IMF-结构改革促进绿色增长（英）VIP专享VIP免费

下载本文档

阅读 216
格式 pdf
大小 2.17 MB
约30页
2023-06-25
收藏
点赞(0)
海报
举报

2023

JUN

Restructuring Reforms

for Green Growth

Serhan Cevik and João Tovar Jalles

WP/23/120

IMF Working Paper

European Department

Restructuring Reforms for Green Growth

Prepared by Serhan Cevik and João Tovar Jalles1

Authorized for distribution by Bernardin Akitoby

June 2023

IMF Working Papers describe research in progress by the author(s) and are published to elicit comments

and to encourage debate. The views expressed in IMF Working Papers are those of the author(s) and do

not necessarily represent the views of the IMF, its Executive Board, or IMF management.

Abstract

Policymakers across the world are striving to tackle the century-defining challenge of climate

change without undermining potential growth. This paper examines the impact of structural

reforms in the energy sector (electricity and gas) on enviromental outcomes and green growth

indicators in a panel of 25 advanced economies during the period 1970-2020. We obtain striking

results. First, while structural reforms so far failed in reducing greenhouse gas emissions per

capita, there is some evidence for greater effectiveness in lowering emissions per unit of GDP.

Second, although energy reforms are not associated with higher supply of renewable energy as a

share of total energy supply, they appear to stimulate a sustained increase in environmental

inventions and patents per capita over the medium term. We also find strong evidence of

nonlinear effects, with market-friendly energy reforms leading to better environmental outcomes

and green growth in countries with stronger environmental regulations. Looking forward, therefore,

structural reforms should be designed not just for market efficiency but also for green growth.

JEL Classification Numbers: D31; L43; L51

Keywords: Structural reforms; environment; green growth; panel data; local

projection; environmmental policy

Author’s E-Mail Address: scevik@imf.org; joaojalles@gmail.com

1 The authors would like to thank Geoffroy Dolphin, Gianluigi Ferrucci, Clara Galeazzi, Antung Liu, Christine

Richmond, Gregor Schwerhoff, and Johannes Wiegand for helpful comments and suggestions.

I. INTRODUCTION

Climate change is the defining challenge of our time, with significant risks to environmental

sustainability and socioeconomic wellbeing.2 The global mean surface temperature has already

surged more than 1.1 degrees Celsius (°C) compared with the pre-industrial average, and

projections indicate an acceleration in climate change with global temperature rising by as much

as 4°C over the next century. This will increase the risk of weather-related natural disasters and

cause greater damage to the environment, lives, and livelihoods (Stern 2007; IPCC 2007, 2014,

2019; 2021). The 2015 Paris Climate Accord, ratified by 194 countries including the European

Union (EU), seeks to contain global warming below 2°C compared to the preindustrial level

through Nationally Determined Contribution (NDC) commitments to reduce emissions.

According to the latest Emissions Gap Report, however, carbon dioxide (CO2) emissions

continued to increase since the Paris Agreement by more than 3 percent across the world, and

greenhouse gas (GHG) emissions will decline by only 7.5 percent by 2030, whereas keeping

global warming below 1.5°C requires a reduction of 55 percent (UNEP, 2021).

Economic growth tends to lead to higher emissions and environmental degradation, but it is

possible to achieve “green growth” by shifting the energy matrix away from fossil fuels and

increasing efficiency in the distribution and use of energy. These objectives, in turn, require

structural reforms and policies designed to alter behavior throughout the economy. In this paper,

we strive to close an important gap in the literature by investigating how structural reforms in

the energy sector (electricity and gas) can contribute to climate change mitigation, help guard

against threats associated with climate change, and thereby promote green growth defined as

environmentally sustainable economic growth. This is not a clear-cut question to answer since

product market reforms can have conflicting effects simultaneously on energy demand and the

supply and composition of energy sources. Furthermore, the extent of which structural reforms in

the energy sector affects environmental outcomes and the composition of economic growth

depends on the design of structural reforms and the country’s environmental policies and

institutional capacity to successfully implement structural reforms.

In this paper, we use the local projection (LP) method proposed by Jordà (2005) to investigate

how structural reforms in electricity and gas sectors—based on a narrative database of product

market reforms looking at public ownership and market access and structure—influence

alternative measures of environmental performance and green growth indicators in a panel of 25

countries during the period 1970–2020. We also explore the possibility of nonlinear effects of

these electricity and gas sector reforms by taking into account the stringency of initial

environmental policies at the time of a reform. We obtain somewhat mixed, but striking results.

First, while structural reforms so far failed in bringing about a reduction in CO2 and GHG

emissions per capita, there is some evidence for greater effectiveness in lowering GHG emissions

per unit of GDP. Second, although market-oriented electricity and gas sector reforms are not

2 There is a growing literature on economic and financial effects of climate change (Nordhaus, 1991, 1992; Cline,

1992; Dell et al., 2012; Acevedo et al., 2018; Burke and Tanutama, 2019; Kahn et al., 2019; Cevik and Jalles, 2020,

2021, 2022, 2023).

