedbelow.
InvestmentsinCarbonCaptureandStorage
Thetransitioningto a lowcarbonenergysystemwilllikely requirea diverse setof lowcarbon
technologies. Carbon capture and storage (CCS) technology offers the opportunity to reduce
carbondioxideemissionsofexistingpower plantsandindustrialfacilities,rapidlyincreasingthe
speed of decarbonization. CCS works by capturing carbon dioxide emission from large point
sourcesofemissions,transportinggasestosubsurfacerockformation,andpermanentlystoring
the carbon to prevent its release into this atmosphere. Equipping a power plant with CCS
technologycanreducecarbonemissionsby80to90percent.Infact,theIPCC(2005)estimates
thatthereisatechnicalpotentialofatleast2,000GtCO
2
storagecapacityworldwideingeological
formations, and predicts CCS will contribute 15 to 55 percent of worldwide mitigation effort
through2100.
ARRAauthorized$3.4billioninactivitiesinCCSresearchanddesign,commercialdemonstration,
implementation,andeducation.WhileCCStechnologiesaresuccessfullybeingdeployedinthis
countryandaroundtheworld,thedevelopmentofanynewtechnologyisadifficultendeavor.
Oftheinitialallocation, theDOEisreturning$1.3billiontotheU.S.DepartmentofTreasuryfor
fourCCSprojectsthatwerefundedbyDOEundertheARRAandwerenotabletoadvancetothe
point that the
ARRA funding could be spent within the timeframe specified by statute. DOE’s
return of these ARRA funds to Treasury is a reflection of the significant challenges faced by
businessesthatareintroducinginnovative,earlystageenergytechnologiestomarketsandnot
necessarilyanegativereflectiononthereadinessofCCStechnologies.
60
OnesuccessfulprojectsupportedthroughARRAfundsincludestheAirProductsandChemicals
Inc.(APCI)hydrogenfacilityinPortArthur,Texas.A$280millionARRAgranttosupportsa$430
million dollar project combining CCS technology with enhanced oil recovery (EOR) to use the
captured carbon to extract untapped
fossil resources before it is stored underground. At full
scale operation, theproject captures over 90 percentof the carbon dioxide from the product
steam of two methane steam reformers,translatingtoapproximately1million metric tons of
carbon dioxide delivered to sequestrations per year.On May 15, 2015, the DOE and APCI

60
DOE(2016)estimatesCCSspendingthrough2015yearendat$1.84billion.
48
announcedthattheprojecthadsuccessfullycaptureditssecondmillionthmetric tonofcarbon
dioxide.
GreenInnovationandJobTraining
Greeninnovationandcreatingjobsinthesustainable21
st
centuryeconomyaretwoofthekey
elements of the Recovery Act. Many of the areas described above involve research and
developmentofnewtechnologies.Thegreeninnovationandjobtrainingcategorycompletesthe
suite of clean energy research, again helping to address innovation market failures. The job
trainingalsoallowsourworkforcetoshifttobecompetitiveinthe21
st
centuryeconomy.
TheRecoveryActdirected$3.5billiontowardsprogramstosupportresearchanddevelopment
of advanced biofuels, clean energyrelated information and communications technology, and
enhanced geothermal systems. In addition, $400 million of these funds was allocated to the
Advanced Research Projects Agency‐Energy (ARPAE) program, which funds new, creative
researchideas aimedatacceleratingthe pace of innovationinadvancedenergytechnologies.
While the ARPAE was created by statute in 2007, ARPAE’s first projects were funded by the
RecoveryAct. Funding in ARPAE was focused on early stage innovations with very highsocial
value.
ARPAEhassponsoredover400energytechnologyprojectsatthecuttingedgeofcleanenergy
markettransformation sinceits initialfunding in2009.Someexamplesinclude a1 MWsilicon
carbide transistor, engineered microbes that use hydrogen and carbon dioxide to make liquid
transportationfuel,andanearisothermalcompressedairenergystoragesystem.
61
Othergrant
awardssupportarangeofdemonstrationprojectsincludingasolarconversiontowertoprovide
dispatchable solar energy, developing plants that produce vegetable oils in leaves and stems,
algaeharvesting frombiofuels,andvariousprojectsworkingtowardimprovementinbatteries
andelectricitystorage.
TheRecoveryActalsoincluded$500millionforcompetitivegrantstostateagenciesandnon
profitstosupportprogramsthattrainworkersforjobsintheenergyefficiencyandcleanenergy
industriesofthefuture.Itincludedanother$100million fortrainingandhiringworkersinthe
utilityandelectricalmanufacturingsectors.
This ARRA funding for
greenjobs supported 25 Energy Training Partnership green job training
grantstoprovidetrainingforworkersinenergyefficiencyandrenewableenergyindustries.The
grants support partnerships among labor organizations and public and private employers to
design and distribute training methods, as well as build an understanding of green industries.
Specific grants included support for electrical apprenticeship training, programs targeting
unemployed dislocated workers, women, minorities and veterans, and community groups
servingunemployedanddislocatedworkers.AnexampleistheSolar InstructorTrainingNetwork
(SITN), which was originally funded under ARRA, and hasprovidedinstructor training to more

