IMF-主權(quán)財富基金會降低財政政策的順周期性嗎？使用非參數(shù)方法的新證據(jù)（英）

Do

Sovereign

WealthFunds

Reduce

FiscalPolicy

Pro-cyclicality?New

Evidence

Using

aNon-Parametric

ApproachAliJ.Al-SadiqandDiegoAlejandroGutiérrezWP/23/133IMF

Working

Papers

describe

research

inprogress

by

the

author(s)

and

are

published

toelicit

comments

and

to

encourage

debate.Theviewsexpressed

inIMFWorkingPapersarethoseoftheauthor(s)anddo

notnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement.2023JUL?2023InternationalMonetaryFundWP/23/133IMF

Working

PaperWesternHemisphereDepartmentDo

Sovereign

Wealth

Funds

Reduce

Fiscal

Policy

Pro-cyclicality?

New

Evidence

Using

a

Non-Parametric

ApproachPrepared

by

Ali

J.

Al-Sadiq

and

Diego

Alejandro

Gutiérrez*AuthorizedfordistributionbyPatriziaTumbarelloJune

2023IMF

Working

Papers

describe

research

in

progress

by

the

author(s)

and

are

published

to

elicitcomments

and

to

encourage

debate.

TheviewsexpressedinIMFWorkingPapersarethoseoftheauthor(s)anddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,

orIMFmanagement.ABSTRACT:Theheightenedvolatilityof

commoditypricesinrecentyears,reflecting

theeffectsofthepandemicandthewarinUkraine,begsthelongstanding

questionoftheoptimalfiscalpolicyresponsetocommoditypriceshocks.Fiscalperformanceinmost

commodity-exportingcountriesis

typicallyshapedbyshiftsincommoditypricesandeconomicactivity,oftenresultinginprocyclicalfiscalpolicy.Onewaytominimizetheprocyclicalityof

fiscalpolicyistosetupa

stabilizationSovereign

Wealth

Fund(SWF).Whilesuchfundscanhelpsmoothgovernmentconsumptioningood

andbadtimes,theempiricalevidenceoftheirvaluesofarhasbeeninconclusive.

However,usinganunbalancedpaneldatasetfor182countriesduring1980-2019,withtwoeconometricmethodsthataddresstheselection-biasproblem,weproviderobustevidencethatstabilization

SWFsdoindeed

helpsmoothgovernmentconsumptionbyreducingfiscalpolicyvolatilityassociatedwithcommoditypricefluctuations.RECOMMENDED

CITATION:

Al-Sadiq,AliandDiegoAlejandroGutierrez,2023.“Do

SovereignWealthFundsReduceFiscal

PolicyPro-cyclicality?NewEvidenceusingaNon-ParametricApproach”,IMFWorkingpaper23/133.Washington,DC:InternationalMonetaryFund.JELClassificationNumbers:Keywords:C33;E62;E63;H10;H30;O13;Q38SovereignWealthFunds;

FiscalPolicyProcyclicality;MatchingAnalysis;

Commodity-exportingCountriesAuthors’E-MailAddress:aalsadiq@,

dgutierrez@*TheauthorswouldliketothankPatrizia

Tumbarello,AlinaCarare,Olusegun

AyodeleAkanbi,Veronique

Salins,OliverBasdevant,Jose

LuisDe

Aro,andIrene

Yackovlevfor

their

veryhelpfulcommentsand

suggestionsontheearlierversionof

thispaper.

WewouldlikealsotothankSoungbeCoquillat

forherassistanceinpreparing

theworkingpaper.Allremainingerrorsareours.WORKING

PAPERSDo

Sovereign

Wealth

FundsReduce

Fiscal

Policy

Pro-cyclicality?

New

Evidence

Using

aNon-Parametric

ApproachPreparedbyAli

J.

