外援的宏观经济效应综述

DepartmentofEconomicsUniversityofSt.GallengTheMacroeconomicEffectsofForeignAid:ASurveyPhilippHarmsandMatthiasLutzSeptember2004DiscussionPaperno.2004-11Editor:Prof.JörgBaumbergerUniversityofSt.GallenDepartmentofEconomicsBodanstr.1CH-90[e-mail]:ElectronicPublication:ForschungsgemeinschaftfürNationalökonomieanderUniversitätSt.GallenDufourstrasse48CH-9000St.GallenPhone+41712242300Fax[phone]:ASurvey*PhilippHarmsMatthiasLutzAuthors’addresses:PhilippHarmsStudyCenterGerzenseeP.O.Box213115GerzenseeSwitzerlandTel.:[phone]Fax:[phone]Email:[e-mail].:[phone]Fax:[phone]Email:[e-mail]acroeconomiceffectsofaidhasexpandedrapidlyinrecentyears.Inthispaper,weprovideasurveyofthisliterature.Westartbyreviewingsometheoreti-calmodelsthatsuggestapositiveimpactofaidoninvestmentandgrowth.Wethendiscusstheempiricalevidence,givingparticularattentiontotheroleofinstitutionsandpoliciesindeterminingaideffectiveness.Asageneralconclusion,wesuggestadoptingamoredisaggregateperspectivewithrespectbothtodifferenttypesofaidandtovari-ousaspectsofgovernance.Keywordsforeignaid,economicgrowth,institutions,governanceJELClassificationsF35,F43,O16,O191.IntroductionForeignaidflowsfromDACcountriestothedevelopingworldstagnatedduringthe1990s,reachingalowpointin1997at$48.5bn(WorldBank2004).1Asthesefiguresareinnominalterms,thetrendinrealtermshasbeenevenworse,whetheradjustedforinflationorcalculatedrelativetorecipientcountries’populations.Particularlystrikingisthedropinaidflowsrelativetodonorcountries’GDP.Onthismeasure,richcountriesreducedtheiraidcontributionsfromaround0.34%to0.23%oftheiroutputbetween1990and2002(WorldBank2004).The‘aid-fatigue’reflectedinthesefigurescanbetracedbacktoanumberofeconomicandpo-liticalchanges(Hopkins2000,RobinsonandTarp2000):changesinindustrialisedcountries’foreignpolicyprioritiesaftertheendofthecoldwar,afurtherweakeningofoldcolonialties,lowerpay-offsforspecialinterestgroupsduetothechangingregionalfocustowardsthecom-merciallylessinterestingAfricancountries,tighterbudgetsindonorcountries,andagrowingdistrustofgovernmentsandinternationalorganisationsinindustrialisedeconomies.Inadditiontotheseforces,akeyreasonforthedryingupofaidflowshascertainlybeentheper-ception–evenamonggroupstraditionallysupportiveofforeignaid–thataidhasfailed,atleastpartly.Therehavebeenreportsofcorruptionandpooradministration,withaidmanagementty-ingupvaluableresourcesinrecipientcountries(Kanbur2000)andquestionableaidallocationdecisionsamongdonors.Althoughmanyaidprojectsweredeemedtobesuccessfulconsideredontheirown(orbetter,withrespecttotheirpre-definedobjectives),thereistheperceptionthattheoverallimpacthasbeenlessthanthesumofitsparts,somethingthatMosley(1987)referredtoasthe‘micro-macroparadox’.AveryillustrativeexampleoftheobservationsthathavefuelledaidscepticismisgivenbyEast-erly(1999).Predictingtheimpactthataidshouldhavehadonoutputonthebasisofthestillwidely-usedtwo-gapmodelhecomparedthiswiththeactualperformanceofalargesetofcoun-tries.Inhispaperandsubsequentbook(Easterly2001)hepresentsthecorrespondingfigureforZambia,acountrywherethepredictiondivergestoaparticularlystrikingextentfromactualper-formance.Whilewehavenotfounditpossibletocompletelyreplicatehisfigurewithnewerdata,thevisualisationofthegapbetweenthesupposedaideffectandrealityisstillstriking(seeFigure1).By2001ZambianGDPpercapitawasonlyaboutafifthofwhatwouldhavebeenpredictedhadallaidgoneintoinvestmentandallinvestmentintogrowth.2Suchablatantdiscrepancyisnosurprisetothoseeconomistswhohavealwaysbeenscepticalabouttheabilityofaidtoliftdevelopingcountriesoutofpoverty.Thus,thelatePeterBauerkeptemphasisingthecorruptingandcounterproductiveeffectsofaid:“Becauseaidaccruestothegovernmentitincreasesitsresources,patronage,andpowerinrelationtotherestofsociety.Theresultingpoliticisationoflifeenhancestheholdofgovernmentsovertheirsubjectsandincreasesthestakesinthestruggleforpower.Thisresultinturnencouragesorevenforcespeopletodivertattention,energy,andresourcesfromproductiveeconomicactivitiestoconcernwiththeout-comeofpoliticalandadministrativeprocessesanddecisions”(Bauer1991,p.45).SodoesrecentaidexperienceproveBauerright?Toanswerthisquestion,oneneedstoassesswhethertheZambianexamplecanbegeneralised.Istheapparentfailureofaidinthiscaseanexceptionordoesitapplytotheaveragedevelopingcountry?Isaidperseineffective,orcanweidentifysomefundamentalforcesthatareresponsibleforthefailureofaidinsomecountriesanditssuccessinothers?Thesearethequestionsthatwewanttoaddressinthissurvey.1Duetoarisein2002,theyhavejustcaughtupwiththelevelsseenintheearly1990s(ataround$58bn).2The‘whatmighthavebeen’serieswascalculatedbytakingactualGDPpercapita(inconstantUS$)in1960andprojectingfuturevaluesusingahypotheticalgrowthrateequaltothesumofactualinvestmentandaidinflow(asashareofGDP)d