Thomas Farole, Esteban Ferro, and Veronica Michel Gutierrez

Job Creation in the Private Sector

JOBS WORKING

PAPERIssue No. 5

An Exploratory Assessment of Patterns and Determinants at the Macro, Sector, and Firm Levels

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

©2017InternationalBankforReconstructionandDevelopment/TheWorldBank.

1818HStreetNW,Washington,DC20433,USA.Telephone:202-473-1000;Internet:www.worldbank.org.

Somerightsreserved

ThisworkisaproductofthestaffofTheWorldBankwithexternalcontributions.Thefindings,interpretations,andconclusionsexpressedinthisworkdonotnecessarilyreflecttheviewsofTheWorldBank,itsBoardofExecutiveDirectors,orthegovernmentstheyrepresent.TheWorldBankdoesnotguaranteetheaccuracyofthedataincludedinthiswork.Theboundaries,colors,denominations,andotherinformationshownonanymapinthisworkdonotimplyanyjudgmentonthepartofTheWorldBankconcerningthelegalstatusofanyterritoryortheendorsementoracceptanceofsuchboundaries.

NothinghereinshallconstituteorbeconsideredtobealimitationuponorwaiveroftheprivilegesandimmunitiesofTheWorldBank,allofwhicharespecificallyreserved.

RightsandPermissions

ThisworkisavailableundertheCreativeCommonsAttribution3.0IGOlicense(CCBY3.0IGO)http://creativecommons.org/licenses/by/3.0/igo.UndertheCreativeCommonsAttributionlicense,youarefreetocopy,distribute,transmit,andadaptthiswork,includingforcommercialpurposes,underthefollowingconditions:

Attribution—Pleasecitetheworkasfollows:ThomasFarole,EstebanFerro,andVeronicaMichelGutierrez.2017.“JobCreationinthePrivateSector-AnExploratoryAssessmentofPatternsandDeterminantsattheMacro,Sector,andFirmLevels.”WorldBank,Washington,DC.License:CreativeCommonsAttributionCCBY3.0IGO

Translations—Ifyoucreateatranslationofthiswork,pleaseaddthefollowingdisclaimeralongwiththeattribution:ThistranslationwasnotcreatedbyTheWorldBankandshouldnotbeconsideredanofficialWorldBanktranslation.TheWorldBankshallnotbeliableforanycontentorerrorinthistranslation.

Adaptations—Ifyoucreateanadaptationofthiswork,pleaseaddthefollowingdisclaimeralongwiththeattribution:ThisisanadaptationofanoriginalworkbyTheWorldBank.ViewsandopinionsexpressedintheadaptationarethesoleresponsibilityoftheauthororauthorsoftheadaptationandarenotendorsedbyTheWorldBank.

Third-partycontent—TheWorldBankdoesnotnecessarilyowneachcomponentofthecontentcontainedwithinthework.TheWorldBankthereforedoesnotwarrantthattheuseofanythird-party-ownedindividualcomponentorpartcontainedintheworkwillnotinfringeontherightsofthosethirdparties.Theriskofclaimsresultingfromsuchinfringementrestssolelywithyou.Ifyouwishtore-useacomponentofthework,itisyourresponsibilitytodeterminewhetherpermissionisneededforthatre-useandtoobtainpermissionfromthecopyrightowner.Examplesofcomponentscaninclude,butarenotlimitedto,tables,figures,orimages.

AllqueriesonrightsandlicensesshouldbeaddressedtoWorldBankPublications,TheWorldBankGroup,1818HStreetNW,Washington,DC20433,USA;fax:202-522-2625;e-mail:pubrights@worldbank.org.

ProjectNumber:P164622ReportNumber:AUS22807

Images:©WorldBank.Furtherpermissionrequiredforreuse.

1

ABSTRACT

Whydosomecountriescreatemorejobsthanothers?Toconsiderthisquestion,inthispaperwefocusononeofthemostbasicrelationships,betweengrowthandemployment.Inpractice,theprivatesectorresponds very differently to growth (and decline) across countries. Understanding the patterns anddriversofprivatesectordecisionstoexpandandshedjobsmaybeimportanttoguidepolicyapproachesfor job creation. This paper analyzes the output-employment relationship in the context of businesscyclesatthreelevels:themacro-economy;industry(inmanufacturing);andfirms.Theresultshighlightmajordifferencesinprivatesectorjobcreationresponsivenesstogrowthacrossstagesofdevelopmentand economic structures, but a critical finding is that economies (and firms)where formal sector jobcreationwasmore responsive togrowthcyclesgeneratedmore jobsoverall. Inaddition, results fromboththemacroanalysisandthesectoralanalysissuggestssignificantcomplementaritybetweencapitaland labor. Finally, the findingsmay help to frame a broad policy agenda for job creation, including:macro-economic fundamentals, responsive labor markets, access to finance, competition, and afacilitativebusiness regulatoryenvironment.Thesearenot surprising,butnevertheless framea setofissuesthatcouldbeexploredinfurtherresearch.

2

ACKNOWLEDGEMENTS

This report was prepared by the World Bank Group’s (WBG) Jobs Group. The principal authors areThomas Farole, Esteban Ferro, and Veronica Michel Gutierrez. We are grateful to Alvaro Gonzalez,MarekHanusch,andGonzaloVarelafortheirvaluablecommentsandsuggestions.WearealsogratefultoAlvaroGonzalez,Hari Subhash,andLeonardoLacovone forprovidingaccess to thedatasetused inSection3ofthispaper.

The publication of this report has been made possible through a grant from theWorld Bank’s JobsUmbrellaTrustFund,whichissupportedbytheDepartmentforInternationalDevelopment/UKAID,andthe Governments of Norway, Germany, Austria, the Austrian Development Agency, and the SwedishInternationalDevelopmentCooperationAgency.

ThereportwaspreparedunderprojectFirms,AccesstoMarkets,ValueChainsandJobs(P164622).

3

CONTENTS ABSTRACT............................................................................................................................................1

ACKNOWLEDGEMENTS.........................................................................................................................2

1.ECONOMICGROWTH,THEPRIVATESECTORANDJOBCREATION–FRAMINGTHEISSUE..................6

2.MACRO-LEVELVIEW:THELINKBETWEENBUSINESSCYCLESANDEMPLOYMENTGROWTH..............9

Businesscyclesandemploymentgrowth–descriptiveoverviewhighlightingcountryvariation...............9

Determinantsofemploymentoutcomes...................................................................................................18

3.INDUSTRY-LEVELVIEW:ANALYSISOFGROWTHANDEMPLOYMENTRELATIONSHIPSINMANUFACTURING..............................................................................................................................21

Relationshipbetweengrowthandemployment........................................................................................21

Howareemploymentelasticitiesassociatedwithoveralljobcreation?...................................................26

4.FIRM-LEVELVIEW:FIRMTYPES,REGULATION,ANDJOBCREATION................................................30

Theregulatoryenvironmentandjobturnover..........................................................................................31

5.CONCLUSIONSANDFUTURERESEARCH..........................................................................................39

ANNEXA:OKUN’SLAW......................................................................................................................42

BIBLIOGRAPHY...................................................................................................................................46

4

ABBREVIATIONS Mostusedabbreviationslist:pleasekeeprelevantonesandaddyourown

ADB AsianDevelopmentBankALMP activelabormarketprogram

ASA advisoryservicesandanalyticsCBO community-basedorganizationCCSA cross-cuttingsolutionsareaCCT conditionalcashtransfer

CEO chiefexecutiveofficerCIS CommonwealthofIndependentStatesCPF countrypartnershipframework

DAC DevelopmentAssistanceCommittee(OECD)DfID DepartmentforInternationalDevelopment(UnitedKingdom)EBRD EuropeanBankforReconstructionandDevelopmentECA EuropeandCentralAsia

EU EuropeanUnionFAO FoodandAgricultureOrganizationoftheUnitedNationsFDI ForeignDirectInvestment

FCS fragileandconflictstateFY fiscalyearGDP GrossDomesticProductGIZ DeutscheGesellschaftfürInternationaleZusammenarbeit

GNI GrossNationalIncomeGP globalpracticeHIC high-incomecountryHS householdsurvey

IAT informalapprenticeshiptrainingICLS InternationalConferenceofLabourStatisticiansICR implementationcompletionreport

IDA InternationalDevelopmentAssociationIE impactevaluationIDB Inter-AmericanDevelopmentBankIEG IndependentEvaluationGroup

IGA income-generatingactivityILC InternationalLabourConferenceIFC InternationalFinanceCorporationIILS InternationalInstituteforLabourStudies

ILS InternationalLabourStandardsIMF InternationalMonetaryFundIOE InternationalOrganizationofEmployers

ITUC InternationalTradeUnionConfederationIZA InstitutefortheStudyofLaborICT informationandcommunicationstechnologyILO InternationalLaborOrganization

5

ISIC InternationalStandardIndustrialClassificationKILM KeyIndicatorsofLaborMarket

KM KnowledgeManagementLDC leastdevelopedcountryLFS laborforcesurveyMDG MillenniumDevelopmentGoal

M&E monitoringandevaluationMFI micro-financeinstitutionMoF MinistryofFinanceMIS managementinformationsystem

MSE microandsmallenterpriseMSME micro,smallandmediumenterpriseNIS NationalInstituteofStatistics

n.a. notapplicableN/A,— notavailableNGO nongovernmentalorganizationNEET notineducation,employment,ortraining

ODA officialdevelopmentassistanceOECD OrganizationforEconomicCo-operationandDevelopmentOOP out-of-pocketpayments

PER publicexpenditurereviewPEP publicemploymentprogrammePES publicemploymentservicesPPP purchasingpowerparity

PHC populationandhousingcensusPRSP povertyreductionstrategypaperRCT randomizedcontroltrialSME smallandmedium-sizedenterprises

SDG sustainabledevelopmentgoalSCD strategiccountrydiagnosticSOE stateownedenterprise

STEP skillstowardsemploymentandproductivityTFP TotalFactorProductivityTOR termsofreferenceTVET TechnicalandVocationalEducationandTraining

UN UnitedNationsUNDP UnitedNationsDevelopmentProgrammeUNIDO UnitedNationsIndustrialDevelopmentOrganization

USD UnitedStatesDollarsVAT ValueaddedTaxVET technicalandvocationaleducationandtrainingVC valuechain

WBG WorldBankGroupWDI WorldDevelopmentIndicatorsWDR WorldDevelopmentReportWTO WorldTradeOrganization

6

1. ECONOMIC GROWTH, THE PRIVATE SECTOR AND JOB CREATION – FRAMING THE ISSUE Listentoanypolicymaker’sspeechorpickupanynewspaper,andtheurgencyof jobcreationisclear.TheWorldDevelopmentReport(WDR)onJobs(WorldBank,2013)reportsthatmorethan200millionpeople worldwide are unemployed, while another 2 billion working age adults remain outside theworkforce.Ontopofthis,thereportestimates600millionadditionaljobswillbeneededoverthenext15yearsjusttokeeppacewithnewentrantstothelabormarket.Wherewillthesejobscomefrom?Whilepublicsectoremploymenthasaroletoplay,theWDRarguesthattheprivatesector istheonlysustainableengineof jobcreation inanyeconomy,emphasizingthewidelyreportedstatisticthat9outofevery10jobsareintheprivatesector(WorldBank,2013).Butthisincludesfarmers,otherself-employed,andinformalworkers,inadditiontothosethatarewageworkersinformalenterprises.Indeed,formalprivatesectorjobsaretheexceptionratherthantheruleinmanycountries.Thus,theroleoftheprivatesectorinjobcreationislikelytovarysignificantlydependingonthestructureof theeconomyandthenatureof theprivatesector. It isalso likely tovarysignificantlydependingthepoliciesgovernmentspursue,whichsets theconditionsthat facilitateorhinderprivatesectorjobcreation.Forthisreason,understandingthepatternsofemploymentacrosseconomies,timeperiods,andbroadpolicydomains,isanimportantstartingpointtoconsideringhowtosupportrobustprivatesectorjobcreation.Wecanthinkofthreebroad(butnotmutuallyexclusive)mechanismsthatdeterminejobcreationintheprivatesector.Firstandforemostisgrowth–inresponsetoincreasedaggregatedemandnewfirmsareestablishedorwhenexisting firmsexpand. Expandedoutput is usually anecessary condition, if not asufficient one (output growth may be employment neutral if labor productivity increases at acorrespondingrateoriffirmssubstitutecapitalforlabor).Second,foranygivenlevelofoutput,relativedemandfor labormayvarydependingonthesectoralandenterprisestructureoftheeconomy(somesectorsandenterprisetypesaremorelaborintensivethanothers.Finally,foranygivenlevelofoutputandsector/firmcomposition, the relativeuseofcapitaland labor (factor intensity)willdetermine thenumberofjobscreated.Thispaperisintendedasthefirstinapossibleseriesofanalysesonprivatesectorjobcreation.Wefocusin this paper on the first of the mechanisms outlined above, analyzing one of the most basicrelationshipsineconomics,–betweengrowthandemployment–althoughwealsoexplorebrieflyhowcompositional effects and factor intensitymediates the output-employment relationship. Subsequentpapersinthisseriescouldprobesomeofthefindingsfromthisinitialassessmentaswellasexploretheothermechanismsoutlinedabove.