Al-SadiqandDiegoAlejandroGutiérrezIMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachContentsGlossary...............................................................................................................................................................31.2.3.4.INTRODUCTION...........................................................................................................................................4THE

FREQUENCYAND

IMPACT

OF

COMMODITY

PRICE

VOLATILITY...................................................................6LITERATURE

REVIEW..................................................................................................................................9EMPIRICAL

ANALYSIS...............................................................................................................................114.1TheModel

..............................................................................................................................................114.1.1MeasuringDiscretionaryFiscalPolicyVolatility

...........................................................................114.1.2TheEndogeneityofSWF

.............................................................................................................124.1.3TheTreatmentEffect

Model.........................................................................................................134.1.4PropensityScoreMatching...........................................................................................................144.2DataSources..........................................................................................................................................165.6.EMPIRICAL

Results.................................................................................................................................165.1TheResultsoftheTreatmentEffectModel............................................................................................175.2TheResultsofPropensityScoreMatching

............................................................................................19CONCLUSION...........................................................................................................................................20References.........................................................................................................................................................26FIGURES1.CommodityPriceVariabilityandUnpredictability..............................................................................................72.CommodityPricesandKey

MacroeconomicIndicators....................................................................................83.MacroeconomicVolatilityinCommodity-ExportingCountries...........................................................................94.SWFsAdoptioninthePastDecades..............................................................................................................10TABLES1.SovereignWealthFundsandVolatilyofFiscalPolicy:FixedEffectsModel...................................................172.SovereignWealthFundsandVolatilityofFiscalPolicy:

AverageTreatmentEffects......................................183.StabilizationSovereignWealthFundsandVolatilityof

FiscalPolicy:Non-ParametricEstimations

...............19INTERNATIONALMONETARYFUND2IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachGlossaryATEFEMGDPGFCGMMILSAverageTreatmentEffectFixedEffectsModelGrossDomesticProductGlobalFinancialCrisisGeneralizedMethodsofMomentsInstrumentalLeastSquaresOrdinary-LeastSquaresPropensityScoreMatchingSovereign

WealthFundOLSPSMSWFINTERNATIONALMONETARYFUND3IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproach1.

INTRODUCTION

AND

MAIN

MESSAGESCommoditypricevolatilityhas

increased

inrecentyears—especiallyforoil—followinginlargeparttheeffectsofthepandemicandthewarinUkraine.1

Thisvolatilitypointstotheneedforanoptimalfiscalpolicyresponsetocommoditypriceshocks,especiallyforcommodity-exportingcountries.2Themainpolicychallengefacedbymostofthesecountriesis

copingwith

commoditypricevolatility.Suchvolatility

istransmittedtothe

economythroughseveralchannels,includingfiscalandexchangeratepolicy.3Moreover,highlyvolatile

commoditypricesgenerallyresultinvolatile

commodity-relatedrevenue,especiallyincountrieswheresuchrevenueconstitutesalargeshareoftotalbudgetrevenue.Thisoftentranslatesintovolatilepublicspending(IMF

2015).Largeandpersistentswingsincommodity-relatedrevenuesometimesresultsinaprocyclicalfiscalstance.Fiscalpolicymakerstendto

increasespendingduringperiodsofexpansion(whencommoditypricesarehigh)whiletheyareforcedtoreducespendingwhencommoditypricesandrevenuedecline

(IlzetzkiandVegh,2008;VillafuerteandLopez-Murphy,2010;Erbil,2011).Thisprocyclicalbehaviorcomplicatesmacroeconomicmanagementandleadstoconsiderableoutputvolatilitythatunderminesoverallmacroeconomicperformance.4

Indeed,severalempiricalstudiessuggestthathighergovernmentspendingvolatilityisassociatedwithlowereconomicgrowth(Al-Sadiqandothers,2021;BleaneyandGreenaway,2001;FurceriandKarras,2007;FatásandMihov,2003,2006)whileothersshowthatfiscalpolicyvolatilityactsasatransmissionmechanismforthe

“resourcecurse”(BleaneyandHalland,2009;SachsandWarner,1995).5

Moreover,largecommodityprice

swingshavemajorimpactsonkeysocialindicators(EstradesandTerra,2012;álvarez,García-Marin,andIlabaca,2021).Undertheneoclassicaland

Keynesianframeworks,the

maingoaloffiscalpolicyis

to

smoothoutputvolatilityduringbusinesscycles;thus,

itshouldbecountercyclical.