/30

下载本文档

文本预览下载提示常见问题

2023JUNRestructuringReformsforGreenGrowthSerhanCevikandJoãoTovarJallesWP/23/120©2023InternationalMonetaryFundWP/23/120IMFWorkingPaperEuropeanDepartmentRestructuringReformsforGreenGrowthPreparedbySerhanCevikandJoãoTovarJalles1AuthorizedfordistributionbyBernardinAkitobyJune2023IMFWorkingPapersdescriberesearchinprogressbytheauthor(s)andarepublishedtoelicitcommentsandtoencouragedebate.TheviewsexpressedinIMFWorkingPapersarethoseoftheauthor(s)anddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement.AbstractPolicymakersacrosstheworldarestrivingtotacklethecentury-definingchallengeofclimatechangewithoutunderminingpotentialgrowth.Thispaperexaminestheimpactofstructuralreformsintheenergysector(electricityandgas)onenviromentaloutcomesandgreengrowthindicatorsinapanelof25advancedeconomiesduringtheperiod1970-2020.Weobtainstrikingresults.First,whilestructuralreformssofarfailedinreducinggreenhousegasemissionspercapita,thereissomeevidenceforgreatereffectivenessinloweringemissionsperunitofGDP.Second,althoughenergyreformsarenotassociatedwithhighersupplyofrenewableenergyasashareoftotalenergysupply,theyappeartostimulateasustainedincreaseinenvironmentalinventionsandpatentspercapitaoverthemediumterm.Wealsofindstrongevidenceofnonlineareffects,withmarket-friendlyenergyreformsleadingtobetterenvironmentaloutcomesandgreengrowthincountrieswithstrongerenvironmentalregulations.Lookingforward,therefore,structuralreformsshouldbedesignednotjustformarketefficiencybutalsoforgreengrowth.JELClassificationNumbers:D31;L43;L51Keywords:Structuralreforms;environment;greengrowth;paneldata;localprojection;environmmentalpolicyAuthor’sE-MailAddress:scevik@imf.org;joaojalles@gmail.com1TheauthorswouldliketothankGeoffroyDolphin,GianluigiFerrucci,ClaraGaleazzi,AntungLiu,ChristineRichmond,GregorSchwerhoff,andJohannesWiegandforhelpfulcommentsandsuggestions.I.INTRODUCTIONClimatechangeisthedefiningchallengeofourtime,withsignificantriskstoenvironmentalsustainabilityandsocioeconomicwellbeing.2Theglobalmeansurfacetemperaturehasalreadysurgedmorethan1.1degreesCelsius(°C)comparedwiththepre-industrialaverage,andprojectionsindicateanaccelerationinclimatechangewithglobaltemperaturerisingbyasmuchas4°Coverthenextcentury.Thiswillincreasetheriskofweather-relatednaturaldisastersandcausegreaterdamagetotheenvironment,lives,andlivelihoods(Stern2007;IPCC2007,2014,2019;2021).The2015ParisClimateAccord,ratifiedby194countriesincludingtheEuropeanUnion(EU),seekstocontainglobalwarmingbelow2°CcomparedtothepreindustriallevelthroughNationallyDeterminedContribution(NDC)commitmentstoreduceemissions.AccordingtothelatestEmissionsGapReport,however,carbondioxide(CO2)emissionscontinuedtoincreasesincetheParisAgreementbymorethan3percentacrosstheworld,andgreenhousegas(GHG)emissionswilldeclinebyonly7.5percentby2030,whereaskeepingglobalwarmingbelow1.5°Crequiresareductionof55percent(UNEP,2021).Economicgrowthtendstoleadtohigheremissionsandenvironmentaldegradation,butitispossibletoachieve“greengrowth”byshiftingtheenergymatrixawayfromfossilfuelsandincreasingefficiencyinthedistributionanduseofenergy.Theseobjectives,inturn,requirestructuralreformsandpoliciesdesignedtoalterbehaviorthroughouttheeconomy.Inthispaper,westrivetocloseanimportantgapintheliteraturebyinvestigatinghowstructuralreformsintheenergysector(electricityandgas)cancontributetoclimatechangemitigation,helpguardagainstthreatsassociatedwithclimatechange,andtherebypromotegreengrowthdefinedasenvironmentallysustainableeconomicgrowth.Thisisnotaclear-cutquestiontoanswersinceproductmarketreformscanhaveconflictingeffectssimultaneouslyonenergydemandandthesupplyandcompositionofenergysources.Furthermore,theextentofwhichstructuralreformsintheenergysectoraffectsenvironmentaloutcomesandthecompositionofeconomicgrowthdependsonthedesignofstructuralreformsandthecountry’senvironmentalpoliciesandinstitutionalcapacitytosuccessfullyimplementstructuralreforms.Inthispaper,weusethelocalprojection(LP)methodproposedbyJordà(2005)toinvestigatehowstructuralreformsinelectricityandgassectors—basedonanarrativedatabaseofproductmarketreformslookingatpublicownershipandmarketaccessandstructure—influencealternativemeasuresofenvironmentalperformanceandgreengrowthindicatorsinapanelof25countriesduringtheperiod1970–2020.Wealsoexplorethepossibilityofnonlineareffectsoftheseelectricityandgassectorreformsbytakingintoaccountthestringencyofinitialenvironmentalpoliciesatthetimeofareform.Weobtainsomewhatmixed,butstrikingresults.First,whilestructuralreformssofarfailedinbringingaboutareductioninCO2andGHGemissionspercapita,thereissomeevidenceforgreatereffectivenessinloweringGHGemissionsperunitofGDP.Second,althoughmarket-orientedelectricityandgassectorreformsarenot2Thereisagrowingliteratureoneconomicandfinancialeffectsofclimatechange(Nordhaus,1991,1992;Cline,1992;Delletal.,2012;Acevedoetal.,2018;BurkeandTanutama,2019;Kahnetal.,2019;CevikandJalles,2020,2021,2022,2023).4associatedwithhighersupplyofrenewableenergyasashareoftotalenergysupply,theyappeartostimulateasustainedincreaseinthenumberofenvironmentalinventionsandpatentspercapitaoverthemediumterm.Furthermore,wefindstrongevidenceofnonlineareffects,withmarket-friendlyelectricityandgasreformsleadingtobetterenvironmentaloutcomesandgreengrowthincountrieswithstrongerenvironmentalregulations.Theseresultshaveseveralimportantimplicationsforthedesignofstructuralreformsandpolicies,whichshouldaimnotjustformarketefficiencybutalsoforgreengrowth.First,decouplingeconomicgrowthfromGHGemissionsispossiblethroughcomprehensivereformsandpoliciesaimedatshiftingtheenergymatrixawayfromfossilfuels.3Second,whiletransitioningenergysupplytolow-carbonsourcesiscritical,achievingenvironmentallysustainablegrowthisalsodependentongreaterefficiencyinthedistributionanduseofenergy.Theremainderofthispaperisorganizedasfollows.SectionIIdescribesthedatausedintheempiricalanalysis.SectionIIIintroducesthesalientfeaturesofoureconometricstrategy.SectionIVpresentsanddiscussestheempiricalresults,includingaseriesofrobustnesschecks.Finally,SectionVoffersconcludingremarkswithpolicyimplications.II.DATAOVERVIEWWeconstructapaneldatasetofannualobservationscovering25countriesovertheperiod1970–2020,drawnfromtheOrganizationforEconomicCo-operationandDevelopment(OECD).Thedependentvariablesarealternativeindicatorsofenvironmentalperformanceandgreengrowth.Thefirstsetlooksatemissionsandenergyintensity,whilethesecondsetfocusesonmeasuresofgreengrowth.4Forenvironmentaloutcomes,weconsiderthreeindicators:(i)CO2emissionsinmetrictonspercapita,(ii)GHGemissionsinmetrictonspercapita,and(iii)GHGemissionsperunitofGDP.5Forgreengrowth,weconsiderthreeindicatorstomeasureenvironmentallysustainableeconomicgrowth:(i)theshareofrenewableenergysupply6,(ii)thenumberofenvironment-relatedinventionspercapita,and(iii)thenumberofpatentsforenvironment-relatedtechnologiespercapita.Themainexplanatoryvariablesofinterestarestructuralreformsintheenergysectorbasedonanarrativedatabaseofmajorpolicychangesinproductmarketregulation.Twosectorsareconsideredoutofsevencovered:electricityandgas,whichrepresenttheenergysector.The3SincetheCOP23in2017,theobjectivehasbeen“tomaintaintheglobalmomentumtodecoupleoutputfromgreenhousegasemissions”(Gough,2017).However,theextenttowhichdecouplingistakingplaceremainsamatterofdispute.Cohenetal.(2018;2022)analyzetherelationshipbetweenrealGDPgrowthandCO2emissionsacross178countriesfrom1960to2018andfindsomeevidenceofdecouplinginrecentyears.IMF(2021)andBlacketal.(2022)providedetailedassessments.4Therearealternativemeasuresof“greengrowth”intheliterature.ThemostcomprehensiveframeworkisdevelopedbytheOECDandcoversasetof12indicatorsincludingenergyuseperunitofGDPandGHGemissionsperunitofGDP(OECD,2017).5ThismeasureofGHGemissionsexcludeslanduse,land-usechangeandforestry.6Notethatbefore2010theshareofrenewableswasverysmall.5originaldatabaseofmajorreformsinproductmarketregulationisputtogetherbyDuvaletal.(2018)andupdatedbyWieseetal.(2023)until2020.Thisdatasetwasbuiltintwosteps.First,foreachofcountryandaforementionedpolicyarea,Duvaletal.(2018)andWieseetal.(2023)recordalllegislativeandregulatoryactionsmentionedinallpastOECDEconomicSurveys—theregularcountrysurveyspublishedbytheOECD—publishedovertheperiod1970-2020,aswellasadditionalcountry-specificsources.7Second,amongallthoseactions,theauthorsidentifymajormeasures(liberalizing/deregulatingandtightening/regulatingtypeofreforms)asthosethatmetatleastoneofthreealternativecriteria:(i)anarrativecriterionbasedonOECDstaff’sjudgementonthesignificanceofthereformatthetimeofadoption8;(ii)whetherthereformwasmentionedagaininsubsequentEconomicSurveys,asopposedtoonlyoncewhenthemeasuredisadopted9;(iii)themagnitudeofthechangeinthecorrespondingOECDindicator,whenavailable.10Whenonlythethirdconditionismet,anextensivesearchthroughotheravailabledomesticandnationalsources,includingthroughtheinternet,isperformedtoidentifythepolicyactionunderpinningthechangeintheindicator.Theapproachconsidersnotonlyreformsbutalso“counter-reforms”—i.e.,policychangesintheoppositedirection(increaseinregulationordecreaseinflexibility).Foreachcountry,ourreformvariableineachareatakesvalue0innon-reformyears,1inreformyears,and-1incounter-reformyears.InAppendixTableA1,wepresentaselectedsetofexamplesofidentifiedreformsintheareasofelectricityandgas.AppendixFigureA1showsthetemporaldynamicsofcountry-specificreformsinelectricityandgas.ItshouldbeacknowledgedthatthecriteriaDuvaletal.(2018)andWieseetal.(2023)appliedtoidentifymajorreforms,astransparentastheyare,arenottheonlypossibleoption—thereisnosingle,objectivewaytodistinguishbetweenmajorandminorreforms.Furthermore,theauthorsdonotdistinguishamongdifferentmajorreforms—allofthemaretreatedequally,eventhoughsomehavelikelybeenmoreimportantthanothersinpractice.Finally,bydesign,thedatasetdoesnotattempttomeasureandcomparepolicysettingsacrosscountries,andassuchisnosubstituteforotherpubliclyavailableindicatorsproducedbyotherinstitutions.7ThelistofcountriesinoursampleincludesAustralia,Austria,Belgium,Canada,CzechRepublic,Denmark,Finland,France,Germany,Greece,Iceland,Ireland,Italy,Japan,Korea,Luxembourg,theNetherlands,NewZealand,Norway,Portugal,theSlovakRepublic,Spain,Sweden,Switzerland,theUnitedKingdom,andtheUnitedStates.8TheOECDEconomicSurveyusesstrongnormativelanguagetodefinetheactionatthetimeistaken,suggestiveofanimportantmeasure(forexample,“majorreform”).Inthisrespect,themethodologyisrelatedtothe“narrativeapproach”usedbyRomerandRomer(1989,2004,2010,and2017)andDevriesetal.(2011)toidentifymonetaryandfiscalshocksandperiodsofhighfinancialdistress.9ThepolicyactionismentionedrepeatedlyacrossdifferenteditionsoftheOECDEconomicSurveyforthecountryconsidered,and/orintheretrospectivesummariesofkeypastreformsthatarefeaturedinsomeeditions,whichisalsoindicativeofamajoraction.10Whenavailable,theexistingOECDindicatoroftheregulatorystanceintheareaconsidereddisplaysaverylargechange(inthe5thpercentileofthedistributionofthecumulativechangeintheindicatoroverthreeyears—toaccommodatepossiblygradualphasing-inofotherwisemajorreforms).TheOECDindicatorsusedforthepurposeofthispaper,aretheindicatorsofproductmarketregulationinthegasandelectricitysectors.6Ourempiricalobjectiveinthispaperistoidentifyandtraceouttheenvironmentalperformanceaftermajorproductmarketreformsintheenergysector,namelyelectricityandgas.ThisdatasethasseveralstrengthscomparedtoindirectmethodsusedinotherpapersthatrelyexclusivelyonchangesinOECDpolicyindicators.Thestructuralreformdatabaseusedinthispaper(i)identifiestheprecisenatureandexacttimingofmajorlegislativeandregulatoryactionsinkeyproductmarketpolicyareas;(ii)detectstheprecisereformsthatunderpinwhatotherwiselookslikeagradualdeclineinOECDpolicyindicatorswithoutanyobviousornoticeablebreak;(iii)capturesreformsinareasforwhichOECDindicatorsexistbutdonotcoverallrelevantpolicydimensions;and(iv)documentsanddescribesthepreciselegislativeandregulatoryactionsthatunderpinobservedlargechangesinOECDindicatorsoveralongperiodoftime.Finally,comparedwithalternativedatasourcesdocumentingpolicychangesinenergymarkets,theapproachtakenbyDuvaletal.