innovation policy for climate change - c's...

Innovation Policy for Climate Change

A REPORT TO THE NATION

Energy Innovation From the Bottom Up

A JOINT PROJECT OF CSPO AND CATF Made Possible Through the Support of the National Commission on Energy Policy,

a project of the Bipartisan Policy Center

S E P T E M B E R 2009

This project began with the goal of bringing the lessons of technological innovation to discussions of

climatechangemitigation.Howmightthechallengesofreducinggreenhousegasesbenefitfromahistorical

analysisoftechnologicaladvance,thediffusionofinnovations,andtheirintegrationintotheeconomy?What

could be learned from a deeper look at the way the U.S. government has in the past marshaled national

resources and capabilities—its own and those of the private sector, from which most innovations flow—

tostimulateasystemabletoprovidelastingconditionsforprogressinreducinggreenhousegasemissionsand

energyconsumption?

Inordertotestthathistoricalanalysisagainstthespecificneedsofparticulartechnologiesnowavailable

oronthehorizon,weengagedtheparticipationofawiderangeofexpertsoninnovation,aswellastechnical

experts, in “case studies” of three sample technologies: photovoltaics, post-combustion capture of carbon

dioxidefrompowerplants,anddirectremovalofCO2fromEarth’satmosphere.Inthreeworkshopsweasked:

Whatwouldthesetechnologiesneedinordertocontributesubstantiallytothedecarbonizationoftheenergy

systemandgreenhousegasreduction?

Theresultsofthisstudyofferafundamentallynewperspectiveonorientinggovernmentandtheprivate

sectortowardnear-termgainsinthedevelopmentoftechnologiestoservethepublicgoodofadecarbonized

energysystem.Wehopethatitstimulatesconversation,questions,andaction.



Daniel Sarewitz Co-Director

ConsortiumforScience,Policy&OutcomesArizonaStateUniversity

Armond Cohen ExecutiveDirector

CleanAirTaskForce

A JOINT PROJECT OF CSPO AND CATF Made Possible Through the Support of the National Commission on Energy Policy,

a project of the Bipartisan Policy Center

Foreword Accelerating the development and deployment of advanced, low-carbon technologies will be crucial to

surmountingtheenergyandclimatechallengesofthe21stcentury.Theadoptionofanoverarchingregulatoryregime

tolimitgreenhousegasemissionsconstitutesthenecessaryfirststeptowardcreatingamarketforsuchtechnologies

and is increasingly viewed as environmentally necessary and politically inevitable. But there is also a need for

complementarypoliciesthatspecificallyaddressthetechnologyinnovationanddevelopmentpipeline—aneedthat

hasunfortunatelyreceivedlessattentioninthelong-runningandoftencontentiousdebateaboutU.S.climatepolicy.

Thishastochange.ThetrackrecordofU.S.governmentinvolvementinenergyresearch,development,demonstration,

anddeploymentoverthelastfourdecadesismixedatbest.Inaneraofincreasinglyscarcepublicresources,achieving

thekindsofemissionstargetsfeaturedinrecentlegislativeproposalswhilealsoassuringamplesuppliesofreliableand

affordableenergyforagrowingeconomymeansthatwemustimproveonthatrecord.

Tothatend,theNationalCommissiononEnergyPolicyaskedtheConsortiumforSciencePolicyandOutcomes

andtheCleanAirTaskForcetoundertakeathoughtfulevaluationofpoliciesforadvancinginnovationincleanenergy

technologies.Byexamininghowournationhassuccessfullystimulatedahostoftransformativeinnovationsinother

sectorsinthepast,andbylookingcloselyatcurrenteffortsintheenergyarena—includinginparticularthevarious

programsfundedandmanagedbytheU.S.DepartmentofEnergy(DOE)—studyauthorssoughttoidentifywhatis

working,andwhatisn’t,inthenation’sexistingenergytechnologyinnovationsystem.Overall,theyconcludethatthis

systemisinurgentneedofreform.Amongthekeyrecommendationsputforwardinthisstudy:requiringavarietyof

institutions,includingtheDOE,tocompeteforpublicresearchdollarsandadoptinganewmulti-pathway,multi-agency

approachthatemphasizes targeted improvements in the low-carbontechnologies thatarealreadyonthehorizon

todayoverthequestforfundamentalbreakthroughsinmoredistanttechnologyoptions.

Ourcountryhashadaproudhistoryofrisingtothechallengesofeachnewage.Pastsuccessesinbuildingworld-

class infrastructurenetworks,exploringspace,anddevelopingtheworld’smostsophisticateddefensetechnologies

shouldinformourapproachtooneofthemostdifficultandconsequentialproblemshumanityconfrontsthiscentury—

andperhapshaseverconfronted.Thisstudyrepresentsasignificantsteptowardclarifyingthetypesofpoliciesand

institutional changes that will be needed to catalyze a profound, long-term transformation of our current energy

system.Byofferingasmall,clear,andworkablesetofprinciplesthatcanbeusedtoevaluatetheefficacyof future

energytechnologyinitiativesandprograms,thisreportprovidesaninvaluableresourceforU.S.policymakersincrafting

oneofthekeycomponentsofacomprehensiveandultimatelysuccessfulclimatepolicy.

ExecutiveDirector,NationalCommissiononEnergyPolicy

President,BipartisanPolicyCenter

Acknowledgments This report owes much of its depth and clarity to the erudition and eloquence of John Alic, the principal project consultant.

His insights, range of knowledge, and skills as an author have been most appreciated since the very beginning of this

project. We would like to also gratefully acknowledge the counsel and participation throughout the project of Frank Laird of

the University of Denver, Joe Chaisson of the Clean Air Task Force, Claudia Nierenberg of CSPO, and Catherine Morris of the

Keystone Center. Their expertise and creativity were essential to the project from its earliest days. Frank and Joe contributed

crucially to our analysis and understanding, particularly of photovoltaic technology and post-combustion capture and

storage of carbon. Claudia managed every aspect of the project with a wise and generous spirit that tamed a vigorous

diversity of personalities. Catherine facilitated the workshops with an unerring ability to guide the discussions toward key

questions and insights. We appreciate the participation and frank discussion that we were able to have with all of the

technical experts who launched the workshop discussions, and the policy experts who never shied away from challenging

and provoking the experts—and us; all participants are listed by name in Appendix 1. Many workshop participants also

provided helpful comments on various drafts of this report. The guidance and support we received from Sasha Mackler and

his colleagues, Tracy Terry and Nate Gorence, at the Bipartisan Policy Center’s National Commission on Energy Policy kept us

focused on practical outcomes that would matter in the short term. BPC’s Sara Bronnenkant made the workshops and dinner

meetings something to look forward to. Finally, we could not have completed the project without the efforts of Lori Hidinger,

managing director at CSPO, and Kellyn Goler, who miraculously appeared at CSPO one day; we were lucky to have her with

us for six months. Richard Hirsh of Virginia Polytechnic Institute and Sharlissa Moore of Arizona State University were kind

enough to read through final drafts and offer helpful comments. The project was made possible by the generous support of

the National Commission on Energy Policy, a project of the Bipartisan Policy Center.

Disclaimer Three expert workshops, convened by Arizona State University’s Consortium for Science, Policy and Outcomes and the

Clean Air Task Force, were part of this project on “Energy Innovation Systems from the Bottom Up: Technology Policies

for Confronting Climate Change” sponsored by the Bipartisan Policy Center (BPC) and its National Commission on

Energy Policy (NCEP). The workshops, held in Washington, DC during March and April of 2009, did not aim to achieve

consensus among the experts, but to use the examples of the three technologies to probe and highlight the

complexities and opportunities of energy innovation policy. While this report draws from these workshops, it does not

represent the views, individual or collective, of workshop participants. Nor does it represent the views of BPC or NCEP.

Table of Contents Summary 2

Introduction 4

1 The Need for Innovation 6 Research and regulation will not give energy-climate innovation enough of a boost.

2 Energy Innovation in Systems Context 9 Innovations unfold over time through complex processes, which policy choices influence but do not control. The evolution of the photovoltaic industry illustrates interactions between technical change and market development. 3 Managing Innovation 12 Because technological systems are complex and constantly evolving, managing them is difficult. Post-combustion capture of carbon dioxide presents an opportunity to learn from past, cautionary experiences such as nuclear power, mismanaged by both public and private sectors.

4 Energy Innovation Compared to Information Technology 17 Despite fundamental differences, the IT revolution holds lessons for energy-climate innovation. Perhaps the most important is the role of government as a demanding and technologically sophisticated customer.

5 Where Have All the Breakthroughs Gone? 21 Technological breakthroughs will not radically transform energy systems in the next several decades.

6 Building an Energy Innovation System 23 Complexity and uncertainty cannot be evaded, but understanding how technologies advance and how the strengths and weaknesses of existing institutions affect the process can point toward new rationales, policy approaches, and management priorities. More competition among government agencies, and a particular focus on effective conduct of demonstration projects, deserve the particular attention of policymakers. Specific technologies may demand appropriately tailored policies, as illustrated through the example of direct air capture of carbon dioxide.

7 Conclusions and Recommendations 36 Effective energy-climate innovation policies will be consistent with a small number of guiding principles and take advantage of a small number of highly leveraged points of opportunity.

Principles to Guide Energy Innovation Policies 37

Recommendations for Priority Action 38

Looking Ahead 40

Appendix 1 41 Workshop Participants

Appendix 2 42 Project Planning Team

2

The world is not on a path to reduced greenhouse gas emissions. The suite of commercially viable climate-

friendlyenergytechnologyneedstobeexpandedrapidly,whichinturnrequiresappropriategovernmentpolicies.Thecurrent

highcostofdeliveringlow-carbonenergymeansthatenergy-climatetechnologiesmustbetreatedbygovernmentsasapublic

good,akintonationaldefense,publichealthanddisasterprotection.Doingsocanopenupnewavenuesforenergy-climate

technologypolicysuchasthosethatallowedtheU.S.toleadtheworldininnovationformuchofthe20thcentury.

ThisreportdrawsonthelessonsofpastU.S.governmentpolicyfortechnologicalinnovation,aswellasthreeworkshops

heldinMarchandAprilof2009.Theworkshopsbroughtexpertsoninnovationandgovernmentpolicytogetherwithexpertsin

threetechnologies—solarphotovoltaics(PVs),post-combustioncaptureofcarbondioxide(CO2)frompowerplants,anddirect

removalofCO2fromtheatmosphere(aircapture)—notinsearchofconsensus,buttoprobeandillustratethecomplexitiesand

opportunitiesofenergy-climateinnovationpolicy.*

Thereportconcludesthat:

o To improve government performance, and expand innovation options and pathways, Congress and the

administration must foster competition within government. Competition breeds innovation. That is true in

economicmarketsanditholdsforgovernmenttoo.TheUnitedStatesreliesfartooheavilyontheDepartmentofEnergy

(DOE)forpursuingenergy innovation.Competitiveforcesdrovemilitarytechnological innovationafterWorldWar II—

East-West competition; competition among defense, aerospace, and electronics firms; and competition among the

militaryservices. Inter-agencycompetitionhasbeenaneffective force in innovationacrosssuchdiverse technologies

as genome mapping and satellites. No such competitive forces exist for energy-climate technologies. Expertise and

experienceexisttodayinmanypartsofthepublicsectorotherthanDOE,includingtheDepartmentofDefense(DoD),

theEnvironmentalProtectionAgency(EPA),andstateandlocalgovernments.Andfacingmeaningfulcompetition,DOE

wouldhavetoimproveitsownperformanceorrisklosingresources.

o To advance greenhouse-gas-reducing technologies that lack a market rationale, government should

selectively pursue energy-climate innovation using a public works model. Thereisnocustomerforinnovations

suchaspost-combustioncaptureofpowerplantCO2andaircapture. (Indeed,nomorethanabouttwodozenpeople

worldwideappeartobeworkingonaircaptureatall–anunacceptablysmallnumberbyanystandard.) Recognition

ofgreenhousegas(GHG)reductionasapublicgoodredefinesgovernmentasacustomer, justasit isfor,say,pandemic

INNOVATION POLICY FOR CLIMATE CHANGE



Summary

3

fluvaccines,floodcontroldams,oraircraftcarriers.Thisperspectivepointstonewapproachesforcreatingenergy-climate

infrastructure in support of innovation and GHG management. Some tasks might be delegated to state and local

authorities,whichalreadycollecttrash,maintainwaterandsewersystems,andattempttosafeguardurbanairquality.

o To stimulate commercialization, policy makers must recognize the crucial role of demonstration projects

in energy-climate innovation, especially for technologies with potential applications in the electric utility

industry. Government-sponsored demonstration programs have a long-established place in U.S. technology and

innovationpolicy,butapoorreputationinenergy.Sincetheprimarypurposeofdemonstrationsistoreducetechnical

andcostuncertainties,theprivatesectorshouldbechieflyresponsibleformanagingdemonstrations,withgovernment

providingfinancialsupport,disseminatingresultsopenly,andensuringa levelcompetitiveplayingfield.Well-planned

andconductedprogramscouldpushforwardtechnologiessuchasCO2capturefrompowerplants.While,forexample,

theDOEhassupportedexploratoryR&Donadvancedcoal-burningpowergenerationforseveraldecades,ithaslargely

ignoredtheissuesraisedbycontrollingCO2fromthenation’sexistingcoal-firedpowerplants,whichproduceoverone-

thirdofU.S.CO2emissions.TechnologiesexistforcapturingCO2fromsuchplants,buttheyhavenotbeentestedatfull

plantscale.

o To catalyze and accelerate innovation, government should become a major consumer of innovative energy

technology products and systems. DoDprocurementhasbeenanenormouslypowerfulinfluenceoninnovationacross

importantareasofadvancedtechnologyfromelectronicstoaerospacetoinfo-tech.Incontrast,theU.S.governmenthas

notsystematicallyandstrategicallyuseditspurchasingpowertofosterenergy-relatedinnovations.Yeteachyear,federal,

state,and localgovernmentsspend largesumsongoodsandserviceswith implications forGHGreleaseandclimate

change, including office buildings, motor vehicles, and transit systems. Government can be a “smarter customer” for

energy-climateinnovations,helpingtocreateearlymarkets,drivingcompetitionamongfirms,andfosteringconfidence

inadvancedtechnologies,includingthosethatarenotyetprice-competitiveintheopenmarket.

LikeotheraspectsofU.S.energyandclimatepolicy, thenation’sapproachtoenergy-climate innovationhas lackeda

clear mission and strategy. Most attention and discussion has focused on advanced research, yet most innovation in the

coming decades will depend much less on frontier research than on other available and proven tools. (Indeed, in none

of our workshops did “more research” surface as the major concern—not even for air capture, which, though radical in

concept,isbasedonwell-understoodconceptsandprocesses.)Weknowwhatworks,basedonthepast60yearsandmore

ofexperience,butsofarwehavenotusedwhatweknowtoaddressenergytechnologiesandclimatechange.Weknow,

forexample,thattechnologicaladvancescomelargelyfromindustry—butthatgovernmentcancatalyze,andevencreate,

newwavesofindustrialinnovationbysupportingthetechnologybase,providingincentives(suchasthosethathavebeen

so effective in expanding the market for PV systems), and deploying its purchasing power. By treating climate mitigation

as a public good and GHG reduction as a public works endeavor, the United States can rapidly strengthen the linkages

betweenpublicinvestmentandprivatesectorinnovation,andbegintoleadothercountriestowardbuildingenergy-climate

technologiesintothefabricoftheirinnovationsystems,theireconomies,andtheirsocieties.

*Whilethisreportdrawsfromtheworkshops,itdoesnotrepresenttheviews,individualorcollective,ofworkshopparticipants;nordoesitrepresenttheviewsoftheproject’ssponsors,theBipartisanPolicyCenter’sNationalCommissiononEnergyPolicy,ortheworkshops’facilitator,TheKeystoneCenter.

SUMMARY

Reductions in carbon dioxide (CO2) and other green-

housegases(GHGs)willdependontechnologicalinnovation.

While a great deal is known about technological advance,

the diffusion of innovations, and their integration into the

economy,littleofthathasyetbeencapturedindiscussions

ofenergy-climatepolicy.Mitigationofclimatechangeisnot

aboutfinding“asolution,”orreplicatingaparticularsuccess

liketheManhattanprojectortheApollomoonlanding,ora

particularinstitutionalinnovation,liketheDefenseAdvanced

Research Projects Agency (DARPA). Rather, what is needed

isapatient,multifacetedapproachthatinvolvesmanydiffer-

ent organizations, competing as well as cooperating,

workingwithindustryanduniversities,andsharing,moreor

less,acommonobjective:toreduceglobalGHGemissions.

Inparticular,aswewillemphasizethroughoutthisreport,

innovationprocessesandenergysystemsaretoocomplexto

yieldtostrategiceffortsatcomprehensivereorientation.Nor

arenarrowlyfocusedpoliciesaimedatachievingaparticular

objective likely to yield predictable consequences. Effective

policy making, rather, will encourage innovation simultane-

ously along multiple pathways, build in opportunities for

continuedlearningandcoursecorrection,andseekparticular

pointsofinterventionwherefocusedattentionmayleverage

biggains.

Innovations come mostly from private firms, but only if

governmentcreatestheappropriateconditions.Aneffective

U.S.innovationsystemforconfrontingclimatechangewillde-

pendontwocentralpillars:first,regulationsstringentenough

Thisreportpresentstheresultsofaprojecton“EnergyInnovationSystemsfromtheBottomUp:TechnologyPoliciesfor

ConfrontingClimateChange”conductedbyArizonaStateUniversity’sConsortiumforScience,PolicyandOutcomesand

theCleanAirTaskForcewithsponsorshipbytheNationalCommissiononEnergyPolicy(NCEP),aprojectoftheBipartisan

PolicyCenter(BPC).Thereportbuildsonthelargebodyofknowledgeconcerningtechnologicalinnovation,aswellas

threeworkshopsheldinWashington,DCduringMarchandAprilof2009.Theseworkshops,facilitatedbyTheKeystone

Center, brought experts on innovation and government policy together with experts in three technologies: solar

photovoltaics;post-combustioncaptureofcarbondioxidefrompowerplants;anddirectremovalofcarbondioxidefrom

theatmosphere.Thereportdrawsfromtheseworkshopsbutdoesnotrepresenttheviews, individualorcollective,of

workshopparticipants(seeparticipantlistings,p.41).NordoesitrepresenttheviewsofBPCorNCEP.



4 INNOVATION POLICY FOR CLIMATE CHANGE

Introduction

5

(or incentives powerful enough) to induce innovation by

business and industry and pull both existing and new tech-

nologiesintowidespreaduse;second,technologypushfrom

policiesandprogramsthatcontributenewtechnicalknowl-

edgeas inputsto innovation.Ataglobalscale,aneffective

innovationsystemwillcombineandintegratethecontribu-

tionsofnationsatvariousstagesofeconomicandtechno-

logicaldevelopment,especiallycountriesnowexperiencing

rapid growth in energy consumption and GHG emissions

such as China and India, tapping their differing resources,

cost structures and market conditions. An effective world

innovationsystemwillalso,necessarily,relyheavilyonlarge

andtechnologicallysophisticatedmultinationalfirmsthatdo

business in many countries. Those, however, are concerns

thatgobeyondthescopeofthisreport,whichaddressesU.S.

institutionsandU.S.policies.

Indeed, because the United States is still the world’s

largest and most innovative economy, as well as the largest

energyuser,gettingU.S.energy-climateinnovationrightwill

be essential for success at the global scale. The past three

decades of U.S. experience suggest that without both

regulatorypullandtechnologypushtherewillbelittleenergy-

climate innovation. Neither of these forces has much

prominence today. Industry may expect GHG regulations at

some point in the future, but without knowing what form

theywill takehas little incentive to innovate.Marketpull for

energy technologies remains generally weak, and in the

UnitedStatesnomarketsexistatallfortechnologiestocontrol

GHGemissions.Thefederalgovernmenthasspentsome$60

billiononenergyR&Dsincethefirstoilembargoin1973,but

hasboughtmostlyresearch,withmodestimpactsoverall.

Our three workshops addressed only three specific tech-

nologies: solar photovoltaic (PV) systems, post-combustion

capture (PCC) of CO2 from fossil fuel-burning power plants,

and air captureofCO2 (direct removal fromtheatmosphere).

Thesecaseswereselectedbecausetheyrepresenttechnolo-

gies of varying maturity under development in different

industries and facing quite different market conditions.

Despitethesecontrasts, innoneofourworkshopsdid“more

research”surfaceasamajorconcern.Inallthree,participants

pointedtotheneedforastablestructureofpoliciestospur

applications development. Innovations in solar PV systems

spring from a vibrant, competitive world industry that has

grown largely as a result of artificial markets created by

governments.TechnologiesforCO2captureandstorageexist

andhavebeencommerciallyappliedatrelativelysmallscales.

Theyrestonreasonablywell-understoodtechnicalandscien-

tificfoundations.Theneedisfordemonstrationunderrealistic

conditions,scale-up,andthekindofongoinginnovationthat

comeswithoperatingexperience.Aircapturedrawsonsimilar

scientificandtechnologicalprinciplesbutisfarlessdeveloped.

These are starting points similar to those that bred military

innovationsafterWorldWarII, innovationsincomputersand

information technology in thecivilianeconomyduring the

latterpartofthetwentiethcentury,andalsotheinnovations

thatrevolutionizedagricultureearlierinthatcentury.Weknow

whatworkedtogeneratethosepreviouswavesofinnovation.

So far, we have not used what we know to address energy

technologiesandclimatechange.

Thisreportstartsoffwithabriefdiscussionoftheneed

forenergy-climate innovations,andusesthethreecasesto

help illustrate the challenge of developing appropriate

policies to stimulate such innovations (Section 1). Next we

presenttheideaofaninnovationsystemasawayofunder-

standing the complexity of innovation—complexity with

which policies must contend (Section 2). In Section 3 we

briefly discuss the implications of this complexity for the

management of innovation. To help make clear the many

questionsassociatedwithenergy-climateinnovationpolicy,

we continue in Section 4 by discussing in some detail an

examplefromanotherdomainthat,althoughverydifferent

intermsoftechnologiesandmarkets,iswidelyviewedasan

innovation policy success: information technology. Section

5 deals with the role of research and technological break-

throughsininnovationandininnovationpolicy,andSection

6 focuses in on a set of issues and opportunities that we

believedeserveparticularattentionifthegovernment isto

significantlyimproveitscapacitytocatalyzeenergy-climate

innovation.Thereportconcludesbyofferingfouroverarch-

ing principles and four policy recommendations to guide

governmentdecisionmakers.

INNOVATION POLICY FOR CLIMATE CHANGE

Substantial cuts in emissions of carbon dioxide, the

majorcontributortoclimatechange,willrequireatransfor-

mation,analogousinscaleandscopeifnotintechnological

or market fundamentals, to the information technology

revolutionthatbeganaround1950.Afterfluctuatingforcen-

turiesintherangeof280partspermillion(ppm),theglobal

average concentration of CO2 in Earth’s atmosphere has

reached 385 ppm. Concentration is continuing to rise at

about0.6percentannuallyandmayexceed500ppmbefore

mid-century.1Theconsequences,althoughuncertainintheir

specifics,willalmostcertainlyincludelong-termimpactson

ecosystems,humansettlements,andtheworldeconomy.