InBarro’s(1979)smoothingmodel,agovernmentisexpectedtorunsurplusesin

goodtimesandfiscaldeficitsinbadtimes(Céspedesand

Velasco,2014).UndertheKeynesianapproach,iftheeconomyis

inrecession,

agovernmentshouldraisespendingand/orlowertaxestosimulateeconomicactivity;whileduringeconomicbooms,thegovernmentshouldsavesurplusesgeneratedbytheoperationofautomaticstabilizers.Hence,fiscalpolicyisexpectedto

followcountercyclicalpatternsthroughautomaticstabilizersanddiscretionarychannels(Erbil,2011).1

Oilprices–asimpleaverageof

threespotprices:DatedBrent,

West

TexasIntermediate,

andDubaiFateh–increasedfromUS$23.4abarrelinApril2020toUS$114.7abarrelinJune2022,its

highestlevelsince

April2012(US$113.8).2

Commodity-exportingcountries

are

thosewithfuel/miningexportsaccountingforanaverageof20percent

oftheirexports

overadecade.3

Largefluctuationsinresourcerevenuemaygiveriseto

realexchangeratevolatility

andincreasesintheserevenuesmayleadto“Dutchdisease”(Davis

andothers2001).4

Outputvolatilitycanbemanifestedinabrupt

orunexpectedchanges

ineconomicgrowth.5

Thenaturalresourcecurseis

associatedintherelatedliteratureto

thosenatural-resource-richcountries

whoselevelsofwealthhavebeenhistoricallylowercomparedwithnon-naturalresource-richcountries.INTERNATIONALMONETARYFUND4IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachForcommodity-exportingcountries,anoptimalfiscalpolicyresponseto

commoditypricefluctuationsiscountercyclicalpolicybehavior:savecommodity-relatedrevenueincreasesduringboomsandusethesebuffersduringdownturns(IMF,2012).Thissuggeststhat

commoditypricevolatilitycreatesanincentivetosavesomecommodity-relatedrevenueasself-insuranceto

smoothconsumptionspending

whenpricesdrop.6

Thisinturnwouldhelpreducethemacroeconomicvolatilitystemmingfromcommodityprice

fluctuations.Thisargumentissupportedbyseveralempiricalstudiesthatfindthatlimitingprocyclicalpoliciesis

associatedwithimprovedmacroeconomicperformance(MedinaandSoto,2007;Rodriguezandothers,2007;Schmidt-Hebbel,2012;IMF,2012).Howshouldcommodity-exportingcountriesinsulatetheir

economiesfromthenegative

terms-of-tradeshocksinducedbylargeandunpredictablecommoditypricefluctuations?Fromatheoreticalperspective,oneapproachtoshieldordelinkpublicexpendituresfromresourcerevenuevolatilityisthroughtheestablishmentofastabilizationSWF.Thisinstrumentisdesignedmainlyto

accumulateresourceswhenthecommoditypriceexceedsacertainreference

priceandtodisbursewhenthepricefallsbelowanotherreferenceprice(Davisandothers,2001).Thus,astabilizationSWFwould

constituteself-insurancetohelp

smoothfluctuationsinbudgetresourcesbyreducingoreliminatingtheuncertaintyandvolatilityofresource-relatedrevenueflowingintothebudget.Overthepastfewdecades,thenumberofSWFsestablishedhasincreasedsignificantly,particularlybycommodity-exporting

countries.Thesefundshavebeencreatedformanyreasons,includingas(IMF,2008):??short-termstabilizationfunds,toinsulatetheeconomyfromswingsin

commodityprices;long-termsavingsfunds,totransformincomefromnaturalresourcesintoadiversified

portfolioofassets,accumulatingsavingsforfuturegenerations;???reserveinvestment

corporations,toincreasethereturnonforeignexchangereserves;developmentfunds,tohelpfundinfrastructuralprojectsthatexpandthecountry’spotentialgrowth;andcontingentpensionreservefunds,tocomplementresourcesfromindividualpensioncontributionsthatprovideforpensionliabilitiesonthegovernment’sbalance

sheet.WhetherstabilizationSWFscontributetolowerfiscalpolicyvolatilityisanempiricalquestionthathasrecentlyreceivedsomeattention

intherelatedeconomicliterature,butempiricalresultshavebeenmixed.WhilesomefindanegativerelationshipbetweenthepresenceofSWFsandthevolatilityoffiscalspendingpolicy,othersfailtofinda

significantrelationship.6

Throughoutthis

paperweusefiscalspendingandgovernmentconsumption

interchangeably.INTERNATIONALMONETARYFUND5IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachThus,thispaperseekstoempiricallyreassesswhetherstabilizationSWFshelpsmoothgovernmentconsumptionandwhetheracountrywithsuchan