7

Thispaperanalyzestheoutput-employmentrelationshipinthecontextofbusinesscycles1,whichallowsfor comparing across a large set of countries by measuring employment elasticities to growth anddecline.Asubstantialliteratureexistsjobcreationoverbusinesscycles.Mostpapersfocusonwhichfirmtypesdrivejobcreation,includingthelong-standingdebateonwhetherlargeorsmalland/oryoungorestablished firms createmore jobs (Gertler andGilchrist, 1994; Fort et al, 2013; Decker et al, 2014).Analysis taking thebusiness cyclesapproachalso tend tobe single-country focused,mostoftenusingdata from the United States or other high income countries, with relatively little analysis done oncountriesoutsideoftheOECD(foranexampleofmulti-countryanalysisoutsidetheOECD,seeHanusch,2012).Among the few large cross-country studies that followabroadly similar approach, FreundandRijkers (2014) study episodes of sharp unemployment reduction in 94 rich, middle-income, andtransitioncountries.Theyfindthatwhilethebusinesscycleisthebiggestfactorshapingunemploymenttrends,countrieswithabetterregulatoryenvironmentbetterarelesslikelytohavehighunemploymentandmorelikelytorecoverrapidlyfollowingaspikeinunemployment.Themainobjectiveofthispaperismodest–weaimtoassesstherelationshipbetweengrowthandjobcreationacrossabroadcross-sectionofeconomies.Thefocusisonidentifyinganddescribingpatternsintherelationshipbetweengrowthandjobcreationoverbusinesscyclesindifferentcountrycontexts.Butwealsoaskwhyisthatsomecountriescreatemorejobsthanothersduringperiodsofgrowth.Andsowebegintoexplorestructuralandpolicy-relatedfactorsthatmayexplainthepatternsofoveralljobcreationandemploymentelasticitytogrowth.Wecanthinkofseveralfactorsthatmayberelevant indetermining these patterns: structural and compositional effects; relative prices of capital and labor;marketrigidities(businessregulations, financialand labormarkets);anduncertaintiesemanatingfrompolicy,institutions,and/orhistoricalvolatility.The hypothesis is that factors such as income levels, sectoral composition, capital intensity, and thestructureoftheenterprisesectorandoflabormarketswillhavesignificantimpactsonhowtheprivatesectorrespondstocreate(shed)jobsinperiodsofgrowth(decline).However,controllingforstructuralfactors,thehypothesisisthatcountrieswithmoreeffectiveenablingenvironments(broadlydefined)tosupporttheprivatesectorwillcreatemorejobsoverallandexhibitemploymentoutcomesthatrespondefficientlytobusinesscycles(i.e.intermsofpatternsoflaborsheddingandabsorption).Thishypothesisis explored through a combination of descriptive and regression analysis of cross-country data onemployment outcomes at three levels: in themacro economy; at themeso (sector) level; and at themicro(firm)level.Thepaperisorganizedinfoursectionsfollowingthis introduction.Section2takesamacrolenstotherelationshipbetweenbusinesscyclesandemploymentoutcomes,introducingtheOkuncoefficientasakeymeasureoftheemploymentelasticitytogrowth.TheOkuncoefficientisderivedfromOkun’sLaw,which refers to the empirical regularity found in the United States that seems to hold between theoutputandunemploymentgaps,orbetweencyclicalunemploymentandcyclicaloutput.WethentestOkun’sLawacrosstheglobaleconomy,wherewefind itholdswell forhighanduppermiddle-incomecountries.Butwefindthat jobcreationisnotassociatedwithoutputchanges in lower-middle income

1Weusetheterm‘businesscycles’torefertopatternsofeconomicgrowthordecline;notethatwhiletheterm‘businesscycle’isoftenassociatedwithquarterlyperiods,thispaperwehavedataonlyinone-yearintervals.

8

and low-income countries. This is perhaps not surprising, given that low-income countries arecharacterized by high levels of self-employment, and so labor markets may adjust to growth morethroughearnings than through jobcreationper se.Nevertheless, the findingsare relevantaswealsofind that countries in which Okun’s law holds, have on average higher employment growth. Thus,understanding what determines Okun’s law may also help us discover what determines higheremploymentgrowthacrosscountries.The third sectionmoves to the sector level and carries out amore descriptive analysiswith a cross-country dataset limited to the manufacturing sector. This helps to control for issues of sectoralcompositionandinformalitythatintroducecomparabilityproblemsinthemacroanalysis.Wefindthat,controlling for these factors, job creation patterns in lower income countries respond equallywell tooutputchangesasinhigherincomecountries(infact,evenmoreso).Thiscorroboratesthehypothesisthat structural and compositional factors determine significantly the patterns observed in themacrodata.Thefindingsfromthesectoralanalysisalsosuggestthatcapitalintensityisnotnecessarilyabarrierto jobcreation. Inthissection,wealsotesttosee if there isevidenceofanysystematicdelaystotheadjustment process at the sector of country-group level and find no evidence. Finally, we test forwhether volatility in output impacts adjustment (as measured by contemporaneous elasticity ofemployment)andfindsupportingevidenceat thesector levelbutnotat thecountry level,suggestingthat countries may follow an explicit or implicit ‘portfolio strategy’ in sectoral specialization thatsmoothsvolatility.Section4moves the lens to the firm level, exploiting firmmicrodata tounderstandhow firms createjobsinresponsetooutputchanges.Weexplorethelinksbetweentheregulatoryenvironmentinwhichfirmsoperateandjobturnoverbyexploitingtheobservedindustry-sizevariationsthroughadifference-in-difference approach. Additionally, we explore firm characteristics that are associated with higheremploymentgrowthratesintheshortrunandthelongrun,andidentifythefirmtypeswiththelargestimpactonemployment growth.Understanding thedeterminantsof thenumberof high growth firmsacross countries canhelpusunderstand that is needed to create theenvironment for these typesoffirmstoflourish.Wefindmixedresults,butoverallmacro,structural,andfirmcharacteristicsaremostassociatedwithjobcreation,whileregulatoryfactorsandinstitutionsalsomediateoutcomes.Section5concludesandoutlinespossiblepathsforfutureresearch.

9

2. MACRO-LEVEL VIEW: THE LINK BETWEEN BUSINESS CYCLES AND EMPLOYMENT GROWTH BUSINESSCYCLESANDEMPLOYMENTGROWTH–DESCRIPTIVEOVERVIEWHIGHLIGHTINGCOUNTRYVARIATIONThe responsiveness of the labormarket over business cycles has important implications forworkers.Shocks to the economy cause output to fluctuate around the economy’s potential. These outputmovements cause firms to hire and fire workers, changing employment; changes in employment, inturn, move the unemployment rate in the opposite direction. But the nature and strength of thisrelationshipvaries.Forexample,intheUS,thelastthreerecessions(1990-91,2001,and2007-09)havebeen followed by "jobless recoveries". Jobless recoveries are periods following the end of recessionswhenoutputgrowth resumesbutemploymentdoesnotgrow. Incontrast, cyclical recoveriesprior to1990wereaccompaniedbypromptrecoveriesinemploymentanddeclinesinunemployment.Whilelabormarketrelationshipswithgrowthvaryovertimewithincountries,theyvaryevenmore,andmore systematically, across countries. Figure 1 shows examples of the relation between GDP andemploymentgrowthforsixdifferentcountries.Inthetoppanel,wecanseethatinSwedenandtheUSthere is very close relation between GDP growth and employment growth rates. It is also worthhighlightinghowthesetwoeconomiesaremorestablethantheothers,particularlyemploymentdoesnot havemajor swings in either direction,with the exceptionof 2009when theworld financial crisisaffected most developed countries. On the other hand, these two economies also show substantialdifferences in their responsetothewild fluctuations inoutputthataccompaniedthe2009crisis,withemploymentfallingmuchlesssharplyinSwedenin2009,butalsoreboundingmuchlesssharplyin2010.This suggests thatwhile theupper incomestatus ispartof theexplanation forGDPandemploymentgrowthpatterns,otherfactorsalsoplayasignificantrole,inthiscasepotentiallythelevelofsafetynetsavailable in Sweden versus the US, the relative share of private versus public employment, or thewillingnessoftheprivatesectortomaintainworkersduringadownturn.Themiddlepanelshowstwodevelopingcountries,EcuadorandVietnam.ItisclearthatbothGDPandlabormarkets aremuchmore volatile than the countries in the top panel. Ecuador is an interestingexample, which shows a low correlation between GDP and employment growth between 1995 and2000;between2000and2005thereisahighcorrelation,andbetween2005and2010thereappearstobeanegativecorrelationbetweenthetwoindicators. InVietnam,therealsoappearstobeanegativecorrelationbetweenGDPandemploymentgrowth.ThefinalpanelshowstherelationbetweenGDPandemploymentgrowthforBurundiandChina.Inthesecountries,thereappearsasifthereisnocorrelationbetweenGDPandemploymentgrowth.InthecaseofBurundithisispartiallyexplainedbythefactthatnearly90percentofthelaborforceisself-employedorotherwiseinformal.Inthiscontext,employmentresponsestogrowth(ordecline)areunlikelytoregisterassignificant(thisissuewillbediscussedinmoredetaillaterinthissection).

10

Figure1RelationbetweenGDPandEmploymentGrowth

Okun’sLawTypically, growth slowdowns coincide with rising unemployment, and vice versa. This negativecorrelation between GDP growth and unemployment has been named “Okun’s law,” after theeconomist Arthur Okun who first documented it in the early 1960s. Part of the enduring appeal of

-10

-5

0

5

Gro

wth

Rat

e (%

)

1995 2000 2005 2010 2015Year

GDP Growth Employment Growth

SwedenGDP and Employment Growth (detrended)

-6

-4

-2

0

2

Gro

wth

Rat

e (%

)

1995 2000 2005 2010 2015Year

GDP Growth Employment Growth

United StatesGDP and Employment Growth (detrended)

-2

-1

0

1

2G

row

th R

ate

(%)

1995 2000 2005 2010 2015Year

GDP Growth Employment Growth

VietnamGDP and Employment Growth (detrended)

-10

-5

0

5

10

Gro

wth

Rat

e (%

)

1995 2000 2005 2010 2015Year

GDP Growth Employment Growth

BurundiGDP and Employment Growth (detrended)

-2

0

2

4

Gro

wth

Rat

e (%

)

1995 2000 2005 2010 2015Year

GDP Growth Employment Growth

ChinaGDP and Employment Growth (detrended)

11

Okun’s law is itssimplicity,since it involvestwo importantmacroeconomicvariables.Additionally, therelationshipappearstoenjoyempiricalsupport.