These consequences could be mitigated by some

combination of three alternatives: (1) reducing releases of

CO2 at the source, which implies capture and storage

(probablyundergroundsequestration)ofCO2especiallyfrom

coal-burningpowerplants,orlarge-scalereductionsinfossil

fuel consumption (e.g., by replacement with renewable

energysourcessuchassolarPVsystemsandconservationin

fuel use), or both; (2) continuing to release CO2 from at

leastsomesourceswhileremovingcompensatingtonnages

from Earth’s atmosphere, for example through air capture,

which amounts to chemical scrubbing of ambient air; (3)

attemptingtodirectlyregulateEarth’stemperaturewithout

regard to CO2 (and other GHGs), using “geoengineering”

approachessuchasinjectionofradiation-reflectingparticles

intotheatmosphere.

Noneofthesepathslookseasy.Revampingtheenergy

system to increase efficiency and reduce GHG emissions

would carry heavy upfront costs—to remodel homes and

commercial buildings by the millions worldwide, replace

thousandsofcoal-burningpowerplantsor retrofit themfor

carboncaptureandstorage,andreplaceahugeCO2-emitting

transportation fleet—while many of the gains would come

yearslaterandmightbeimperceptibletotaxpayersandvoters.

LikecaptureandstorageofCO2fromthecoal-burningpower

plantsthatproducenearlyhalfofU.S.electricalpower,and

emiteachyearsome2½billiontonsofCO2,directremovalof

CO2fromtheairwouldbecostly.Aircapture,moreover,has

yet to be demonstrated outside the laboratory. Geoengi-

neeringispoorlyunderstoodandhighlycontroversial.

Little in the three workshops or in our broader under-

standing of innovation suggests that reducing GHG

emissions and controlling atmospheric CO2 is the sort of

problem susceptible to radical technological innovation, at

leastradicalinnovationbasedonnewscience(radicalinno-

vationsalsostemfromcontinuoussmaller-scaleinnovations,

which over time crystallize into something fundamentally

new and different, as in the case of the Internet). Rather,

climatechange isasystemsproblem,andsolutionswillbe

multipleandlargelyincremental.Manyenergytechnologies

are mature. So are energy markets. Potentially transforma-

tional discoveries could appear, quite unexpectedly, as

high-temperature superconductivity (HTS) did in 1986. But

the HTS story also shows the fallacy of expecting research

breakthroughstodealwithproblemssuchasclimatechange.

While HTS is a flourishing subfield of solid-state physics,

the practical applications to energy systems so widely


1. The Need for Innovation Research and regulation will not give energy-climate innovation enough of a boost.

1International Energy Outlook 2008(Washington,DC:DepartmentofEnergy,EnergyInformationAdministration,September2008),p.90.CO2accountsforoverfour-fifthsofU.S.GHGemissions(82.6percentin2007onaCO2equivalentbasis).AlmostalltheCO2emissionsstemfromburningfossilfuels—coal,naturalgas,petroleumproducts—primarilytoproduceenergyforheating(includingindustrialprocessheat),transportation,andelectricpower.Bythemselves,coal-burningpowerplantsaccountforover35percentofthenation’sCO2release.Emissions of Greenhouse Gases in the United States 2007(Washington,DC:DepartmentofEnergy,EnergyInformationAdministration,December2008)andElectric Power Annual 2007(Washington,DC:DepartmentofEnergy,EnergyInformationAdministration,January2009).

anticipatedinthelate1980shaveyettoappear(thereasonsarenot-

edinalatersection).

Research is important, not because it can be expected to solve

today’s problems (although sometimes it does), but because it lays

foundations for solving tomorrow’s problems. (See the sidebar for

distinctions between research, development, demonstration, and

commercialization.) Moreover, there is no predictable or controllable

chainofactivitiesleadingfromresearchtoreal-worldapplication.Even

when innovations follow from research and discovery—and many

innovationsdonot,includingthoseasfundamentalasthemicroprocessor

(the outcome of engineering design and development without new

research)—two or three decades commonly separate discovery and

widespread application. That would be a long time to wait, even if

researchoutcomescouldbepredicted.Buttheycannot.

Discussion at all three workshops converged on a somewhat

differentperspective.Thethreetechnologiesexamineddiffergreatlyin

theirmaturity.Thefirstsolarcellswerefabricatedin1954,earlyapplica-

tions in spacecraft were followed by commercial production for niche

markets inthe1970s,andtodayaglobal industryflourishes—butonly

because of government subsidies in countries including the United

States,sincecostsforPV-generatedelectricityremainwellabovethoseof

alternatives.Therehavebeenmanyquitefundamentaldevelopmentsin

PVcellsover thepasthalf-century,butnonehave radicallyalteredthe

evolutionary pattern of innovation and growth, which over time looks

likearelativelysmoothprogression.

Post-combustion capture technologies are likewise available and

demonstrated, although not on utility scale (e.g., in the range of 500

megawatts [MW]). Further discoveries are possible, but cannot be

promised,andthereisnogoodreasontoawaitthemprovidedthecosts

of proceeding with available PCC technologies are judged acceptable.

Demonstrationandscale-up,somethinglikePVtechnologiesexperienced

afterthe1970senergyshocks,areattheforefrontoftheagenda.

Air capture too is technically possible. As for PCC, the science is

relativelywellunderstood:bothtechnologiesmakeuseofgasseparation

chemistry, a staple of the chemical industry for a century. Practical

problem-solving, scale-up, and demonstration lie ahead, along with

explorationofworkablebusinessmodels for implementing,operating,

and financing these and other technologies for carbon capture and

storage(CCS).

Research, Development, Demonstration, and Commercialization

Research, often subdivided into basic (or fundamental) research and applied research, aims to generate and validate new scientific and technical knowledge—whether in physics, in chemical engineering, or in organizational behavior. Development refers to a broad swath of activities that turn knowledge into applications. Generally speaking, development differs from research in being a matter of synthesis—envisioning and creating something new—rather than analysis in search of understanding. Design and development, the core activities of technical practice and hence the source of much technological innovation, apply knowledge in forms such as technical analysis based on mathematical models and methods. These can be used, for instance, to predict how PCC processes will scale up to larger sizes and for estimating their energy consumption. Over the past several decades, computer-based modeling and simulation have complemented and sometimes substituted for costly and time-consuming testing of prototypes. Demonstration is a particular type of development activity intended to narrow or resolve both technical and business uncertainties, as by validating design parameters and providing a sound basis for cost estimates. Demonstra-tion has considerable importance for some energy-climate innovations, notably carbon capture and storage, as discussed in later sections of this report. (Accounting and budgeting conventions normally treat demonstration as a form of R&D.) Commercialization, finally, marks the introduction into economic transactions of goods or services embodying whatever is novel in an innovation. Commercialization does not imply widespread adoption, which, if it does occur, may still take decades. The definitions of R&D used by the National Science Foundation in compiling its survey-based estimates have become widely accepted; they appear on p. 4-9 of Science and Engineering Indicators 2008, Vol. 1 (Arlington, VA: National Science Board/National Science Foundation, January 2008).

7THE NEED FOR INNOVATION

The basic difference in the innovation environment for

carboncaptureandPVtechnologiesissimpleenough:private

firms have little incentive to innovate in PCC or air capture

sincethereisnoCO2market(CO2issoldasagasinquantities

that are small relative to power plant emissions), while PV

manufacturershavebeeninnovatingforyearsbecauseofthe

incentives created by government-subsidized markets. PV

firms have focused in considerable part on bringing down

costs, while government-funded research has continued to

support technological foundations. Workshop participants

particularlycreditedtheSolarEnergyResearchInstitute(now

the National Renewable Energy Laboratory) for helping

advancePVtechnologyduringthe1980s.

Taken together, then, the CSPO/CATF work-shops reinforce the historical lessons of technological innovation: that the response to climate change will take the path of multiple small-scale innovations, manyofthem

next to invisible except to those directly involved, in many

loosely related energy-climate technologies. “Breakthrough”

discoveriesmayappear.Butasaparticipant in thePVwork-

shop put it, “We’ve been waiting for breakthroughs for

30 years; the portfolio approach is best.” Indeed, the only

post-World War II innovation in energy-climate technology

withtrulylarge-scaleimpactshasbeennuclearpower,andit

teachesequivocallessons,tobeexploredinlatersections.

Thepolicyquestion,then,ishowtomovedecisivelyalong

multiple pathways with a portfolio of policies and projects

to integrate superior technologies into the socio-economic

infrastructuresofmanynations,maximizing(speakingloosely)

GHG reductions while minimizing (again, speaking loosely)

disruption and costs.To accomplish this requires a systemic

view of innovation and innovation-enhancing policies,

includingpoliciesthatspeeddiffusion,applications,andnew

learning not only in the United States but globally. To the

extentthatallcountries,includinglargenon-Westerncountries

suchasChinaandIndia,becomeactivecentersofindigenous

innovation, the world will be better placed to find and

implementaneffectivesetofresponsestoclimatechange.

ThosewhoareconcernedaboutGHGstypicallyconsider

theroleoftechnologyintwoways.First,theyhaveemphasized

the need to displace fossil-fuel burning technologies with

currentlyavailabletechnologiesthataremoreefficient(such

as hybrid automobile powertrains), more energy-conserving

(such as better designed buildings), and cleaner (such as

renewable energy sources). Second, they have supported

investment in research that they hope will lead to radical

technologicalbreakthroughsoverthelongterm.

But these two perspectives, the former typically

advancedthroughregulation,thelatterthroughR&Dspend-

ing, fail to encompass the innovation activities that will be

necessary for transformationofenergyproductionanduse

to achieve significant emissions reductions. Innovation in

modern economies is continuous and feeds off both

researchandapplicationsexperience.Existingandavailable

technologieswillcontinuetoimprove,butnoonecanknow

how far and how fast.2 Breakthroughs are unpredictable,

andexpectingthemtomaterializeandsolvetheproblemis

unrealisticand irresponsible.What remains is the key challenge for energy innovation for the next 10-20 years and beyond: to implement policies that can ensure continual improvement in the performance of the broad suite of energy technologies upon which society depends. An important, but little-appreciated, reason why

innovationtoreduceGHGsmustbepursuedalongmultiple

pathways is the uncertainty endemic to innovation.

Uncertaintiesattachnotonly to technicalperformance (e.g.,

rates of improvement over time), but to trends in costs,

compatibility with other technologies already embedded in

the economy, the outcomes of competition among

technologies with similar applications (e.g., nanobatteries

relative to nanocapacitors for compact energy storage), and

acceptancebycustomersandsocietyatlarge.Thus,theproper

waytothinkaboutcontrolofatmosphericCO2isintermsof

systems—technological and innovation systems, energy

systems, the climate system itself, policy systems, social

systems,andtheworldsystemofinter-relatedeconomiesand

nation-states.Itistothisideathatwenowturn.


2Forexample,seeRogerPielke,Jr.,TomWigley,andChristopherGreen,“DangerousAssumptions,”Nature,Vol.452,April3,2008,pp.531-532.

2. Energy Innovation in Systems Context Innovations unfold over time through complex processes, which policy choices influence but do not control. The evolution of the photovoltaic industry illustrates interactions between technical change and market development.

An innovation system is a social system in which newapplications of technical knowledge take shape and diffuse.As such, an innovation system includes both organizationsthat generate and apply knowledge, and institutions thatguide,shape,andregulatethoseactivities.Theorganizationsinclude business firms, government agencies and govern-mentlaboratories,anduniversities.Whilegovernmentpaysformuchnewknowledge, inpartthroughfundingforresearch,privatefirmsgeneratemostinnovations.3

The system is not static, and not just because oftechnologicaladvance.Companiescomeandgo,forinstancethroughacquisitionofentrepreneurialstartups inrenewableenergy by larger firms, government policies change, and sodoes the innovation system and its subsystems. Industrialfirms aim at markets that shift over time. Some design andmanufactureequipment,suchasPVcellsandmodules,whileothers install them.ChemicalfirmsandequipmentsuppliersdevelopandsellproprietarytechnologiesforseparatingCO2or sulfur dioxide from power plant flue gases. Consultantsprovideengineeringservicesandfinancialintermediarieshelparrangefinancingforlarge-scaleprojects.Thesystemisdiverse. Institutionschangeovertimeintheirinformaldimensionstoo. This means that the context for innovation, includingincentives such as intellectual property rights, is not static.Patent law must be interpreted, and enforcement dependson legal precedents and on the practices of the Patent andTrademark Office. Engineers and scientists draw on aknowledge base that is only partially codified. Someknowledge has been reduced to textbook form, appears in

thetechnicalliterature,orissetdownincompany-proprietarydatabasesanddesignmanuals.Otherpartsoftheknowledgebase, such as rules-of-thumb widely known in the relevanttechnicalcommunity,maynothavebeenformalized(exceptperhaps in proprietary company documents) but arenonetheless crucial to practice. Successful innovators havelearnedtotapavailablesourcesofknowledge,someofitonoccasionnewbutmuchofitwidelyknown,forthedesignanddevelopment of marketed products and systems: dye-sensitizedPVcellsascommercializedin2003,theiPod™,andhybridautomobiles,anoldideaonlybroughttomarketinthelate 1990s. Business incentives drive these activities, andeffective innovation policies recognize that technology byitselfisnotnearlyenough. The innovation systems framework also has a place forinstitutions such as technical communities (e.g., of nuclearreactor engineers, of specialists in the design of chemicalprocesses) that serve as repositories of skills and expertise.Most of the people in these communities are industrialemployeesthoughsomeworkinuniversitiesorgovernment,andmuchofwhattheydo isonly loosely relatedtoscienceand research (e.g., because the tests of practical usefulnesshave precedence in innovation over the tests of fidelity tonatural phenomena that govern science). Largely throughinformal mechanisms, technical communities generate,validate, and disseminate knowledge and methods deemed“best practices” in their specialized areas. These practicesunderlie, for example, the design of mass-produced PVmodules,nuclearreactorsafetysystems,andsteamturbines.

9ENERGY INNOVATION IN SYSTEMS CONTEXT

3Formoreontheinnovationsystemsframeworkadoptedforthisproject,includinganoverviewofgovernmentpoliciesaffectinginnovation,seeJohnAlic,“EnergyInnovationSystemsfromtheBottomUp:ProjectBackgroundPaper,”CSPO/CATF,March2009,<www.cspo.org/projects/eisbu/>,andthecitationstherein.

Theinnovationsystem,inotherwords,isdynamicaswell

asdiverse.Considerphotovoltaics:TheworldwidePVindustry

is intenselycompetitive, thecompetitioncenteringonprice

and technical performance, notably efficiency (as measured

by the fraction of the energy in incident sunlight converted

to electricity). Although government-sponsored R&D (and

purchases,includingprocurementsofadvancedcelltypesfor

spacecraft)hasbeenamajorlong-termstimulustoinnovation,

inrecentyearssubsidieshavebeenamoreprominentfeature

of the PV innovation system (Box A). With sales increasing

thanks to policies including tax credits and feed-in tariffs

(guaranteed prices for PV electricity supplied to the grid, a

policyadoptedbyseveralcountriesinEurope),manufacturers

haveinvestedinR&D,soughttodeveloplower-costfabrication

methodsforhigh-efficiencycelltypes,andbuiltnewfactories.

Even so, PV-generated electricity remains more costly than

alternatives,absent subsidies.4Because incentivescomeand

go, and PV sales respond, manufacturers must factor policy

uncertainties into their R&D and investment plans. Such

uncertainties are an inherent part of the innovation system:

nearly500billsconcernedwithenergyefficiencyorrenewables

were introduced in the 110th Congress, dozens on solar

energyalone.5

Politicscreatesadditionaluncertainties.Boththefederal

government and many states have offered tax credits for

businessandhouseholdpurchasesofsolarPVsystems.But

thePVindustryishardlyalone:ineffect,theU.S.government

subsidizesallformsofenergy,andhasdonesofordecades.

Somethinglikehalfofthenationwideincreaseinelectrical

generating capacity during the 1930s came from dams

built and still operated by public agencies (flood control

and economic development were primary motives).

ProddedbyCongress’s JointCommitteeonAtomicEnergy,

the Atomic Energy Commission (AEC, ancestor of the

Department of Energy, DOE) provided direct financial

assistanceforutilityinvestmentsinnuclearpowerbeginning

inthe1950sandthefederalgovernmenthascontinuedto

support R&D, manage nuclear fuel supplies, and insure

against liability claims under the Price-Anderson Act. As

Table 1 shows, subsidies for nuclear power do not match

thoseforcoal,whilerenewableshaverecentlybeenatthe

topofthe list intermsofsubsidyvaluerelativetoelectrical

output.(Treatingcoal-basedsyntheticfuelsseparately,rather

thanlumpedwithothersubsidiesforcoal,yieldsafigureeven

higherthanthatforsolarenergy.)

The U.S. innovation system is messy and complicated.

Noonecanknowhowadecisionmade,say, tosubsidizea

particulartechnologywillinteractwithothersuchdecisions

(not to mention continually evolving knowledge and

institutions)toinfluencediffusionandimprovementofthat

technology. The global innovation system is even more

complex. Countries including China, for instance, are

beginning to innovate incrementally in technologies for

coal-burningelectricalpowerplants,andoftencandosoat

costs lower than in the United States. Yet decisions must

continually be made, by public officials, by business

executives,engineeringconsultants,andmore,withaneye

towardtheoutcomeseachseeks.


4SincethereisnomarketpriceforPV-generatedelectricity(powerisgenerallysoldintheformofutilitybuy-backs),costestimatesnecessarilyreflectaccountingassumptionsfordepreciationandusefullife,oftentakentobearound30years,andcapacityfactor,afunctionofavailablesunlight,hencegeographiclocation.ThemarketresearchfirmSolarbuzzputs2009costsintheU.S.Sunbeltatabout37½centsperkilowatt-hour(¢/kWh)forresidentialinstallations,severaltimesutilitypricingofabout10½¢/kWh(businessandindustrialcustomerspayless).“SolarElectricityPrices,”February2009,<www.solarbuzz.com>.5FredSissine,LynnJ.Cunningham,andMarkGurevitz,Energy Efficiency and Renewable Energy Legislation in the 100th Congress,ReportRL33831(Washington,DC:CongressionalResearchService,November13,2008).

ENERGY SOURCE

SolarPVandThermalb $14 $24.30

Wind 720 23.40

Nuclear 1270 1.60

Coalc 3010 1.50

Biomass/Biofuels 36 0.89

Hydroelectric 170 0.67

OilandGas 230 0.25

Allsourcesd $6750 $1.70

Table 1. U.S. Government Subsidies for Electricity Production, 2007 a

TOTAL PER UNIT OF OUTPUT (MILLIONS OF DOLLARS) (DOLLARS PER MEGAWATT-HOUR)

aIncludesbothdirectfederalspendingandtaxexpenditures.Doesnotincludestateorlocalgovernmentsubsidiesornon-electricityenergysubsidiesestimatedat$9.8billion,orsubsidiesfortransmissionanddistribution,estimatedat$1.2billion.bPVandthermalnotavailableseparately.cIncludingsubsidiesforgasificationandliquefaction,whichaccountfor3½percentofcoal-generatedelectricityand72percentofsubsidiesforcoal.dOtherrenewables,includinggeothermal,landfillgas,andmunicipalsolidwaste.

Source:Federal Financial Interventions and Subsidies in Energy Markets 2007(Washington,DC:DepartmentofEnergy,EnergyInformationAdministration,April2008),Table35,p.106.

ESTIMATED SUBSIDY VALUE

11ENERGY INNOVATION IN SYSTEMS CONTEXT

Box APhotovoltaics

Technology

PVcellsarefabricatedfromsemiconductingmaterialssuchassilicon,asaretransistorsandintegratedcircuits.Government-

sponsoredsatelliteprogramssupportedtheearlydevelopmentofsolarcells,whichfoundtheirfirstapplicationin1958

aboard Vanguard I. Interest in terrestrial power blossomed with surging energy prices in the 1970s, and production

continuestogrowatdouble-digitrates,albeitfromwhatisstillasmallbase.Sincethelate1970s,theU.S.governmenthas

spentperhaps$3billiononPV-relatedR&D.a

Globally,morethan250companiesnowmakeorsellPVcellsorassemblies;oneintendoessointheUnitedStates.bMany

countrymarketshavebeenheavilysubsidizedinrecentyears.Germany,hardlyfavoredforsolarpowerincloudynorthern

Europe,accountedfornearlyhalfofworldPVsalesin2007,thankstolucrativefinancialincentives,followedbySpainand

Japan,thentheUnitedStates,withabout8percentoftheworldmarket.cWiththe2008-2009recession,manyfirmshave

cutbackonexpansionplansandreducedpricestokeepproductionlinesrunning.d

Overthelongterm,cellandmodulecostshavecomedownasaresultofinnovation(cellsormodulestypicallyaccount

foraroundhalfthecostofaninstalledPVsystem),andfurtheradvancesinPVtechnologiescanbeexpected.Thereare

many possible pathways, through advances in materials (e.g. compound and organic semiconductors), fabrication

processes,andcellconfigurationstailoredthroughmicrostructuralandnanostructuralengineering.Universityresearch

continues to be important in exploring exotic cell structures such as quantum dots. With rapid market expansion,

companies have moved new technologies into production quite rapidly.Thus while familiar crystalline solar cells still

accountforaboutfour-fifthsofthemarket,thinfilmsofmoreexoticmaterialsdepositedonsubstratesofsteel,glass,or

plasticdoubledinsalesfrom2006to2007asnewproductioncapacitycameonline.

Policy

Intensecompetitioninworldmarkets,manyofthemheavilysubsidized,hasfedinnovation,arguablyreducingtheneed

forcontinuinggovernment-supportedresearchexceptforlong-term,highpotentialpayoffworkthatprivatefirmsdeem

toouncertainorrisky.Thisissmallscience,typically,andassuchisbestconductedinanacademicsettingwithfunding

fromagenciesthatrelyonpeerreviewforselectingamongcompetingproposals,suchastheNationalScienceFoundation

(NSF), since even the better government laboratories tend to be more subject than NSF to political and bureaucratic

considerationsthatcaninfluenceresearchprioritiesanddirections,tothedetrimentofinnovation.