instrumenthaslessvolatilefiscalpolicy.Thepapercontributestotheliteraturein

severalways.First,weuse

alargesampleof182advancedanddevelopingcountriesovertheperiod1980-2019.Second,giventhat

acountry’sdecisiontoestablishastabilizationSWFisnotarandomdecision,weaddresstheself-selectionproblemofhavingsuchafundby

usingtwodifferenteconometricapproaches:i)a

two-stepestimationmethod

basedonatreatmenteffectmodel,andii)anon-parametricestimationmethodbasedonaPropensityScoreMatching(PSM)technique.Hereinliesthemethodologicalnoveltyofourapproach.Webelievethisis

thefirst-timefiscalpolicyhas

beenevaluatedinthisway.OurstudyfindsanegativerelationshipbetweenthepresenceofstabilizationSWFsandthevolatilityofgovernmentconsumption.Empiricalresultssuggestthat

countrieswithstabilizationSWFstendtoexperienceless

volatilefiscalpolicy.Theestimatedaveragetreatmenteffectsshowsthata

countrywithastabilizationSWFhasabout14percentlessvolatilegovernmentconsumptionrelativetoacountrywithoutanSWF.Thisresultisrobustunderdifferentspecificationsandsample

periods.Therestofthepaperbrieflydescribesthefrequencyand

theimpactofcommoditypricevolatility(Section2);surveystheexistingliterature(Section3);presentstheempiricalmodel,theestimationmethods,anddatasources(Section4);discussestheempiricalresults(Section5);andconcludes(Section6).2.

THE

FREQUENCY

AND

IMPACT

OF

COMMODITYPRICEVOLATILITYCommoditypricecycleshavebeencharacterizedbyremarkableboomsandslumps,

aswellashighlyuncertainforecasts,

inpastdecades(Figure1).Theboom

episodesoccurredduringmuchofthe1970s,2002-08,andthepost-COVID-19

period,whilethesharpdeclinestookplaceduring1979-86,theglobalfinancialcrisis,andduring2012-16.Boomsandbustscanmeanpricesmovingbyas

muchas

40-80percentoveradecade(IMF,2015).Moreover,overthepastfewyears,oilpriceshavefluctuatedwidely—betweenUS$23.4abarrelinApril2020toUS$114.7abarrelinJune2022.7Apartfromcyclicalfluctuations,

commoditypricesdisplaylong-termtrends.Whilethesetrendsaredifficultto

forecast,theyshowthatsomepriceshocksmayhaveapermanentcomponent(IMF,2012).7

Oilprices

areasimpleaverageofthreespot

prices:DatedBrent,WestTexasIntermediate,andDubaiFateh.INTERNATIONALMONETARYFUND6IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachFigure

1.

Commodity

Price

Variability

and

Unpredictabilitya.

Commodity

Price

Index(Index,2010=100)b.

Oil

Price

Forecasts($US/barrel)Sources:

WorldBank’sPinkSheet,

BloombergLP,

andIMF,WorldEconomic

Outlook.1/

Totalcommodity

pricesmeasuredbyBloomberg’scommodity

priceindex.Macroeconomicperformanceincommodity-exportingcountriestendstomovewith

commodityprice

cycles.Infact,historicaldatashowhowtheperformanceofkeymacroeconomicindicatorscan

beinfluencedbyunexpectedfluctuationsininternationalcommodityprices.

Theglobalpriceincreaseobservedpriortotheglobalfinancialcrisiscoincidedwithcurrentandfiscaldeficitresultsandanotable

improvementincommodity-exportingcountriesinthe

sameperiod(Figure2).INTERNATIONALMONETARYFUND7IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachFigure

2.

Commodity

Prices

and

Key

Macroeconomic

Indicators

1/a.

Economic

Growthb.

Inflation

2/(PercentageChange,Index)(PercentageChange,Index)c.

Current

Account

Balanced.

Overall

Fiscal

Balance(InpercentofGDP,Index)(InpercentofGDP,Index)Sources:IMF,WorldEconomicOutlook,BloombergLP,andauthors’calculations.1/

Keyvariables

areexpressedas

theweightedaverageofcommodity

exportingandnon-commodity-exportingcountriesbyusingtheirnominalGDPdenominatedinUS

dollars

adjustedbyPurchasingPowerParity.2/Refersto

endofperiodinflation.3/Totalcommodity

pricesmeasuredbyBloomberg’s

commodity

priceindex.