Ever since Okun (1962) tested it empirically, the relationship has been examined by a number ofeconomists including, inter alia, Smith (1975), Gordon (1984), Knoester (1986), Kaufman (1988),Prachowny (1993), andWeber (1995).Whilemost these studies examined its validity for the UnitedStates economy, the relationship has also been tested for other countries, e.g., Knoester (1986)Kaufman (1988), andMoosa (1997).Also, Ball, Leigh and Loungani (2016), show thatOkun’s Lawhasheldupwell forasetof20advancedeconomies.Althoughtheresultsgenerallysupporttheempiricalvalidityoftherelationshipinthesenseoffindingasignificantlynegativecoefficientoncyclicaloutput,theso-calledOkuncoefficient;nevertheless,theauthorsfindsignificantcross-countrydifferencesinthemagnitudeofthecoefficients,withlargercoefficientsfoundfortheUnitedStatesandCanadathanforEuropeandJapan.Theauthorsattributethedifferenceintheelasticityacrosscountriestodifferencesintherigidityofthelabormarkets.WeexpecttothatOkun’slawwillbeapplicableindevelopingcountries;however,wecanexpecttofindthat theOkun coefficientwill be significantly different across countries. Since labormarkets differ inregulations,structure,andeachcountryhasaverydifferenteconomicstructureintermsofwhattheyproduce and in the income level of their populations we also expect that there are significantdifferences in theway labormarkets react to changes in business cycles. For example, inmany low-income countries the large majority of the working age population is self-employed, either insubsistenceagricultureor some informalnon-farmactivity.While theseareaccounted foras ‘jobs’ intheemploymentdata,therealityisthattheywillnotcomeandgoinperiodsofeconomicgrowthanddecline; rather, the adjustment to business cycles is likely to be observed through earnings, workinghours,and/orshiftsinemploymentwithintheinformalsector.Inthissection,weexamine—amongotheremploymentoutcomes—howtheOkuncoefficientbehavesfor developing and developed countries. Since Okun’s Law has not been studied extensively in thecontextofdevelopingcountries(althoughseeHanusch,2012,foroneexample),andisalesswellknownconcept, much of this subsection explains how the Okun coefficient is estimated and how it differsbetween countries of different income levels. We then explain why some countries’ labor force aremoreresponsivetochangesinoutcomethanothers.Sinceweareparticularlyinterestedinhowprivatesectorcontributestojobcreation,weexploreemploymentintheformalsectoraswellasprivatesectordeterminantsoftheOkuncoefficient.Themotivationistwo-fold.First,developingcountriesaccountfora large,andgrowing, shareof theglobal labor force.Hence,understanding thedeterminantsof labormarketoutcomesinthesecountriesisimportant.Thereisampleevidencethatjobcreationcontributestoindividualandsocialwelfare,whereasunemploymentandjoblossareassociatedwithpersistentlossofincome,healthproblems,andbreakdownoffamilyandsocialcohesion(seetheWorldBank’sWorldDevelopmentReporton“Jobs” (2013)).A secondmotivation is toprobe the commonperception thatlabormarketoutcomes indeveloping countries reflectmostly structural factors rather than short-runcyclical fluctuations. Whether this perception is correct has important policy implications. If cyclicalfluctuationsaccountforasubstantialpartof labormarketdevelopments,macroeconomicstabilizationpolicies—suchascentralbankactions,countercyclicalfiscalpoliciesandprudentialpoliciestomitigatefinancialcrises—gaininimportancerelativetostructuralpolicies(e.g.improvingeducationandskillsofthelaborforce).

12

TherearedifferentversionsonhowtoestimateOkun’scoefficient.However,Okun’stworelationships,arise from the observation thatmore labor is typically required to producemore goods and serviceswithinaneconomy.More laborcancomethroughavarietyof forms,suchashavingemployeesworklongerhoursorhiringmoreworkers. To simplify theanalysis,Okunassumed that theunemploymentratecanserveasausefulsummaryoftheamountoflaborbeingusedintheeconomy.

Ahighrateofunemployment,Okunreasoned,wouldtypicallybeassociatedwithidleresources.Insuchacircumstance,onewouldexpecttheactualrateofoutputtobebelowitspotential.Averylowrateofunemployment would be associated with the reverse scenario. The original version of Okun’s lawcompares the gap between actual and potential (or natural) rate of unemployment and the gapbetweenactualoutputandpotentialoutput.Alternatively,weestimatethedifferenceversionofOkun’slawwhichhas theadvantage that there isnoneed todetermine thenaturaloutputandemploymentwhichanon-trivialexercise.ThisversionofOkun’sLawpositsarelationshipbetweenthechangesintheunemploymentrateandthegrowthrateofoutputandisdefinedasfollows:

∆"#,% = '∆(#,% + *#,% (1)where∆"#,%isthechangeintheunemploymentrateincountryiinyeart.∆(#,%isthechangeinoutputincountryiinyeart,and*#,%istheerrorterm.Equation (1)maybemodifiedby includingaconstant termand introducingsome limiteddynamics toobtain

∆"#,% = + + ,∆"#,%-. + '∆(#,% + /#,% (2)The additionof lagged change in unemployment is required to remove serial correlationwhichoftenarisesifthestaticequation(3)isused.2ThismodelisestimatedusingarollingOLSwithawindowof10years. A rolling regression estimates a specific relationship overmany different sample periods. Eachregression produces a set of estimated coefficients. If the relationship is stable over time, then theestimated coefficients should be relatively similar from one regression to the next. Variations in therelationshipwillappearassizablemovementsintheestimatedcoefficients.Each rolling regression is estimatedbasedon10 yearof datapoints. Thus, the first rolling regressionwouldestimatethevaluesof+and'fromequation(2),usingthesampleperiodforthefirst10yearsofavailabledataforcountryi.forexamplefrom1991to2000.Thesampleperiodisthenmovedforwardoneyear,andtheregressionisre-estimatedtoproduceasecondsetofestimatesof+and',usingdata

2Itmustbeemphasized,however,thatfollowingHendryandMizon(1978)and,morerecently,Mizon(1995)serialcorrelationisviewedasindicatingmisspecification,andsothemodelisrespecifiedinsteadofadoptingthefaultyalternativeofcorrectingfor serial correlation. In a paperwith theprovocative title ‘‘A simplemessage for autocorrelation correctors:Don’t,’’Mizon(1995)assertsthatcorrectionforautocorrelationmayyieldinconsistentestimates,thatitisgenerallyinvalid,andthatitcannotbejustifiedonthegroundsof‘‘robustifying’’estimationagainstthepresenceofresidualserialcorrelation.InastudyofOkun’slaw,Weber (1995,p.440)also found thestatic regression toexhibit first-order serial correlationwhichheconsidered tobe‘‘evidenceformisspecifieddynamics,’’andsoheintroduceddynamicequations.Inourcase,thelimiteddynamicsturnsouttobeadequatebecauseoftheuseofannualdata.

13

from1992to2001.Thisprocessisrepeateduntilthefinalestimatesaremadeusingtheentiresampleperiodavailableforcountryi.AsarobustnesscheckonourresultswealsoestimatetheoriginalversionofOkun’slawusingdifferentmethodstoestimatethenaturalrateofunemploymentandoutput.MoredetailsontheseestimationscanbefoundintheAnnex.ApplyingOkun’sLawtoTheRestofTheWorldTheideaistoexplorehowOkun’slawbehavesindifferentcountriesandtolearnhowthisrelationshipcan help us understand private sector job creation around theworld. Countrieswith higher absoluteOkun coefficients will be able to absorbmore workers when the economy is expanding, whereas ineconomicdownturns,firmsincountrieswithhighabsoluteOkuncoefficientareabletolayoffworkerstoadjusttothenewmacroeconomicconditionsinthemarket.Ontheotherhand,anOkuncoefficientcloser to zero would imply that the labor market is unresponsive to economic expansions orcontractions.

WeestimatetheOkuncoefficientusingunemploymentrateandGDPdatafromtheWorldBank’sWorldDevelopmentIndicators.Thesampleconsistsof176countriesfromdifferentincomegroupsandregionsfortheperiod1993until2015.Tomotivatethisresearch,Figure2shows,therelationbetweenchangesin the unemployment rate and real output growth for several countries, which belong to differentincome groups, regions, and economic structure. It is evident that Okun’s law is not stable acrosscountries.EventhoughOkun’s lawholds—forthemostpart—intheUnitedStatesacrosstime,Okun’scoefficientisnotconstantacrosscountries.

Figure 2 shows that the Sweden’s and United States’ (two high income countries) labormarkets aremoreresponsivetochangesinrealGDPthanEcuador’sandThailand’slabormarkets(twouppermiddleincomecountries).Honduras(lowermiddle-incomecountry)andHaiti(low-incomecountry)havelabormarkets thatarebarely responsivetoGDPgrowthnevertheless, theslopesof theregression linesarenegative.On theotherhand, labormarkets inCameroon (lowermiddle-incomecountry)andRwanda(low-incomecountry)arenon-responsivetoGDPgrowth,theregressionlineevenhasaslightlypositiveslope.Furthermore, intheUSandThailandforexample,theobservationsarewellrepresentedbythefittedlinewhereasfortheothercountriestheobservationsaremoredistanttothefittedline.

14

Figure2Okun’slaw:examplesofcountryvariation

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-5 0 5 10GDP Growth

SweedenOkun's Law - GDP Growth and Unemployment

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-5 0 5 10GDP Growth

United StatesOkun's Law - GDP Growth and Unemployment

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-10 -5 0 5 10GDP Growth

ThailandOkun's Law - GDP Growth and Unemployment

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-5 0 5 10GDP Growth

EcuadorOkun's Law - GDP Growth and Unemployment

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-5 0 5 10GDP Growth

HondurasOkun's Law - GDP Growth and Unemployment

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-5 0 5 10GDP Growth

CameroonOkun's Law - GDP Growth and Unemployment

15

Intheseexamples,apatterndevelopswherecountrieswithhigherlevelsofdevelopmenttendtohavemore responsive labor markets to changes in GDP growth compared to those in less developedcountries.Inthefollowingsection,weexploretherelationbetweenchangesintheunemploymentrateand GDP growth by country groups including income and economic structure. Figure 3 shows howOkun’slawperformsincountrieswithdifferentincomelevels.WecanclearlyseefromthegraphsthattheOkuncoefficientdecreases(absoluteterms)withthelevelofincome.Unemploymentratesinhigh-income countries improve on average by 0.21 for every percentage point increase in GDP.3 TheresponsivenessoftheunemploymentratestochangesinGDPdecreasesbyincomegroups.TheaverageOkuncoefficient forupper-middle incomecountries is -0.08, -0.03 for lowermiddle-incomecountries,and-0.005forlowerincomecountries.ThestatisticalsignificanceofthecoefficientsandtheR2fortheregressionsalsodecreasesasthelevelofincomedecrease.4

As suggested previously, these findings on income groupings are not at all unexpected. Inmost low-incomecountries,veryfewareunemployed–lackofsufficientsafetynetsmeanstheycannotaffordtobe.Butratheralargeshareofworkersareself-employed,eitherinsubsistenceorbasicagriculture,orinnon-farm self-employment. In these environments, adjustment to upturns and downturns in theeconomyarelikelytotakeplacethroughearningsratherthanemploymentperse.

On the other hand, Figure 3 highlights significant dispersion around the fitted lines. Thismeans thateven among countries within similar income groupings, there exist large differences in theresponsivenessof jobcreationtogrowth.Forexample,theOkuncoefficient is0.85inPoland,but just0.03 inHungary; it is 0.43 inBolivia and0.38 inChile but just 0.03 in Peru; similarly,while theOkuncoefficientisnearlyzeroinVietnamitis0.32inIndonesia.

3TheOkun’scoefficientfortheUnitedStatesis0.44butwithaconstanttermequalto1.2.ThisishigherthantheonereportedbyKnoteck(2007);however,thesampleperiodisdifferent.4AnnexFigure1showsthedistributionofOkun’scoefficientsbyincomegroups.