ContinuinginnovationinPVsystemswillalsodependonlearningthroughexperienceasinstallationscontinuetogrow

inscaleandfeedelectricityintothegrid.Regulatorsneedtomakesureutilitiesdonotputunnecessaryobstaclesinthe

wayofgrid-connectedPVinstallations,whichhavebeengrowingveryrapidlyinsomestates,especiallyCalifornia,and

plannersmusttakesuchinstallationsintoaccountinrenovatingtheelectricpowergridanddeveloping“smartgrid”

systems for the future. Policymakers should also continue to encourage government procurements, particularly of

advancedPVtechnologiesthatprivateinvestorsmightavoid,asaspurtocontinuinginnovationandlearning.

aEstimatedbasedonRenewable Energy: DOE’s Funding and Markets for Wind Energy and Solar Cell Technologies,GAO/RCED-99-130(Washington,DC:GeneralAccountingOffice,May1999)andFederal Electricity Subsidies: Information on Research Funding, Tax Expenditures, and Other Activities That Support Electricity Production,GAO-08-102(Washington,DC:GovernmentAccountabilityOffice,October2007).DOEhassuppliedthebulkofR&Dfundinginrecentyears;theDefenseDepartment,theNationalScienceFoundation,andtheNationalAeronauticsandSpaceAdministrationalsosupportPVR&D.ForfurtherdiscussionofPVmarketsandtechnology,see“WorkshopBackgroundPaper:Photovoltaics,”CSPO/CATF,March2009,<www.cspo.org/projects/eisbu/>,whichincludesadditionalcitations.b“SolarPhotovoltaicCell/ModuleManufacturingActivities2007,”DepartmentofEnergy,Energy InformationAdministration,Washington,DC,December2008,<www.eia.doe.gov/cneaf/solar.renewables/page/solarreport/solarpv.pdf>.cPVsales inGermanymadeup47percentofareported2007worldtotalof$17.2billion.“Marketbuzz™2008:AnnualWorldSolarPhotovoltaic IndustryReport,” March 17, 2008, <www.solarbuzz.com>.With falling government revenues caused by recession, some countries have reduced their incentives;Spain,forinstance,whichofferedfinancialbenefitstothefirst2400MWofnewPVinstallationsin2008willsubsidizeonly500MWofsuchprojectsin2009.dSomereportspredictthataveragepricingmayfallbyhalf,fromanaverageofnearly$4perwattin2008toaround$2perwattin2009.“SolarPowerIndustryDeclines,LeadstoDropinPrices,”Wall Street Journal,May11,2009.

The energy system, like the innovation system justdescribed, is large and complicated, but in its own way.Successful technologyand innovationpoliciesmustattendto interactions within and between these two systems.Planning,design,andconstructionofpowerplantsdrawona wide range of skills that must be pulled together andintegrated, an organizational task that may involve thecoordinated activities of dozens or perhaps hundreds offirms. Utilities commonly hire engineering firms for thesetasks, much as overall responsibility for a military systemsuchasatelecommunicationsnetworkmightbeassignedaprimecontractoroverseenbyaPentagonsystemsofficeorafederally-fundedresearchanddevelopmentcenter(FFRDC). If utility-scale technologies, such as post-combustioncaptureofCO2atcoal-firedpowerplants,aretoplayapartinGHG reduction, then systems integrations and project

management will be major tasks, much more difficult than,say, adding sulfur dioxide scrubbers to meet Clean Air Actmandates.(Sulfurdioxidefromcoal-burningpowerplants,thechief cause of acid rain, has been regulated since the 1970sundertheCleanAirActandamendments.) As indicated in Box B, carbon capture and storage willinvolve power companies, equipment vendors, perhapschemicalfirmswithproprietaryCO2separationtechnologies,andengineeringconsultantsandcontractors.Mostutilitiesdonot maintain large engineering staffs, since they build newplants irregularly.They rely more heavily than firms in manyother industries on outside technical expertise. This can beproblematic,assuggestedbymisstepswithnuclearpowerinthe 1960s and 1970s, an episode with important lessons forenergy-climateinnovation.


Box BPost-Combustion Capture of CO2 from Coal-Burning Power Plants

Technology

CoalconsistsprimarilyofcarbonandburningatonofcoalreleasesabouttwotonsofCO2.In2007,coal-firedplants—

fewerthan1500boiler-turbine-generatorunitsonperhaps500sites—generated48.5percentofU.S.electricalpowerand

morethan35percentofthenation’sCO2emissions.aLiketheUnitedStates,ChinaandIndiahaveabundantcoalreserves

thatcanbecheaplyminedforproducinglow-costelectricalpower.Chinaisputtingupnewcoal-burningplantsatahigh

rate and India seems poised to follow within the next decade. Unless PCC technology is reduced to practice and

implementedsoon,itwillbeverydifficulttostabilize,muchlessbringdown,atmosphericconcentrationsofCO2.

Therearetwobasicwaysof reducingoreliminatingtheCO2producedwhencoalburns.Thecoalcanbegasified, for

instanceinanintegratedgasificationcombinedcycle(IGCC)plant,withtheCO2removedpriortocombustion.OrCO2can

beremovedaftercoalisburned.Thesecondrouteistechnologicallystraightforwardandatleastinprinciplewouldpermit

3. Managing Innovation Because technological systems are complex and constantly evolving, managing them is difficult. Post-combustion capture of carbon dioxide presents an opportunity to learn from past, cautionary experiences such as nuclear power, mismanaged by both public and private sectors.

13MANAGING INNOVATION

existingcoal-firedpowerplantstoberetrofitted.(IGCCwouldalmostcertainlyrequirenewconstruction;so,mostlikely,

wouldathirdalternative,oxyfuelcombustion,whichburnscoalinnearlypureoxygensoastoleavefluegasesconsisting

ofnearlypureCO2tofacilitateseparation.)Anyofthesepathswouldbecostly.AddingPCCtoanaverage-sizeU.S.power

plantwouldprobablyrequireaninitialinvestmentintherangeof$500million.Operatingcostswouldincreasesubstantially,

inpartbecauseaconsiderablefractionoftheelectricitygeneratedwouldbeconsumedinseparatingouttheCO2and

compressingitfortransportandsequestration.

Separationprocessesofasortwidelyusedinindustryforotherpurposesandwellunderstoodbychemistsandchemical

engineerscanremove90percentormoreoftheCO2influegas.Gasseparationisastandardprocessinthechemical

industry,withmanythousandsofplantsoperatingworldwidetoproduceindustrialgasesforsale,includingCO2,which

hasvalueinusesthatrangefromcarbonatingbeveragestoshieldingweldingarcsandenhancedoilrecovery.Because

thesemarketsaresmallrelativetoanthropogenicCO2emissions,experiencetransfersonlypartially,andprocessessuchas

scrubbingfluegaseswithamines(compoundsrelatedtoammonia,whichbindtheCO2forlaterseparation)haveyetto

bedemonstratedonthescaleoftypicalpowerplants.Long-termsequestrationofhighlycompressedCO2wouldlikewise

needfurtherdemonstrationforanyCCSoption.Nonetheless,themajorobstaclestoPCCappeartolieinthecosts,notin

technologiesforeithercaptureorstorage.Expensivenewequipmentwouldbeneeded,costlytooperateaswellasto

build.Electricitycostswouldrise.Indeed,theymightdouble.

Proprietaryamine-basedprocessesforseparatingCO2fromnitrogen(theprincipalconstituentinair,andhenceinCO2-

heavyfluegases)havebeenavailablefordecades.Theyworksomethinglikesulfurdioxidescrubbing.Allcoalcontainsup

toafewpercentsulfur,whichcombineswithoxygenduringcombustiontoformsulfurdioxide.Inthescrubber,sulfur

dioxide reacts chemically with another substance to form a solid that can be disposed of. Power companies began

installingsulfurdioxidescrubbersseveraldecadesago;withexperience,costshavecomedownandperformancehas

improved.Aminescrubbers,somewhatsimilarly,passfluegasesthroughasolutionofanaminecompound,generallyin

water,toabsorb(i.e.,dissolve)CO2.Inadownstreamstage,theCO2isreleased(“stripped”),leavingarelativelypuregasto

becompressedfortransportandstorage,withtheaminesolutionregeneratedforreuse.A500MWplantthatproduces

10 tonsperhourofsulfurdioxidemightemitsome500tonsperhourofCO2.Thusequipmentofmuch largersize is

neededandbothfirstcostsandoperatingcostswillbemuchgreater.

Therearetwoprimaryreasonsforincreasedoperatingcosts.InmostofthePCCprocessessofarenvisioned,steamfrom

theboilerwouldbebledoffforprocessheat(e.g.,tostriptheCO2fromsolution).Energythatwouldotherwisedrivethe

turbine to generate electricity will be lost. (In retrofits, moreover, the turbine may have to be operated off of design

conditions,resultinginfurtherlosses.)Second,electricityequivalenttoasignificantportionoftheplant’selectricaloutput

willbeconsumedfordrivingcompressorsandpumps,notablyforraisingthepressureoftheCO2toperhaps2000pounds

persquareinch(over100timesatmosphericpressure)priortotransportandstorage.These“parasitic”lossescouldamount

to30percentoftheelectricaloutputotherwiseavailable.

In the absence of utility-scale demonstrations, cost estimates are uncertain. Engineering studies prepared by the

DepartmentofEnergy(DOE)forrepresentativecasesofretrofitstoanexistingcoal-firedplantandforanew“greenfield”

plant,withandwithoutwhatisdescribedas“advancedamine-basedcapturetechnology,”yieldanestimatedincremental

costof6.9¢perkilowatt-hour(kWh)fortheretrofitcaseand5.5¢perkWhforanewplant.bThesecanbecomparedwith

generatingcostsforatypicalpulverizedcoalplant,putbyDOEat6.4¢perkWh.

Costswouldprobablydeclinesomewhatovertime,butgasseparationisarelativelymaturetechnologyandnoneofthe

alternativestoamineseparationunderinvestigationappeartoholdsubstantialpromiseofmajor,ratherthanincremental,

gains.Thesealternativesincludedifferentaminecompoundsandcombinationsofamines,ammoniaasasolventinstead

ofanamine,distillation,membranesthatpassCO2preferentially,andporoussolidstoadsorbit.

Inadditiontocostincreases,retrofittingofexistingpulverizedcoalplantswouldsharplyreducegeneratingcapacity,while

retrofittingmaybe impossibleatsomesites,perhapsaconsiderablenumber, for lackofspace(groundareaoccupied

mightnearlydouble).Someofthetechnicalcompromisesnecessaryinretrofitscouldbeavoidedfornewplants,butnot

thefundamentalissueofhighinvestmentandoperatingcosts.GasifyingcoalandremovingtheCO2beforecombustion,

ratherthanatthe“endofthepipe,”holdsmorepromiseforgreenfieldconstruction.

Policy

Coal-fired power plants emit huge tonnages of CO2. Equipping such plants to control CO2 emissions will drastically

diminishtheircostadvantagesoverothergeneratingtechnologies,perhapsraisingthecostsabovesomealternatives.

Governmenttechnologyandinnovationpoliciesshouldsupportlong-termR&Danddemonstrationaimedatsubstantial

improvementsinPCCandCCS(e.g.,pre-combustiongasification),butwithouttheexpectationofbreakthroughs(which

arepossiblebutbynomeansassured),andathigheroverallthermalcycleefficiencies,whichmoderateemissionssince

lesscoalmustbeburnedtogenerateagivenamountofelectricalpower.

Tothispoint,businessandfinancingarrangementsforimplementingPCChavehardlybeenexplored.Chemicalcompanies

andequipmentsuppliershavehadlittleincentivetopushforwardwithengineeringdevelopmentanddemonstration.

Alternativesforgovernmentincludesimplypayingsomeorallofthecostsormandatinginstallationandallowingthe

markettodeterminehowcostswouldbeapportionedandrevenuesraised(e.g.,throughhigherratesforelectricity).

aElectric Power Annual 2007 (Washington,DC:DepartmentofEnergy,EnergyInformationAdministration,January2009).Nuclearandnaturalgas-firedplantsproducemostoftheremainderofU.S.electricalpower,some41percentaboutequallydividedbetweenthetwo.Renewablesourcestrailwellbehind,at8½percent,andmostofthisishydropower.PVinstallationsaccountforperhaps0.05percentofU.S.electricityconsumption.ForfurtherdiscussionofPCC,see“WorkshopBackgroundPaper:Post-CombustionCapture,”CSPO/CATF,April2009,<www.cspo.org/projects/eisbu/>.b“CarbonDioxideCapturefromExistingCoal-FiredPowerPlants,”DOE/NETL-401/110907,November2007,DepartmentofEnergy,NationalEnergyTechnol-ogyLaboratory,<www.netl.doe.gov>.


Firmsinalmostanyindustrymustcopewithtechnological

changeinordertosurvive,eveniftheydonotseektoinnovate

themselves. Businesses engage in R&D both to generate

innovations internallyandtosustaintheircapacity to locate,

evaluate,andexploittechnologiesavailableexternally—from

otherfirms,fromuniversityresearchgroups,fromgovernment

laboratories.Regulatorychangesandsubsequentrestructuring

oftheelectricutilityindustryduringthe1990sevidentlyledto

declines in R&D spending and technical employment, as

shown in Figure 1. According to power company managers

and industry analysts, regulatory shifts made it harder for

utilitiestorecoverR&Dexpenditures.Infact,theutilityindustry

hasneverspentmuchonR&D.Intheearly1990s,whenR&D

acrossallU.S.industriesaveraged3.2percentofsales,utilities

spent less than 0.2 percent of revenues. More recently, the

industry-wide average increased somewhat, to about 3.4

percentofrevenues,whilethefigureforutilityR&Ddroppedto

only0.1percent.6

Figure 1. Utility R&D Spending and Technical Employment

Notes:Series start in1991 (noemploymentfigures reported for1996and1998);2004latest year available. Non-federal R&D only. Data covers all utilities (SIC [StandardIndustrialClassification]49,“Electric,gas,&sanitaryservices,”through1999;NAICS[NorthAmerican Industry Classification System] 22,“Utilities,” for later years), however electricpowercompaniesaccountforaboutthree-quartersofemploymentinthelargerutilityindustry.Mostutilitycompanyemployeesoccupationallyclassifiedasengineersworkinoperationsandmaintenance,notinR&D.

Source:NationalScienceFoundation,DivisionofScienceResourcesStatistics,SurveyofIndustrialR&D,variousyears.

6Research and Development in Industry: 1993,NSF96-304(Arlington,VA:NationalScienceFoundation,1996),TableA-2,p.12;Research and Development in Industry: 2004,NSF09-301(Arlington,VA:NationalScienceFoundation,December2008),Table21,p.72.

500

1000

1500

2000

0

100

200

300

400

1990 1995 2000 2005

R&D

Spe

ndin

g (M

illio

ns o

f yea

r 200

0 do

llars

)

R&D

Per

sonn

el

Number of R&D Scientists and EngineersReal (Inflation-adjusted) R&D Spending

Box CNuclear Power: Mismanaged by Government and Utilities

America’snuclearpowerplants, anoutgrowthofColdWargeopoliticsand themilitary-centeredColdWar innovation

system,haveperformedremarkablywellinrecentyears.The100-plusplantsinoperationproduceaboutone-fifthofthe

nation’selectricity,over20timesthecontributionofwindandsolar,makingnuclearpowerthesolepost-WorldWar II

energyinnovationoftrulylargescale.Nuclearplantsnowoperatewithcapacityfactors—aprincipalmeasureofreliability,

correspondingtothefractionoftimeaplantremainsonlinedeliveringelectricity—averaging92percent,wellabovethat

forcoal-firedgeneration.Overthedecadeofthe1970s,bycontrast,capacityfactorsfornuclearplantsaveragedonly60

percent,emblematicofthetroubledearlyyears.a

Untilthelate1950s,utilitiesshowedlittleinterestinnuclearelectricity.Without,asyet,restrictionsonairpollutionand

givenlargeU.S.reservesofcoal,allprojectionsshowedthatfossilfuelscouldproducelower-costpowerthannuclear

plantsfordecadestocome.Yetinjustafewyearsexcitementovernuclearelectricityhadreplacedthedisinterestof

earlieryears,andthesalesmanshipoftheAtomicEnergyCommission,Congress,andequipmentsuppliersspurreda

waveofinvestment.

Firms fund internal R&D so they have the capacity tomake smarter decisions.When companies buy technology,whetheraPVsystemfortheroofofanofficebuilding,orPCCequipment,shouldthatbemandated,theymusttrytojudgesupplier claims; and it may take considerable technicalexpertise to be a“smart customer.” Can a newly developedthinfilmPVmodulebeexpected to last for threedecades,thetypicallifetimeofcrystallinecells?Orwillitsperformance,as measured by efficiency losses over time, degrade morerapidly? Might a superior cell module reach the market inanother six months? Should the recommendations of aconsulting firm to buy amine separation technology fromonechemicalcompanyratherthananotherbetrusted?Howreliablearetheconsultant’scostestimateslikelytobe?Whatarethechancesconstructioncontractorswillcompletetheirworkontime?Thesewillbecriticalquestionsforutilities inthe future, just as they were for nuclear power in the past.Manyutilities(notall)wentbadlywrongatthattime,anditappearstheindustryhasevenlesstechnicalcapacitynow. Asymmetries intechnicalknowledgebetweensuppliersofspecializedgoodsorservicesandtheircustomersarenor-mal:thatisthebasisformarkettransactionsintechnology.7Atthe same time, technologically naive customers may makebadchoicesorinviteexploitation.8Therearenolistpricesforcomplexpiecesofcustom-designedhardware,suchassteam

turbinesornuclearreactors,muchlessfor“know-how.”Termsmustbenegotiated,andsuppliersbeginwiththeadvantageofdeepunderstandingoftheirowntechnologies.Ifintellec-tual property protection is strong, customers may not beprivytodetailsbeforepaying;ifdisappointed,theymayhavenorecourse.Shouldregulatorymandatesforcepowercom-paniesto invest inCCS,theywillhavetoassessproprietarytechnologies available for licensing and select capablesuppliers. Inthe1960s,utilitiesplungedintonuclearpowerdespite evidence that such investments would not payoff. In effect, utility company managers were gullible andthe AEC and suppliers of reactors, technical services, andconstructionmanagementsoldthemabillofgoods. Ithastakendecadestoovercometheseearlymistakes(BoxC). The presumptive decline in internal technical capacitiesofutilitiessuggestedbyFigure1couldcreatesimilardifficultiesformanagementofPCCorotherapproachestoGHGcontrol.

If policymakers—and utilities— cannot do a better job than during the heyday of nuclear power in the 1960s, decarbonization could be slowed substantially.

15MANAGING INNOVATION

7AshishArora,AndreaFosfuri,andAlfonsoGambardella,Markets for Technology: The Economics of Innovation and Corporate Strategy(Cambridge,MA:MITPress,2001).8Notafeworganizationshavecontractedforinformationtechnologysystemspoorlymatchedtotheiroperations,especiallyhospitalsandclinics,lesssophisticatedcustomersthantypicalpurchasersinothereconomicsectors.Forrecentexamples(expressedcircumspectly),seeChadTerhune,KeithEpstein,andCatherineArnst,“TheDubiousPromiseofDigitalMedicine,”Business Week,May4,2009,pp.31-37.

BothGeneralElectric(GE)andWestinghousehaddevelopedreactorsfortheU.S.Navy’snuclear-poweredsubmarines

underAECcontract,andAdmiralHymanG.Rickoveroversawdesignandconstructionofthefirstcommercialplant,a

heavilysubsidizedsmall-scaledemonstrationatShippingport,Pennsylvania.TheShippingportplant, incorporatinga

Westinghousereactor,wenton lineat theendof1957andfunctionedwell.Yetnothingmuchhadchanged inthe

economicsofnuclearpower:objectiveestimatescouldonlyshowitwouldremainmorecostlythanalternativesfor

manymoreyears.bNonetheless,utilitiesrushedtoinvestinthemiddle1960s,orderingmorethan60nuclearplantsin

justtwoyears:1966and1967.Eventhesmallestweredesignedtoproducefarmorepowerthanexistingreactors,and

thoseonthedrawingboardscontinuedtogroweventhoughutilitieshad,asyet,accumulatedhardlyanyoperating

experience:in1967,thecapacityofreactorsonorderexceededthecapacityofthosethathadbeencompletedby25-

30times.cConstructionschedulesslipped,sometimesbyyears,andcostsrose,sometimesbytwoorthreetimesover

initialestimates(inconsiderablepartbecauseofescalatingsafetyrequirements).Whencompleted,thebestnuclear

plantsperformedaswellasthebestfossil-fueledplants.Theworstweretrulyabysmal.Thegapincostsandreliability

between the best plants and the worst did not represent growing pains for an immature technology: it reflected

miscalculationsand inadequateoversightbypowercompaniesandpolicy failuresbygovernment,whichpusheda

complexnewtechnologyintoamarketplaceinwhichitcouldnotcompetewithoutsubsidies.

TheAECfumbledthemoveintonuclearpowertechnicallybecausetheCommissionersandstaff,undergreatpressure

fromCongressandthearmedforcestoexpandthe inventoryofatomicbombsanddevelopahydrogenbomb,were

almost entirely consumed by those tasks. From the beginning, the military had opposed civilian control of nuclear

warheads,whichtheyconsideredweaponslikeanyothers;seriousmisstepsbytheAECwouldhavehelpedtheservices

maketheircasetotheWhiteHouseandCongress.Surpassingallexpectations,theAECbuiltupthestockpilefromafew

hundredwarheadsatthebeginningofthe1950stoover20,000attheendofthatdecade,testedahydrogenbombin

1952andconductedthefirstairdropofthenewweaponin1956.Meanwhile,reactorR&Dsufferedfromlackofattention

andresources.Ofmanypossiblereactorconfigurations,onlyafewwereexplored,andtheseeitherbecausetheypromised

toboostproductionoffissionablematerial forbombsorbecauseof lobbyingby laboratoryscientistswhowantedto

explorereactorphysicsasaresearchtopicratherthanforpowergeneration.Astheofficialhistoryputsit,whiletheAEC’s

reactorR&Dprogram“appearedrationalandcomprehensive,”it“lackedfocus”and“offerednosimple,direct,andpredictable

routetonuclearpower.”dAsaresult,whentheAEC,aftermultiplefailurestostiruputilityinterest,receivedanacceptable

proposal fromDuquesneLightCompanyfortheShippingportplant, ithadnorealalternativebuttoadopttheNavy’s

light-waterreactortechnology.TheAECpersuadedRickovertotakeontheShippingportproject,settingtheU.S.industry

on course for near-universal adoption of light-water reactors, not necessarily a good choice for commercial power

generation—butavailable.Intheyearsthatfollowed,bothWestinghouseandGEreportedlysetpriceswellbelowtheir

owncostsineffortstoestablishdominanceinwhattheybelievedwouldbecomealucrativeinternationalmarket.Soon

therewashardlyanyincentivetopursuealternativeapproachestocommercialnuclearpower.

aElectric Power Annual 2007(Washington,DC:DepartmentofEnergy,EnergyInformationAdministration,January2009),TableA6,p.102;Nuclear Power in an Age of Uncertainty(Washington,DC:OfficeofTechnologyAssessment,February1984),p.89.Capacityfactorsforcoal-firedplantsaveragedabout74percentin2007,butthisfigureincludesplantsrunonlyduringperiodsofhighdemand,whereasnuclearplantssupplybase-loadpowerandoperatemore-or-lesscontinuallyunlessshutdownforscheduledorunscheduledmaintenanceorrepair.bEvenAECchairmanLewisStrauss,onceashoesalesman,whoisrememberedfordeclaringthatnuclearelectricitywouldbecome“toocheaptometer,”wascarefultoputthetimewellinthefuture.BrianBalogh,Chain Reaction: Expert Debate and Public Participation in American Commercial Nuclear Power, 1945-1975(Cambridge,UK:CambridgeUniversityPress,1991),p.113.cIrvinC.BuppandJean-ClaudeDerian,Light Water: How the Nuclear Dream Dissolved(NewYork:Basic,1978),p.74.Elsewheretheseauthorswrite:“Wefoundnoindicationthatanyoneraisedthe…fundamentalpointoftheuncertaintyinmakingcostestimatesfornuclearplantsforwhichtherewaslittlepriorconstructionexperience….”(p.46).“[W]hatwasmissing…wasindependentanalysisofactualcostexperience”(p.76).“Thedistinctionbetweencostrecordsandcostestimates…eludedmanyingovernmentandindustryforyears”(p.71).dRichardG.HewlettandJackM.Holl,Atoms for Peace and War, 1953-1961: Eisenhower and the Atomic Energy Commission(Berkeley:UniversityofCaliforniaPress,1989),p.251.Earlierintheirexhaustiveaccount,coveringpowerreactorR&Danddemonstrationalongwithmanyothertopics,theauthorswritethat“Thedivisionofreactordevelopment…hadbeenforcedtoconcentrateitseffortsalmostentirelyonproduction[offissionablematerialforbombs]andmilitarypropulsionreactors.Notmuchmorethanone-tenthoftheoperatingfundsforreactordevelopmentweregoingdirectlyintopowerreactorprojects”(p.23).