4/

Yearsinwhichinflationincommodity-exportingcountriesreachedmorethan50

percent

wereremovedfromthesample.Moreover,thereisaconsensusintheliteraturethatinternationalcommodityprice

volatilityleadstomorevolatilemacroeconomicconditions,mitigatingeconomicstability(Figure3).Usingannualdatafor1990-2007,Cavalcanti,Mohaddes,and

Raissi(2011)showthatthe

negativegrowtheffectsofcommodityterms-of-tradevolatilityoffsetthepositive

impactofcommoditybooms,

contributingtothe“resourcecurse”paradox.Similarly,Blattmanandothers(2007)estimatetheimpactofterms-of-tradevolatility,theresultof

abruptcommoditypriceINTERNATIONALMONETARYFUND8IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachfluctuations,

onforeigninvestmentandeconomicgrowthperformanceforapanelof35

countries.Thisnegativerelationshipbetweenterms-of-tradevolatilityandeconomicgrowthisalsoassessedbyBleaneyandGreenaway(2001).Figure

3.

Macroeconomic

Volatility

in

Commodity-Exporting

Countriesa.

Current

Account

Balance

and

Growthb.

Overall

Fiscal

Balance

and

Growth(StandardDeviation,2002-2019)(StandardDeviation,2002-2019)Sources:IMF,WorldEconomicOutlookandauthors’calculations.3.

REVIEW

OF

THE

LITERATURESWFsaregovernment-ownedinvestmentfunds,establishedforavarietyofpurposes.

TheIMF

(2008)distinguishesseveraltypesof

SWFsbasedontheirmain

objectives:i)stabilizationfunds,setuptoinsulatetheeconomyagainst

swingsincommodityprices;ii)

savingsfunds,whichtransformtheincomefromcommodity-relatedresourcesintoadiversifiedportfolioofassets,accumulatingsavingsforfuturegenerations;iii)reserveinvestment

corporations,establishedtoincreasethereturnonforeignexchangereserves;iv)developmentfunds,whichhelpfund

infra-structuralprojectsto

increasethecountry’spotentialgrowth;andv)contingentpensionreservefunds,whichcomplementresourcesfromindividualpensioncontributionstoprovideforpensionliabilitiesonthegovernment’sbalancesheet.Overthepastfewdecades,thenumberofSWFsestablishedhasincreasedsignificantly,particularlyby

commodity-exportingcountries(Figure4).INTERNATIONALMONETARYFUND9IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachFigure

4.

SWFs

Adoption

in

the

Past

Decades

1/a.By

Commodity-Exporting

Status(Numberofcountries)b.By

Type

of

SWFs(Numberofcountries)Sources:GlobalSWFDatabase,Sovereign’s

WealthFundInstitute,andIMF,WorldEconomic

Outlook.1/

Thecountry

numberis

basedon

thesampleusedinthispaperwhichcovers

182countries.AstabilizationSWFisamechanismdesignedtoreducetheeffectsofvolatile

commodity-relatedrevenueongovernmentconsumptionandtheoveralleconomy’sperformance.Itsobjectivesmayalsoincludesupportingfiscaldisciplineandprovidinggreatertransparencyinthespendingofrevenue.StabilizationSWFscanhelpagovernmentdelinkitsexpenditurefromthenegativeimpactsofcommodity-relatedrevenuevolatilitycausedbytheboomsandbustsofcommodityprices.Whenthecommoditypricesexceeda

certainreferencelevel,resourcesareaccumulated.

Whenpricesarebelowanotherreferencelevel,governmentcanutilizepartoftheseresourcestosmoothoutitsspending

intheshort-term(Davisandothers,2001).Intheabsenceoffinancingopportunities,whenthecommodity-relatedrevenuesarelow,governmentstendtocutexpendituresorseektoraisenon-commodityrevenue.Sincethissometimescouldbedifficulttodo

intheshortterm,agovernmentthathasastabilizationSWFcouldsmoothitsspendinginshort-termthroughdrowningdownresourcesfromitsfund.Thus,astabilizationSWFservesasaself-insuranceinstrumenttohelpsmoothfluctuationsintheresourcesavailabletothebudget,byreducingtheuncertaintyand