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-5 0 5 10GDP Growth

RwandaOkun's Law - GDP Growth and Unemployment

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-5 0 5 10GDP Growth

HaitiOkun's Law - GDP Growth and Unemployment

16

Figure3Okun’slawbyIncomeGroup

Okun’s law also varies greatly by the economic structure of each country (Figure 4). The lesssophisticated an economy is the less responsive is the labor force to changes in GDP. The negativerelationshipbetweenunemployment ratesandGDPgrowth is stronger in industrial, service intensive,tourismintensive,andhightechproductintensiveeconomies.Incontrast,economies,whichdependonnaturalresourcesandagriculture,havearegressionlinewithaslopethatisnearlyzero.5Again,someofthisislikelyrelatedtothestructureofemployment(relianceonself-employment),atleastinthecaseofagriculturally-dominatedeconomies, ormorebroadly in limited shareof formal employment in thesecountries.

5Todefinetheeconomicstructureofacountryweusethepercentageofnaturalresources,agriculture,industry,servicesasapercentageofGDPandtourismandhigh-techexportsaspercentageoftotalexports.Foreachofthesemeasureswecomputethe80thpercentileperyearandestimatethedifferencebetweeneachcountry’smeasureandthe80thpercentile.Wethenrankthesedifferences;thelargestvaluedeterminesthegrouptowhichthecountrybelongsin.

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-10 0 10 20GDP Growth

High Income CountriesOkun's Law - GDP Growth and Unemployment

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-20 -10 0 10 20GDP Growth

Upper-Middle Income CountriesOkun's Law - GDP Growth and Unemployment

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-20 -10 0 10 20GDP Growth

Lower-Middle Income CountriesOkun's Law - GDP Growth and Unemployment

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-20 -10 0 10 20GDP Growth

Low Income CountriesOkun's Law - GDP Growth and Unemployment

17

Figure4Okun’slawbyEconomicStructure

RelationshipBetweenOkun’sCoefficientandAggregateJobCreationThattherearedifferencesintheresponsivenessofprivatesectorjobcreationtoeconomicgrowthonlymattersifthisrelationshipisassociatedwithgreateroveralljobcreation.Inthissection,weexplorethecorrelation between employment growth and the Okun coefficients estimated for each country.Weexpect to find that a larger negativeOkun’s coefficient is associatedwithhigher employment growth

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-20 -10 0 10 20GDP Growth

Natural Resource Intensive CountriesOkun's Law - GDP Growth and Unemployment

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-20 -10 0 10 20GDP Growth

Agricultural CountriesOkun's Law - GDP Growth and Unemployment

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-20 -10 0 10 20GDP Growth

Industrial CountriesOkun's Law - GDP Growth and Unemployment

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-10 -5 0 5 10 15GDP Growth

Service Intensive CountriesOkun's Law - GDP Growth and Unemployment

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-10 0 10 20GDP Growth

Tourism Intensive CountriesOkun's Law - GDP Growth and Unemployment

-4

-2

0

2

4

Cha

nge

in U

nem

ploy

men

t Rat

e

-10 -5 0 5 10 15GDP Growth

High Tech CountriesOkun's Law - GDP Growth and Unemployment

18

sincethelaborforcemovesmorefreelywithGDPmovements.Weconfirmourhypothesiswithasetofregressionswherethedependentvariablesareemploymentgrowthrateintheformalsectoraswellasthe number of formal jobs added from year to year.We control for country and time fixed effects.Results are displayed in Table 1. We use the three different estimates for unemployment Okun asdescribedintheprevioussectionandintheAnnex.Okun’scoefficientdoesappeartobeassociatedwithhigher levels of formal employment growth. The results highlight the importance of this relation andsupportstheideathattheOkuncoefficientshouldbeanalyzedasarelevantemploymentoutcome.

Table1RelationbetweenOkun’sCoefficientandEmployment

DETERMINANTSOFEMPLOYMENTOUTCOMESThus farwehavedocumented importantdifferences inOkun’s coefficientsacross incomegroupsandacross different economic structures, and have shown that these differences matter from theperspectiveofoverall jobcreation.WhatdeterminestheresponsivenessofemploymenttochangesinGDP?Isthereacommonpatternacrosscountriesthatcanhelpusexplainemploymentoutcomesacrosscountries? We identify possible determinants of employment outcomes, including indicators of thelabor market structure, macroeconomic indicators that measure economic stability and investment,indicatorsoflaborregulation,indicatorsofqualityofinstitutions.TheseindicatorscomefromtheWorldDevelopment Indicators (WDI) database.Wealso include several indicators thatmeasure the easeofdoingbusiness in the country. These indicators arebasedon theWorldBankGroup’sDoingBusinessreport. For thesemeasures tobe comparable across countries and time, theyare recomputedas thedistancetothefrontier.Ameasureof100,isthebestorthefrontier,andameasureof0isthefurthestfromthefrontier.6

We expect that macroeconomic variables such as GDP per capita, GDP growth, gross fixed capitalformation, foreign direct investment, openness, and domestic credit to the private sector should allhave a positive associationwith employment growth (negative relationwith theOkun coefficient). Abetterbusiness environment (i.e., time to start abusiness and the globalmeasureofdoingbusiness)

6Forthissectiononthedeterminantsofemploymentoutcomes,weomitfromthesamplelowincomecountriesbecausefromtheprevioussectionwelearnedthereisnostatisticalrelationbetweenchangesinunemploymentratesandtheratesofgrowthofGDP.Includingthesecountriesinthesampleonlyaddsnoisetoourregressionswithoutprovidinganyadditionalbenefit.

VARIABLES

Estimation Method OLS GMM OLS GMM OLS GMM OLS GMM OLS GMM OLS GMM

1 2 3 4 5 6 7 8 9 10 11 12

okun (difference) -0.041 -0.093 -0.616 -0.950

[0.012]*** [0.029]*** [0.192]*** [0.407]**

okun (gap - detrend) -0.026 -0.057 -0.407 -0.858

[0.015]* [0.027]** [0.210]* [0.305]***

okun (gap - moving average) -0.020 -0.033 -0.421 -0.687

[0.017] [0.029] [0.178]** [0.246]***

formal employment growth (t-1) -0.385 -0.397 -0.397

[0.038]*** [0.037]*** [0.040]***

formal jobs added (t-1) -0.269 -0.268 -0.247

[0.044]*** [0.046]*** [0.043]***

Constant 0.020 0.046 0.023 0.037 0.042 0.058 10.844 13.850 10.899 13.422 11.027 13.611

[0.006]*** [0.043] [0.007]*** [0.016]** [0.039] [0.045] [0.122]*** [0.577]*** [0.123]*** [0.587]*** [0.136]*** [0.506]***

Observations 1,711 1,549 1,555 1,400 1,593 1,433 1,272 814 1,154 729 1,182 750

R-squared 0.148 0.144 0.143 0.820 0.822 0.824

Number of id 136 136 136 127 125 126

formal employment growth formal jobs added

Robust standard errors in brackets *** p<0.01, ** p<0.05, * p<0.1

19

and institutions (i.e., ruleof law)shouldhaveapositiveeffectonemploymentgrowthandanegativeeffectontheOkuncoefficient.Amongthemeasuresoflabormarketrigidity,weexpectamoreflexibleprobationtimefornewhireswillhaveapositiveeffectofemploymentgrowth(negativeontheOkuncoefficient)whereasahigherseverancepayforredundancydismissalwillnegativelyaffectemploymentgrowthandpositivelyaffecttheOkuncoefficient.Finally,asmentionedearlier,thebiggertheinformalsectorofacountry,measuredbythepercentageofthelaborforcethatisself-employed,theslowerweexpectemploymenttogrow(negativecoefficientonemploymentgrowth)aswellasaslowerreactionofthelaborforcetoreacttochangesinbusinesscycles(positivecoefficientontheOkunregression).

Table2showstheresultsofregressionswherethedependentvariablesaretheemploymentoutcomesof interest, includingemploymentgrowthrate,thechangeinthenumberof jobs intheformalsector,andtheOkuncoefficientsfoundforeachcountry.Theindependentvariablesaredescribedabove.Wemakeuseof two specifications. The first includes country andyear fixedeffects,whereas the secondspecificationonlyincludesyeareffects.Insteadofcountryfixedeffects,weuseadditionalcontrolsthatonlyvaryacrosscountriesbutdonochangeovertime,suchastheinitiallaborforceandtheinitiallevelofdevelopmentmeasuredbyacountry’sGDPpercapita.Weincludethissecondspecificationbecausewewant toexploit the variationacross countries andmuchof this variation is absorbedby the fixedeffectsusedinthefirstspecification.

ThemostrobustresultssuggestthatbetteremploymentoutcomesareassociatedwithhigherratesofGDPgrowth,higher levelsof economicdevelopment, and lower levelsof labor informality. TheOkuncoefficient isnegativeandstatistically significant in the first setof regressions, indicating thatamoreresponsive labor force results in higher employment growth, and a higher number of created jobs.AmongothercountrycharacteristicsthataresignificantlyassociatedwithhigherjobcreationarehigherFDIinflowsand,counterintuitively,amoreclosedeconomyasmeasuredbyitsopenness(thesumofacountry’s exports and imports as percentage of GDP) (see columns 2 and 5). Better institutions,measuredbytheruleoflawofacountry,isassociatedwithamoreresponsivelaborforcetochangesinbusinesscycles(columns3and6).

Surprisinglyour results show that theease tohire and fireworkersdoesnot appear to influence theemployment outcomes. Our proxies for the rigidity of the labor market, the maximum time for aprobationperiodduringthehiringprocessandtheseverancepayforredundancydismissalfor1yearofservice, are not statistically significant except for severance pay variable in column 6, which has theopposite sign thanexpected.The results suggest thathigher severancepay is associatedwithamoreresponsive labor force, even though a higher severance paymeans that it is more burdensome andexpensiveforfirmstofireworkers.Previousliteraturesuggestthattheresponsivenessoflabormarketscoulddependonregulationsgoverninglaborandproductmarkets.Forinstance,indiscussinghiringandfiringregulationsinMiddleEasternandNorthAfricancountries,Ahmed,Guillaume,andFurceri(2012)arguethatsuchregulationscandiscourage“firmsfromexpandingemploymentinresponsetofavorablechangesintheeconomicclimate.”Thatis,greateremploymentprotectioncandampenhiringandfiringasoutputfluctuates,reducingtheemploymentresponsiveness.Wefind littleassociationbetweentheOkuncoefficientandaggregatemeasuresoflabormarketflexibility.Ourresultsalsosuggestthatprivatesectorinvestmentandprivatesectoraccesstocreditdonotinfluenceemploymentoutcomes.

In this section, we learned that Okun’s law is applicable to other countries; however, the Okuncoefficientvariesgreatlyacrosscountries.Thelevelofincomeandtheeconomicstructureofacountry

20

determinehow responsive the labor force is to changes along thebusiness cycles. Because theOkuncoefficient iscorrelatedtohigheremploymentgrowth, it is importanttoanalyze itasanemploymentoutcome as well as a determinant of employment growth. Unfortunately, we find that neither jobcreationnortheelasticityofunemploymenttoGDPisdeterminedbyprivatesectorcharacteristicssuchas private sector investment and private sector access to credit but, mainly by macroeconomic andstructuraldeterminantssuchasGDPgrowthandtheleveloflaborinformalityinacountry.