61
DOE(2016c).
49
than 1,000 qualified and credentialed solar PV instructors—leading to over 30,000 students
throughouttheUnitedStatesreceivingtrainingincareersinsolarenergy.
CleanEnergyEquipmentManufacturing
To promote the development of a domestic supply chain to support a growing clean energy
industry, $2.3 billion in ARRA funds were allocated toward a 30 percent tax credit (under IRS
Section 48C) to be awarded on a competitive basis. Eligible projects included investments in
advanced clean energy manufacturing for a wide range of clean energy products such as
equipment for renewable energy generating facilities, energy storage, energy conservation,
carboncaptureandsequestration,fuelcells,andtherefiningandblending ofrenewablefuels.
Since nearly all of theseinvolve new technologies with possible innovation spillovers, ARRA is
helping to address innovation market failures in multiple areas through these competitively
awardedtaxcredits.
The $2.3 billion program was highly oversubscribed, with interest vastly exceeding DOE
expectations. There were over 500 applications, which requested over $8 billion in credits
combined. Credits were awarded based on commercial viability, domestic job creation,
technological innovation, and speed to project completion.
62
Awards were provided to 183
domestic clean energy manufacturing facilities until the $2.3 billion available was exhausted.
Thesecreditswerematchedby privatesector fundingofupto$5.4billion.Thesuccess ofthe
programledtoanotherroundofcompetitivefunding(PhaseII)announcedinFebruary2013for
$150millioninremainingtaxcreditsnotfullyutilizedbypreviousgranteesforprojectstobeput
inserviceby2017.
Oneindustrysignificantlysupportedbythegrantswaswindturbinemanufacturing.Whilemany
factors may have contributed, there was a dramatic increase in the share of domestically
producedwindturbinesinstalledintheUnitedStatesafterARRA,from25percentin20062007
to72percentin2012.
63
Thegrantsalsosupportedenergystorageprojects.Justasthebattery
awards described above spurred new battery technologies, Section 48C complemented these
awards. Section 48C Phase I and II awards included at least $36 million in credits for battery
related manufacturing including mobile charging, advanced battery packs, and intermediate
materialsandcomponents.