The complexities of innovation systems and their

management must be grasped and mastered to develop

effective energy-climate policies. In past episodes of fast-

paced innovation, government policies have been crucial

catalysts.Whatcanbelearnedfromcasessuchasinformation

technology?TheITrevolutionstemmedfromapairoftruly

radicaltechnologies:electroniccomputersrunningsoftware

programsstoredinmemory,andthesolid-statecomponents,

transistorsandintegratedcircuits(ICs)thatbecameaprimary

source of seemingly limitless performance increases.These

spawnedcountlessfurtherinnovationsthattransformedthe

products of many industries and the internal processes of

businessesworldwide.

4.1 Why Energy Is Not Like IT The needed revolution in energy-related technologies

willnecessarilyproceeddifferently.Therearetwofundamental

reasons.First,thelawsofnatureimposeceilings—impenetrable

ceilings—on all energy conversion processes, whereas

performancegainsinITfacenosimilarlimits.Second,digital

systems were fundamentally new in the 1950s. To a minor

extent they replaced existing “technologies”—paper-and-

pencil mathematics, punched card business information

systems.Tofargreaterextent,theymadepossiblewhollynew

end-products, indeedcreatedmarkets for them.Bycontrast,

energy isacommodity,new“products”willconsistsimplyof

newwaysofconvertingenergyfromoneformtoanother,and

costsaremore likelytorisethandecline,at least inthenear

term,asaresultofinnovationsthatreduceGHGemissions.

ITperformancegainshaveoftenbeenportrayedinterms

of Moore’s Law, the well-known observation that IC density

(e.g.,thenumberoftransistorsperchip,nowinthehundreds

ofmillions)doubleseverytwoyearsorso.Sinceper-chipcosts

have not changed much over time, increases in IC density

translatedirectlyintomoreperformanceperdollar.Andwhile

ICdensitywillultimatelybe limitedbyquantumeffects, the

ceiling remains well ahead, even though performance has

alreadyimprovedbyeightornineordersofmagnitude.

Forconversionofenergyfromoneformtoanother,by

contrast,whethersunlightintoelectricityorchemicalenergy

storedincoal intoheat(e.g., intheboilerofapowerplant)

andthenintoelectricity(inaturbo-generator),fundamental

physicallawsdictatethatsomeenergywillbelost:efficiency

cannot reach 100 percent. Solar energy may be abundant,

butonlyafractionoftheenergyconveyedbysunlightcan

beturnedintoelectricalpowerandonlyintheearliestyears

of PV technology was improvement by a single order-of-

magnitudepossible (asefficiencypassed10percent).Even

though both PV cells and IC chips are built on knowledge

foundations rooted in semiconductor physics, PV systems

operateunderfundamentallydifferentconstraints.Todaythe

bestcommercialPVcellsexhibitefficienciesintherangeof

15-20 percent (considerably higher figures have been

achieved in the laboratory). After more than a century of

innovation, the best steam power plants reach about 40

percent, somewhat higher in combined cycle plants (in

which gas turbines coupled with steam turbines produce

greater output). Limited possibilities for performance gains

translate into modest prospects for cost reductions, and in

some cases innovations to reduce, control, or ameliorate

GHGsimplyreductionsinperformanceonfamiliarmeasures,

such as electricity costs, as already recounted for power

plantsfittedforcarboncapture.

Whileenergyisacommodity,andPVsystemscompete

withotherenergyconversiontechnologiesmore-or-lessdi-

rectly(generatorsforoff-gridpower,windturbinesandsolar

17ENERGY INNOVATION COMPARED TO INFORMATION TECHNOLOGY

4. Energy Innovation Compared to Information Technology Despite fundamental differences, the IT revolution holds lessons for energy-climate innovation. Perhaps the most important is the role of government as a demanding and technologically sophisticated customer.

thermal for grid-connected applications), successive waves

of IT products have performed new tasks, many of which

wouldearlierhavebeenallbutinconceivable.Semiconduc-

torfirmsdesignedearlyICchipsinresponsetogovernment

demand for very challenging and very costly defense and

spaceapplications,suchasintercontinentalmissilesandthe

Apolloguidanceandcontrolsystem.Withinafewyears,they

weresellinginexpensivechipsforconsumerproducts.Sales

to companies making transistor radios paved the way for

salestoTVmanufacturersatatimewhencolor(commercial-

izedinthelate1940sandslowtofindamarket)wasreplacing

black-and-white(colorTVsales intheUnitedStatesdoubled

duringthe1970s).ICchipsledtomicroprocessorsandmicro-

processors led to PCs, mobile telephones, MP3 players, and

contributed to the Internet. Innovations in microelectronics

made possible innovations in many other industries.That is

why, although the semiconductor and PV industries began

at about the same time, sales of microelectronics devices

grew much faster, by 2007 reaching $250 billion worldwide

comparedwithPVrevenuesof$17billion.

Likeothertechnologicalrevolutions,therevolutioninIT

reflectedresearchconductedinearlieryears,atfirstexploiting

foundations laid before World War II when quantum

mechanicswasappliedtosolid-statephysicsandchemistry.

Much relatively basic work now finds its way quite rapidly

into applications: like many others, the semiconductor

industry lives off what might be termed just-in-time (JIT)

research.JITresearch,conductedinternally,bysuppliers,by

consortiaoffirmssuchasSematech,andinuniversities,has

helpedfirmssustaintheMoore’sLawpace.Generallysimilar

processes have characterized developments in PV tech-

nology, but natural limits on efficiency gains, and the

commodity-like nature of electricity, keep this technology

fromfollowingapathsimilartoIT.

4.2 Energy-Climate Innovation Policy Choices Plainly, the needed revolution in energy-climate tech-

nologieswillbeverydifferent fromthat in IT,even though

the sources will be similar: technological innovation taking

placewithinprivatefirms,withassistsfromgovernmentand,

for GHG reduction, either a strong regulatory prod or the

governmentasdirectpurchaser.Thetechnologyandinnova-

tionpoliciesonwhichtheU.S.governmentcancallcomein

manyvarietiesandworkinmanyways(Table2).Competition

both inR&Dand inprocurement, forexample,werehighly

effectiveinITbuthavenotbeenverysignificantinenergy-

climatetechnologies,whichhavebeenmonopolizedbythe

DepartmentofEnergy(DOE).

Governmentpurchasesofearly ICchipswereat leastas

important in stimulating innovation as government R&D

contracts.Morebroadly,withmultipleindependentsourcesof

supportforinformationtechnologies,radicalideasmightgeta

hearinginoneplaceifnotanother,ifnotwithintheDepartment

ofDefense(DoD),perhapsattheNationalScienceFoundation

(NSF), the National Aeronautics and Space Administration

(NASA), or the National Security Agency.9 Within DoD,

moreover, each service has extensive dedicated R&D

capabilities, and because the armed forces sometimes resist

innovationsthatmightseemtothreatentheirorganizational

culturesandmissions,asballisticmissilesappearedtothreaten

Air Force pilots at a time when the Strategic Air Command

dominated that service, civilian officials created an agency

managedbycivilianstooperateindependentlyoftheservices,

now known as the Defense Advanced Research Projects

Agency (DARPA). Recognized especially for incubating the

Internet,andwithnumerousmorepurelymilitarytechnologies

to its credit, DARPA became the model for the Advanced

Research Projects Agency-Energy, ARPA-E, which Congress

created in 2007. (Section 6.4, below, compares DARPA in its

institutional setting to ARPA-E.)This is a key point, to which

wewill return:competition and diversity among government agencies was an essential element of the IT story. So far, such competition is virtually absent in government efforts to foster energy innovation. Policies to enhance private sector competition, notably

antitrust, have also fostered innovation, although not

necessarily because they are focused on technology (that is


9Funding a Revolution: Government Support for Computing Research(Washington,DC:NationalAcademiesPress,1999).

whytheyarenotincludedas“technologypolicies”inTable2).

Pentagon officials insisted on multiple sources of supply for

vital system components such as ICs. They also pressed for

nonproprietarysoftwarestandards,helpingfueltheexpansion

of wide-area computer networks and the Internet.10 While

DoDsupportfor IThasbeenlessvisiblesincethe1970s,the

mainreasonisthatcommercialapplicationshavebecomeso

muchmoreprominent.

Therehasbeenplentyofduplication,waste,backtracking,

andmismanagementinmilitarytechnologydevelopment.As

Secretary of Defense in 1991, Richard Cheney canceled the

Navy’sA-12attackplane,a$50-plusbillionprogramthatafter

three years had overrun its budget by half. We reluctantly

observethatthecomplexityofinnovation,combinedwiththe

temptationsofpolitics,maymakesuchinefficienciesinevitable

whengovernmentactstoaccelerateinnovationtoaddressan

urgent societal need. Even so, at least while the Cold War

continued, thedeeply felt threatposedby theSovietUnion

keptthemilitaryinnovationsystemlargelyontrack,creatinga

compelling sense of mission in government and, if less

urgentlyfelt,indefensefirmsanduniversities.Thethreatwas

noabstraction:near-disasterduringtheopeningstagesofthe

KoreanWar,whenpoorlyequippedU.S.troopswerepushed

almostintothesea,galvanizedWashingtonpolicymakerswho

directedmassivefundinginfusionsintodefensefirmsandalso

universities.Thedefensebudgethadbeenslashedafter1945

tofreeresourcesforthecivilianeconomy.In1950,whenthe

Korean War began, DoD’s outlays for new weapons totaled

19ENERGY INNOVATION COMPARED TO INFORMATION TECHNOLOGY

Table 2. Technology and Innovation Policies a

POLICY COMMENTS

aForfurtherdiscussionofthepolicyclassificationandentries,see“EnergyInnovationfromtheBottomUp:ProjectBackgroundPaper,”CSPO/CATF,March2009,<www.cspo.org/projects/eisbu/>.

1. R&Dcontractswithprivatefirms(fullyfundedorcostshared). Normallysupportgovernmentmissions,suchasdefense.

2. R&Dcontractsandgrantswithnonprofits. Mostlyuniversities,mostlybasicresearch.

3. IntramuralR&Dingovernmentlaboratories. Widerangeofactivities,dependingonagency.Somelaboratories muchmorecapablethanothers.

4. R&Dcontractswithconsortiaorcollaborations. ProprietaryinterestsofparticipatingorganizationsmaylimitR&D togeneric,pre-competitivework.

5. R&Dtaxcredits. Unlikelytoalterfirms’risk/rewardassessments.Difficultor impossibletotarget.

6. Patents. Thestrongertheprotection,theweakertheincentivesfor diffusionthroughimitationorcircumvention.

7. Taxcreditsorproductionsubsidiesforfirmsbringingnew Tendtopushtechnologiesintothemarketplacefromsupplyside. technologiestomarket.

8. Taxcredits,rebates,orpaymentsforpurchasers/usersof Createdemandpull,incontrasttotechnologypush(above). newtechnologies.

9. Procurement. Powerfulstimuluswhengovernmentsalesasubstantialfractionofthetotal.

10.Demonstrationprojects. Intendedtovalidatetechnologiesviewedastooriskyfor privateinvestment.

11.Monetaryprizes. Administrativelysimple,onceruleshavebeenset.

12.Educationandtraining. Manyestablishedchannelsslowacting(e.g.,universitydegreeprograms).

13.Codificationanddiffusionoftechnicalknowledge Usuallymustawaitacceptanceasvalid,useful(e.g.,informationand (e.g.,screening,interpretation,andvalidationofR&Dresults, knowledgegeneratedthroughdemonstrationprojects). supportfordatabases).

14.Technicalstandards. Dependsonconsensus;compromisesamongcompetinginterests maylock-ininferiortechnologies.

15.Technology/industryextension. Timeconsuming,costlytoreachlargenumbersoffirms,individuals.

16.Publicity,persuasion,consumerinformation. Competinginterestsmayattenuate,perhapsdistort,themessage.

I. DIRECT GOVERNMENT FUNDING OF KNOWLEDGE GENERATION

II. DIRECT OR INDIRECT SUPPORT FOR COMMERCIALIZATIONAND PRODUCTION

III. DIFFUSION AND LEARNING

10JohnA.Alic,DavidC.Mowery,andEdwardS.Rubin,U.S. Technology and Innovation Policies: Lessons for Climate Change(Arlington,VA:PewCenteronGlobalClimateChange,November2003),pp.37-39.Bythetimethelocusofinnovationshifted,around1970,fromdefense(andspace)tocommercialIT,arobusttechnologybasehadbeenputinplace,largelybyDoD.Electricalengineeringdepartmentsinthenation’suniversities,forinstance,shiftedtheiremphasisfromanalogtodigitallogicandcircuitry,andcomputerscienceandengineeringprogramsbeganaperiodofrapidexpansion.Americanuniversitiesgraduatedfewerthan100peoplewithdegreesincomputersciencein1966;adecadelater,thenumbersexceeded6000andby1980hadreachedwellover10,000.Science and Engineering Degrees: 1966-2006: Detailed Statistical Tables,NSF08-321(Arlington,VA:NationalScienceFoundation,October2008),Table34,p.37.

Table 3. The Cold War Innovation Policy Portfolio

POLICY OBSERVATIONS AND OUTCOMES




1. R&Dcontractswithprivatefirms DoDR&Dfundsflowpredominatelytoengineeringdesignanddevelopment (fullyfundedorcostshared). ofweaponssystems.Somecontractorsalsomaintainresearchlaboratories thatcontributemoregenericformsofknowledgetothetechnologybase.

2. R&Dcontractsandgrantswithnonprofits. UniversitiesgetmorethanhalfofDoDbasicresearchdollars,butbasic researchcomprisesonlyabout2percentofDoDR&D.Not-for-profitFFRDCs (federally-fundedresearchanddevelopmentcenters)sometimestakeon systemintegrationtasksfortheservicesandintelligenceagencies.

3. IntramuralR&Dingovernmentlaboratories. ThearmedforcesoperatedozensofR&Dlaboratoriesoftheirown.Mostof theseconductrelativelyappliedwork,suchasextendingacademicresearch orverifyingtechnicalmethods.DoDengineersandscientistssometimes workdirectlywithcontractorsoratfront-linemilitarybases.

4. R&Dcontractswithconsortiaorcollaborations. Littleactivity,althoughCongressgavetheDefenseAdvancedResearch ProjectsAgencyoversightofthepioneeringR&Dconsortium,Sematech, atthetimeitapprovedcost-sharedfederalfunding.

5. R&Dtaxcredits. Rarelysignificantindefense.

6. Patents. DoDgenerallyfavoredopenratherthanproprietarytechnologies,asillustrated byitsinsistenceonmultiplesourcingofsemiconductordevices,whichrequires technologysharing,andopenprotocolsforwide-areacomputernetworks culminatingintheInternet.

7. Taxcreditsorproductionsubsidiesforfirms Notapplicable,sinceDoDisthecustomer. bringingnewtechnologiestomarket.

8. Taxcredits,rebates,orpaymentsforpurchasers/users Notapplicable. ofnewtechnologies.

9. Procurement. Powerfulstimulusfordual-useinnovationsinmicroelectronics,computer languages,jetengines,avionics(e.g.,fly-by-wire),andmaterials(fiber- reinforcedcomposites).

10.Demonstrationprojects. InDoD’sseven-tieredR&Dbudgetclassification,category6.5,“system developmentanddemonstration,”accountsforaboutone-quarterofallR&D, around$20billioninrecentyears.

11.Monetaryprizes. Notapplicable(althoughcontractorsmayviewprocurementcontractsasthe rewardforsuccessfulR&D).

12.Educationandtraining. DefenseofficialsrecognizedafterWorldWarIIthatnationalsecuritydepended onahighlycapabletechnicalworkforce,abletoinnovateasscientistsand engineershadinnovatedduringtheManhattanProject.DoDhassupported graduatestudentsthroughfellowshipsandresearchandalsosponsoredprograms thatplaceengineeringandsciencefacultyinmilitarylaboratoriesfortemporary assignments,alongwith“summerschools”focusedonparticularproblems.

13.Codificationanddiffusionoftechnicalknowledge DoDengineersandscientistsparticipateinandsometimesorganizeprograms (e.g.,screening,interpretation,andvalidationofR&D tovalidateandtestnewtechnicalmethods(computerizedfinite-elementanalysis, results,supportfordatabases). testproceduresforfiber-compositematerials).

14.Technicalstandards. Defenseagencieshavebeenactiveparticipantsinstandards-setting.Insome cases,militarystandardshaveprovidedabasisforcivilianstandards;theFederal AviationAdministration,forexample,baseditsairworthinessrequirementsfor structuraldesignonthoseoftheAirForce.

15.Technology/industryextension. DoDsometimesprovidestechnicalassistancetosuppliers,especiallysmaller firms,toensuretheyuserequiredmethodsorsimplybestcommercialpractices.

16.Publicity,persuasion,consumerinformation. DoDofficialsandmajorcontractorsoften(notalways)joininstatementson, forexample,educationandtrainingofengineersandscientists.


only$2.4billion(about$21billionintoday’sdollars).By1952,

thefigurehadclimbedto$16.7billion($133billionintoday’s

dollars) and both R&D and procurement remained high

thereafter. Table 3 summarizes policies underlying military

technologicalinnovationduringtheColdWar.Whilethebasis

was laid during World War II and in 1946, when Congress

established the AEC and the first of the military research

organizations, the Office of Naval Research, the structure

expandedgreatlyaftertheKoreanWaraspartoftheeffortto

counter,throughtechnology,thethreatposedbyanumerically

superiorfoe.

Table 3. The Cold War Innovation Policy Portfolio

1. R&Dcontractswithprivatefirms DoDR&Dfundsflowpredominatelytoengineeringdesignanddevelopment (fullyfundedorcostshared). ofweaponssystems.Somecontractorsalsomaintainresearchlaboratories thatcontributemoregenericformsofknowledgetothetechnologybase.

2. R&Dcontractsandgrantswithnonprofits. UniversitiesgetmorethanhalfofDoDbasicresearchdollars,butbasic researchcomprisesonlyabout2percentofDoDR&D.Not-for-profitFFRDCs (federally-fundedresearchanddevelopmentcenters)sometimestakeon systemintegrationtasksfortheservicesandintelligenceagencies.

3. IntramuralR&Dingovernmentlaboratories. ThearmedforcesoperatedozensofR&Dlaboratoriesoftheirown.Mostof theseconductrelativelyappliedwork,suchasextendingacademicresearch orverifyingtechnicalmethods.DoDengineersandscientistssometimes workdirectlywithcontractorsoratfront-linemilitarybases.

4. R&Dcontractswithconsortiaorcollaborations. Littleactivity,althoughCongressgavetheDefenseAdvancedResearch ProjectsAgencyoversightofthepioneeringR&Dconsortium,Sematech, atthetimeitapprovedcost-sharedfederalfunding.

5. R&Dtaxcredits. Rarelysignificantindefense.

6. Patents. DoDgenerallyfavoredopenratherthanproprietarytechnologies,asillustrated byitsinsistenceonmultiplesourcingofsemiconductordevices,whichrequires technologysharing,andopenprotocolsforwide-areacomputernetworks culminatingintheInternet.

7. Taxcreditsorproductionsubsidiesforfirms Notapplicable,sinceDoDisthecustomer. bringingnewtechnologiestomarket.

8. Taxcredits,rebates,orpaymentsforpurchasers/users Notapplicable. ofnewtechnologies.

9. Procurement. Powerfulstimulusfordual-useinnovationsinmicroelectronics,computer languages,jetengines,avionics(e.g.,fly-by-wire),andmaterials(fiber- reinforcedcomposites).

10.Demonstrationprojects. InDoD’sseven-tieredR&Dbudgetclassification,category6.5,“system developmentanddemonstration,”accountsforaboutone-quarterofallR&D, around$20billioninrecentyears.

11.Monetaryprizes. Notapplicable(althoughcontractorsmayviewprocurementcontractsasthe rewardforsuccessfulR&D).

12.Educationandtraining. DefenseofficialsrecognizedafterWorldWarIIthatnationalsecuritydepended onahighlycapabletechnicalworkforce,abletoinnovateasscientistsand engineershadinnovatedduringtheManhattanProject.DoDhassupported graduatestudentsthroughfellowshipsandresearchandalsosponsoredprograms thatplaceengineeringandsciencefacultyinmilitarylaboratoriesfortemporary assignments,alongwith“summerschools”focusedonparticularproblems.

13.Codificationanddiffusionoftechnicalknowledge DoDengineersandscientistsparticipateinandsometimesorganizeprograms (e.g.,screening,interpretation,andvalidationofR&D tovalidateandtestnewtechnicalmethods(computerizedfinite-elementanalysis, results,supportfordatabases). testproceduresforfiber-compositematerials).

14.Technicalstandards. Defenseagencieshavebeenactiveparticipantsinstandards-setting.Insome cases,militarystandardshaveprovidedabasisforcivilianstandards;theFederal AviationAdministration,forexample,baseditsairworthinessrequirementsfor structuraldesignonthoseoftheAirForce.

15.Technology/industryextension. DoDsometimesprovidestechnicalassistancetosuppliers,especiallysmaller firms,toensuretheyuserequiredmethodsorsimplybestcommercialpractices.

16.Publicity,persuasion,consumerinformation. DoDofficialsandmajorcontractorsoften(notalways)joininstatementson, forexample,educationandtrainingofengineersandscientists.

21WHERE HAVE ALL THE BREAKTRHOUGHS GONE?

5. Where Have All the Breakthroughs Gone? Technological breakthroughs will not radically transform energy systems in the next several decades.

11Commercializing High-Temperature Superconductivity (Washington,DC:OfficeofTechnologyAssessment,June1988),pp.6,24.