Table2DeterminantsofEmploymentOutcomes

1 2 3 4 5 6

VARIABLES

employment

growth rate

change in

formal jobs

okun

(difference)

employment

growth rate

change in

formal jobs

okun

(difference)

okun (difference) -0.063 -0.915 -0.025 0.050

[0.038]* [0.403]** [0.022] [0.298]

polulation growth rate -0.040 0.355 -0.016 0.008 0.368 0.017

[0.020]* [0.336] [0.033] [0.006] [0.072]*** [0.023]

GDP growt rate 0.005 0.069 0.001 0.005 0.055 0.000

[0.001]*** [0.016]*** [0.002] [0.001]*** [0.014]*** [0.004]

log GDPpc -0.119 -2.014 -0.203 -0.022 0.308 0.039

[0.052]** [0.799]** [0.092]** [0.020] [0.204] [0.049]

informality rate -0.011 -0.090 -0.003 -0.018 -0.032 0.003

[0.003]*** [0.034]*** [0.003] [0.005]*** [0.006]*** [0.002]*

domestic credit to private sector (% of GDP) 0.000 0.008 0.001 0.000 -0.011 -0.000

[0.000] [0.008] [0.001] [0.000] [0.005] [0.001]

gross fixed capital formation, private sector (% of GDP) 0.002 -0.010 0.001 0.000 -0.010 0.002

[0.001] [0.016] [0.002] [0.001] [0.014] [0.002]

foreign direct investment, net inflows (% of GDP) 0.000 0.027 0.001 0.001 0.019 -0.001

[0.001] [0.016]* [0.002] [0.001] [0.015] [0.002]

openess -0.001 -0.019 -0.001 -0.000 -0.002 0.000

[0.000] [0.006]*** [0.001] [0.000] [0.002] [0.000]

time to start a business (distance to frontier) 0.000 0.006 0.000 0.000 0.000 0.000

[0.000] [0.004] [0.001] [0.000] [0.003] [0.001]

rule of law 0.021 -0.035 -0.102 -0.004 -0.306 -0.162

[0.031] [0.538] [0.049]** [0.023] [0.235] [0.062]***

average Doing Buiness global (distance to frontier) 0.001 0.010 0.004

[0.002] [0.016] [0.003]

log max length of probation period (months) -0.009 -0.257 0.035

[0.010] [0.123]** [0.024]

-0.002 -0.034 -0.010

[0.002] [0.019]* [0.004]**

log (employment at t0) 0.141 1.997 0.139

[0.138] [1.513] [0.307]

log (GDP at t0) 0.002 -0.291 0.105

[0.025] [0.178] [0.085]

log (GDPpc at t0) -0.020 -0.289 -0.131

[0.033] [0.274] [0.114]

log (labor force at t0) -0.135 -0.805 -0.236

[0.137] [1.541] [0.310]

Constant 1.464 32.210 1.672 0.193 0.902 -0.962

[0.460]*** [7.368]*** [0.843]** [0.143] [1.614] [0.409]**

Fixed effects c, y c, y c, y year year year

Observations 405 308 413 309 231 317

R-squared 0.417 0.822 0.660 0.145 0.779 0.163

Severance pay for redundancy dismissal (for a worker

with 1 year of tenure, in salary weeks)

Robust standard errors in brackets *** p<0.01, ** p<0.05, * p<0.1

21

3. INDUSTRY-LEVEL VIEW: ANALYSIS OF GROWTH AND EMPLOYMENT RELATIONSHIPS IN MANUFACTURING

Giventhesignificantdifferencesineconomicstructuresacrosscountries,someofthekeyfindingsfromthe previous section may be explained by the fact that the employment responses to growth ineconomies dominated by agriculture and informality cannot be compared with those where mostworkers are wage employed. Therefore, in this section we provide some initial descriptive analysiscarriedoutatthesectoral level,withinmanufacturing.Thishelpstocontrolbothforsectoralvariationandmitigatesissuesrelatedtoinformality.Ontheotherhand,byrestrictingtheanalysistowithinthesector,weareunabletocapturespillovereffects(e.g.multipliersarisingfromsupplychainlinkagesandotherindirecteffects).

ThedatasetusedfortheanalysisisdrawnfromtheIndustrialStatisticsDatabaseoftheUnitedNationsIndustrialDevelopmentOrganization(UNIDO).Itprovidesindustrialstatisticsreflectingcharacteristicsoftheunits engaged in a class of industrial activities,which are arrangedaccording to the InternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC)summarizedto2digits.Dataisreportedbycountry,yearandISICcode. It includes170countriesand52years(1963-2014),althoughcoveragevaries by country. Specifically, this analysis comprises an unbalanced panel that uses 23 classes ofindustrialactivitywithinthemanufacturingsector,definedintermsoftheISICRevision3.Itmakesuseofthreestatistics:numberofemployees,valueadded,andgrossfixedcapitalformation,anddrawsonpreviousanalysisfromGonzalez,Iacovone,andSubhash(2013).

RELATIONSHIPBETWEENGROWTHANDEMPLOYMENTAs a first step,we assess correlations between employment growth and growth in output and valueadded.AsshowninTable3,moderatecorrelationisevidentbetweenemploymentgrowthandoutputgrowthinthemanufacturingsector.Notsurprisingly,thiscorrelationincreaseswhenusingvalueaddedratherthansimpleoutputasthemeasureforsectoralgrowth.Wealsoassesswhethertheemploymentrelationshipwithoutputandvalueaddedisstronger inacontemporaneousmeasure(sameyear)or ifemploymentoutcomesarelaggedbyoneortwoyears.AsTable2shows,therelationshipdiminishestoinsignificance as soon aswe start to lag the impact, suggesting the contemporaneous effect ismoreappropriate for considering sectoral responses to output and value added growth. For the remaininganalysis inthissectionwe,therefore,focusonlyonthecontemporaneousrelationshipofemploymentgrowthtovalueaddedgrowth.

Table3Correlationcoefficientofmanufacturingsectoremploymentwithoutputandvalueadded

Contemporaneous One-yearlag Two-yearlagOutput 0.230 0.0034 0.0035ValueAdded 0.357 -0.0037 0.0051

22

Howdotheserelationshipsvaryby incomegroups?Table4shows interesting results.Moststriking istheextremelyhighcorrelationbetween jobgrowthandvalueaddedgrowth in low incomecountries.This result is striking not only because the correlation is so high (this might, in fact, be expected ineconomieswhicharelikelytoberelativelylowwageandlaborintensive)butbecausetherelationshipisso dramatically different from what we saw with the macro look in Section 2, where low-incomecountriesexhibitedvirtuallynocorrelationbetweenGDPgrowthandemploymentgrowth.Thefindinghere suggests that themacro result is indeed shaped largely by the dominance of self-employment,particularlyinagriculture,intheseeconomies.Whereasoncewefocusonasectorthatislargelyformal,thetightrelationshipbetweengrowthandjobcreationbecomesclear.TheotherinterestingfindinginTable4 is the fact thathigh incomecountriesalsoshowa relatively strongcorrelationbetweenvalueaddedgrowthandjobcreation.Onereasonforthismaybethatalthoughhigh incomecountrieshaverelatively limited direct employment creation through value added growth, they have much higherindirectjobcreationthroughsupplychainspillovers,whilelowandmiddleincomecountriesmayhavelessintegrateddomesticsupplychains,limitingpotentialemploymentmultipliersthroughgrowth.

While employment-output relationships are relatively strong in high income countries, they are, bycontrast,muchweaker in lowermiddle incomeand(especially)uppermiddle incomecountries.Whilewe have no data to indicate why this may be the case, it may be that the middle-income categorycontainsadiversesetofeconomicstructuresandcontexts.Itmayalsohavesomethingtodowiththenatureoftheindustrialtransformation(seeresultsbelowonhowsimilarpatternsemergewhenwegofromlowtomid-leveltohighlevelsofcapitalintensity).

Table4Correlationcoefficientofmanufacturingsectoremploymentandvalueaddedbyincomelevel

Incomelevel Valueadded(contemporaneous) Correlationcoefficient NLow 0.86685 4238Lowermiddle 0.23576 12,833Upper-middle 0.17430 16,214High 0.46981 36,020

Figure 5mirrors the descriptive analysis from Figure 1, illustrating annual trends in value added andemploymentgrowth (detrended)– in this casewithdata fromthemanufacturing sector.TheanalysiscomparesmanyofthesamecountriesasinFigure1,althoughmanufacturingsectordataisnotavailablefor all countries presented previously.While the relationship between value added and employmentgrowth in Sweden’s manufacturing sector matches almost exactly with the overall economy pictureshowninFigure1,thepatternsintheUSareremarkablydifferent(bothfromSwedenandfromtheUSoverall economy).Herewe see job creation trendingwith valueaddedgrowth in the two tails of theperiod,butdivergingsharply(andinexplicably)fromthemid-1990sthroughthemid-2000s.Asmightbeexpectedfromnaturalresources-basedeconomies,bothEcuadorandCameroonshow,formostoftheperiod, relatively modest shifts in employment growth despite much sharper swings in value addedgrowth. Finally, we look at manufacturing-oriented economies in latter stages of structuraltransformation.InthecaseofTurkey,jobcreationappearstorespondfluidlytochangesinvalueadded,withgreaterresponsivenessontheupsidethanonthedownsidesince2000.Formostoftheperiod(atleastsince1995),Chinasimilarlyshowsstrongelasticityofemploymenttomanufacturingvalueadded,

23

whichstandsincontrastthemuchweakerrelationshipbetweenjobsandgrowthintheoveralleconomy(Figure1).Figure5Manufacturingvalueaddedandemploymentgrowth–countryexamples

In Figure 6, we illustrate differences in the growth-employment relationship in manufacturingsubsectors;thetableisorganizedaccordingtothestandardcategorizationofISIC2-digitsectors.Wesee

-40

-20

0

20

Gro

wth

Rat

e (%

)

1995 2000 2005 2010 2015Year

Value Added Growth Employment Growth

SwedenValue Added and Employment Growth (detrended MVA)

-5

0

5

Gro

wth

Rat

e (%

)

1990 1995 2000 2005 2010Year

Value Added Growth Employment Growth

USAValue Added and Employment Growth (detrended MVA)

-100

-50

0

50

100

Gro

wth

Rat

e (%

)

1990 1995 2000 2005 2010Year

Value Added Growth Employment Growth

EcuadorValue Added and Employment Growth (detrended MVA)

-40

-20

0

20

40

Gro

wth

Rat

e (%

)

1980 1985 1990 1995 2000Year

Value Added Growth Employment Growth

CameroonValue Added and Employment Growth (detrended MVA)

-40

-20

0

20

40

Gro

wth

Rat

e (%

)

1995 2000 2005 2010 2015Year

Value Added Growth Employment Growth

TurkeyValue Added and Employment Growth (detrended MVA)

-40

-20

0

20

40

Gro

wth

Rat

e (%

)

1990 1995 2000 2005Year

Value Added Growth Employment Growth

ChinaValue Added and Employment Growth (detrended MVA)

24

verylargevariationsacrosssectors,whichmightbeexpected.Sectorswiththehighestcorrelationsare:furniture,woodproducts, andmachinery andequipment,which tend tobe relatively labor intensive,althoughalsosomewhatcapitalintensive.Bycontrast,sectorswherejobcreationisleastresponsivetogrowthpatterns includeoneof themost labor intensive sectors (wearing apparel) alongwith sectorsthattendtobecapitalintensive,including:radio,televisionandcommunicationequipment;petroleum,andchemicals.Figure6Correlationcoefficientofmanufacturingsectoremploymentandvalueaddedbyincomelevel

Onefactorthatmayexplaincountrydifferences,aswellassubsectoraldifferencesistherelativecapitalintensityofmanufacturingactivities.Wewouldassumethatmorecapital-intensiveproductionwouldbeassociatedwitha lowercorrelationbetweenvalueaddedgrowthand jobsgrowth.Table5 shows theresultsofcorrelationsusingtwodifferentmeasuresofcapitalintensity(capitaltolaborratioandcapitaltovalueaddedratio).Theresultsconfirmtheexpectednegativerelationshipbetweencapitalintensityandemploymentresponsivenesstogrowth,withbroadlysimilartrendsusingbothmeasures.However,itthedeclineincorrelationbetweenlowandhighlevelsofcapitalintensityisrelativelyminimal.

Table5Correlationcoefficientofmanufacturingemploymentandvalueaddedbycapitalintensity

We also check again whether it may be the case that some country groups or sector types (capitalintensityof sectors)mayadjustdifferently if, forexample, structural factorsdelaybutdonotprevent

corr_coeff _N corr_coeff _N

Low 0.457431 25107 Low 0.577714 26461Middle 0.354886 28447 Middle 0.466883 25972High 0.35204 16585 High 0.357352 17706

Capitaltolaborratio Capitaltovalueaddedratio

25

theadjustmentprocess. Figure7presents the correlationbetweenvalueaddedgrowth in time t andemploymentgrowth indifferentstages:contemporaneous(timet)aswellas laggedby1-3years.Thebasic correlation analysis does not give any hint of systematic adjustment delays. In fact, all countrygrouping and capital intensity grouping react in broadly the same way, with high contemporaneouscorrelation,rapiddegradationofthecorrelation(andturningnegative)timet+1,lowbutpositiveintimet+2,andthenfurtherfadingoftherelationship.