volatilityofcommodity-relatedrevenues.However,itisimportanttonotethatintheabsenceofliquidityconstraints,a

stabilizationSWFmightnothavedirectroleinstabilizinggovernmentexpendituressincegovernmentscouldmeettheirfinancingneedsthroughborrowing.Fromanempiricalperspective,thereisnoconsensuson

whetherthepresenceofastabilizationSWFhelpreducefiscalspending

volatility.SomestudiesfindthatstabilizationSWFscontributeto

smoothinggovernmentspendingwhileothersfindthatthesefundsleadtohigher

volatilityinthefiscalpolicyandothersdonotfindstatisticallysignificantresults.Bagattini(2011)findsthatthepresenceofastabilizationSWFhelpsreduceINTERNATIONALMONETARYFUND10IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachgovernmentspendingvolatilityalthoughtheimpact

is

verysmall.Sugawara(2014),usingpaneldatafor68resource-richcountriesover

1988–2012,findsthatstabilizationSWFscontributetosmoothinggovernmentexpenditure.CrainandDevlin(2002)usingdatafor71countriesovertheperiodof1970–2000,showthatnaturalresourcefundsincreasefiscalvolatility,particularlyinoil-exportingcountries.

Ossowskiandothers(2008)andBovaandothers

(2016)didnotfindsignificant

evidencethatSWFsreducefiscalvolatility.8Whilethese

studiesemployeddifferentmethodologies,differentsample-selections,differentsetsofexplanatoryvariables,andanalyticaltools,theyfailedtoproperlycontrolfortheendogeneityoftheSWFs.

SettingupaSWFisnotarandomdecisionbutratherdeterminedbyeconomicandnon-economicfactors,includingpastcommoditypriceboomandbustcycles.Thispapercontributestothe

literatureinseveralways.First,weuselargesampleof

advancedanddevelopingcountriesovertheperiod1980-2019.Second,giventhat

acountry’sdecisiontoestablishastabilizationSWFisnotarandomdecision,weaddresstheself-selectionproblemofhavingsuchafundthroughusingtwodifferenteconometricapproaches:i)inthefirstapproach,weuseatwo-stepestimationmethodbasedonatreatmenteffectmodelandii)

inthesecondapproach,weuseanon-parametricestimationmethodbasedonaPropensityScoreMatching(PSM)technique.4.

EMPIRICAL

ANALYSIS4.1

The

ModelWebeginourempiricalanalysisby

notingthattheeffects

ofastabilizationSWFongovernmentconsumptionvolatilitymay

beformalizedinthefollowinglinearrelationship:′,(1)=+++,,,whereσ

isameasureofdiscretionaryfiscalpolicyvolatilityinacountry(i)attime(t).

x

isa

vectorofi,texogenousvariables.SWFis

adummyvariableequalsto

oneifcountry(i)hasastabilizationSWFattimetandzerootherwise.β

and

areparameterstobeestimated,η

istime

invariantcountry-specificerrorterm,andisatimedependentrandomdisturbanceterm.4.1.1

Measuring

Discretionary

Fiscal

Policy

VolatilityTheliteratureonfiscalpolicyusesseveralapproachesto

measuringthediscretionaryfiscalpolicyvolatility,noneisclearlysuperiortoany

other.Thefirstapproachconsistsofcalculatingthestandarddeviationoftheannualgrowthrateofrealgovernmentconsumption.However,themainproblemofthisapproachisthatit8

Severalempiricalstudiesalsolookedattherelationshipbetweenoilfundsandmacroeconomicstability

(ShabsighandIlahi,

2007;Mehrara,

KarsalariandHaghiri,

2012).INTERNATIONALMONETARYFUND11IMFWORKINGPAPERSSovereignWealth

Fundsand

Fiscal

PolicyProcyclicality:

Non-ParametricApproachignoresthecyclicalstateoftheeconomy.Thesecondapproach—whichisbasedonthepioneerworkofFatásandMihov(2003,2006)—measuresvolatilitybasedonthestandarddeviationoftheresidualsfromaregressionofafiscalreactionfunction.AccordingtoFatásandMihov(2003),fiscalpolicyconsistsofthreemaincomponents:(i)automaticstabilizers;(ii)discretionarypolicythatreactstothestateoftheeconomy;and(iii)dis