For energy-climate technologies, research-basedbreakthroughscanpromiseonlymodestaidduring thefirsthalf of the twenty-first century.There are two main reasons.First, energy-related technologies have become deeplyimbedded in the world economy, in the form of electricalpower networks, the huge and ever-growing fleet of motorvehicles,alreadynumberingover500millioncarsandtrucksworldwide,plusshipsandplanesalsorunningonfossilfuels,andtheheating,airconditioning,andprocessenergysystemsinhomes,officebuildings,andindustrialfacilities.Sunkcostsare huge compared to the sunk costs of, say, the officetypewriters,addingmachines,andotherbusinessequipmentreplacedbyIThardwareandsoftwareduringthe1980s.ManyotherIT-basedinnovationsmovedintoavacuum.Bycontrast,replacementofoldergenerationsofenergytechnologieswillbecostlyandslow.Second,althoughtheresultsofscientificandengineeringresearchfindtheirwayintomarketedgoodsandservicesfastertodaythaninthepast,inmostcasesyearsif not decades elapse between discovery and applications,meaning that fundamental research on energy-climatetechnologies today (as opposed to applied research anddevelopment)maynotpayoffuntilthe2040sorlater.High-temperaturesuperconductivityillustrates. The initial discovery of HTS in 1986 generated greatexcitement in the scientific community, in business, and ingovernment.Afloodofresearchfollowed,asscientistsintheUnitedStates,Japan,andEuropesynthesizednewHTSmateri-alsandsoughttounderstandtheunderlyingphenomena.Tofuturists,R&Dmanagers,andbuddingentrepreneurscourtingventure capitalists, these materials promised ultra-highefficiencyelectricalequipment, loss-freepowertransmission,and superconducting storage rings in which powerfulcurrents could circulate until needed, overcoming one ofthechiefobstaclestorelianceonintermittentenergysources

suchassunlightandwind.Inanunprecedentedappearancein July 1987 at the Federal Conference on Commercial Ap-plications of Superconductivity, President Ronald Reagan,attendedbythreecabinetsecretaries,announcedhisadmin-istration’splanforhurryingthenewtechnologytomarket.11

More than two decades later, no applications haveappeared (althoughat leastonesmall-scaledemonstrationisinprogress).TheinitialdiscoveryearnedtwoIBMphysiciststhe 1987 Nobel Prize—the shortest interval betweendiscoveryandawardinhistory—andHTSsciencecontinuestoflourish(unlikecoldfusion,announcedin1989andquicklydiscredited). But the practical problems have simply beentoogreat,whileknottytheoreticalpuzzles,asyetunresolveddespite a great many scientific advances, have left experi-mentalists (and process designers) with little guidance forsynthesizing better materials and microstructures to meetcommercialrequirementswithoutcompromisingsupercon-ductingperformance.

Thepointhereissimpleenough.Research mustcontinue, else innovation will at some point dry up. But the path of innovation is uncertain and policymakers cannot assume that today’s research will pay off in near-to-medium term energy-climate technologies. That may happen, or it

maynot;noonecanknow.Thecallssocommonlyheardforanew Manhattan project or Apollo program,“big science” or“bigtechnology”(orboth)tospurenergy-climateinnovation,reflect misunderstandings not only of how innovationnormally occurs, but the objectives, organization, andaccomplishmentsofthoseundertakings,asrelatedinBoxD.


Box DCrash Development of Large-Scale Technology: The Manhattan Project and Apollo Program

TheWorldWarIIManhattanprojectandtheApollomoonlandingdifferedinmanyrespects.Theatomicbombsprang

fromhighlyesotericsciencewhilealsodependingonamassiveengineeringandproductionefforttoproduceenough

material,uranium-235orplutonium,forlaboratoryexperimentstounderstandthephysicsofanexplosionand,onceit

seemedthepathtoafunctionalweaponwasopen,tomakeahandfulofwarheads.Howmuchmaterialwouldbeneeded

was at first unknown: it might be pounds, it might be tons.The entire effort exemplified just-in-time research. In the

beginning,scientistshadlittleideahowtoproceed,literallyfromweektoweek;onseveraloccasionsPresidentRoosevelt

madelargefundingcommitmentsbasedonnothingbeyondthescientificandengineeringintuitionofhisadvisers.The

Manhattanproject literally threwmoneyatuncertainty. Intheend, thebombprogramtooksome45percentof total

wartimeR&Dfunds(intoday’sdollars,about$21billionof$47billion).a

Inmarkedcontrast,theApolloprogramventuredintounknownareasonlywhenabsolutelynecessaryandthenwithgreat

caution.Technologicalconservatismwasthewatchword,sincehumanswouldbesentalong,theworldwaswatching,

andtheUnitedStateswasseekingapropagandavictoryinaraceitwasboundtowin,absentdisaster,sincetheSoviet

Union,althoughhappytogoadtheUnitedStatesalong,hadnodesiretospendthemoneyneededtosendhumansto

themoon,somethingthatultimatelycosttheUnitedStatesnearly$100billion(intoday’sdollars).Apollomanagersspent

moneytoensuresafetyandreliability.

Asasymbolofnationalcommitmenttoenergy-climatetechnologies,itmayservetocallforanotherManhattanorApollo

program.Otherparallelsarefew.UnlikeApollo,whichhadawell-definedendpoint,andtheManhattanproject,justified

bythebeliefthattheUnitedStateswasengagedinalife-and-deathracewithaGermanbombprogram,energy-climate

technologyisanissuewithtimehorizonsofdecadestocenturiesandobjectivesthatareboundtochangeinfutureyears.

Theresponsewillhavetobeglobalandwillalmostcertainlyinvolvethousandsoftechnologicaladvancesandmillionsof

technologicalchoicesinresponsetomanypoliciesandmanygovernmentagenciesinmanycountries.Innovationswill

comefromtensofthousandsoffirmsemployinghundredsofthousandsofpeople.Nosingleinventionordiscovery,no

matterhowdramatic, is likely toprovidemorethanonesmallpieceof thepuzzle.TheApolloandManhattanproject

metaphorsarenotveryhelpfulasplanningguides.

aSpendingtotalsfortheManhattanprojectandApollo(nextparagraph)arefromDeborahD.Stine,The Manhattan Project, the Apollo Program, and Federal Energy Technology R&D Programs: A Comparative Analysis, RL64343(Washington,DC:CongressionalResearchService,September24,2008).

23BUILDING AN ENERGY INNOVATION SYSTEM

New technologies emerge over time, typically severaldecades,shapedbydemand(fornewgoodsandservices)aswellassupply(ofknowledgeandideas).Innovatorslearnfromusers (with failures to learn interspersed along the way).Knowledgeable customers—textile manufacturers that buysyntheticfibersfromchemicalfirms,companiesinanyindustrythatpurchasecomputerstoautomatebusinesstransactions,computer manufacturers exploiting ICs to improve theperformance of their hardware—push their suppliers forinnovationsand,whentheypurchasenewproducts,helppayfor further innovation. In contrast to these examples,innovation has been desultory in most energy-relatedtechnologiessinceWorldWar II.Thechiefexception,nuclearpower, rests on military foundations, as do other exceptionsincludingjetenginesandgasturbines.

6.1 Lacking Policy, Lackluster Innovation Inrecentdecades,politicalfigures,businessleaders,andinterest groups have called repeatedly for a national energypolicytogoalongwiththeenvironmentalpolicythatbegantotakeshapeinthe1960s.Toomanygroupshavewantedtoomanydifferentthings,andnoconsistentpolicyhasresulted.Researchbecameafallback,symbolizedinthenamegiventheEnergyResearchandDevelopmentAdministration(ERDA) in

1974, when the AEC was split to separate out its regulatoryfunctions.Thatapproachhasnotbeenverysuccessful.12The$60billionthatERDAandthenDOEhavespentonenergyR&Dsince the mid-1970s has not been disciplined by anythingresembling a strategy, and the zigs and zags have impededinnovation. As one workshop participant recalled, “WhenReagancutsolaritputusintheweeds;we’dhavecomealotfurtherbynowotherwise.” ERDA,andafter1977DOE,continuedtosupportnuclearpoweralongwithasweepingscientificagendaalsoinheritedfrom the AEC, while absorbing several programs fromelsewhereingovernment(e.g.,theInteriorDepartment’scoalresearch,fundedatafewmilliondollarsannuallyduringmostofthe1960sandrisingtonearly$125millionin1974afterthefirst energy shock). Nuclear weapons (and naval reactors)remainedtheprimaryconcern,untilquiterecentlyaccountingfor three-quarters of the DOE budget, now down from thatbut still over 60 percent (including remediation of sitescontaminatedbybomb-makingactivities).Sincethenucleardeterrentdependedentirelyonthenationallaboratories(BoxE),scientistsandlaboratorymanagershavehadgreatleveragewith Washington since the 1940s. So long as the warheadskeptcoming,thelaboratoriescoulddomuchastheywished.

6. Building an Energy Innovation System Complexity and uncertainty cannot be evaded, but understanding how technologies advance and how the strengths and weaknesses of existing institutions affect the process can point toward new rationales, policy approaches, and management priorities. More competition among government agencies, and a particular focus on effective conduct of demonstration projects, deserve the particular attention of policymakers. Specific technologies may demand appropriately tailored policies, as illustrated through the example of direct air capture of carbon dioxide.

12Forafair-mindedanalysis,seeEnergy Research at DOE: Was It Worth It? Energy Efficiency and Fossil Energy Research 1978-2000(Washington,DC:NationalAcademiesPress,2001).

Box EManaging Innovation in the U.S. Government, or, Why Swords are Easier than Ploughshares

Nominally runbypoliticalappointeeswhocomeandgowhilecivil servantscarryon,and lacking the incentivesand

metricsfoundintheprivatesector,governmentagenciesposesingularmanagementdilemmas.TheDefenseDepartment

isamongthemostdifficulttomanage,becauseofitssheersizeandbecausetheservicesguardtheirindependencewith

martialfervor.AsPresidentFranklinD.Rooseveltoncetoldanadviser,“TochangeanythingintheNa-a-vyislikepunching

afeatherbed.”aTheDepartmentofEnergy’slaboratoriesmaynotbetheNavy,buttheydohaveadegreeofautonomy

uniqueamongfederaltechnologyagencies,onereasonforthedisappointingrecordofDOEanditslaboratoriesinenergy-

relatedinnovation.

AfterWorldWarII,thenewlyestablishedAtomicEnergyCommission(DOE’sancestor),soughttoretaintheservicesofa

core of experienced bomb designers, many of them elite scientists with distinguished prewar careers to which they

expectedtoreturn.ContractorsrantheManhattanproject’slaboratoriesfortheArmyCorpsofEngineers.GeneralLeslieR.

Grovesandtheofficerswhoreportedtohimkeptclosecontrolofcriticaldecisions(onspendingespecially),listenedto

theirscientificadvisorsontechnicalissues,andleftmostday-to-daymatterstoexperiencedindustrialmanagersdetailed

byfirmssuchasDuPontthatservedaswartimecontractors.Scientistsaccustomedtodoingwhattheywantedchafed,

butsolongasthewarcontinuedallsidestoleratedtheresultingfrictions.OnceJapanhadbeendefeated,leadingscientists

beganwarningtheAECthatthebombproject’sbestpeoplewouldleaveunlessfreedfromtightsupervision,sparedthe

ignominyofcivilservicestatus,andallowedtospendpartoftheirtimeonresearchoftheirownchoosing.b

ThiswasapowerfulargumentgiventhepressureontheAECtobuildmoreandbetterweapons,andtodosoquickly.

TensionswiththeSovietUnionwererisingandtheU.S.stockpilenumberedunderadozenatomicbombs,hardlyenough

todeteramanlikeStalin,whohadexecutedtheRedArmy’sbestofficersevenwhileanticipatingawarthatkilledmany

millionsofSovietcitizens.Barelyconsideringwhattheconsequencesmightbe,theAECtookovertheArmy’scontractual

arrangements for the weapons laboratories—although at the time the Commission had hardly any staff qualified to

overseethem,someoftheindustrialcontractorswerewithdrawing,andthosethatagreedtocontinueinsistedonrecalling

their best managers to peacetime duties. With the atomic bomb acclaimed as the weapon that had forced Japan’s

surrenderandphysicsthemasterkeytothemysteriesofnature,scientists—nowfreeofbothGrovesandhisdeputiesand

the strong-willed managers temporarily detailed by industrial contractors—got much of what they wanted from

Washington,includingmoneyforbasicresearchanddesignationas“national”laboratoriestosignifyspecialstatus,setoff

fromlaboratoriesattachedtootheragencies,suchastheU.S.DepartmentofAgriculture(USDA).c

Ontheweaponsside,theAECinthe1940sandDOEsincethe1970sworkeddiligentlytosatisfythearmedforces.Failure

todosowouldhave risked lossofdefenseprograms to thePentagon.Nosuchdisciplining forcesexisted forenergy

technologies,evennuclearpower(GEandWestinghousewerealreadybusyonNavycontractsandtheutilitieshadtobe

drawnin).DOEtodayhasfewindustrialclientscomparabletotheagribusinessfirmsthattheUSDAlaboratoriessupport

orthepharmaceuticalcompaniesthatliveoffresearchfundedbytheNationalInstitutesofHealth(NIH).Therearenot


evenpeergroupsinuniversitiesakintotheagricultureschoolsandmolecularbiologyfacultiesagainstwhichUSDAand

NIH scientists can benchmark their work (a few universities do retain nuclear engineering departments).While DOE’s

laboratories rightly claim many research accomplishments, cooperation with industry has gone in and out of favor

depending on budgetary allocations and demands from Washington to demonstrate relevance. (To be sure, the

laboratoriesdiffer: theNationalEnergyTechnologyLaboratory, forexample,which isgovernment-operatedaswell as

government-owned,spendsmostofitsR&Ddollarsonexternalprojects.)DOE’scooperativeR&Dagreements(CRADAs)

withindustryrosefromafewhundredannuallyatthebeginningofthe1990stooverathousandinthemiddleofthe

decade,astheendoftheColdWarandnucleartestingraisedquestionsconcerningthefutureofthelaboratorysystem

(detenteandthehalttoatmospherictestinghadgeneratedsimilarquestionsearlier),butthenfellalmostasfastasanxiety

(andCongressionalattention)faded.d

Since World War II, R&D management in the federal government has reflected an often-awkward mix of top-down

directives from political appointees and bottom-up decisions by civil servants, originally imprinted byVannevar Bush,

wartimeczarofmilitarytechnologydevelopment.Bushbelievedthatprioritiesshouldbesetandkeydecisionsmadeby

technicalandscientificelites,meaninghispeersinwhatwerethenasmallhandfulofresearchuniversities—Bushhimself

had been a professor and administrator at MIT—and implemented at working levels with considerable discretion by

personnelwithfine-grainedknowledgeinspecializedfieldssuchasradarengineering.Bush’stwinbeliefs,intop-down

guidanceandbottom-upautonomy,wereinconsiderableconflict,andthepersistenceofthepatternheestablishedhas

contributed to managerial difficulties in many federal technology agencies.e Most have reached some sort of

accommodation. DOE has not. Not that long ago,“a DOE employee at Oak Ridge [told a committee of the National

ResearchCouncilthat]‘Werecognizenoauthorityoutsidethe[DOE]OfficeofScience,’”oneofthemanyillustrationsofthe

insularcultureofthe laboratories.f Whilethesefacilitiesdifferamongthemselves,DOEasawholecomesperhapsthe

closestofanypartofthefederalscienceandtechnologystructuretoRoosevelt’sfeatherbed.

aRoosevelthadbeenAssistantSecretaryoftheNavyduringWorldWarIandafterward.AsheexplainedtothemanheappointedtoheadtheFederalReserve:“TheTreasuryissolargeandfar-flungandingrainedinitspracticesthatIfinditalmostimpossibletogettheactionandresultsIwant…ButtheTreasuryisnottobecomparedwiththeStateDepartment.Youshouldgothroughtheexperienceoftryingtogetanychangeinthethinking,policy,andactionofcareerdiplomatsandthenyou’dknowwhatarealproblemwas.ButtheTreasuryandtheStateDepartmentputtogetherarenothingascomparedwiththeNa-a-vy.Theadmiralsarereallysomethingtocopewith—andIshouldknow.TochangeanythingintheNa-a-vyislikepunchingafeatherbed.Youpunchitwithyourrightandpunchitwithyourleftuntilyouarefinallyexhausted,andthenyoufindthedamnbedjustasitwasbeforeyoustartedpunching.”MarrinerS.Eccles,Beckoning Frontiers: Public and Personal Recollections,SidneyHyman,ed.(NewYork:Knopf,1951),p.336.

bOnManhattanprojectmanagement,seeRichardG.HewlettandOscarE.Anderson,Jr.,The New World, 1939/1946, Volume I: A History of the United States Atomic Energy Commission (University Park: Pennsylvania State University Press, 1962). Richard G. Hewlett and Francis Duncan, Atomic Shield, 1947/1952, Volume II: A History of the United States Atomic Energy Commission(UniversityPark:PennsylvaniaStateUniversityPress,1969)coverstheAEC’seffortsafterthewartoworkoutasatisfactorysetofrelationshipswiththelaboratoriestakenoverfromtheArmy.

c“AECstaffandcommissionersbought into [the] theory” that“‘scientificpersonnelconsiderdirectemploymentby theGovernmenthighlyundesirable’”eventhough“theAECitself…managedtoavoidcivilserviceregulations.”“TheAECagreedtosupportworkonthemarginsofitsmission;thesepaste-onsthengrewtobecomeamajorcomponentofitsprogram.Bythe1950shigh-energyphysicswoulddominatetheresearchbudget….”PeterJ.Westwick,The National Labs: Science in an American System, 1947-1974(Cambridge,MA:HarvardUniversityPress,2003),pp.49and151.

dTechnology Transfer: Several Factors Have Led to a Decline in Partnerships at DOE’s Laboratories,GAO-02-465(Washington,DC:GeneralAccountingOffice,April2002).

eLarryOwens,“TheCounterproductiveManagementofScienceintheSecondWorldWar:VannevarBushandtheOfficeofScientificResearchandDevelopment,”Business History Review,Vol.68,1994,pp.515-576.

fProgress in Improving Project Management at the Department of Energy: 2003 Assessment(Washington,DC:NationalAcademiesPress,2004),p.46.



WhatDOEhasnotdoneisworkconsistentlyandwellwithindustry,aspartners inenergy-related innovation. (Congress,in turn, has been inconsistent in its demands for suchcooperation.) The accomplishments of DOE scientists andengineersinfieldssuchasparticlephysicshavehadrelativelyfew counterparts in areas of practical technological interest,notwithstandingahandfulofnotableaccomplishmentssuchas energy-saving low-emissivity windows.This is the legacy,indeedthedirectconsequence,ofhastydecisionsbytheAECin theearlypostwaryears,madetokeepweaponsscientistsattachedtolaboratoriesinheritedfromtheManhattanproject.Thecultureofmostofthelaboratorieshasnotbeenamenabletothecollaborative,incremental,disciplinedneedsofenergytechnologyinnovation.Lackofcompetitionforresourceswithotherfederalagencieskeepsthisculturelockedin. The inertia caused by history, politics and culture is onfurtherdisplayinDOE’senergyR&Dportfolio,whichcontinues

to feature nuclear energy and coal, as indicated in Table 4.Congressional opportunism is also apparent, in the highpriority afforded to biomass (a reflection of the power ofagriculturalinterests),alongwithcoalandotherfossilfuels,inthe allocation of stimulus funds under the 2009 AmericanRecoveryandReinvestmentAct. Asthisreportemphasizesthroughout,itwilltakeabroadportfolioofpolicies,notjustR&D—eveniftheR&Disbetter-balanced than in the past—to build an effective energyinnovationsystem.The fact is thatmanytechnologieswithpotentialforreducingGHGemissionsexistandarereasonablywellunderstood.Theyneedjust-in-timeresearch,demonstra-tion,andthekindofongoingincrementalimprovementsthatarecoreactivitiesofnormalinnovation.Theexamplesincludeaircapture,whichisinitsinfancybutevensoappearstoneeddevelopmentanddemonstrationmorethannewresearch,assummarizedinBoxF.

Nuclearenergy $630million $515million $515million $403million

Fossilenergyc 625 876 4280 618

Biomass 225 217 1000 235

Solar 156 175 175 320

Wind 53 55 173 75

Geothermal 30 44 444 50

OfficeofScience 4310 4760 6360 4940

Table 4. Department of Energy R&D Budget a

aFiscalyears.Excludesdefenseprograms.bStimulusfundsprovidedundertheAmericanRecoveryandReinvestmentActof2009tobespentoverfiscalyears2009and2010.cMostlycoal.

Sources:2009request–“AAASR&DFundingUpdateonR&DintheFY2009DOEBudget,”March3,2008,AmericanAssociationfortheAdvancementofScience,<www.aaas.org/spp/rd/doe09p.htm>;allotherfigures–DepartmentofEnergy,<www.energy.gov/media/Steve_Isakowitz_2010_Budget_rollout_presentation.pdf>,May27,2009.

2009 APPROPRIATION BUSH OBAMA ADMINISTRATION EXCLUDING INCLUDING ADMINISTRATION 2009 REQUEST STIMULUS b STIMULUS 2010 REQUEST


Box FDirect Removal of Carbon Dioxide from Earth’s Atmosphere

Technology

CO2canbeextractedfromtheatmospherethroughgasseparationprocesses,asitcanfrompowerplantfluegases.The

primarydifferenceisconcentration:CO2isdiluteintheatmosphere,lessthan0.04percentbyvolume,comparedwith12-15

percentCO2inthestackgasesofcoal-burningpowerplants.Foreconomicreasons,separationforsaleasanindustrialgas

beginswithrichsourcesofCO2,suchasnaturalgas.Hence,incontrastwithpre-orpost-combustioncapture,thereisno

experiencewithaircapturetoprovideastartingpointforpracticalimplementation.a

Lowconcentrationmeansthatlargevolumesofairwouldhavetobetreated,sothatcostswillbehigherthanforPCC.

Otherwise,mostoftheschemessofaradvancedwouldworkingenerallysimilarfashion,byabsorbingCO2inaliquid(e.g.,

water containing sodium hydroxide) for subsequent removal, compression, and sequestration. (Low concentration will

probablycallforadifferentsorbentthanforPCC.)RegenerationandCO2compressionwould,again,consumeconsiderable

energy.Compensatingsomewhat,whilepre-orpost-combustioncaptureequipmentwouldhavetobeinstalledatthe

powerplant(andtailoredtoeachplant’soperatingcharacteristics),aircaptureunitscouldbemassproducedandsitedto

takeadvantageofprevailingwinds(tominimizethepowerconsumedbyfansforcirculation),otherwiseunusableenergy

resources (e.g., natural gas for process heat from oil fields that now flare gas for lack of market access), or geological

formations suited to sequestration. As one workshop participant observed,“With air capture you can do sequestration

where it’s [geologically]easy—inWyoming,not theOhioRiverValley.”Aircapturealsoappears tobe theonlypractical

means, other than fuel switching, for managing CO2 from small-scale and mobile sources, including road vehicles and

aircraft.AspartofaconcertedefforttocontrolatmosphericCO2,itmightbecomeacomplementtopowerplantCCS.

Althoughuniversitygroupsandstartupfirmshavebuilt laboratory-scaleunits. thereareno industrialprecedents forair

capture,unlikePCC.Norhavegovernments, intheUnitedStatesorabroad,conductedorfundedmuchtechnicalwork.