Figure7Correlationcoefficientofmanufacturingsectoremploymentgrowthintimet,t+1,t+2,andtimet+3tovalueaddedgrowthintimet

Finally,weaskwhethervolatilityinoutputmayaffectadjustment.Tothedegreetherearerestrictionsto the flexibility of hiring and firing or even just transaction costs involved in the process (whichwillalwaysbethecase),wecanimaginefirmswillbemorehesitanttohireinsituationswhereoutputtendstovarysignificantlyyear-on-year.Bycontrast, incountriesandsectorswheredemandandoutputarerelativelystable,firmsmaybeexpectedtoadjustquickly.Scatterplotsofthisrelationshipatthesectorand country level are shown in Figures 8 and 9, respectively. We find a strong relationship at thesectoral level. In general, manufacturing sectors that have greater volatility in output have weakeremploymentelasticitiestogrowth.However,noclearpatternisevidentatthecountrylevel(Figure9).Onepossiblereasonforthisisthatcountriesendupadopting(maybeimplicitly)aportfolioapproachtotheirindustryspecialization,thatbalanceshighvolatilitysectorswithothersthatarelowerinvolatility.

26

Figure8Relationshipbetweensectoraloutputvolatilityandemploymentelasticity(contemporaneous)

Figure9Relationshipbetweennationaloutputvolatilityandemploymentelasticity(contemporaneous)

HOWAREEMPLOYMENTELASTICITIESASSOCIATEDWITHOVERALLJOBCREATION?In this section, we ask if the employment elasticities we observed in the previous subsection areassociatedwithaggregatejobcreationoveraperiod.Thepointhere,aswasalsodiscussedinsection2,is that having a robust relationship between growth and job creation is only useful if it is associatedsomehow with aggregate job creation. As a first step to exploring this issue, in Figure 10 we mapcountries in the dataset based on their average annual job creation during periods of growth anddecline.Weseesignificantvariationacrosscountries,althoughthemostcommonsituation ispositivejobcreation ingrowthperiodsand jobshedding inperiodsofdecline.Herewehavecountries likeSri

27

Lanka, Libya, and to a lesser degree Jordan andMalaysia that experienced large jobs growth duringgrowth periods but also relatively sharp job loss during declines. On the other hand, oil producingcountries (UAE, Qatar) along with countries that have experienced strong structural transformation(China,Turkey,Honduras)experiencedalmost thesame(large)shiftboth inperiodsofoutputgrowthand decline. Transition countries like Ukraine and Croatia show relatively strong downsideresponsiveness during decline but much weaker upside responsiveness during growth (this patternholdsformanytransitioncountries).

Figure10Relationshipbetweensizeofemploymentmovementsduringperiodsofgrowthanddecline

Figure11Relationshipbetweenelasticityandaggregatejobcreation

28

Figure11looksatlongtermjobcreationanditsassociationwithjobselasticityofvalueaddedgrowthinthemanufacturingsector.Herewemeasureoverall jobcreation(growth%)overaperiodof20years7andtheaverageemploymentelasticitytovalueaddedgrowthoverthatperiod.Overallitshowsastrongcorrelation between employment elasticity to growth and overall job creation. Again, we see the oilproducers (UAE, Qatar, Kuwait, possibly Jordan) with very high overall job creation, despiteresponsivenessofemploymenttogrowth.Bangladesh,wheregrowthand jobcreationhasbeenrapidbut where informality still dominates, looks similar. Finally, somewhat disparate set of countries,includingThailand,Ethiopia,Mexico,andBolivia,showrelativelystrongresponsivenesstogrowthalongwithrobustjobcreationinthemanufacturingsectorFinally,we assess how the employment elasticity to growth varies by interacting the country incomelevel and capital intensity of the subsector. Figure 12 shows that for low capital intensive activities,elasticity of jobs to growth is highest in low-income countries (as might be expected given theircomparativeadvantageinlaborcosts.Similarly,theoppositeistrueforhighincomecountries.However,asonemovestomid-levelcapitalintensity,therelationshipchangesquicklywithlowandlower-middleincome countries facing much lower employment elasticity, while is upper middle-income and highincome countries experience a sharp jump in elasticity. Moving to highly capital intensive activities,however,weseeelasticityjumpalmostallincomegroups,includingthelowestincomegroup.Thisdipinmid-tier levelsofcapital intensityand jumpagainathigher levelsofcapital intensity suggests the jobcreationrelationshipweareseeingmaybelinkedtoprocessesofstructuraltransformationthatemergein the mid-levels of capital intensity. Yet at upper levels of capital intensity, activities may be ofsufficientlyhighervalueaddedorpartofsufficiently largedomesticclusterstomaketheemploymentimpactofgrowthmuchstronger.ThesefindingsaresupportedinFigure13,whichshowsthelong-termemployment impacts are actually highest (across just about all income groups) in the most capitalintensiveactivities.

7Wetakethelatest20yearsavailablefromeachcountry,whichresultsissignificantdifferencesintheperiodscoveredacrosscountries.

29

Figure12Elasticityofemploymenttogrowthbyincomeandcapitalintensitylevel

Figure13Longtermjobcreationbyincomelevelandcapitalintensity

30

4. FIRM-LEVEL VIEW: FIRM TYPES, REGULATION, AND JOB CREATION

In section 2, we explored the relation between changes in unemployment and GDP growth anddetermined that in high income countries there exists a statistically significant, negative associationbetween these two variables. The higher the absolute value of the Okun’s coefficient, the moreresponsive is the labor market to changes in GDP. We also documented that countries with highernegativevaluesofOkun’scoefficient,onaverage,alsocreatemorejobsduringtheperiodanalyzed.Insection 3, we tried to get a better perspective on these findings by considering the manufacturingsector, and found that the relationshipbetweengrowthand job creationwas strongest in the lowestandhighestincomecountries,andintheleastandmostcapitalintensivesectors.Finally,inthissection,asacomplementtothepreviousanalyses,weturntothefirm-leveldeterminantsofprivatesectorjobcreationandcarryoutthreeseparate,butrelatedanalyses.

Firmentryandexit–Schumpeterian“creativedestruction”– iscriticalforthecontinueddynamismofthemoderneconomy.Thoughevidencehaslinkedentrepreneurshipandeconomicgrowthindevelopedcountries,wehave scarceevidence that sucha relationshipexists indeveloping countries. New firmsenterthemarketandcreatenewjobs,whileotherunprofitablefirmsexitthemarketcontributingtojobdestruction(seee.g.Sutton(1997),PakesandEricson(1998),Geroski(1995)).Incumbentfirmsareinacontinuousprocessofadaptationinresponsetothedevelopmentofnewproductsandprocesses,thegrowth and decline in markets and changes in competitive forces (Davis and Haltiwanger (1999)).Marketconditionsandinstitutionalfactorsplayamajorroleinshapingthemagnitudeofjobflowsandtheircharacteristics(Davisetal.(1996)).

In the firstanalysis in thissection,using firm leveldata,westudyhowregulationsaffectemploymentgrowth across countries. We explore the links between the regulatory environment in which firmsoperate and job turnover by exploiting the observed industry-size variations through a difference-in-difference approach. Additionally, we explore firm characteristics that are associated with higheremployment growth rates in the short run and the long run and identify those firms that create thelargest impact on employment growth. We exploit cross-country differences to control for howmacroeconomic conditions, andbusinessand labor regulationsaffect job creation.Note thatbecausewedonothaveaccesstocross-countrypaneldataatthefirmlevel,inthissectionwemoveawayfromthe focus on responses to business cycles, althoughwe do incorporate theOkun coefficient into theassessment.

Tocarryouttheanalysis,weusesurveyscarriedoutbytheworld-wideEnterpriseSurveyprojectoftheWorldBankGroupbetween2006and2016.Thefirmssurveyedareforthemostpartregisteredfortaxpurposes;theyemployatleastfiveemployeesandarelocatedinurbanareas.Themanager/ownerofthefirmisinterviewedindetailaboutbasicfirmcharacteristics(age,legalstatus,numberofemployees,etc.),aswellasspecificinvestmentclimatequestions,forinstancewhetherthefirmhasexperiencedpoweroutages,whatthedelayshavebeenwhenitrequestedapublicservice,howdifficultitistodeal

31

withpublicofficers,etc.TheEnterpriseSurveysuseastratifiedsamplingmethodologythusindividualobservationsneedtobeproperlyweightedwhenmakinginferencesaboutthepopulation.8

THEREGULATORYENVIRONMENTANDJOBTURNOVERWiththesedata,weexploreindetailtheindustryandsizedimensionsofthejobflows,andrelatethemto institutional differences across countries. Two papers exploit job flows across industries withincountries to investigate the role of employment protection: Haltiwanger, Scarpetta, and Schweiger(2008)andMiccoandPages(2006).MiccoandPages(2006)exploitsindustrylevelgrossjobflowsdatafor manufacturing for 18 countries and uses a difference-in-difference. Haltiwanger, Scarpetta, andSchweiger(2008)useaharmonizedfirm-leveldatabasethatcoversallfirmswithatleastoneemployeefor both manufacturing and non-manufacturing sectors for 7 European countries, 5 Latin Americancountries, and 4 transition economies, and exploit country, industry and firm size variation in theiranalysis.Theanalysishereafter followsclosely theworkofHaltiwangeretal. (2006)butweapply themethodology to 164 countries ranging from high-income countries to low-income countries. As inHaltiwanger et al. (2006), we find that firm size is by far the most important factor accounting forvariation in the job flows across country, industry, and firm size classes. This suggests that exploitingdatabyfirmsizeisimportanttoprovidegreaterwithincountryvariationinjobflowsforourempiricalidentificationstrategy,butalsothatdistortionstojobflowsacrosscountriesmayverywellinteractwiththeflowandfirmsizerelationship.Specifically,weexplore the linksbetween the regulatoryenvironment inwhich firmsoperateand jobturnoverbyexploitingtheobservedindustry-sizevariationsthroughadifference-in-differenceapproach(see Rajan and Zingales (1998)). The test is constructed as follows: we identify an industry-sizepropensity for job reallocation from thePolishdata. Poland is the countrywith the largest (negative)OkuncoefficientascalculatedintheanalysisinSection1,meaningthatPolandhasalabormarketthatadjustseasilytomovementsinbusinesscycles.9Undertheassumptionthatregulationsinthelaborandgoods markets in Poland are among the least restrictive in our sample, variation in job reallocationacrossindustry-sizecells inPolandshouldproxyforthetechnologicalandmarketdrivendifferencesinjob reallocation in the absence of policy induced adjustment costs. Under the additional assumptionthatthesetechnologicalandmarketdrivendifferencesinthedemandforjobreallocationcarryovertoothercountries,weassesswhetherindustry-sizecellsthathaveagreaterpropensityforjobreallocationaredisproportionallyaffectedbyregulationsthatraiseadjustmentcosts.Thiswouldimplythat,ceterisparibus, industry-size cells with more volatile idiosyncratic profit shocks and more frequent factoradjustments should be more strongly affected by regulations raising adjustment costs than thoseindustry-size cells with less volatile idiosyncratic profit shocks and less frequent adjustment. Theadvantageofthisapproachcomparedtostandardcross-country/cross-industryempiricalstudiesisthatwe exploit within country differences between industry-size cells based on the interaction between

8Understratifiedrandomsamplingunweightedestimatesarebiasedunlesssamplesizesareproportionaltothesizeofeachstratum.With stratification, the probability of selection of each unit is, in general, not the same. Consequently, individualobservationsmustbeweightedbytheinverseoftheirprobabilityofselection(probabilityweightsorpwinStata.)9Haltiwangeretal.(2006)usetheUnitedStatesasthebenchmarkcountryfortheirsample.SincetheenterpriseSurveysmostlycoversdevelopingcountriesanddoesnotincludetheUnitedStateswehadtochooseandalternativebenchmark.AsarobustnesscheckweuseChile(bestscoreinEconomicFreedomIndexinoursample)asabenchmarkandresultsareconsistenttothosereported.Theseresultsareavailablefromtheauthor.