62
MundacaandRichter(2015).
63
EERE(2013).
50
AppendixII:CEACalculationofARRACleanEnergySpending
This appendix provides an overview of CEA’s calculation of the amount of ARRA clean energy
funding that was spent by the end of 2015. This spending estimate is used to calculate the
numberofjobyearssupportedbythecleanenergyinvestmentsinARRA.
Forthecalculation,CEAusesthebestevidenceavailableonactualARRAspendingfromDOE,the
U.S. Treasury (UST), and the Office of Management and Budget (OMB) to develop a spending
path from 2009 to the end of 2015. All nonARRA spending is also removed, including any
extensionsofprogramsaftertheARRAperiod.
64
Whereactualspendingisnotavailable,themost
recentavailableforecastsareused.Theapproachusedfollowsthreesteps:(1)useforecastsof
tax outlays from the UST for tax incentives, (2) add in known spending from the respective
sources,and(3)allocatethefundsspentinremainingprograms.
Step1.UseForecastsofTaxOutlaysfromTreasury
First,CEAusestheforecastofannualgovernmentoutlaysfromARRAtaxincentivesprovidedby
theUSTOfficeofTaxAnalysisforFY2012.
65
Thisforecastisthenmodifiedasfollows:
1. 1603CashGrantinLieuofITC
Forthe1603program,theUSTexpectedoutlaysarereplacedwithactualawardsprovidedin
theyears2009to2015. Followingthestaticbudgetscoringmethodology,theCEAanalysis
alsoadjusts forreceiptsthatrepresentaforecastedreductioningovernmentreceiptsthat
occurselsewhereasaresultofthe1603program.Inotherwords,totaloutlaysoverthe2009
to2015periodarereducedbyforecastedreceiptsfrom2009to2015.
66
2. CleanEnergyManufacturingSection48c
ThisprogramwasforecastedbyUSTwithoutlaysof$1.6billion.However,updatedspending
estimatesindicatethatthisprogramactuallyawarded$2.3billioningrantstocleanenergy
manufacturers.Thus,thespendingpathforecastfromUSTisadjustedfortheactualspending
of$2.3billion.Inotherwords,thespendingpathfromUSTwasproratedsuchthatthetotal
outlayssumto$2.3billion.

64
Theonlyexceptionwhereanextensionfromtheoriginal2009allocationisincludedisthecaseofthe1603Cash
Grant.In2010thisprogramextendedtheconstructionstartdaterequiredforprojectsto be eligibleforthecash
grantbyoneyear.Projectsthattookadvantageofthis extension
areincludedinthespendingestimates.
65
TheCEA (2010) 3
rd
QR ARRA Updateuses Treasury forecast for FY 2011. The FY 2012 shows largeroutlays for
residential energy efficiency credits. Clean energy subcategories with OTA forecasts of tax outlays include: PTC
extension;extensionofITCeligibility;1603CashGrantinLieuofITC(modifiedwithprogramspecificdetails);removal
of
the ITC cap for small wind; expansion of the residential energy efficiency tax credit; Clean Renewable Energy
Bonds; Qualified energy conservation bonds; refueling tax credit; alternative motor vehicle credit; tax parity for
transitandparkingcredits;andSection48cManufacturing(modifiedwithprogramspecificdetails).
66
ReceiptsforecastedfromUSTFY2011areusedhereas1603outlayswerenotincludedinUSTFY2012estimates
oftaxbenefits.
51
Step2.AddInKnownSpending
For categories with records of funds spent by a given date, the total spending is distributed
equally across years before that date. These programs include: the Weatherization Assistance
Program (updates from DOE), 1705 Loan Program (updates from DOE), high speed rail and
FederalTransitAdministrationgrantspending(updatesfromFRAandDOT),andtheincreasein
borrowing authority to the Bonneville Power Administration and Western Area Power
Administration(updatesfromOMB).
67
Step3.AllocateFundsSpentinRemainingPrograms
The majority of the categories remaining after considering the spending categories with tax
forecasts and known spending are managed by the DOE. As of May 2013, 82 percent of DOE
allocatedfundshadbeenspent.Asofyearend2015,90percent ofallocated funds had been
spent.Assuch,forprogramswithspendingpathsnotfilledinbyStep1andStep2above,apath
isdistributedequallyacrossyearssuchthat82percentofthefundsarespentbymid2013.Next,
spendingisdistributedequallyacrossfrommid2013toyearend2015isdevelopedsuchthat
in
total90percentoffundsarespentbyyearendof2015.FundsleftoverafterSteps1through3
areassumednotspentoverthe2009to2015period.

67
ForWAPthespendingpathisnotuniform:aspendingpathidentifiedfromtheDOEPAGEdatabaseisusedtopro
ratetotalWAPexpendituresreportedbyDOE(2016a)todevelopaspendingpathfrom2009to2013.