Companieswiththeresourcesandexperiencetoundertakedetailedengineeringstudieshaveshownlittleinteresttothis

point,nodoubtbelievingcommercialopportunities,ifany,remainwellinthefuture.Itwouldalmostcertainlytakeeither

directfederalexpendituresusingapublicworksjustification,orahighandpredictablystablepriceoncarbon,probablyin

excessof$100pertonofCO2,toattractprivateinvestment.

Policy

Participants at the CSPO/CATF workshop agreed that no more than perhaps two dozen people worldwide have been

activelyexploringaircapturetechnologies,aminusculelevelofeffort.Majortechnicaluncertaintiesappeartohavemore

todowiththeconceptualdesignofalternativesthanwithresearch,andcostestimatesareuncertainevencomparedto

thoseforPCC.Engineersroutinelygeneratemanyalternativesystemconfigurationsforinfanttechnologies,evaluatethem

throughcalculations,modeling,andsimulation,supportedasnecessarybyexperimentandtesting,andchoosethemore

attractiveforfurtherstudy(e.g.,topindownkeyvariablessuchasoperatingtemperaturesandpressures,andratesoffluid

flowandheattransfer),withthehelpofjust-in-timeresearchasappropriatetoresolvecriticalproblemsorsimplytoreduce

uncertainties.Thissortofworkhasbarelybegunforaircapture.Itwillbetimeconsumingandcouldbequiteexpensive,the

moresoastheagendamovesontoprototypesandfieldtests.Thusaconsiderableperiodoftechnicalexploration lies

aheadbeforeareasonablyclearpictureofthecostscanbeexpected.ItseemslikelythatanaircaptureR&Dprogramcould

productivelyabsorb20to100timesmoreresourcesintheneartermthancurrentefforts.

Thebenefitsofaircapture—slowingtheriseofatmosphericCO2andperhapseventuallystabilizingitatatolerablelevel—

wouldstementirelyfromtheavoidanceoffuturesocietalcosts.Sinceitisnotitselfanenergytechnology,R&Dneednotfall

under DOE’s purview. Indeed, policymakers might be wise to avoid DOE involvement, since competition with other

agenciesshouldhelpimproveDOE’soverallperformanceandthetechnicalagendaforaircapturecouldquitesooninclude

demonstrations,whichDOEhassometimesmismanaged(mostrecentlyinthecaseoftheFutureGenprogram,asdiscussed

inthemaintext).

a“WorkshopBackgroundPaper:AirCapture,”CSPO/CATF,April2009,<www.cspo.org/projects/eisbu/>.

Table 5. Policy Priorities for Photovoltaics, Post-Combustion Capture, and Air Capture of CO2 a

TECHNOLOGY b

Relatively basic research, for instance in new PVmicro-ornanostructures,couldleadtosubstantialperformance gains. Such work is best performedbyspecializedgroupstypicallyfoundinacademicorsimilarsettings.Fundingmechanismsbasedonpeer review are most likely to identify promisingresearchdirections.

PVmanufacturersandequipmentsuppliersunder-takeprocessR&Dintendedtoreducecostsforex-istingcelldesigns.Sincesuccessfuloutcomescon-tributedirectly tofirms’competitiveability,directgovernment fundingneednotbeahighpriority.Longer-termprocess-orientedR&D—e.g.,asmightbenecessarytofabricateinnovativecellstructuresatproductionscale—mightinsomecasesbenefitfromgovernmentsupport.

PCC implementation with existing technologiessuch as amine absorption awaits demonstrationatutilityscaletogeneratereliableinformationonfield performance and costs. For greatestcredibility with power companies, directgovernment participation should be minimizedand participation by private firms maximized.Whilesuppliersofproprietarytechnologywouldprobably be unwilling to reveal process details,potentialadoptersnonethelessneedassurancesthat resultscanbe trusted.Demonstrations thatprovide direct comparisons of competingtechnologies could be desirable. In sum,government should provide some or all thenecessary funding, act as arbiter to ensure thatutilities get the information they need and thatequipment suppliers cannot tilt demonstrationsto show their technologies to unfair advantage,but should not directly plan or manage PCCdemonstrations.

MorebasicR&DaimedatunprovenmethodsofCO2 capture also merits support. Selection ofresearch avenues should be based on peerreview by engineers and scientists withexperience and accomplishment in industrialand academic settings (as distinguished,for example, from internal DOE reviews).Diversityinproposedresearchpathwaysshouldbe encouraged. Competition among fundingagenciesaswellasresearchgroupsalsoshouldbeencouraged.

Continued development of air capturetechnologies requires further laboratory research,along with prototype construction and testing,explorationofconceptualalternatives,andscale-updemonstrations—allatmuchgreater levelsofactivity than currently supported. As with PCC,private firms have little incentive to engage insuch work; unlike PCC, government has as yetshown little interest. Since air capture is notfundamentally an energy technology, there is nonecessary reason why DOE should have the leadrole.Giventheneedtoestablishcredibilityforthisarea of research, and cultivate a much largercommunity of involved scientists and engineers,the National Science Foundation could be anappropriatesponsor.

SOLAR PV PCC AIR CAPTURE

PV sales since the 1970s have depended onsubsidies, including state government incentiveprograms. Greater predictability would helpboth PV firms and their customers plan futureinvestmentsinR&Dandproductioncapacity.

Operators of fossil fuel-fired power plants haveno reason to explore PCC at present except as ahedge against the prospect of future regulatorymandatesorfinancialincentivesfromwhichprof-its could be wrung. Firms with proprietary PCCtechnologies available or in development havegreaterincentivesbutfewmeanstoenticepartici-pationbypowercompanies indevelopmentanddemonstration.Alongwithregulations,thefederalgovernment could consider extending financialsupportfordemonstrationandscale-upormakingdirect investments in PCC installations.The latteralternative would require negotiated agreementswith power companies covering issues such aspossible revenue losses under various contin-gencies, but a substantial commitment of fundsprovided either directly or indirectly (e.g., bybuying the captured CO2) would at a minimumtelegraph serious policy intent and encouragepower companies to act on their own, if only tomaintaincontroloveroutcomesattheirfacilities.

Some years of R&D (including demonstrations)would necessarily precede implementation, forwhich government would have to pay the costs,either directly or indirectly (e.g., by buying thecapturedCO2).

Household and small business purchaserscould benefit from reliable information on theadvantages and disadvantages of availablePV technologies, along with more transparentexplanations of available forms of purchasingassistance and their financial implicationsanduncertainties.

The purpose of demonstrations is technologicallearning, and a major policy objective should beto diffuse results to nonparticipating firms. Thatrequires openness and credibility, which are alsoprerequisitesforbuildingpublicconfidenceinlong-termCO2sequestration.

There is little chance that PCC will be deployedon a large enough scale to make a difference foratmospheric concentration of CO2 unless politicalleadersandthepublicacceptthesimplerealitythatelectricbillswillhavetoriseortheincreasesinthecostsofelectricitybeotherwisecoveredbypublicexpenditures.

If implemented, air capture units would probablybe manufactured in volume to a standard design.Arriving at low-cost dependable standardizeddesigns will probably take a considerable periodof experiential learning, during which operatingproblems would be identified and resolved anddesign improvements implemented. Prematurecommercialization and large-scale production,suchasoccurredwithnuclearpowerinthe1960s,arepredictabledangerssometimesencouragedbypublic officials anxious to proclaim the success oftheirpoliciesandprograms.Theyarebestavoidedby programs planned and implemented so as tolimit unrealistic expectations from the beginning.ThemodelshouldbemorelikeruralelectrificationthanamegaprojectliketheGrandCouleeDam.




aTheentriesinthistablerepresentthejudgmentoftheCSPO/CATFprojectstaffandconsultants,whichmaydifferfromthatofworkshopparticipants.b While these technologies can be subdivided among competing approaches (e.g., crystalline and thin film PV cells, amine and ammonia PCC separation), and policies maydifferamongthealternatives,thetablegeneralizesacrossfamilieswithoutattemptingtodistinguishamongsuchalternatives.


Table 5 summarizes findings for all three technologies

addressed in theworkshops.Theyneed incremental innova-

tion and improvement.They do not necessarily need to be

replaced, although if ongoing research leads to something

better, that would help the overall energy-climate under-

taking.Demonstration isespecially important forPCCand

aircapture.

Box GDemonstration Projects: Managing Uncertainty

Research aims to uncover new knowledge that meets the tests for scientific acceptance or systematically explorestechnological alternatives. Demonstrations, although normally budgeted as R&D, fit a different template.The primaryintentistoreduceuncertaintiesconcerningoperationalperformanceandcostsbeforecommitmenttoafullydetailedsystemdesignandtoproduction.Theseareroutinetechnicalactivities inmanyindustries.Manufacturersofsolarcellsconductacceleratedagingteststoestimatetherateatwhichefficiencywilldegradeovertwoorthreedecadesofoutdoorexposure.RickoverandtheNavyinsistedonextensiveengineeringtrialsondrylandbeforesettlingonthedesignsforsubmarinereactors.Autocompanies“demonstrate” theperformanceofhybridpowertrains inAlaskanwintersandtheArizonadesert.


6.2 Where to Intervene? Effective Demonstrationof Practical Technologies The most pressing near-term policy issues raised at the

three CSPO/CATF workshops concerned the planning and

conduct of demonstrations, especially capture and storage

ofCO2 fromcoal-burningpowerplants,andalsoaircapture.

We emphasize the importance of this issue from several

perspectives. First, public discussions and policy proposals

sometimesmisunderstandoroverlookthekeyroleofdem-

onstration projects. Yet demonstration marks an essential

transition point on the pathway to widespread acceptance

and deployment for any costly large-scale technology. Sec-

ond, given the complexity of innovation, the demonstration

phasemarksacritical interventionpoint,wheregoodpolicy

andpracticecanmakeadecisive,discernibledifference.And

third, effective demonstration programs are not easy to

develop and manage. Well conceived and well managed

demonstrations, appropriately designed for the particular

technology,willbeanessentialelementofsuccessfulenergy-

climateinnovationpolicies.

Ourworkshopshighlightedtheneedfordemonstrations

of pre-combustion and post-combustion capture combined

withlong-termsequestrationofCO2.Pre-combustioncapture

seems best suited to new coal-burning plants, while post-

combustioncapture,whichiswell-provenatsub-utilityscale,

appearstobetheonlypracticalmeansofretrofittingexisting

plants.Althoughaircapture is likelytobemorecostly, ithas

a singular advantage: if proved practical, it could be

implemented independently of power plants and other

sourcesofCO2emissions.Intheabsenceofbindingregulations,

power companies have no reason to venture into CCS and

there will be little incentive for private firms to invest in air

capture, unless the government simply pays them to do so,

justifying the expenditure as provision of a public good, or

until amarketexists inwhichcapturecreditscanbesoldat

sufficientlyhighandpredictableprices.

Government-sponsoreddemonstrationprogramshavea

long-established place in U.S. technology and innovation

policy.Theyhavebeenmoresuccessfulinsomeagenciesand

industries than others and have an especially clouded

reputation,althoughperhapsnotentirelydeserved,inenergy-

relatedtechnologies.Whileindustryparticipationisessential,

since the primary purpose, as summarized in Box G, is to

reducetechnicalandbusinessuncertainties,DOE,aswehave

already noted, has an inconsistent record of cooperating

effectively with industrial partners.13The agency’s reputation

wasfurtherdarkenedbythe2008restructuring,amountingto

abandonment,of itsflagshipCCSdemonstration,FutureGen.

Bycontrast,aerospacefirmsknowwhattoexpectwhenthey

cooperatewithDoDorNASA.Whenpoliticalandbureaucratic

squabbles occur, they do not necessarily disrupt technical

agendas (DoD procurement contracts are much more

likely than R&D contracts to result in formal appeals and

legalproceedings).

13Exceptionsinclude,e.g.,DOE’spowerturbineprogram,whichhasreceivedgenerallysatisfactoryreviews.“Thecommitteeconsidersthe[AdvancedTurbineSystems]programtobeagoodexampleofasuccessfulindustry-governmentRD&Dpartnership.”Energy Research at DOE: Was It Worth It?,pp.121-127;quotationfromp.126.

Technical and Political Objectives

Government-sponsoreddemonstrationsservebothtechnicalandpoliticalends.Theygeneratedataandinformationtoreduceuncertainties,asinprototypereactordemonstrationsconductedfortheNavybyWestinghouseandGE.FortheShippingportplant,politicalconsiderationsweredominant.Whilethereweretechnicalquestionstoanswer,especiallyconcerningcostsforbuildingandoperatingacommercialplant,ShippingportenabledtheAtomicEnergyCommissiontoshowCongress,especiallytheJointCommitteeonAtomicEnergy,thatitwasseriousaboutcommercialpower,toshowutilitiesthatsuchplantswerepractical,andtoshowtheworld,particularlyunalignednationsintheThirdWorld,thattheUnitedStatescoulddeployatomicenergyforpeaceaswellaswar.TheAECwentontosponsortwofurtherdemonstrationrounds with larger reactors.While scale-up was part of the rationale, other projects were already underway withreactorslargerthanShippingport.TheAEC’schiefconcernwastoencourageinvestmentanddrawinmoreutilitiesasparticipants.a

TheAEC’sdemonstrationswerenothingnew.Afterall,Congresshadappropriated fundsso thatSamuelMorsecoulddemonstrate his telegraphy invention (by building a line between Baltimore and the Capitol) and the AgricultureDepartment,alongwithstategovernments,begandocumentinggainsfrom“scientificagriculture”ontestplotsandtestfarmsoveracenturyagotoencourageadoptionbyriskaversesmall farmers.AfterWorldWar I, theNationalAdvisoryCommitteeforAeronautics(NACA)cooperatedwithaircraftfirmsintestinginnovationsincludingcowlingsdesignedtoimproveenginecooling;sinceabsorbingNACAin1958,theNationalAeronauticsandSpaceAdministrationhascontinuedtocollaboratewithindustry(andDoD),forinstanceontheX-seriesofdemonstrators.Inboththeseexamples,demonstrationwaspartofalargerpolicy:raisingincomelevelsforfarmfamiliesatatimeofruralpoverty,andnationaldefenseatatimeofrapidgainsinaircraftperformance,manyoftheminpotentiallyhostileforeigncountries.

Theoriginof thedifficultiesencounteredbyenergydemonstrations liesnot indemonstrationasapolicy tool,but ingovernment-industryrelationships.Governmenthasbeenasometimesheavy-handedregulatorofenergyindustries(andsometimes co-opted). It has also been a competitor. Publicly-owned utilities, ranging from small cooperatives to theTennesseeValleyAuthority (TVA)andBonnevillePowerAdministration,makeupaboutone-sixthof thenation’s3000power companies and account for a similar share of revenues.b Privately-owned utilities have opposed public powervehemently, labeling theTVA socialistic and seeing in some of the AEC’s early proposals a stalking horse for furtherencroachmentundertheguiseofsafeguardingthe“secrets”oftheatom.The1970CleanAirActcreatedanewsourceofconflict, and power companies have resisted buy-back provisions intended to encourage non-utility investments inrenewablegeneratingcapacitysuchasPVsystems.Onanotherfront,collusionandprice-fixingintheelectricalequipmentindustrywereopensecrets,apparentlytoleratedbyutilitiessincetheycouldexpecttorecovertheaddedcostsunderrate-of-returnregulation(indeed,higherpricedequipmentmightmeanhigherprofits)andpubliclyrevealedonlywhentheTVAcomplainedtotheJusticeDepartmentafterreceivingnear-identicalbidsforamulti-milliondollarturbo-generator.cMentionofenergytechnologydemonstrations,finally,stillbringstomanymindstheSyntheticFuelsCorporation(SFC),establishedin1980onlytobecloseddownafterfiveyearsandthecollapseofworldoilprices.d

Carbon Capture and Storage

CCSdemonstrationspromisetobeuniqueinseveralrespects.First,theinitialcapitalcostsforutilityapplicationswillbehigh(e.g.,intherangeof$500millionforPCCata500MWplant)andofferpowercompaniesnothinginreturnexceptcompliance with regulatory mandates that have yet to be put in place.To this point, utilities have little incentive toparticipate. Second, while Rickover insisted on standardized reactor designs for reasons of costs, reliability, and crewtraining,utility-scalecoal-burningandnuclearpowerplantshavenormallybeendesignedandbuilttoorder,sothateachplant differs, meaning that PCC installations are likely to differ too, even though incorporating many off-the-shelfcomponents. (Plant-to-plant differences will probably be considerably greater than, for example, with sulfur dioxidescrubbers,whicharerelativelysimpleadd-onsbycontrastwithPCCandevensoexhibitawiderangeofcosts.)Lackofstandardizationwillincreasetechnicaluncertaintiessomewhatandwillalsoraisecapitalcosts.Third,noonecompanywillbeincharge.Fewutilitiesmaintainlargeengineeringstaffs.Theyarealsoriskaversetothepointthatmany“prefertowaituntilafellowcompanyhasoperatedaplantusingnewtechnologybeforeorderingonethemselves.”Moreover,“[m]ostofthemdonotcountgovernment-fundedplantsusingadvancedtechnologyaseffectivedemonstrations.”eAlthoughitis


their money, utilities do not necessarily try to act as“system integrators,” or even to exercise close control over workperformedbycontractorsandsuppliers.Delegationmeansthattechnicalissuessometimesgounrecognizedorunresolveduntillateinaproject,drivingupcosts,andlaxoversightcanleaveopeningsforself-dealingandopportunismofasortalltoofamiliaronconstructionprojects.f

Technically, chemical processes such as amine separation methods applied to PCC scale somewhat unpredictably.Chemicalcompanies themselves,despite theiraccumulationsofspecializedknowledgeandexperience, tendtobecautiousaboutmovingtoorapidlyfromlaboratory-scaleprototypestolargerinstallations.OneparticipantinthePCCworkshoparguedthat,“Ifyoureallyunderstandyourprocess,scale-upshouldbestraightforward.”Othersdisagreed:“PCCwon’tworkrightoutofthebox;youshouldmakeyourmistakesondemonstrators.”Whentheexpertsdisagreein this way, it seems fair to conclude that some sort of publicly-funded demonstration will be needed to helpcharacterizeuncertainties.

Onsequestration,theworkshopgenerallyseemedinaccordwiththestatementthat,“Todemonstratethetechnology,youhavetogetitallthewayuptothepointofsendingit[CO2]downthehole;otherwiseyouarenotreallydemonstratingthewholeprocess.”Allpresentagreedthatdemonstrationsoflong-termsafetywereessentialforpublicacceptance.

aDuquesneLightaloneoperatedShippingport,a60MWplant.TenutilitiesjoinedwiththeAEConthe175MWYankeeRowedemonstrationplant,builtinRowe,MA,during1956-1961,andtwelveonthe575MWConnecticutYankeeplant,completedin1967atHaddomNeck,CT.RichardG.HewlettandJackM.Holl,Atoms for Peace and War, 1953-1961: Eisenhower and the Atomic Energy Commission(Berkeley:UniversityofCaliforniaPress,1989),pp.205ff.bElectric Power Annual 2007(Washington,DC:DepartmentofEnergy,EnergyInformationAdministration,January2009),Tables8.1,8.3,and8.4,pp.69-70.cWestinghouseandGEquotednearlyidenticalpricesof$17.4millionfortheTVA’s500MWturbo-generator,whiletheBritishfirmParsonsreturnedabidof$12.1million.BoththeseU.S.firmsandmorethantwodozenothersweresubsequentlyconvictedofriggingbidsonequipmentsoldtoover60utilities.TheJusticeDepartment’sinvestigationrevealedsecretmeetingseveryfewmonthsovermanyyearsatwhichtheconspiringfirmssetpricesanddividedupmarketsforstandardpiecesofoff-the-shelfequipment,suchasmetersandtransformers,aswellasspecially-designed,one-of-a-kinditemssuchasturbo-generators.(Normally,thefirmsmanagedtokeepupappearancesbysubmittingbidsthatdifferedbyatleastafewpercent,unlikeintheTVAcase.)See,e.g.,ClarenceC.WaltonandFrederickW.Cleveland,Jr.,Corporations On Trial: The Electric Case(Belmont,CA:Wadsworth,1964).dTheSFCdoesnotfittheusualpatternofademonstrationprogram.Still,itgrewoutofextensiveR&Danddemonstrationoncoalgasificationbyfederalagencies,asexploredin,e.g.,LindaR.CohenandRogerG.Noll,The Technology Pork Barrel(Washington,DC:Brookings,1991),pp.259-319.Foramoregener-allypositiveperspectiveondemonstrations,basedonagreaternumberofcasestudies(albeitolder)thanincludedinThe Technology Pork Barrel,seeWalterS.Baer,LelandL.Johnson,andEdwardW.Merrow,Analysis of Federally Funded Demonstration Projects: Final Report,R-1926-C(SantaMonica,CA:RAND,April1976),summarizedinWalterS.Baer,LelandL.Johnson,andEdwardW.Merrow,“Government-SponsoredDemonstrationsofNewTechnologies,”Science,Vol.196,May27,1977,pp.950-957.eJayApt,DavidW.Keith,andM.GrangerMorgan,“PromotingLow-CarbonElectricityProduction,”Issues in Science and Technology,Vol.23,No.3,Spring2007,pp.37-43,quotationsfromp.42.f“AccordingtoTransparencyInternational...theconstructionindustryisthemostcorruptintheworld.”MinaKimes,“Flour’sCorporateCrimeFighter,”Fortune,February16,2009,p.26.Alsosee,onrecentmisdeedsintheelectricalequipmentindustry,“BavarianBaksheesh,”Economist,December20,2008,p.112,re-portingthattheGerman-basedfirmSiemens“pleadedguiltytochargesofbriberyandcorruptionandagreedtopayfinesof$800minAmericaand€395m($540m)inGermany,ontopofanearlier€201m.”

FutureGenbeganin2003asademonstrationofcommer-

cial-scaleIGCCcombinedwith(pre-combustion)CO2capture

andsequestration.14DOEhadbeenpushingcoalgasification

since the late 1970s, as a way to reduce dependence on

importedoilandgas;whentheagencystartedtofundR&Don

CO2captureandstorage in1997, IGCCbecame itspreferred

route to separation.15 The 2008 restructuring replaced the

IGCC/CCS demonstration with a potpourri of smaller-scale

undertakingsunderthesamename.Whiletheactualreasons

for the restructuring remain unclear, the episode further

damagedDOE’sreputationasareliablepartnerindemonstra-

tions. The Obama administration stated in June 2009 that

FutureGenistoberesurrectedagain,butdetailshadnotbeen

announcedasourreportwasbeingcompleted.

Demonstration, of course, is just a prelude. As one

workshop participant put it,“We need to get people to do

whathastobedone,notjustprovethatitcanbedone.”Rather

than expecting research to map out pathways to new

technology, a demonstration-centered strategy would press

forwardwiththeaidofjust-in-timeresearchtosolvetechnical

14TheoriginalFutureGenprogramwasclassifiedas“R&D”underthe2006EnergyPolicyAct,sothatprivate-sectorcostsharingcouldbelimitedto20percent.Asrestructuredin2008,FutureGenbecamea“commercialdemonstration,”requiring50percentcostsharingandquashingindustryinterest.15Energy Research at DOE: Was It Worth It?,pp.174-177;Prospective Evaluation of Applied Energy Research and Development at DOE (Phase Two) (Washington,DC:NationalAcademiesPress,2007),pp.132-149.



problemsastheyappeared. Indeed,this isthetruelessonof

the Manhattan Project: that the central, near term role for

research in accelerating innovation is to answer questions

(often unexpected ones) that emerge during design,

development, testing, and demonstration. Such questions

maybequitefundamental,asweremanyofthoseencountered

andansweredduringtheManhattanProject.