32

countryandindustry-sizecharacteristics.Thus,wecanalsocontrolforcountryandindustry-sizeeffects,therebyminimizingproblemsofomittedvariablebiasandothermisspecifications.Thecoremodelspecificationusedinourempiricalanalysiscanbesummarizedasfollows:

012345#6 = '7 + ,5#85# + ,68696:. + '.(<32012345#×>6)@×A

5#:. + /5#6 (8)whereDsiareindustryxsizesi(si=1,….,IxS)dummies,Dcarecountryc(c=1,….,C)dummies,PolJFlowsiis Poland’s job flow variable in size class s and industry i, and /is the iid error term. Controlling forcountry effects sweeps out any country-specific variation, controlling for industry-size effects sweepsoutthelargecommonfactorsassociatedwithindustryandsize,andthekeyinteractiontermbetweenPoland’s flow in the industry-size class and the country regulation allows us to identify how themeasured regulatory environment affects the variation across industry-size classes within countries.Poland’s job flow is used to quantify the propensity for the industry-size class to reallocate and, asdiscussed, reflects the fundamental driving forces underlying job reallocation across industry-sizeclasses.Inwhatfollows,themeasureofjobflowsusedintheempiricalanalysisisthesumofjobcreationand

jobdestructionrates(sum).JobcreationrateisdefinedasC3D5#6% =E∆FGHIJH∈LM

7.O(FGHIJEFGHI,JPQ)andjobdestruction

rate asRST5#6% =E∆FGHIJH∈LM

7.O(FGHIJEFGHI,JPQ) , where, i represents industry, s represents size class, c represents

country, t represents time and E denotes employment. Capital letters S and C refer to a set of sizeclasses or countries, respectively. The symbol ∆ denotes the first-difference operator, ∆U% = U% −U%-..10 In addition to the core specification, we consider some closely related specifications. As arobustnesscheck,weestimateanaugmentedmodelthatalsoconsidersbusinesssectorregulations.Weconsider threemeasuresof the regulatoryenvironment foreachcountry:oneacomposite index,oneaimedatmeasuringlaborregulations,andathirdthatmeasuresbusinessregulations.ThesourceofthesemeasuresistheEconomicFreedomoftheWorld(EFW)measurescreatedbytheFrasierInstitute.TheindexpublishedinEFW,measuresthedegreetowhichthepoliciesandinstitutionsofcountriesaresupportiveofeconomicfreedom.Studieshavefoundthatcountrieswithinstitutionsandpoliciesmoreconsistentwitheconomicfreedomhavehigher investmentrates,morerapideconomicgrowth,higherincomelevels,andamorerapidreductioninpovertyrates.Eachoftheseindexesismeasureonascaleof0to10,with0beingthemostrestrictiveenvironment.Theaverageofthecompositeindexishighestfor high-income countries (7.18) and lowest for low-income countries (6.11)whereas the average forlower and upper middle-income countries falls in between these two groups. The monotonicallyincreasingaveragesamongincomegroupsarealsofoundforlaborandbusinessregulations.Previous research (see, e.g., Caballero et al. (2004), Heckman and Pages (2004)) suggests thatregulations affect outcomes in the economy in the degree in which those regulations are enforced.Availableindicatorssuggestasignificantvariationintherulesoflawandthedegreeofenforcementoflawsandregulationsinoursample.NotonlyaresomefirmsandjobsnotregisteredinLatinAmericabut10AsinHaltiwangeretal.(2006),wetakeaveragesofposandnegamongfirms,andthencalculatesum.

33

alsoregisteredfirmsmayalsonotfullycomplywiththeexistingrulesandregulationswherecorruptionisrampant—suchasinsomeAfricancountriesinoursample.Asanindicationofthedifferentdegreeofenforcementoflawsandregulations,weconsidertheruleoflawindicatorfromtheWorldBank’sWorldGovernanceIndicators.Theindicatorrangesbetween-2.5and+2.5,with+2.5beingthebestperformer.The indicator shows high-income countries have the highest compliance with laws and regulations(averageof0.68),followedbyuppermiddle-incomecountries(averageof-0.37),thenbylowermiddle-incomecountries(averageof-0.67),andfinallybylow-incomecountries(average-0.83).To control for possibly differing degrees of enforcement of laws and regulations we adjust ourregulatoryvariablesasfollows:

>6,WXY =Z[\]^_\W`Ea.O

O×>6 (9)

Table6presentstheempiricalresultsoftheestimationofourcorespecification(8)andvariousvariations.Recallthatinequation(8)wespecifyadifference-in-differenceanalysisthatidentifiestheimpactofregulationsviatheinteractioneffectPoland’sjobflowsintheindustry×sizeclasswiththecountry-specificregulation.TheestimatedcoefficientsontheinteractionbetweenPoland’sjobflowandthedifferentregulationmeasures(Table6)isstronglysignificantoverall,andineachoftheincome-groupswhenweallowthecoefficientoftheinteractiontovary.Consistentwiththeworkinghypothesis,morevolatileindustriesandfirmsizeclassesareimpactedmoreincountrieswithmorestringentregulations,includingbusinessandlaborregulations.Theresultsarerobusttothedifferentregulationvariablesadjustedforthedegreeofenforcementineachcountry.

Table6JobFlowsandtheRoleofRegulations(DifferenceinDifferenceAnalysis)

DeterminantsofEmploymentGrowthatTheFirmLevelInthissubsection,weexplorespecificfirmcharacteristicsthatareassociatedwithhigheremploymentgrowth rates in the short run and the long run.Wealso exploit cross-countrydifferences todiscoverhowmacroeconomicconditions,andbusinessandlaborregulationsaffectjobcreation.

FollowingLaPortaetal. (1999),Boteroetal. (2004),andFoxandOviedo (2013)weexamine the linkbetween regulation and job growthusing a reduced-formmodel.Weuse robustOLS to estimate the

VARIABLES(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

jflow_POL x regulation 0.064 0.037 0.039 0.029 0.043 0.029[0.021]*** [0.011]*** [0.018]** [0.014]** [0.014]*** [0.015]*

low income x jflow_POL x regulation -0.019 -0.016 -0.019 -0.042 -0.037 -0.044[0.008]** [0.008]** [0.008]** [0.019]** [0.018]** [0.020]**

lower-middle income x jflow_POL x regulation -0.014 -0.011 -0.013 -0.033 -0.030 -0.034[0.006]** [0.006]* [0.007]* [0.016]** [0.017]* [0.017]*

upperr-middle income x jflow_POL x regulation -0.007 -0.003 -0.006 -0.012 -0.005 -0.010[0.006] [0.006] [0.007] [0.013] [0.013] [0.015]

Constant -0.026 0.083 0.019 0.079 0.019 0.081 0.050 0.081 0.043 0.078 0.053 0.079[0.051] [0.040]** [0.043] [0.040]** [0.046] [0.040]** [0.041] [0.040]** [0.041] [0.040]* [0.041] [0.040]**

Observations 8,118 8,118 8,118 8,118 8,118 8,118 8,118 8,118 8,118 8,118 8,118 8,118R-squared 0.229 0.229 0.229 0.228 0.228 0.228 0.228 0.228 0.229 0.228 0.228 0.228

regulation = adj. business index

Robust standard errors in brackets *** p<0.01, ** p<0.05, * p<0.1. The depedent variable in all regressions is jflowsic. All regressions include sector x size effects and country effects.

regulation = EF index

regulation = labor index

regulation = business index

regulation = adj. EF index

regulation = adj. labor index

34

relationship between overall business and labor regulation and job growth, controlling for firm andcountrycharacteristics.Weestimatethefollowingequation:

SbC_Td#% = +6 + +e + +% + 'fghS#%i + 'jTS# + 'k4RSdDℎgC# + mknℎSd#

WhereSbC_Td#% isfirm i’sjob(employment)growth;Sizeistheinitialsizeofthefirm,dividedintosixemployment categories (5 or less, 6–10, 11–20, 21–50, 51–100 andover 100); Age is oneof five agecategories (less than5,5–9,10–19,20–39and40andolder);Ownershipdenoteswhether the firm isprivate-domestic,private-foreign,public,orother.Other representsother firmcharacteristics,namelyexporter status, access to credit, technologyuse, andhumancapital.Wehave two specifications, thefirst includescountry-yearandindustryfixedeffects.Thesecond, includes,country, industry,andyeareffects. We include these effects to minimize the potential for omitted variable bias. The secondspecificationallowsustocontrolforcountry-yearvaryingdeterminantssuchasGDPgrowth,therollingOkuncoefficient,andlaborandbusinessenvironmentmeasuredbyEFWindicatorsdescribedabove.Weclustertheerrorsatthecountrylevel.Finally,weestimatedequation(1)usingfirmweightsprovidedbythesurveys.

ThedependentvariableSbC_Td#%ismeasuredeitherasannualaveragejobgrowthoverthelast3years(shortterm)orastheaverageannualgrowthratefromstartuptocurrentsize(longrun)forfirmsthatareatleast5yearsold.Inbothcases,wehaveatruncationproblem.Thesamplecontainsdataonlivefirms only; therefore, changes in employment over the period considered for job growth excludechangescomingfromfirmexit.Obviously,the longertheperiodconsideredfor jobgrowth,theworsethetruncationproblemislikelytobe.Thus,forthelong-runjobgrowthvariable,thetruncationproblemmaybesevere.Forexample, ifemploymentismorevolatileincountrieswithalowEFWscorethanincountrieswithahighEFWscore (there ismorerapidexpansionbutalsomanymore firmdeaths),wewouldobserveahigherexpansionincountrieswithalowEFWscoreeventhoughlong-runjobgrowthisactually higher in countries with a high EFW score because there are fewer deaths. Thus, theseestimatesshouldberegardedwithcare.

Table7presentstheOLSresultsforthetwodependentvariables.Theresultsfortheshort-termgrowthconfirmthatsmallerandyoungerfirmsgrowfaster,sincealloursizeandagecoefficientsarenegativeand significant (the excluded categories are the smallest and youngest firms, respectively). Thedifferencesaregreaterbetweenageclassesthanbetweensizeclasses.Annualjobgrowthseemstobecloselyrelatedwithexporterstatusandtechnologyuse,asmeasuredbytheuseofe-mail.Also, firmswithaccesstocreditandthatinvestinhumancapitaltendtohavehighergrowthrates.Theseresults—exportersandtechnologyusers that invest inhumancapitalgrowfaster—are likelydue inpart tothefactthatthesetypesoffirmsaremoreproductive.Regardingthecountryenvironmentcontrols,wefindthat a more hassle-free labor environment is associated with higher employment growth, whereascountries withmore hassle-free business regulations are related to lower employment growth. Eventhough institutions has a positive coefficient it is not statistically significant. We also control formacroeconomicvariablesincludingGDPgrowth,FDI,grosscapitalformation,andofparticularinterest,theOkun’scoefficientfoundinsection2.ResultsshowthathigherGDPgrowth,investment(i.e.,grossfixed capital formation), and the level of development are all associated with higher levels ofemployment growth in the short run. FDI is found to have a negative and significant coefficient;however,itisnotstatisticallysignificant(columns2and3).Amoreresponsivelaborforce,givenbythe

35

Okuncoefficient, is associatedwithhigheremploymentgrowthat the levelof the firm;however, theeffectisnotstatisticallysignificant.

Results fromusing long-termgrowth as thedependent variable differ and reveal further elements ofemployment dynamics. Micro-firms still grow faster (from a small base), but there are also largedifferences in growth between small, medium, and large firms. However, we find that state-ownedenterprises have a negative effect on employment growth and the effect is statistically significant. Infact,allthefirmcharacteristicsincludedintheregressionshowpositiveandsignificantcoefficients;thisincludes whether a firm provides training to their workers, suggesting that firms that invest in theirworkershavehigherjobgrowthinthelongerterm.Overthelong-term,themacrovariables, includingthe Okun coefficient, largely lose their statistical significance, except for openness which becomesstatisticallysignificant.Inthelong-run,itappearsasahassle-freelaborenvironmentdeteremploymentgrowth.