(2018)allowsidentifyingaratherlimitedsetofmajorlegislativeandregulatoryreforms,asopposedtojustalonglistofactionsthatinsomecaseswouldbeexpectedtohavelittleornobearingonmacroeconomicoutcomes.Thisisparticularlyusefulforempiricalanalysesthatseekstoidentify,andthenestimate,thedynamiceffectsofreformshocks.Table1presentsstylizedfactsonstructuralreforms(takingthevalue1)intheenergysector—thatis,decreasesinregulationorincreaseinmarketflexibility—andcounter-reforms—thatis,increasesinregulationordecreaseinmarketflexibility.Thelatterarerelativelyrareeventsinproductmarkets(whiletheycanaccountforupto25percentoftotalshocksinthelabormarket).Figure1andFigure2providethenumberanddistributionofreformsidentifiedinthesample,respectively,andillustratetheheterogeneityofreformeffortsacrossproductmarketregulatoryareasandcountries.Thesehavebeenmorefrequentlyimplementedintelecommunicationsandairtransport.Thevastmajorityofproductmarketreformsinoursamplewereimplementedduringthe1990sandthe2000s.11Intermsofgeographicaldistribution,EUcountriestookmoreactionsthannon-EUcountriesonaverage,reflectingthegreaterscopeforactionintheformergroup.Table1.StructuralReformCategories,1970-2020ReformtypeNumberofreformsNumberofcounter-reformsReforms(%oftotal)Counter-reforms(%oftotal)Productmarketreforms235498.31.7OfwhichElectricitysector480100.00.0Gassector490100.00.0Source:Duvaletal.(2018);Wieseetal.(2023);author’scalculations.11Exceptionsarereformsintheareaofrailtransportundertakeninthe1980s,whicharebeyondthefocusofthispaper.Notealsothatitwouldbeverysurprisingifsuchproductmarketreformsatthatperiod(particularlyinthe1990s)wouldreduceemissionsAtthetime,renewableenergywasinitsinfancyandreformswerelikelytargetedatincreasingenergysupplyandreducingprices.7Figure1.NumberofStructuralReformsbyArea,1970-2020Source:Duvaletal.(2018);Wieseetal.(2023).Figure2.NumberofElectricityandGasReformsbycountry,1970-2020Source:Duvaletal.(2018);Wieseetal.(2023)Wearealsointerestedinwhetheracountry’senvironmentalpoliciesatthetimeofintroducingproductmarketreformsintheenergysectoraffectstheimpactofenvironmentaloutcomes.Tothisend,weusethecountry-specificEnvironmentalPolicyStringency(EPS)IndexcreatedbytheOECDanddefinedasthedegreetowhichenvironmentalpoliciesputanexplicitorimplicitpriceonpollutingorenvironmentallyharmfulbehavior(BottaandKozluk,2014).Thesedataarethemostcomprehensiveavailablesourceforpolicymeasurescovering28OECDcountriesand6012345678AustraliaAustriaBelgiumCanadaCzechRepublicDenmarkFinlandFranceGermanyGreeceIcelandIrelandItalyJapanKoreaLuxembourgNetherlandsNewZealandNorwayPortugalSlovakRepublicSpainSwedenSwitzerlandUnitedKingdomUnitedStatesnumberofreformselectricityreformsgasreforms8emergingmarketeconomiesovertheperiod1990–2020.12TheEPSindexallowsustoinvestigatetheimpactofdifferentpolicyinstruments—scaledfrom0(notstringentatall)to6(verystringent)—relativetoanoverallaggregateindexconsistingofbothmarket-basedandnon-market-basedmeasures.Inthiscontext,market-basedmeasuresincludeinstrumentssuchasacarbontax,emissiontradingschemesandfeed-intariffs,whilenon-market-basedindicatorscapturelegislationonemissionlimitsandR&Dsubsidies,amongothers.Therearealsotechnologysupportpoliciesthatincludesthosethatsupportinnovationincleantechnologiesandtheiradoption.Figure3presentsthebreakdownoftheEPSindexbreakdownin2021.Figure3.The2021OECDEnvironmentalPolicyStringencyIndexSource:Kruseetal.(2022).III.ECONOMETRICMETHODOLOGYStructuralreforms—inanyarea—tendtohaveevolvingeffectsoveranextendedperiodoftime.Inthispaper,weestimatetheimpulseresponsefunctions(IRFs)ofenvironmentaloutcomesandmeasuresofgreengrowthtostructuralreformsinelectricityandgassectorsbyapplyingLPmethod.ThisapproachhasbeenadvocatedbyAuerbachandGorodnichenko(2012,2013)andRomerandRomer(2019)asaflexiblealternativetovectorautoregressions(VAR)and/ordistributedlagmodels.13TheLPmethodisalsoflexibletoaccommodateapanelstructureanddoesnotconstraintheshapeofIRFs,therebyallowingtoanalyzedifferenttypesofpolicyshocks(AuerbachandGorodnichenko,2013;JordàandTaylor,2016;RameyandZubairy,2018;RomerandRomer,2019;Bornetal.,2020).Inthispaper,giventhepanelsetting,weadopttheLP12OneconcernmightbethatenvironmentlegislationisadoptedatthesupranationallevelsuchastheEuropeanUnion.Thiswouldbeproblematicforourempiricalanalysisasnationalgovernmentsmaythennotdirectlyresponsibleforthestricterenvironmentalregulation.Despitethispotentialconcern,itisnoteworthythatsubstantialcross-countryvariationexistswithintheEUandenvironmentalpolicymakingtakesplaceatthenationallevel.13Plagborg-MollerandWolf(2021)furtherdiscussthepropertiesoflocalprojections,aswellastherelationshipbetweentheseandVARestimationofimpulseresponses.9methodovercommonlyusedVARmodelsforthefollowingspecificreasons.First,ourestimationentailsalargepaneldatasetwithaconstellationoffixedeffects,whichmakesadirectapplicationofstandardVARmodelsmoredifficult.Second,theLPmethodobviatestheneedtoestimatetheequationsfordependentvariablesotherthanthevariableofinterest,therebysignificantlyeconomizingonthenumberofestimatedparameters.Third,theLPmethodisparticularlysuitedtoestimatingnonlinearities(forexample,howtheeffectofenergyreformshocksdiffersincountrieswithhighorlowEPS),asitsapplicationismuchmorestraightforwardcomparedtonon-linearstructuralVARmodels,suchasMarkov-switchingorthreshold-VARmodels.14Moreover,itallowsforincorporatingvarioustime-varyingfeaturesofsource(recipient)economiesdirectlyandallowfortheirendogenousresponsetoenergyreformshocks.Lastly,theerrorterminthefollowingpanelestimationsislikelytobecorrelatedacrosscountries.ThiscorrelationwouldbedifficulttoaddressinthecontextofVARmodels,butitiseasytohandleintheLPmethodbyeitherclusteringstandarderrorsorusingtheDriscoll-Kraay(1998)standarderrors,whichallowsforarbitrarycorrelationsoftheerrorsacrosscountriesandtime.Accordingly,toaccountforthecumulativeresponsesofelectricityandgassectorreformsoverafive-yearhorizon,weusethefollowingbaselinespecification:𝑦,−𝑦,=𝛼+𝜏+β𝑆𝑅,+𝜃𝑋,+ε,(1)inwhich𝑦denotesaproxyofenvironmentalperformancemeasuredas:(i)CO2emissionsinmetrictonspercapita,(ii)GHGemissionsinmetrictonspercapita,and(iii)GHGemissionsperunitofGDP;andgreengrowthmeasuredas:(i)theshareofrenewableenergysupply,(ii)thenumberofenvironment-relatedinventionspercapita,and(iii)thenumberofpatentsforenvironment-relatedtechnologiespercapita.;thecoefficients𝛼and𝜏arecountryandtimefixedeffects,respectively,accountingforcross-countryheterogeneityandglobalshocks;𝛽denotesthecumulativeresponseofenvironmentaloutcomesineachkyearaftertheimplementationofaproductmarketreform;𝑆𝑅,denotesstructuralreformsinelectricityandgassectorsasdescribedintheprevioussection.Weincludetreatmentlagsinourmodels.Itisanempiricalissuehowlongtheeffectofprogressivityshockspersistsinthedata.