6.3 Beyond Energy and the Energy Department Wehavefirstnotedthedearthofgovernmentcompetitors

to DOE in the energy innovation domain, and second the

agency’sspottyrecordincooperatingwiththeprivatesector,

which reflectsanagencycultureandstructureof incentives

that rewards scientific achievement more than technical

management (with exceptions for weapons programs),

perhaps reinforced by a lack of consistent attention from

Congress. Because DOE has not sought to develop, reward,

andretainpersonnelthatunderstandindustryconcernsand

engineering considerations for complex, large-scale energy

systems, it too often embarks on demonstration projects

without fully comprehending the priorities of private-sector

participants,losingcredibilityandgoodwill.16

Theseobservations leadtoa third,perhapsunexpected

point: There is no necessary reason why DOE should be

chargedwithresponsibilityfordemonstrationsoftechnologies

toreduceCO2emissions.Attheleast,policymakersmightseek

tocreatecompetitionforDOEanditslaboratories,giventhat

competition in defense has been such a powerful force in

stimulating military innovation. DoD, indeed, is one of the

obviouscandidatestohelpdriveenergyinnovation,asithas

done for years with jet engines/gas turbines. Policymakers

could also move to a new model for confronting climate

change,treatingGHGreductionasapublicgood,something

liketheprovisionofwatersupplyandwastewatertreatment

servicesasmattersofpublichealthandwelfare,apolicywith

enormous benefits for human health and longevity (the life

expectancyofAmericansatbirthrosefrom48yearsin1900to

60by1930).

For years, the Pentagon has been a major customer for

energy technologies and energy services. The largest PV

system operating in the United States supplies electricity to

NellisAirForceBase,Nevada.DoDnowspendsover$1billion

annuallyonalternativeenergyR&D, inadditiontosome$10

billionannuallytofuel itsships,planes,groundvehicles,and

generators, and another $4 billion for electrical power.17

Twelve retired generals and admirals comprising the CNA

MilitaryAdvisoryBoardrecentlyconcluded:“Byaddressingits

ownenergysecurityneeds,DoDcanstimulatethemarketfor

newenergytechnologies....[T]heDepartment’shistoricalrole

astechnologicalinnovatorandincubatorshouldbeharnessed

tobenefitthenationasawhole.”18Thejetenginehasbeena

prime example. Once proven afterWorldWar II, jet engines

movedfromfighterstobombers,civilianairliners,and,inthe

formofgasturbines, totransports,helicopters,smallernaval

vessels, and the Army’s M-1 Abrams tank. By about 1980,

efficiency, initially very low, had improved to the point that

utilitiesbeganbuyinggasturbinestogeneratepeakingpower.

Major contributions to continuous innovation came from

military R&D and procurement and from the lessons of

operationalexperiencefedbacktomanufacturersbyboththe

armedforcesandcommercialairlines.

Liquid fuels account for more than three-quarters of

DoD’s energy spending (jet fuel, also used for many non-

aviation purposes such as generators, predominates).

Supplying operational forces, especially in combat zones, is

both costly and dangerous (as it has been in Iraq and

Afghanistan)andreducingconsumptionthroughconservation

andefficiency(e.g.,ingroundvehicles)hasbecomeapriority.

DoDalsomanagesbuildingswithovertentimesthefloorarea

ofthenon-militarybuildingsoverseenbytheGeneralServices

Administration.IncludingthethousandsofbasesintheUnitedStatesandabroad,somethesizeofsmallcities,DoDspends

16Inunusuallypointedlanguageforsuchabody,aNationalResearchCouncil(NRC)committee,attheendofamulti-yearstudythatincludedmanysitevisitsandmuchinterviewing,reportedthat“DOEinvestslittleinhumanresourcedevelopmentforprojectmanagementcomparedwiththeeffortsofotherfederalagenciesorprivatecorporations....TherearesimplytoofewqualifiedDOEprojectdirectorsandprojectmanagementsupportstaffforthenumberandcomplexityofDOEprojects.”Progress in Improving Project Management at the Department of Energy: 2003 Assessment(Washington,DC:NationalAcademiesPress,2004),p.3.DOE’srelianceoncontractpersonnelincriticaldecisionmakingpositionsdrewparticularcriticisminseveralCSPO/CATFworkshops.Whileotherfederalagencies,includingDoD,havelostsomeoftheirbestpeopleoverthepasttwodecadesorsotoretirementsorreplace-mentofpermanentstaffbycontractorsinthenameofprivatization,DOE’sresponse,bytheevidencetheNRCcommitteeassembles(whichcoversonlyaportionofDOEactivities,primarilythosedealingwithcapitalinvestmentprojectsandcleanupofnuclearsites),evidentlycomparesunfavorablywithotherpartsofgovernment.17Report of the Defense Science Board Task Force on DoD Energy Strategy: “More Fight – Less Fuel” (Washington,DC:OfficeoftheUnderSecretaryofDefenseForAcquisition,Technology,andLogistics,February2008).18Powering America’s Defense: Energy and the Risks to National Security(Alexandria,VA:CNA,May2009),pp.viii.CNAistheparentoftheCenterforNavalAnalyses,anFFRDCthatadvisestheNavyandotherpartsofDoD.


about $2½ billion annually just for electricity. Many vitalsecurityfunctions,beginningwithnationalcommand,control,and communications, depend almost entirely on thecommercial grid. DoD has both means and motivation tocontribute to robust “smart grid” technologies and to grid-independent generation, which is especially important inremote locationswhereoperationalunitsnowrelyondiesel

generators.Inshort,DoD is positioned to be the type

of discerning customer for energy innovation

that it has been in the past for IT, jet engines,

and other once-novel technologies. As we havenoted,whileDoDhasbeenfarfromperfect,itshistoricalroleincatalyzinginnovationhasoftenledtorapidimprovementsinbothpriceandperformanceandhencetomarketspillovers.OurpointisnottoidealizeDoD,buttohighlightitsinstitutionalcapabilitiesandtheirrelevanceforenergy-climateinnovation. Fromadifferent(andneglected)perspective,governmentcould treat GHG reduction as a public works undertaking,aimedatprotectingfuturesociety.CustomersforGHG-reduc-ing systems and equipment might include public agenciesor a government enterprise (theTennesseeValley Authority,TVA, is one sort of government-sponsored enterprise),public-private partnerships, or private firms operating undergovernment contract. The public works model opens newapproachestoGHGmanagementandtechnologyinnovation.Sometasks,forexample,mightbedelegatedtostateandlocalauthorities, which already collect trash, maintain water andsewer systems, and attempt to safeguard urban air quality(eventhoughpollutantstravellongdistances).Similarprinci-ples could be applied to GHGs and could strengthen thepracticalandpoliticalcasefornationalstandards(andreducethelikelihoodofdisparatestateorregionalstandardsforGHGcontrol). The public works model also opens the way forassigning tasks to DOE competitors, possibly including theEnvironmentalProtectionAgencyoreventheArmyCorpsofEngineers.19Publicly-ownedutilitieswithmanypowerplantsand extensive operating experience such as TVA couldbecometestsitesandearlycustomersforinnovativeenergy-climate technologies. The federal government currentlybudgets over $60 billion annually for infrastructure invest-ments,exclusiveofhighways,andstateandlocalgovernments

spend several times as much. Policymakers could approachGHGcontrolinsimilarfashion. Ourintentisnottopointtoparticularagenciesascandi-dates for undertaking CO2 management, simply to illustratethe larger point that organizations other than DOE may bebetter suited for vital energy-climate innovation tasks.Therearealsopossibilitiesoutsidegovernment,including(non-DOE)federallyfundedresearchanddevelopmentcenters(FFRDCs),severalofwhichspecializeinlarge-scalesystemsplannedandbuiltforthefederalgovernment.TheAerospaceCorporation,forinstance,wasestablishedin1960asanonprofitorganiza-tion to coordinate the efforts of defense firms on complexaircraft,missile,andspaceprogramsthatthreatenedtoover-whelm the management capabilities of the Air Force. GiventhatDOEseemstohavebeenoverwhelmedbymanagementofprogramssuchasFutureGen,asimilarapproachmightbeappropriateforlarge-scaleenergy-climateprojects.

6.4 DOE and Innovation Energyinnovation,quitesimply,hasnotbeencentraltothe mission of DOE (or its antecedents, the AEC and ERDA),withtheexceptionofnuclearpower,whichwasmismanagedbecause there was little agreement within government onwhatthenationshouldbetryingtoaccomplish,andwhy.Un-likenuclearweaponsprograms,theenergysideofAEC/ERDA/DOEhasnever,fromtheverybeginning,hadanagreed-uponsetofobjectives.Thenationallaboratoriesfoughtfor,gained,and continue to exploit a level of autonomy considerablygreaterthanfoundintherestoffederallaboratorystructure.Incontrast, DoD laboratories, staffed and managed by civilianswho answer to military officers, must satisfy the services ofwhichtheyarepart. DARPA,whichisindependentoftheservices,avoidedthedifficulties of laboratory management by avoiding laborato-ries;allwork isconductedexternally.NASAlearnedarelatedlessonfromtheNationalAdvisoryCommitteeforAeronautics(NACA),whichhaditsownlaboratoriesandavoidedriskyproj-ectstoavoidcriticismintheeventoffailure.DuringWorldWarII,policymakersdeclinedtotrustthoselaboratorieswithcriti-cal,perforcerisky,projectssuchasjetenginedevelopment,inwhich the United States had fallen far behind Britain andGermanyduelargelytoNACA’scaution.WhileNASAcontinuesto operate laboratories it took over from NACA, it does not

19Sinceabout1970,CongresshasassignedtheCorpsofEngineersnewtasksinbothenvironmentalprotectionandremediation.Beforeaddingothers,policymakerswouldbewisetofirstassurethemselvesthatagency’srecentmanagerialdeficienciescanbeovercome.SomeofthesearesummarizedinCorps of Engineers: Observations on Planning and Project Management Processes for the Civil Works Program,GAO-06-529T(Washington,DC:GovernmentAccountabilityOffice,March15,2006).


dependsoheavilyonthemandcontractsoutamuchhigherproportion of work to industry. The National Institutes ofHealth (NIH) operate extensive in-house laboratories, butthey engage in basic research almost exclusively, whichexposes them to scrutiny by the academic biomedicalresearch community. DARPA’s shortcomings are revealed ifnoneofthearmedforcesadoptsatechnologyithaspushed,NASA’swhenamissionfails,NIH’swhencitationanalysisre-vealsunproductiveresearchgroups.Overtheyears,theAEC/ERDA/DOElaboratoriesproducedinnovativenuclearwarheaddesignsand research thatpassed the tests forgoodscienceimposedbydisciplinarycommunities.ButDOE’scontributionsto innovationinenergytechnologieshaveoftenbeenques-tioned, and commentary during our workshops reinforcedskepticism about DOE’s capacity to drive the innovationsystemindirectionsneededtoeffectivelygrapplewithGHGemissionsandclimatechange.20 Dissatisfaction with the pace of energy innovation andwithDOE’scontributionsledCongress(whosehistoricallackoffocus has, to be sure, been an important contributor to theproblem) to establish ARPA-E in 2007. To succeed, ARPA-E,

which did not receive an appropriation until 2009 ($415million,allbut$15millioninstimulusfunding),willnowhavetoplotacourse.Ithasannouncedambitiousifnotgrandiose-soundinggoals:to“enhancetheeconomicandenergysecurityoftheUnitedStatesthroughthedevelopmentofbreakthroughenergy technologies; reduce the need for consumption offoreign oil; reduce energy-related emissions, includinggreenhouse gases; improve the energy efficiency of alleconomicsectors;andensurethattheUnitedStatesmaintainsa technological lead indevelopinganddeployingadvancedenergytechnologies.”21

What ARPA-E must do first is find clients and researchperformers, build constituencies, and, most important of all,defineamissionforitselftoguideoperationaldecisions,justasDARPAdidinitsearlyyears.DARPAdidnottrytobeallthingstoallmilitaryspecialties,asBoxHexplains. If ithad, itwouldneverhavebecomethemodelforARPA-E.Andoverthepastdecade or so, DARPA’s performance has come underconsiderable criticism and the agency could be facing ashakeup of its own even as ARPA-E managers try to decidewhichfeaturesmeritemulation.

Box HARPA-E and DARPA: Mission and Management

OrganizationalaspectsofDARPA,suchasafamouslyleanandqualifiedstaffandanabsenceofinternallaboratoriestosoakupfundsbetterspentontheoutside,havesometimesbeenmistakenforitsessence.Infact,thesefeatureswereputinplaceasmeanstoanend,thatendbeingareasonablechance,first,ofsurvivinginthefaceofpowerfulmilitaryoppositionintheearlyyears,whenthearmedforcesfoughttransferofevenasmallportionofR&Dfundstociviliancontrol.aOnceDARPA managers had won that battle, they had to keep it from flaring up again by delivering results in the form ofcontributionstofunctionalweaponssystems.Pressuretoperformshapedtheagency.

Although DARPA has always funded some relatively basic work, other agencies and subagencies within DoD nurtureresearchinmanyfieldsofpotentialinteresttothemilitary,rangingfromcombustionkineticsinsupersonicramjetstothesocialstructuresofterroristgroups.DARPA’sprimaryjobhasnotbeenresearch,butthedevelopmentanddemonstrationofadvancedtechnologiesthatcanrelativelyquicklybe incorporated intofieldedsystems. Itsmission isuniquewithinDoD,quitedeliberatelycuttingacrossthetraditionalroles,missions,andself-imagesoftheservices,whichcanconstraininnovation.Withoutitsnow-acceptedmission,DARPAwouldbesuperfluous,andwouldnothaveachievedthetechnicalsuccessesininformationtechnology,sensorsandsurveillance,directedenergyweapons,andaerospacethatmadeitthemodelforARPA-E.b

Putsomewhatdifferently,DARPAhasanidentifiablesetofclients:high-rankingofficersintheArmy,AirForce,Navy,and

20AlookatthemanyR&DroadmapsandrelatedplanningdocumentspreparedbyDOE,suchasthereportsissuedsince2002intheseriesentitledBasicResearchNeeds,runningtowellover2500totalpages,makesplainthatmostoftheagency’sresearchhaslittletodowithtechnologieslikelytoshowupcommerciallyoverthenexttwodecadesorso.Assum-marized,forinstance,in“The‘BasicResearchNeeds’WorkshopSeries,”OfficeofBasicEnergySciences,OfficeofScience,U.S.DepartmentofEnergy,April2007,<www.er.doe.gov/bes/reports/files/brn_workshops.pdf>,theresearchoutlinedinthesemanyreportsmaywellbeneededtolayfoundationsforinnovationinthelonger-termfuture.Buttheverydensityoftheseroadmaps,andtheiremphasisonexoticresearchtopicsopaquetoallbutdisciplinaryspecialists,pointstothevacuumthatexistsinplanningfornearer-termworkthatmighthelpdealwithmoreimmediateenergy-climateproblems.21“FactSheet:AHistoricCommitmenttoResearchandEducation,”WhiteHouse,OfficeofthePressSecretary,Washington,DC,April27,2009,p.3.


MarineCorps.Tosurvive,theagencymustkeepthemhappy,orat leastkeepthemfromgrumblingtoomuch.DARPAmanagershavealwaysunderstoodthat.(Generallyspeaking,militaryleaderswanttoseemoneyspentfortoday’sweapons,nottomorrow’s,justascorporateexecutiveswanttoseeproductsthatwillbringinrevenuesnextfiscalyear,ratherthanR&Dthatmight,ormightnot,payoffaftertheyhaveretired.)DARPAproceededmuchasdidVannevarBush’sOfficeofScientificResearchandDevelopment(OSRD)duringWorldWarII.Bushandhisdeputiespushedconstantlytogeteffectiveweaponsintoproductionandintothehandsofcombatforcesintimetomakeadifference.DARPAmanagersalsoworkedtogettechnologiessuchaselectronicimaging(inplaceofoptics)outofthelaboratoryandintothefield.OSRDsteeredthebulkofitsfundstouniversity-basedorganizationsandafewindustriallaboratories,suchasthoseofAT&T:thedefenseindustryassuchhadhardlyanyresearchcapabilityuntilafterWorldWarII.Butinthe1950sthischangedandDARPAcouldtakeadvantage:mostoftheagency’sfundinggoestoindustry.

Insettingobjectivesandfindingaviablestrategy,DARPAhadapowerfulsourceofleveragethatwill,atleastinitially,beunavailabletoARPA-E.DefensefirmsviewR&Dasawedgeintotheprocurementcontractsonwhichtheyexpecttoearntheirprofits.Asaresult,DARPAproposalrequestsgettheattentionoftopmanagers,whooftenassigntheirbestengineersand scientists to DARPA-funded programs. ARPA-E has no such carrots to motivate industrial contractors, who havesometimesbeenreluctanttolettheirmoretalentedpeopleworkonDOE-sponsoredprograms(muchlessrevealtheirbestideas).IfasaconsequenceARPA-EfindsitselfsteeringtoomuchofitsR&Dfundingtothenationallaboratories,oreventouniversities, itwillnotmovemuchbeyondtheoldmodelsofenergyR&Dandmaynot realize itspotential. (FromtheestablishmentofDOEuntilitwasreorganizedoutofexistencein2000,theAdvancedEnergyProjectsDivisionoftheOfficeofBasicEnergySciencesfundedR&DwithobjectivesostensiblysimilartothoseannouncedforARPA-E.)

DARPAhasalwaysunderstood that itsprojectsmust,over time, show results that satisfy thearmed forces.ARPA-E’smanagerswillhavepoliticalappointeeslookingovertheirshoulders,notcareermilitaryofficerswhoknowwhattheywantandwhy,andonlythe“moralequivalentofwar”todisciplinedecisions.cThatmaybethetoughestproblemofall.What energy-climate innovation needs is goal-directed R&D—work that falls into what has been called Pasteur’squadrant—pursuedconsistently,withoutwanderingoffintoresearchforthesakeofresearch.dSuchresearchhasnotbeenDOE’sstrength.

aRobertFrankFutrell, Ideas, Concepts, Doctrine: Basic Thinking in the United States Air Force, 1961-1984,Vol. I(MaxwellAirForceBase,AL:AirUniversityPress,December1989),pp.477-504,recountstheeffortsoftheservicestostallthecreationofDARPA.Theagency’srecentbudgetshavebeenintherangeof$3billion,muchlargerthanARPA-E’sappropriationbutlessthan5percentofallDoDR&D.bOnDARPA’sR&Dportfolio,seeRichardH.VanAtta,SeymourJ.Deitchman,andSidneyG.Reed,“DARPATechnicalAccomplishments,VolumeIII:AnOverallPerspectiveandAssessmentoftheTechnicalAccomplishmentsoftheDefenseAdvancedResearchProjectsAgency:1958-1990,”IDAPaperP-2538,InstituteforDefenseAnalyses,Alexandria,VA,July1991.cThephraseisofcourseJimmyCarter’s:“ThePresident’sProposedEnergyPolicy,”Vital Speeches of the Day,April18,1977,pp.418-420.dTheallusionistoDonaldE.Stokes,Pasteur’s Quadrant : Basic Science and Technological Innovation(Washington,DC:Brookings,1997),thetitleofwhichreferstodirectedbutnonethelessrelativelyfundamentalworkofthesortthatledtothetransistoraswellaspasteurization.

WhileDOE’slimitationsmayappeartostandinthewayofmore effective energy innovation policies, the systemsperspective we take in this report does not lead us towardrecommendations to “reform” the agency. The politicalobstacles to achieving such a task are daunting, and haveproven insurmountable in the past. Indeed, the question ofwhat a“reformed” DOE would look like would be endlesslydebatedandprobablyneverresolved.Similarly,ARPA-Emayormay not prove to be an effective work-around to some ofDOE’s limitations.Yet we want to emphasize that if energy-climate innovation is viewed largely as a problem for DOE,then the problem itself has been misunderstood, for the

simplereasonthatmostinnovations,nomatterthetechnologydomain,comefromindustry. Policymakers will have to decide how to act so thatenergy-climatetechnologiesadvancealongmultiplefrontsinmultiple institutional settings, enlisting the appropriatestrengthsofthepublicandprivatesectorsandaccountingforthe risks and uncertainties that accompany innovation. Thedeliberationsinourworkshops,combinedwiththelargebodyofknowledgeandexperiencethatnowexistsconcerningthepost-WorldWarII innovationsystem,pointustowardasmallset of principles and recommendations that can help guidepolicy,andweturntotheseinourfinalsection.


The U.S. economy has been fundamentally transformed since World War II by technological change, international

competition,andshiftinggovernment regulatorypolicies.Some industries,notablycomputersand informationtechnology,

haveflourished,innovatingfuriously.Othershavefaltered.Innovationinenergy-relatedtechnologieshasbeenhaltingandslow

forfourmainreasons:

o Market pull has been weak and customers undemanding.Theexperienceofutilitieswithnuclearpowerdeepened

theiraversiontoriskandregulatoryshiftscontributedtodeclines in internal technicalcapacity. Industrialpurchasersof

energy,withexceptionsforenergy-intensivesectorssuchasprimarymetals,havemostlyviewedthisasaminoritemintheir

coststructures.Individualandhouseholdcustomerscaremoreaboutfirstcoststhanlife-cyclecosts(iftheyareevenaware

of the latter), and often choose personal vehicles and home appliances for reasons that have little to do with energy

consumption.Energyitself,especiallyelectricity(anundifferentiatedcommodity),offerslimitedscopeforinnovationsthat

affectproductattributes.IntheUnitedStates,demandforGHGmitigationhasyettobetranslatedthroughgovernment

actionintoeffectivedemandforinnovation.

o Industry has often been uncomfortable working with the Energy Department.FewofDOE’slaboratories,focusedon

weaponsandonbasicscience,havemaintainedclosetieswiththeprivatesector.R&Ddirectionshavebeenchosenwith

limitedattentiontonationalinterestsandtoooftenhavebeenshapedbytheinterestsofparticularindustriesandresearch

communities,andthusreflectvaryingcombinationsofpoliticalforces(coal)andlaboratorypredilections(laserfusion).Asa

partnerindemonstrations,DOEhasbeeninconsistent.

o Government has not systematically used purchases to foster energy-related innovations.The military, while a

majorcustomerformanytechnologies,hasbeenprimarilyinterestedinperformance—faster,morepowerfultanksfor

theArmy,forexample,despiteheavyfuelconsumptionpenalties.Onlyinthepastseveralyearshavethearmedforces

beguntostressenergysecurityandfuelefficiency.TheGeneralServicesAdministrationhasbeenreluctanttouse its

leveragetoencourageenergyconservationingovernment-ownedor-leasedbuildings.‘BuyAmerica’preferenceshave

trumped purchases of hybrid vehicles—until recently only available from foreign-owned automakers—and publicly-

ownedpowercompanies,afterdecadesofattacksfromprivatepowerinterests,havebeennomorewillingthantherest

oftheutilityindustrytoinnovate.

o The nation’s energy innovation policy has lacked a clear mission and strategy.Whileextendingsubsidiestorenewables

inthe1970sandagaininrecentyears,governmentotherwisehasreliedheavilyonR&Dfunding,neglectingotherpolicies

that,aspartofaportfolio,wouldfosterinnovation.DOE,unlikeDoD,hasnocompellingmissiontodisciplineitsplanning:it

isnotthe“customer”forendproducts,haslittleincentivetofosterinnovationbyindustrialfirms,andhasnotorganizeditself

7. Conclusions and Recommendations Effective energy-climate innovation policies will be consistent with a small number of guiding principles and take advantage of a small number of highly leveraged points of opportunity.