𝑋,isavectorofadditionalcontrolvariables.WeuseAkaike’sinformationcriteriontodeterminethelaglength:weemploy2lagsofthestructuralreformshock,2lagsofrealGDPgrowth,inflationandthedependentvariable.Forrobustness,weintroduceadditionalcontrolsfortwolagsofotherdeterminantsofenvironmentalperformanceinsomespecificationsofthemodel–seebelow.WeestimatetheequationusingtheOrdinaryLeastSquares(OLS)method.15WecalculateSpatial14SeeChoietal.(2018)andMiyamotoetal.(2019)fortherecentapplicationoflocalprojectionstotheestimationofnonlinearitiesandinteractioneffectsofshocksusingalargepaneldataset,asitisthecasewithoursample.15Anotheradvantageofthelocalprojectionmethodcomparedtovectorautoregression(orautoregressivedistributedlag)specificationsisthatthecomputationofconfidencebandsdoesnotrequireMonteCarlosimulationsorasymptoticapproximations.Onelimitation,however,isthatconfidencebandsatlongerhorizonstendtobewiderthanthoseestimatedbyVARs.10CorrelationConsistent(SCC)standarderrorsasproposedbyDriscollandKraay(1998).16βdenotesthe(cumulative)responseofthevariableofinteresthyearsaftertheenergyreformshock.Impulseresponsefunctions(IRFs)arethenobtainedbyplottingtheestimatedβfork=0,1,…,5with90(and68)percentconfidencebandscomputedusingthestandarddeviationsassociatedwiththeestimatedcoefficientsβ.Todevelopamoregranularanalysis,wealsoexplorewhetherinitialenvironmentalpolicies,asmeasuredbytheEPSindexatthetimeofthereform,influencetheimpactofstructuralreformsonenvironmentaloutcomesandgreengrowth.17AsdiscussedinAuerbachandGorodnichenko(2012,2013),theLPapproachtoestimatingnon-lineareffectsisequivalenttothesmoothtransitionautoregressive(STAR)modeldevelopedbyGrangerandTeräsvirta(1993).Theadvantageofthisapproachistwofold.First,comparedwithamodelinwhicheachdependentvariablewouldbeinteractedwithameasureoftheEPSindexconvertedintoadummyvariableforhighandlowvaluesaccordingtosomeadhoccriterion,itpermitsadirecttestofwhethertheeffectoftheenergyreformshockvariesacrossadifferentregimes.Second,comparedwithestimatingstructuralVARforeachregime,itallowstheeffectofenergyreformstochangesmoothlybetweenlowandhighlevelsofEPSbyconsideringacontinuumofstatestocomputetheimpulseresponsefunctions,thusmakingtheresponsemorestableandprecise.Accordingly,theaugmentedLPmodeltotestfornon-lineareffectstakesthefollowingform:𝑦,−𝑦,=𝛼+𝜏+𝛽𝐹(𝑧,)𝑆𝑅,+𝛽(1−𝐹(𝑧,))𝑆𝑅,+θ𝑋,+𝜀,(2)with𝐹(𝑧)=()(),𝛾>0inwhich𝑧istheEPSindexthatisnormalizedtohavezeromeanandunitvariance.Theweightsassignedtoeachregimevarybetween0and1accordingtotheweightingfunction𝐹(.),sothat𝐹(𝑧)canbeinterpretedastheprobabilityofbeinginagivenstateoftheeconomy.Thecoefficients𝛽and𝛽capturetheimpactofstructuralreformshocksonenvironmentalperformanceandgreengrowthateachhorizonkincasesoflowEPS(𝐹(𝑧)≈1whenzgoestominusinfinity)andhighEPS(1−𝐹(𝑧)≈1whenzgoestoplusinfinity),respectively.Wechoose𝛾=1.5.18IV.EMPIRICALRESULTSThemainvariableofinterestinthisanalysisisthecumulativechangeinenvironmentaloutcomesandmeasuresofgreengrowthinresponsetostructuralreformsintheenergysectorasdescribedintheprevioussection.InFigure3,wepresenttheresultsofourbaselinespecification16Thisisanonparametrictechniqueassumingtheerrorstructuretobeheteroskedastic,autocorrelateduptosomelag,andpossiblycorrelatedacrosscountries.17ThereareotherstudiesusingsuchaSTARfunctioninthecontextofLP,suchAbiadetal.(2016),FurceriandLi(2017),GuptaandJalles(2022),JallesandKarras(2022).18Ourresultshardlychangewhenusingalternativevaluesoftheparameter𝛾,between1and4.11includingcontrolvariables.Eachchartshowsthecumulativeeffectsinresponsetoanenergy-sectorreformoneachofoursixalternativedependentvariablesinoursampleof25countriesoverafive-yearhorizon,where0indicatestheyearinwhichthestructuralreformoccurs.Theshadedareasindicatethe90percentand68percentconfidencebandsbasedonDriscoll-Kraay(1998)robuststandarderrorsclusteredatthecountrylevel.First,wefocusonmeasuresofenvironmentalperformanceandfindthatstructuralreformsinelectricityandgasmarketsleadtohigherCO2andGHGemissionspercapita,buttheseunconditionalresultsarestatisticallyinsignificantandsurroundedbygreatuncertainty.GHGemissionsperunitofGDP,ontheotherhand,respondstostructuralreformsintheoppositedirection,decliningbelowtheinitialleveloverthefive-yearperiodandshowingsomesignsofdecouplingbetweeneconomicgrowthandemissions.Second,wefocusonindicatorsofgreengrowth—definedasenvironmentallysustainableeconomicgrowth—andfindthatstructuralreformslowerthesupplyofrenewableenergyasashareoftotalenergysupply,butthisnegativeeffectisnotpersistentoverthelongrun.Furthermore,structuralreformsintheenergysectorstimulatesasustainedincreaseinthenumberofenvironmentalinventionsandpatentspercapitabeyondtheinitiallevel.Althoughthisresultseemscontradictorytopreviousstudiesthatfindanegativerelationshipbetweenproductmarketreforms(i.e.,liberalization)intheenergysectorandR&Dspending(Sirin,2011;JamasbandPollitt,2011),wethinkthatouranalysisbasedonthelatestdataandalargersetofcountriescapturesthesurgeofrenewableenergytechnologiesoverthepastdecade.Wedevelopamoregranularanalysisbyfocusingontypesofstructuralreformsinelectricityandgassectorsandobtainsimilarresponsepatternsfor“marketaccessandstructure”reforms(Figure4)and“publicownership”reforms(Figure5).Inthecaseofenergy-sectorprivatization,itshouldbenotedthatreformsleadtohigheremissionsacrossallmeasures,includingGHGemissionsperunitofGDP.InAppendixFiguresA2-A3,wepresenttheIRFsforstructuralreformsinelectricityandgasmarketsseparately,whichareconsistentwiththebaselineresults.Wearealsointerestedinwhetherthestrengthofinitialenvironmentalpoliciesatthetimeofanenergyreforminfluencesitsimpactonenvironmentalperformanceandgreengrowthbyestimatingastate-dependentversionofthemodelthatallowsdynamicresponsestovarywiththeEPSindex.Theseresults,presentedinFigure6,showstrikingdifferencesintheimpactofstructuralreformsonmeasuresofemissionsandgreengrowth.First,incountrieswithstrongerenvironmentalregulationsandpro-climatepolicies,structuralreformsintheenergysectordeliversasignificantandpersistentdeclineinCO2andGHGemissionspercapitaandGHGemissionsperunitofGDPinthefirstyearandoverthelongrun,whereasemissionscontinuetoincreaseincountrieswithlowenvironmentalstandards.Second,electricityandgassectorreformsmakeagreatercontributiontoincreasingtheshareofrenewableenergyandthedevelopmentofenvironmentaltechnologiesincountrieswithhigherEPSindex,whiletheimpactisoppositeinlow-EPScountries.ThisisinlinewiththeworkbyEugster(2021)whofindsthattheestimatedeffectofclimatechangemitigatingpoliciesoninnovationincleanenergytechnologiesispositiveonnet,meaningthatincreasedinnovationincleanandgreytechnologiesisnotoffset12byadecreaseininnovationindirtytechnologies.Weobtainsimilarresultswhenweestimatethestate-dependentmodelfor“marketaccessandstructure”reforms(Figure7)and“publicownership”reforms(Figure8).Inparticular,weobservethatenergy-sectorprivatizationresultsinsignificantlybetterenvironmentaloutcomesandgreengrowthincountrieswithstrongerenvironmentalregulations.Figure3.ImpactofEnergySectorReforms:BaselineModelNote:x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.-1012012345yearGHGpc(%)-1012012345yearCO2pc(%)-1.5-1-.50.51012345yearemissionperunitGDP(%)Impactonfromelectricityandgasreforms-.6-.4-.20.2.4012345yearrenew.En.sup.(%toten.-10-50510012345yearEnv.Inventionspc(%)-5051015012345yearPatentsEnv-rel.tech.(%)Impactonfromelectricityandgasreforms13Figure4.ImpactofEnergyMarketAccessandStructureReforms:BaselineModelNote:x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.-1012012345yearGHGpc(%)-1012012345yearCO2pc(%)-1.5-1-.50.51012345yearemissionperunitGDP(%)Impactonfrommarketaccess&structureenergyreforms-.8-.6-.4-.20.2012345yearrenew.En.sup.(%toten.-15-10-50510012345yearEnv.Inventionspc(%)-5051015012345yearPatentsEnv-rel.tech.(%)Impactonfrommarketaccess&structureenergyreforms14Figure5.ImpactofEnergyPrivatization:BaselineModelNote:x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.Weconductseveralrobustnessexercises.Forreasonsofparsimony,wefocussolelyontwodependentvariables,whichwefindtobethemostrepresentativeofgreengrowth—GHGemissionspercapitaandthenumberofpatentsforenvironment-relatedtechnologiespercapita.0246012345yearGHGpc(%)02468012345yearCO2pc(%)-20246012345yearemissionperunitGDP(%)Impactonfrompublicownershipenergyreforms-1-.50.5012345yearrenew.En.sup.(%toten.-20-1001020012345yearEnv.Inventionspc(%)-100102030012345yearPatentsEnv-rel.tech.(%)Impactonfrompublicownershipenergyreforms15First,weknowthatapossiblebiasfromestimatingequation(1)usingcountry-fixedeffectsisthattheerrortermmayhaveanon-zeroexpectedvalue,duetotheinteractionoffixedeffectsandcountry-specificdevelopments(TeulingsandZubanov,2014).Thiswouldleadtoabiasoftheestimatesthatisafunctionofk.Toaddressthisissue,equation(1)wasre-estimatedbyexcludingcountryfixedeffectsfromtheanalysis.Theseresults,showninAppendixFigureA4,suggestthatthisbiasisnegligible.Second,toestimatethecausalimpactofenergyreformshocksonenvironmentalandgreengrowthoutcomes,itisimportanttocontrolforprevioustrendsinreformdynamics.Inthebaselinespecification,weattempttodothisbycontrollingforuptotwolagsinthedependentvariable.19Tofurthermitigatethisconcern,were-estimateequation(1)byincludingcountry-specifictimetrendsasadditionalcontrolvariables.Theseresults,presentedinAppendixFigureA4,remainqualitativelyunchanged.Third,sinceelectricityandgassectorreformsmaybeimplementedaspartofbroaderpackages,wealsore-estimateourmainregressioncontrollingforreformsinadditionalareas(suchasroadandrailway,unemploymentbenefitsreplacementrate,andEPLforregularcontracts),whicharedrawnfromthesamestructuralreformdataset.Theseresultsalsoremainconsistentwithourbaselinefindings.Fourth,whilethepreviousrobustnesschecksgoalongwaytowardmitigatingendogeneityconcerns,wealsoestimatethemodelbyusingadditionalcontrolvariablesandtheinstrumentalvariable(IV)approach.Theliteraturehasputforwardseveraltheoriestorationalizewhyandwhenreforms(donot)happen.Wefocusononebroadfactorexaminedintheliterature:politicalinstitutions.20Specifically,weusethefollowingsetofpoliticaleconomyvariablesasexternalinstruments,whichwedivideinfourcategories:(i)ideologyofthegoverningparty/ies,usingadiscretevariabletodistinguishbetweenleft,centerandright(3,2and1,respectively)(Parties);(ii)politicalsystem,usingadiscretevariableforparliamentary,assembly-electedandpresidentialformsofgovernments(2,1and0,respectively)(System);(iii)partyfragmentation,usingacontinuousvariableboundedbetween0(nofragmentation)and1(maximumfragmentation)tocapturethenumberofpoliticalpartiesinthelowerhouseofthelegislativeassembly(Fragmentation);(iv)thestrengthofdemocraticinstitutionsasmeasuredbythePolityIVindex,whichisnormalizedbetween0and1(Democ).WeobtainthesefromtheWorldBankDatabaseofPoliticalInstitutionsdatabase.Bymeansofatwo-stageleastsquaresestimator,were-estimateequation(1)usinguptotwolagsofthefourpoliticaleconomyexogenousinstrumentsdescribedabove.21Theseresults,19Similarresultsareobtainedwhenusingalternativelagparametrizations.Resultsforzero,oneandthreelags(notshown)confirmthatpreviousfindingsarenotsensitivetothechoiceofthenumberoflags.20Duval,FurceriandMiethe(2018)providesarecentcontributioninthisarea.21Tocheckthevalidityofourinstrumentsandassessthestrengthofouridentification,werelyontheKleibergen-PaapandHansenstatistics.Theunderidentificationtestteststhattheexcludedinstrumentsare"relevant"(meaningcorrelatedwiththeendogenousregressors).OurobtainedstatisticsgenerallyrejectthenullhypothesisthatthedifferentequationsareunidentifiedaccordingtotheStock-Yogocriticalvalues.Then,theHansenteststatisticsrevealthattheinstrumentsetscontainvalidinstruments(i.e.,uncorrelatedwiththeerrorterm,andthattheexcludedinstrumentsarecorrectlyexcludedfromtheestimatedequation)isnotrejected.16reportedinAppendixFigureA4,arebroadlysimilartoourbaselinefindings,confirmingthatendogeneityisnotaseriousconcerninourcase.Figure6.ImpactofEnergySectorReforms:State-DependentModelNote:estimationofequation2usingEPSaszinF(z).x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.Thebluedottedlinedenotestheunconditionalbaselineresultfromestimatingequation(1).18Figure7.ImpactofMarketAccessandStructureReforms:State-DependentModelNote:estimationofequation2usingEPSaszinF(z).x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.Thebluedottedlinedenotestheunconditionalbaselineresultfromestimatingequation(1).19Figure