37CONCLUSIONS AND RECOMMENDATIONS

broadly to do so. Unlike DARPA and NSF, DOE relies heavily on long-term government employees (even if technically

employedbytheFFRDCsthatadminister16ofDOE’s21laboratoriesandtechnologycenters)andhasnotbeenveryopen

to outside technical influence. Unlike the National Institute for Standards andTechnology, DOE employees do not see

themselves primarily as supportive of private sector agendas and endeavors. The end result has been ineffective and

inefficientR&Dspending.

Withoutstringentregulations,incentivesforadoptionofcostlynewtechnologieswillbetooweaktospuraction.Without

moreR&D,moreeffectivelymanaged,thefoundationsforfutureinnovationwillbeporousandweak.Nonetheless,tohasten

energy-climateinnovation,thefederalgovernmentwillhavetodomorethanregulate,andmorethanboostR&Dspending.The

UnitedStates(andtheworld)willhavetoconfrontclimatechangewithtechnologiesthatarealreadyinsight.Theseexist,asthe

CSPO/CATFworkshopsrevealed,evenforatechnologyasseeminglyvisionaryasaircapture.Theyneedjust-in-timeresearchto

help guide and steer development. But there is no point expecting, much less waiting for, breakthroughs.They cannot be

predicted,andsometimesgounrecognizeduntilwellaftercommercialization(thecaseforthemicroprocessor).

Weconcludebyofferingfouroverarchingprinciples,andfourspecificpolicypriorities,thatcanprovideaframeworkfor

assessingenergy-climatepolicyandlegislativeproposals,andguidegovernmentdecisionmakersintheireffortstoenhancethe

nation’senergy-climateinnovationenterprise.

Principles to Guide Energy-Climate Innovation Policies 1. The energy system is highly complex, as are energy-climate innovation processes.Theenergysystemhasgrown

andevolvedovermorethanacentury;onlyrecentlyhaveclimateconsiderationsbeguntoplayanypartinitsdevelopment.

Policies that influence energy technology innovation form an incomprehensible thicket of subsidies, incentives, intellectual

propertyregimes,environmentalregulations,andsoon.Understandingofthesystemisdistributedamonglargeanddiverse

groupsofpeopleworking inamultiplicityof institutional settings.Theyknowdifferent things,believedifferent things, and

respondtodifferentsetsofincentives.Organizationalfeatures,suchasthetechnologicalconservatismoftheutilityindustry,

persistoverdecades.Suchfactorsshouldbeunderstoodassystemconditions.Changingthemtoyielddesiredoutcomesmay

notbepossible.

At the same time, the daunting diversity of institutions, of technologies, of decision makers, and of policies should be

recognizedasanopportunity.Thearrayoftechnologiesnowinuseorbeingactivelyexploredtogenerateandconserveenergy

meansthatmanyplausiblepathsareopenalongwhichinnovationcanflourish.Failureinonecornerofthesystemtoeffectively

solvesomeproblemassociatedwithclimatechangedoesnotimplyfailureofthesystemasawhole.

Systemcomplexityalsomeansthattherewilloftenbegreatuncertaintyattachedtoinnovationpolicydecisions,whichin

turndemandsawillingnesstotakesensiblerisksandaccepttheinevitablefailuresandinefficiencies.Improvisationandlearning

willberequired.Thisperspectivesuggeststwocomplementaryapproachesforpolicymakers.Thefirstistoencourageinnovation

alongmultiplepathways,withaportfoliooftechnologiesandinstitutionalsettings.Thesecondistolookforpointsofintervention

wherefocusedactioncanleveragebiggains.Ourrecommendations,below,addressbothapproaches.

2. Energy innovation over the next 10-20 years, and perhaps much longer, will be dominated not by “breakthroughs,”

but by incremental advances in technologies that are either in use now, or that are ready to be scaled up for use.The

roleofresearchinthisprocessisoftenmisunderstoodbypolicymakersandscientistsalike.The government supports research

aimedatincrementalinnovation(forexample,insomeareasofagricultureandmedicine)butitalsosupportsresearchinthe

hopethattheresultswilleventuallyleadtomajortechnologicalbreakthroughs.Butnoonecanknowwhatwillemergefrom


either sort of research. Sometimes work with modest initial objectives spurs major innovations.This was the case with the

microprocessor,whichdidnotentailnewresearch,simplythedesignandfabricationofatypeofintegratedcircuitthathad

beenwidelydiscussedwithinthesemiconductorindustrybutnotpreviouslycommercialized.Oftenworkpursuedinhopesof

somesortofbreakthroughleadstolittleofpracticalconsequence,asillustratedbymuchhigh-temperaturesuperconductivity

researchsince1986.Genuinebreakthroughs,then,canneitherbepredictednorplanned.

Policymakers,moreover,havefewtoolstouseinsearchofbreakthroughs:primarily,basicresearchfundingandintellectual

propertyprotection.Bothareblunt instruments,sincetherewillneverbeenoughmoneytofundall theresearchforwhich

scientistsandengineerscanconstructplausiblejustifications,andintellectualpropertyrightssometimesstifleinnovationrather

than encourage it. Most innovation, in energy technology and elsewhere, occurs incrementally and in the private sector.

Todependon“breakthroughs”asthecentralstrategyforbringinginnovationtobearonclimatechangeistopromulgatefalse

hopesofeasysolutions to theenergy-climatedilemma.Whilegovernmentmustofcourse invest in researchtoprepare for

innovationsseveraldecadesinthefuture,creatingthetechnologicalcapacitytoachieveGHGreductionsstartingtodayrequires

energy-climateinnovationpoliciesthataccelerateratesofperformanceimprovementforexistingtechnologies.

3. A decarbonized energy system is a public good akin to national defense, individual and community health and

safety, and protection from natural disasters. Inprovidingsuchpublicgoods,theU.S.governmenthasoftenactedtospur

technologicalinnovation.TheKoreanWarcementedU.S.commitmenttoahigh-technologymilitary.Sincethe1960s,recognition

that only government could safeguard the environment has underlain policies for both protection and remediation. If

governmenttreatsenergy-climateinnovationsasapublicgood,thenmajornewavenuesofpublicpolicyandinvestmentopen

up,forexamplethroughpurchasingandprocurement,notjustresearchanddevelopment.

4. Energy-climate innovation policies must be tailored to particular technologies and suites of technologies.There

aremanyprovenpolicytoolsavailableforstimulatinginnovation.Differenttechnologies,atdifferentstagesofdevelopment,are

likely to respond todifferentpolicyapproaches.Yetgovernmenthas reliedheavilyonR&D funding,whileneglectingother

policiesthat,aspartofaportfolio,wouldfosterinnovation.Forexample,thePentagon’sinsistenceoncompetitionandnon-

proprietarytechnologieshadpowerfullong-termeffectsoninformationtechnology,asdiditssupportforacademicprograms

infieldssuchassoftwareengineeringandmaterialsscience.

Recommendations for Priority Action 1. Congress and the administration must foster competition within government to catalyze energy-climate

innovation. In particular, the United States relies far too heavily on the Energy Department. Competition drove military

technological innovation—East-West competition, competition among defense, aerospace, and electronics firms, and also

competitionamongthemilitaryservices.ThenucleararmsraceshapedDOEanditspredecessors,andthethreatoftakeoverby

thearmedforces,whichbelievethatmilitaryprofessionalsshouldhavefullcontrolofallweaponsatallstages,includingresearch,

haskeptthenuclearprogramsontrack.Competitionamonggovernmentagencieshasbeenaneffectiveforceininnovation

across such diverse technologies as genome mapping and satellites. No such competitive forces exist for energy-climate

technologies. Expertise and experience related to energy-climate innovation policy exist today in many parts of the public

sector other than DOE, including DoD, the Environmental Protection Agency, and state and local governments that have

involvedthemselvesinclimatechangefromfrustrationoverstalemateinWashington.Competitionbreedsinnovation.Thatis

trueineconomicmarketsanditholdsforgovernment,too.Forexample,iffinancialassistanceforfirmsdevelopingrenewable

energysystems is tobeprovided through low-interestorguaranteed loans,agencieswithexpertise infinance, suchas the

TreasuryDepartment,shouldhaveagreaterrole,ifonlytohastenlearningelsewhereingovernment.

39CONCLUSIONS AND RECOMMENDATIONS

2. Government should selectively pursue energy-climate innovation using a public works model, especially for

GHG-reducing technologies such as post-combustion capture of power plant CO2 and air capture that lack a market

rationale. OnewaythattheU.S.governmenthashistoricallyfulfilleditsresponsibilitytoprovideimportantpublicgoodslike

nationalsecurityandpublichealthisthroughdirectparticipationinthecreationofappropriateinfrastructure.Recognitionof

GHGreductionasapublicgood,notunlikecleanwaterandwastedisposal,wouldopennewapproachesforcreatingenergy-

climateinfrastructureinsupportofinnovationandGHGmanagement.Thepublicworksmodelopensthewayforassigning

tasks toanewarrayofDOEcompetitors, includingpublicagencies,public-privatepartnerships,andprivatefirmsoperating

undergovernmentcontract.

3. Policymakers must recognize the crucial role of demonstration projects in innovation, especially for technologies

with potential applications in the electric utility industry.Government-sponsoreddemonstrationprogramshavealong-

established place in U.S. technology and innovation policy. Industry participation is essential, since the primary purpose of

demonstrations is to reduce technical and cost uncertainties. Moreover demonstration is a routine part of technology

developmentinmanymanufacturingindustries,somethingthatcompaniesandtheiremployeesknowhowtodo;theirskills

andexpertiseshouldbetappedmoreeffectively.WellplannedandconductedprogramscouldadvancePCC,aircapture,and

othertechnologiesthatpromisehelpinaddressingclimatechange,butdemonstrationshavearegrettablereputationinenergy-

relatedtechnologies.Agencieswithmoreexperienceconductingormanagingdemonstrationsofcomplextechnicalsystems

shouldbecalleduponasappropriate.

4. Direct government purchases of innovative products and systems could greatly accelerate innovation along

some technological pathways.Asmentioned,demonstrationprojectsmeritwellconceivedandmanagedexpenditures,and

deploymentofavailableifnotfullyproventechnologiessuchasCO2captureandstoragefromcoal-burningpowerplantscould

bepursuedusingapublicworksmodel.Morebroadly,governmentbodies,whichspendlargesumseachyearonpurchases

withimplicationsforGHGreleaseandclimatechange(officebuildings,motorvehicles,transitsystems)canbemajorcustomers—

smarter customers—for energy-climate innovations, helping to create early markets and fostering confidence in advanced

technologies, includingthosethatarenotyetpricecompetitive.Procurementpoliciesthatstimulateincrementalinnovation

throughexperience-basedtechnicallearning,sometimesleadingtowhatseemsinretrospecttoberadicalinnovation,hasbeen

apowerfulforceinelectronicsandaerospace,andDoD’suniquelyhugescaleandtechnicalresourcesshouldbeexploitedmore

aggressivelytofosterenergy-climateinnovation.Despiteitsimperfections,nopartofthefederalgovernmentcomescloseto

DoDinbroad-basedtechnologicalcompetence.

The U.S. government has no choice but to rely on the private sector for innovation.That is where the knowledge and

experiencelie.Sofar,ofcourse,theincentiveshavebeenlacking.Butitwilltakemorethanapriceoncarbon.Governmentmust

buildstrongerbridgestoindustrysofarasenergy-climatetechnologiesareconcernedandbecomeasmartercustomer,justas

DoDhasoftenbeenasmartcustomerwithdeeppocketsformilitaryinnovation.Bytreatingclimatemitigationasapublicgood

andGHGreductionasapublicworksendeavor,analogoustopublichealthandsafetyinfrastructure,vaccinestockpiles,dikes,

levees, weather forecasts, and national defense, the United States can begin to show other countries how to build energy-

climatetechnologiesintothefabricoftheirinnovationsystemsandtheirsocieties.

Ourstudyraisesashortlistofpressingpolicy-relatedquestionsthatcannotbeansweredwithoutfurtherconsiderationbyexpertswho,

liketheparticipantsintheCSPO/CATFworkshops,havedeepunderstandingofparticularpartsoftheenergy-climatesystemandtheinnovation

system.Amongtheopenquestionsconcerningenergy-climateinnovation,sixstandout:

Policy Choice. Wehaveemphasizedthewiderangeofpolicytoolsavailableforaddressingenergy-climateinnovation.Thetechnologies

explored in theCSPO/CATFworkshopsexhibitdifferent levelsofmaturityandmarketprospects (forexample, the institutionalcontext for

demonstrationsofPCCseemverydifferentfromthoseforaircapture).OurcallforgreatercompetitionforDOEcouldbeansweredfirstwithan

assessmentofproveninstitutionalcapabilitiesacrossthefederalgovernmentthatarerelevanttoenergy-climateinnovation.Moreover,policy

portfoliosformanyothertechnologies,includingthoseforpowergeneration(nuclear,wind,geothermal)andinsectorssuchastransportation,

remaintobeexplored.

Climate Management as a Public Good.WehavearguedthatgovernmentshouldtreatCO2reductionasapublicgood,akintowater

andsewerservices,implementedthroughdirectgovernmentpurchase,andselectivelyemployingapublicworksmodel.Buttheappropriate

institutionalandfinancialarchitectureforthisapproachneedsfurtherthought.Itisonethingtoenvision,say,modularunitsforaircapture

producedinvolumeandshippedaroundthecountry.Itisanothertodeterminewhoshouldmanufacture,install,operate,andmaintainthem,

andselectthesites.Thereistimefordeliberation,sinceevenwithacrashprogramitwillbesomeyearsbeforeaircapturecouldbereadyfor

deployment,whilepost-combustioncaptureofpowerplantCO2,althoughnearertorealization,awaitslarge-scaledemonstration.

ARPA-E. New agencies typically have considerable programmatic latitude before settling into patterned routines, making “initial

conditions” critical. For ARPA-E these are likely to include: priority-setting mechanisms, norms for selection among competing proposals,

guidelines for sending R&D dollars outside DOE and its laboratories, and personnel policy, namely the extent of reliance on career DOE

employees.Givenpartialandimperfectparallelswithmilitarytechnologydevelopment—indefense,governmentisthecustomerandthat

simplerealitypervadesDARPA—ARPA-EshouldseektolearnnotonlyfromDARPAbutfrombusinessfirms.Astructuredassessment,something

likethebenchmarkingexercisescommonintheprivatesector,couldhelpARPA-Emanagerssetaproductivecourse.

R&D Planning. InsularplanningprocessesseemtobeonereasonforDOE’sspottyrecordindrivingenergy-climateinnovation.These

processeshavenotbeendeeplyexplored,perhapsbecauseenergypolicyspecialistshave focusedsoheavilyon thedecline inDOER&D

spendingsincetheearly1980sastoleavetheimpressionthatmoremoneyisallthatisneeded.DOEappearstoconsultlessfrequentlythan

DoDwithoutsideexperts(bodiessuchastheDefenseScienceBoard,FFRDCssuchasRAND,NationalResearchCouncilboardsandpanels).A

usefulsteptowardimprovementwouldbetobenchmarkDOE’sR&Dplanningprocesses,insomedetail,againstthoseelsewhereingovernment,

inprivatefirms,andinindustrialconsortiasuchasSematech.

Financing Innovation. Publicpolicies that speedmarketpenetrationanddiffusion in theearlyyearsofanewtechnology,whether

throughgovernmentpurchases(aswehaveemphasizedinthisreport)orfinancialsubsidiesforinnovatingfirmsandtheircustomers,actas

“innovationmultipliers,”acceleratingratesoftechnicaladvanceandcostreduction.Butfinancialsubsidiessometimesdistortmarketsindeep

andlastingfashion,anditisnotalwaysclearhowbesttosteerfinancingtoinnovatorswithoutsupporting,ontheonehand,workthatcould

gainprivatefinancingevenwithoutsubsidy,or,ontheother,poorideasbackedbypersuasiveorpowerfuladvocates.Thedizzyingprofusion

ofsubsidiesforPVinvestments,manyofthematthestatelevel,suggeststheneedforafreshlookathowbesttoprovidefinancialsupportfor

energy-climateinnovationwhengovernmentisnottheprimarycustomer.

International Innovation.Theworldinnovationsystemcanbepicturedintermsofnationalinnovationsystems(e.g.,theUnitedStates,

China)plusthetransnationalorganizations(e.g.,corporations)andinstitutions(e.g.,labormarkets,theWorldTradeOrganization)thatlinkand

inter-penetratethem.AnalysisofsuchasystemwillbefarmorecomplicatedthanfortheUnitedStatesalone.Yetclimatechangeisaglobal

problem,andcountriessuchasChinaarealreadypositionedtocontributeasactiveparticipantsininnovation.Forexample,therapidpaceof

powerplantconstructioninChinaessentiallyguaranteesaflowofincremental,experience-basedinnovations,someofwhichmaymigrateto

theUnitedStates.Indeed,theUnitedStateswillknowtheinternationalinnovationsystemismovinginproductivedirectionswhenimportsof

PCCequipmentoriginateinChinaandtheshareofPVmodulesfromMexicorisesabovethatfromWesternEurope.Ananalysissimilartowhat

wehavepresentedherefortheU.S. innovationsystem,butaimedattheglobalpicture,couldhelpclarifymeaningfuloptionsavailableto

nationalandinternationaldecisionmakers.

Looking Ahead

40 LOOKING AHEAD

41APPENDIX 1

INNOVATIONANDPOLICYCOREGROUPJane “Xan” Alexander Independent ConsultantD. Drew Bond Vice President, Public Policy, BattelleDavid Danielson Partner, General Catalyst PartnersDavid Garman Principal, Decker Garman Sullivan and Associates, LLCBrent Goldfarb Assistant Professor of Management and Entrepreneurship, University of MarylandDavid Goldston Bipartisan Policy Center; Visiting Lecturer on Environmental Science and Public Policy, Harvard UniversityDavid Hart Associate Professor, School of Public Policy, George Mason UniversityMichael Holland Program Examiner, Office of Management and BudgetSuedeen Kelly Commissioner, Federal Energy Regulatory CommissionJeffrey Marqusee Director, Environmental Security Technology Certification Program, and Technical Director, Strategic Environmental Research and Development Program, Department of DefenseTony Meggs MIT Energy InitiativeBruce Phillips Director, The NorthBridge GroupSteven Usselman Associate Professor, School of History, Technology, and Society, Georgia TechShalini Vajjhala Fellow, Resources for the FutureRichard Van Atta Core Research Staff Member for Emerging Technologies and Security, Science and Technology Policy Institute

PVTECHNICALEXPERTSDan Shugar President, SunPower Corporation, SystemsTom Starrs Independent ConsultantTrung Van Nguyen Director, Energy for Sustainability Program, National Science FoundationKen Zweibel Director, Institute for Analysis of Solar Energy, George Washington University

PCCTECHNICALEXPERTSHoward Herzog Principal Research Engineer, MIT Laboratory for Energy and the EnvironmentPat Holub Technical and Marketing Manager, Gas Treating, Huntsman CorporationGary Rochelle Carol and Henry Groppe Professor in Chemical Engineering, University of Texas at AustinEdward Rubin Alumni Professor of Environmental Engineering and Science; Professor of Engineering & Public Policy and Mechanical Engineering, Carnegie Mellon University

AIRCAPTURETECHNICALEXPERTSRoger Aines Senior Scientist in the Chemistry, Materials, Earth and Life Sciences Directorate, Lawrence Livermore National LaboratoryDavid Keith Director, Energy and Environmental Systems Group Institute for Sustainable Energy, Environment and Economy, University of CalgaryKlaus Lackner Ewing-Worzel Professor of Geophysics, Department of Earth and Environmental Engineering, Columbia UniversityJerry Meldon Associate Professor of Chemical and Biological Engineering, Tufts UniversityRoger Pielke, Jr. Professor, Environmental Studies Program, and Fellow of the Cooperative Institute for Research in Environmental Sciences, University of Colorado

OTHERPARTICIPANTSKeven Brough CCS Program Director, ClimateWorks FoundationMike Fowler Technical Coordinator Coal Transition Project, Clean Air Task ForceNate Gorence Policy Analyst, National Commission on Energy PolicyMelanie Kenderdine Associate Director for Strategic Planning, MIT Energy InitiativeMichael Schnitzer Director, The NorthBridge GroupKurt Waltzer Carbon Storage Development Coordinator, Coal Transition Project, Clean Air Task ForceJim Wolf Doris Duke Charitable Foundation

Appendix 1List of Workshop ParticipantsThe CSPO/CATF energy-climate innovation project hosted a series of three workshops in the spring of 2009, each organized around a single technology (solar photovoltaics; post-combustion capture of carbon dioxide from power plants; and direct removal of CO2 from the atmosphere). The workshops asked the question, first to a panel of technical experts, “what would it take to accelerate advances in development and diffusion of this technology, to make it operational?” Other participants included experts on innovation and government policy.

While this report draws from these workshops, it does not represent the views, individual or collective, of workshop participants. Nor does it represent the views of the Bipartisan Policy Center or the National Commission on Energy Policy.

Below is the list of individuals who attended the workshops. We are grateful for their time, their insights, and their commitment to realistic options for decarbonizing the energy system. Those listed under “Innovation and Policy Group” were asked to attend the set of workshops. Each workshop also had a separate group of technical experts, listed by their expertise.

42 APPENDIX 2

Appendix 2Planning Team | Energy Innovation From the Bottom Up

Daniel Sarewitz Co-Director Consortium for Science, Policy & Outcomes Arizona State University

Armond Cohen Executive Director Clean Air Task Force

John Alic Consultant

Joe Chaisson Research and Technical Director Clean Air Task Force

Frank Laird Associate Professor of Technology and Public Policy Josef Korbel School of International Studies University of Denver

Sasha Mackler Research Director National Commission on Energy Policy

Catherine Morris Senior Associate Energy Program, Center for Science and Public Policy The Keystone Center

Claudia Nierenberg Associate Director CSPO-DC

Kellyn Goler Intern Consortium for Science, Policy and Outcomes

The Consortium for Science, Policy, and OutcomesArizona State University

1701 K Street, NW Washington, DC 20006

[email protected]

www.cspo.org

Clean Air Task Force18 Tremont Street Boston, MA 02108 617-624-0234 [email protected] www.catf.us

innovation policy for climate change - c's...

Documents