8.ImpactofPublicOwnershipReforms:State-DependentModelNote:estimationofequation2usingEPSaszinF(z).x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.Thebluedottedlinedenotestheunconditionalbaselineresultfromestimatingequation(1).V.CONCLUSIONAddressingclimatechange—thedefiningchallengeofourtime—requiresglobaleffortstoreduceGHGemissions,whichareprojectedtoincreasebyonly7.5percentby2030comparedtotherequiredreductionof55percentjusttokeepglobalwarmingbelow1.5°C.Therefore,whattheworldneedsisanewdevelopmentmodelthatbetterbalancesincomegrowthandenvironmentalprioritiesbymodernizingtheenergymatrixawayfromfossilfuelsandincreasingefficiencyinthedistributionanduseofenergy.Inturn,theseobjectivesrequirestructuralreformsandpoliciesdesignedtoalterbehaviorthroughouttheeconomy.Tothisend,thispaperclosesanimportantgapintheliteraturebyinvestigatinghowstructuralreformsinelectricityandgassectorscancontributetoclimatechangemitigation,helpguardagainstthreatsassociatedwithclimatechange,andtherebypromotegreengrowthdefinedasenvironmentallysustainableeconomicgrowth.WeusetheLPmethodtoestimatethecumulativeimpactofstructuralreformsintheenergysector—basedonanarrativedatabaseofproductmarketreforms—onalternativemeasuresofenvironmentalperformanceandgreengrowthinapanelof25countriesduringtheperiod1970–2020.Wealsoexplorethepossibilityofnonlineareffectsofstructuralreformsbytakingintoaccountinitialenvironmentalpoliciesatthetimeofanenergyreform.Weobtainsomewhatmixed,butstrikingresults.First,whileelectricityandhassectorreformssofarfailedinbringingaboutareductioninCO2andGHGemissionspercapita,thereissomeevidenceforgreatereffectivenessinloweringGHGemissionsperunitofGDP.Second,althoughelectricityandgassectorreformsarenotassociatedwithhighersupplyofrenewableenergyasashareoftotalenergysupply,theyappeartostimulateasustainedincreaseinthenumberofenvironmentalinventionsandpatentspercapitaoverthemediumterm.Furthermore,wefindstrongevidenceofnonlineareffects,withmarket-orientedelectricityandgassectorreformsleadingtobetterenvironmentaloutcomesandgreengrowthincountrieswithstrongerenvironmentalregulations.Theseresultshaveseveralimportantimplicationsforthedesignofstructuralreformsandpolicies,whichshouldaimnotjustformarketefficiencybutalsoforgreengrowth.First,decouplingeconomicgrowthfromGHGemissionsispossiblethroughcomprehensivereformsandpoliciesaimedatshiftingtheenergymatrixawayfromfossilfuels.Second,whiletransitioningenergysupplytolow-carbonsourcesiscritical,achievingenvironmentallysustainablegrowthisalsodependentongreaterefficiencyinthedistributionanduseofenergy.21REFERENCESAcevedo,S.,M.Mrkaic,N.Novta,E.Pugacheva,andP.Topalova(2020).“TheEffectsofWeatherShocksonEconomicActivity:WhatAretheChannelsofImpact?”JournalofMacroeconomics,Vol.65,pp.103207Auerbach,A.,andY.Gorodnichenko(2012).“MeasuringtheOutputResponsestoFiscalPolicy,”AmericanEconomicJournal–EconomicPolicy,4:1-27.Auerbach,A.,andY.Gorodnichenko(2013).“OutputSpilloversfromFiscalPolicy,”AmericanEconomicReview,Vol.103,pp.141–146.Black,S.,J.Chateau,F.Jaumotte,I.WHParry,G.Schwerhoff,S.DThube,andK.Zhunussova.(2022).“GettingonTracktoNetZero:AcceleratingaGlobalJustTransitioninThisDecade.”StaffClimateNotes2022(010).Born,B.,G.J.Müller,andJ.Pfeifer(2020).“Doesausteritypayoff?”,ReviewofEconomicsandStatistics,Vol.102(2),323-338.Burke,M.,andV.Tanutama(2019).“ClimaticConstraintsonAggregateEconomicOutput,”NBERWorkingPaperNo.25779(Cambridge,MA:NationalBureauofEconomicResearch).Cevik,S.,andJ.Jalles,2020,“FeelingtheHeat:ClimateShocksandCreditRatings,”IMFWorkingPaperNo.20/286(Washington,DC:InternationalMonetaryFund).Cevik,S.,andJ.Jalles(2021).“AnApocalypseForetold:ClimateShocksandSovereignDefaults,”OpenEconomiesReview,Vol.33,pp.89–108.Cevik,S.,andJ.Jalles(2022).“ThisChangesEverything:ClimateShocksandSovereignBonds,”EnergyEconomics,Vol.107,10585.Cevik,S.,andJ.Jalles(2023).“ForWhomtheBellTolls:ClimateChangeandIncomeInequality,”EnergyPolicy,Vol.174,113475.Cline,W.(1992).TheEconomicsofGlobalWarming(NewYork:NewYorkUniversityPress).Dell,M.,B.Jones,andB.Olken(2012).“TemperatureShocksandEconomicGrowth:EvidencefromtheLastHalfCentury,”AmericanEconomicJournal:Macroeconomics,Vol.4,pp.66–95.Driscoll,J.,andA.Kraay(1998).“ConsistentCovarianceMatrixEstimationwithSpatiallyDependentData,”ReviewofEconomicsandStatistics,Vol.80,pp.549–560.Duval,R.,D.Furceri,B.Hu,J.Jalles,andH.Nguyen(2018).“ANarrativeDatabaseofMajorLaborandProductMarketReformsinAdvancedEconomies,”IMFWorkingPaperNo.18/19(Washington,DC:InternationalMonetaryFund).Eugster,J.(2021).“TheImpactofEnvironmentalPolicyonInnovationinCleanTechnologies.”IMFWorkingPapers2021(213).Im,K.,M.Pesaran,andY.Shin(2003).“TestingforUnitRootsinHeterogeneousPanels,”JournalofEconometrics,Vol.115,pp.53–74.IMF(2021).“ReachingNetZeroEmissions.”G20Note(Washington,DC:InternationalMonetaryFund).22Jamasb,T.andPollitt,M.G.(2011)."ElectricitySectorLiberalizationandInnovation:AnAnalysisoftheUK'sPatentingActivities,"ResearchPolicy,Vol.40(2),pp.309-324.Jordà,O.(2005).“EstimationandInferenceofImpulseResponsesbyLocalProjections,”AmericanEconomicReview,Vol.95,pp.161–182.Jordà,O.,andA.Taylor(2016).“EstimationandInferenceofImpulseResponsesbyLocalProjections,”EconomicJournal,Vol.126,pp.219–255.Kahn,M.,K.Mohaddes,R.Ng,M.Pesaran,M.Raissi,andJ-C.Yang(2019)“Long-TermMacroeconomicEffectsofClimateChange:ACross-CountryAnalysis,”IMFWorkingPaperNo.19/215(Washington,DC:InternationalMonetaryFund).OECD(2017).GreenGrowthIndicators2017(Paris:OrganizationforEconomicCo-operationandDevelopment).Nordhaus,W.(1991).“ToSloworNottoSlow:TheEconomicsoftheGreenhouseEffect,”EconomicJournal,Vol.101,pp.920–937.Nordhaus,W.(1992).“AnOptimalTransitionPathforControllingGreenhouseGases,”Science,Vol.258,pp.1315–1319.Ramey,V.,andS.Zubairy(2018).“GovernmentSpendingMultipliersinGoodTimesandinBad:EvidencefromUSHistoricalData,”JournalofPoliticalEconomy,Vol.126,pp.850–901.Romer,C.,andD.Romer(2019).“FiscalSpaceandtheAftermathofFinancialCrises:HowItMattersandWhy,”NBERWorkingPaperNo.25768(Cambridge,MA:NationalBureauofEconomicResearch).Sirin,S.(2011)."EnergyMarketReformsinTurkeyandTheirImpactonInnovationandR&DExpenditures,"RenewableandSustainableEnergyReviews,Vol.15,pp.4579-4585.Teulings,C.N.,andN.Zubanov(2014).“IsEconomicRecoveryaMyth?RobustEstimationofImpulseResponses.”JournalofAppliedEconometrics29(3):497−514.UNEP(2021).TheEmissionsGapReport2021:TheHeatIsOn(Nairobi:UnitedNationsEnvironmentProgram).Wiese,R.,J.Jalles,andJ.deHaan(2023).“StructuralReformsandIncomeDistribution:NewEvidenceforOECDCountries,”REMWorkingPaperNo.0256-2022(Lisbon:ResearchinEconomicandMathematics,UniversityofLisbon).23AppendixTableA1.ExamplesofIdentifiedStructuralReformsinElectricityandGasYearAreaContentNormativelanguageMentioninotherreportsLargechangeinOECDindicatorreform/counter-reformElectricityAustralia1996MarketaccessandstructureTheCouncilofAustralianGovernmentsagreedtohavethenecessarystructuralchangesinplacetocreateacompetitivemarketforbulkelectricityinsouthernandeasternAustraliafrom1July1995.(pg.126,1994)Initsreportof30June1997,theNationalCompetitionCouncilattestedonthebasisoftheStatesandTerritories1996-97annualreportsthat:goodprogresshasbeenmadetowardsimplementingtheNationalElectricityMarketineasternandsouthernAustralia,includingcommitmentsforinterconnectionbybothQueenslandandTasmania…(pg.76,1998)…thiswilloffernewscopeforgreatercompetitionintheelectricitymarketwiththecommencementofcross-bordertrading.(pg.76,1998)pg.68,1995pg.76,1998Yes1Finland2000PublicownershipPrivatizationmandatesarebroadenedbyParliament.Thegovernmentnowhasauthorizationtoreducethegovernmentownershipto50.1percentinAltiaGroupandVapo,to20percentinRautaruukki,to15percentinKemiraGroup,10percentinOutokumpuandzeroinInspecta.(pg.151,2002)Productmarketshavebeenrapidlyliberalized…andthetelecommunicationandelectricitymarketsarenowfullyliberalized.(pg.64,2002)Yes1GasBelgium1996PublicownershipAmongthetransactionsthatoccurredthatyear[1996],theinitialpublicoffer(16.60percentofcapital)oftheBelgiangastreatment,transmissionandstoragemonopolyDistrigazistobementioned.TheBelgiangovernmentlatersolditsremainingshareinthecompany,butretainsonegoldenshare...[seehttp://www.privatizationbarometer.com/atlas.php?id=6&mn=PM]Yesin19951SlovakRepublic2007MarketaccessandstructureTheSlovakRepublicimplementedwide-rangingreformstointroducecompetitioninenergymarkets...Managerialseparationhasbeenputinplaceinboththegasandelectricitytransportnetworks.Legalunbundlingofcompaniesoperatinggasandelectricitynetworkisvirtuallycomplete.(pg.105,2007)…wide-rangingreformstointroducecompetitioninenergymarkets...(pg.105,2007)Yes1Source:Duvaletal.(2018);Wieseetal.(2023).24AppendixFigureA100.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020USAelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020UKelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Austriaelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Belgiumelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Denmarkelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Franceelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Germanyelectricitygas-1.5-1-0.500.511.519701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Italyelectricitygas2500.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Luxembourgelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Netherlandselectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Norwayelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Swedenelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Switzerlandelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Canadaelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Japanelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Finlandelectricitygas2600.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Greeceelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Icelandelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Irelandelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Portugalelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Spainelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Australiaelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020NewZealandelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Koreaelectricitygas2700.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020CzechRepublicelectricitygas00.20.40.60.811.219701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020SlovakRepublicelectricitygas00.10.20.30.40.50.60.70.80.9119701972197419761978198019821984198619881990199219941996199820002002200420062008201020122014201620182020Polandelectricitygas28FigureA2.ImpactofElectricityReforms:BaselineModelNote:x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.-1012012345yearGHGpc(%)-1012012345yearCO2pc(%)-1.5-1-.50.51012345yearemissionperunitGDP(%)Impactonfromelectricityreforms-.6-.4-.20.2.4012345yearrenew.En.sup.(%toten.-20-1001020012345yearEnv.Inventionspc(%)-1001020012345yearPatentsEnv-rel.tech.(%)Impactonfromelectricityreforms29FigureA3.ImpactofElectricityReforms:BaselineModelNote:x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreform,darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.-1-.50.51012345yearGHGpc(%)-1-.50.511.5012345yearCO2pc(%)-1.5-1-.50.51012345yearemissionperunitGDP(%)Impactonfromgasreforms-1-.50.5012345yearrenew.En.sup.(%toten.-20-1001020012345yearEnv.Inventionspc(%)-10-50510012345yearPatentsEnv-rel.tech.(%)Impactonfromgasreforms30FigureA4.ImpactofEnergyReforms:RobustnessExercisesNote:x-axisinyears;t=0istheyearofthestructuralreform;t=1isthefirstyearofimpact.Solidblacklinesdenotetheresponsetoastructuralreformunderdifferentrobustnessorsensitivityexercisesasdescribed;darkgreyareadenotes90percentconfidencebandswhilelightgrayareadenotes68percentconfidencebands,basedonstandarderrorsclusteredatcountrylevel.Solidbluelinesdenotethebaselineresponse.ExcludingcountryfixedeffectsIncludingcountry-specifictimetrendsControllingforadditionalreformareasInstrumentalVariableApproach012345012345yearGHGpc(%)-5051015012345yearPatentsEnv-rel.tech.(%)-1-.50.511.5012345yearGHGpc(%)-505101520012345yearPatentsEnv-rel.tech.(%)-10123012345yearGHGpc(%)-1001020012345yearPatentsEnv-rel.tech.(%)-1-.50.511.5012345yearGHGpc(%)-5051015012345yearPatentsEnv-rel.tech.(%)

1、当您付费下载文档后，您只拥有了使用权限，并不意味着购买了版权，文档只能用于自身使用，不得用于其他商业用途（如 [转卖]进行直接盈利或[编辑后售卖]进行间接盈利）。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。
3、如文档内容存在违规，或者侵犯商业秘密、侵犯著作权等，请点击“违规举报”。

碎片内容

碳中和的最新文档