36

Table7DeterminantsEmploymentGrowthbyFirms

1 2 3 4

VARIABLES

exporter 0.013 0.008 0.006 0.009

[0.004]*** [0.004]** [0.004] [0.007]

credit 0.009 0.020 0.005 0.016

[0.003]*** [0.002]*** [0.002]** [0.004]***

training 0.006 0.024 0.015 0.010

[0.004] [0.003]*** [0.003]*** [0.004]**

email 0.006 0.015 0.011 0.008

[0.004] [0.002]*** [0.003]*** [0.004]**

size6-10 -0.040 -0.038 -0.021 -0.009

[0.006]*** [0.009]*** [0.003]*** [0.005]*

size11-20 -0.047 -0.043 -0.010 0.000

[0.006]*** [0.013]*** [0.004]*** [0.006]

size20-50 -0.051 -0.046 0.000 0.004

[0.006]*** [0.011]*** [0.003] [0.006]

size50-100 -0.059 -0.052 0.014 0.019

[0.006]*** [0.015]*** [0.005]*** [0.008]**

size100+ -0.061 -0.070 0.014 0.031

[0.007]*** [0.006]*** [0.005]*** [0.009]***

age5-9 -0.042 -0.078 0.091

[0.011]*** [0.008]*** [0.004]***

age10-19 -0.058 -0.111 -0.050 -0.050

[0.011]*** [0.008]*** [0.001]*** [0.006]***

age20-39 -0.065 -0.127 -0.077 -0.077

[0.011]*** [0.008]*** [0.001]*** [0.006]***

age40+ -0.075 -0.140 -0.106 -0.111

[0.013]*** [0.009]*** [0.001]*** [0.007]***

majorityforeign -0.004 -0.002 0.000 -0.000

[0.005] [0.005] [0.002] [0.009]

majoritystate 0.003 -0.054 -0.021 -0.049

[0.012] [0.013]*** [0.004]*** [0.019]***

majorityother 0.025 -0.010 -0.004 -0.018

[0.019] [0.011] [0.002] [0.009]**

okun_chg_iunemp_rt -0.016 -0.004

[0.025] [0.028]

polulationgrowthrate -0.079 -0.038

[0.024]*** [0.024]

GDPgrowtrate 0.004 0.002

[0.001]*** [0.001]

logGDPpc -0.022 -0.025

[0.066] [0.067]

informalityrate -0.005 -0.001

[0.002]** [0.001]

domesticcredittoprivatesector(%ofGDP) 0.000 -0.000

[0.001] [0.000]

grossfixedcapitalformation,privatesector(%ofGDP) 0.003 0.000

[0.001]** [0.001]

foreigndirectinvestment,netinflows(%ofGDP) -0.004 -0.001

[0.001]*** [0.002]

openess 0.000 0.001

[0.000] [0.000]**

laborregulations 0.030 -0.017

[0.010]*** [0.010]*

businessregulations -0.044 -0.002

[0.013]*** [0.014]

institutions 0.261 0.070

[0.210] [0.073]

Constant 0.238 0.568 0.036 0.471

[0.031]*** [0.559] [0.015]** [0.575]

Observations 92,450 27,890 79,222 23,383

R-squared 0.083 0.104 0.189 0.174

short-runemploymentgrowth long-runemploymentgrowth

Robuststandarderrorsinbrackets***p<0.01,**p<0.05,*p<0.1

37

Determinantsofemploymentgrowthinhigh-impactfirmsIn this final subsection, we focus the analysis on job creation from a small subset of firms – thosereferredtoas‘high-impact’firms.Whilehigh-impactfirmsareseentoplayanespeciallyimportantroleintheprocessof jobcreationovertimecomparedwitheithertheplantsof largeexistingfirmsorthemajority of very small startups that tend not to grow, to date there has been little research on thecharacteristicsofhigh-impactfirms,andevenlessontheirmicroandmacrodeterminants(Acs,Parsons,and Tracy, 2008) With the framework that has been used throughout this study we analyze whatdeterminestheavailabilityofhigh-impactfirmsacrosscountries.Wedefineahigh-impactfirmasafirmthathashadimpressivejobcreationinthethreeyearsleadingupto the survey year. Specifically, a high-impact firm, is a firm that reaches an employment growthquantifiergreaterthan15.Theemploymentgrowthquantifier(EGQ)istheproductoftheabsoluteandpercentchange inemploymentoverathree-yearperiod.TheEGQisusedtomitigatetheunfavorableimpactofmeasuringemploymentchangesolely ineitherpercentorabsolute terms, since the formerfavors small companies and the latter large businesses. A firmwith 10 employees in the initial yearwould have had to increase its number of workers to 23 in order to reach an EGQ higher than 15whereasafirmwith100initialemployeeswouldneedtoincreaseitslaborforceby40workersinorderto reach a EGQof 15. Annex Table 1 shows the number of firms by country-year in our sample thatobtainedanEGQof15,20and50.Figure14showstheimportanceofhighimpactfirmsinaneconomy.Eventhoughthesefirmsrepresentaverysmallshareofthenumberoffirmsintheeconomy,theycreatethemajorityofjobs.FirmswithanEGQof15orhigherrepresent,onaverage,5%ofthefirmsinthesamplewhereastheycreate44%ofpositive jobs in theeconomy.Clearly thesesharesvarybycountryandyear, forexampletheshareoffirmsinoursamplewhichhaveanEGQhigherthan15,variesbetween0and20%whereastheshareofpositivejobscreatedbyfirmswithanEGQhigherthan15rangesfrom0to91%.Thiswidevariationofthenumberofhighperformingfirmsacrosscountriesleadsustoquestionwhy.Whatfactorsdeterminethatacountryproducesahighnumberofhighperformingfirms?Table8showsresultsofnegativebinomialregressionswherethedependentvariableisthenumberoffirmswithEGQgreater than15,20and50,and the independentvariablesarecountry levelvariablesthan aim at exploring the determinants of a proper environment for high-impact firms to flourish.Resultsshowthatahighlycompetitiveenvironment,proxiedbythenumberoffirms,economicgrowth,andaresponsivelabormarketproxiedbytheOkuncoefficientarethethreemostrobustfactorsforanenvironmentthatproduceshighimpactfirms.Investmentbytheprivatesector,openness,flexiblelaborandbusinessregulations,andruleoflawthatprotectsprivaterightsareallpositivelyrelatedwithhighernumber of high impact firms; however, their effect is no statistically significant. Counterintuitively,foreign direct investment and access to credit have a negative effect; however, it is not statisticallysignificant.

38

Figure14HighImpactFirms-AverageShareofFirmsandAverageShareofPositiveJobs

Table8DeterminantsofHighImpactFirms

5.4

44.4

4.3

40.9

2.0

30.3

0

10

20

30

40

share of firms HIMPF 15 share of postive jobs HIMPF 15share of firms HIMPF 20 share of postive jobs HIMPF 20share of firms HIMPF 50 share of postive jobs HIMPF 50

(1) (2) (3)

VARIABLES himpcf_15 himpcf_20 himpcf_50

log number of firms 1.007 1.058 1.079

[0.061]*** [0.067]*** [0.084]***

okun (difference) -0.645 -0.580 -1.325

[0.172]*** [0.197]*** [0.284]***

polulation growth rate -0.031 -0.053 -0.047

[0.023] [0.027]** [0.039]

GDP growt rate 0.025 0.024 0.021

[0.008]*** [0.010]** [0.014]

log GDPpc 0.139 0.051 0.157

[0.106] [0.119] [0.181]

domestic credit to private sector (% of GDP) -0.002 -0.002 -0.006

[0.003] [0.004] [0.005]

gross fixed capital formation, private sector (% of GDP) 0.022 0.022 0.015

[0.013]* [0.016] [0.018]

foreign direct investment, net inflows (% of GDP) -0.039 -0.055 -0.050

[0.026] [0.033]* [0.034]

openess 0.000 0.001 0.002

[0.003] [0.004] [0.005]

labor regulations 0.010 0.006 0.009

[0.048] [0.053] [0.085]

business regulations 0.186 0.152 0.190

[0.082]** [0.094] [0.133]

institutions 0.270 0.156 0.200

[0.157]* [0.178] [0.242]

Observations 90 90 90

Robust standard errors in brackets *** p<0.01, ** p<0.05, * p<0.1

39

5. CONCLUSIONS AND FUTURE RESEARCH

Thispaperisintendedasthefirstinaseriesofassessmentsthataimtounderstandbetterthepatternsanddeterminantsofprivatesector jobcreation. Inthisfirstpaper,wetookanexploratory lookattherelationshipbetweengrowthandjobsbyanalyzingpatternsofjobcreationanddestructioninbusinesscyclesacrossalargenumberofglobaleconomies.Theobjectiveoftheexercisewasmodest–toassessthepatternsofthegrowth-jobsrelationshipbycountries,incomegroups,economicstructures,sectors,and firm types.Three separate,but relatedanalyseswerecarriedout, lookingat themacro, sectoral,andfirmlevels.

Fromthedescriptiveanalysis,wefoundthatwhilegrowth is,ofcourse, fundamental for jobcreation,verydifferent labormarketresponsestogrowthemergedacrosscountriesand incomegroups.Takingonly themacro view,we saw that low-income countries, particularlywhendominatedby agriculturaland other forms of self-employment, appear to have job creation patterns that are unresponsive togrowth. However, the industry-level view, which largely controls for self-employment, suggests thatemployment in low-incomecountriesdoes in fact respondelastically togrowth,at least in the formalsector.These findingshighlight the importantroleofstructural transformation inshapingthegrowth-jobsrelationship.Morebroadly,thesignificantdifferencesinthepatternsofjobcreationacrossincomegroupsandeconomicstructurescallsformattersforconsideringhowandwhenweshouldexpectjobsgrowthatdifferentstagesoftheeconomiccycle.

An important focus of this paper was to test whether labor market responsiveness to growth anddecline (asmeasuredbytheOkuncoefficientorbystandardelasticities) isassociatedwithoverall jobcreation. Here we found that economies where formal sector job creation was more responsive togrowth cycles (i.e. they create jobs in linewith growth and shed jobs in linewith decline) generatedmorejobsoverallthaneconomieswithlessresponsivepatternsofjobcreation.Thiswasalsofoundtobetrueatthefirmlevel–firmswhichcreateandshedjobsmorefluidlycreatemorejobsoverall.Atthesame time, many countries (and presumably, many firms) expanded employment significantly inaggregatedespitearelatively inelasticresponsetobusinesscycles;similarly,somecountriesexhibitedweakoveralljobcreationdespiteahighlyresponsivelabormarket.Thisappearstobedrivenmainlybystructural factors – less responsive, high job-creating countries consist mainly of: i) high income,extractive (mainly oil producing) economies; and ii) high-growth, early-stage transformation (highinformality) economies.By contrasthighly responsive, low job-creating countriesappear tobemainlytransitioneconomiesinCentralandEasternEurope,wherethelong-runanalysispicksupmajordeclinesinemploymentfromthe1990s.

Another important finding in the paper relates to job creation in technology and capital intensiveenvironments.The results fromboth themacroanalysis (groupingof ‘high technology’countries)andthesectoralanalysisindicatethat,farfrombeingasimplesubstitutefromjobs,economiesandsectorsthat are capital intensive tend to be have a responsive growth-jobs relationship. In fact, the sectoralanalysis in manufacturing found highly capital intensive sectors were associated with higheremployment elasticity to growth than medium or low capital intensive sectors across most incomelevels; and for all income levels, highly capital intensive sectors generated faster overall employment

40

growth than less capital intensive sectors. This suggests significant complementarity between capitalandlabor,afindingwhichcertainlyshouldbeprobedmoredeeply.

Finally, while the focus of the paperwas to take a broad look at economic growth and job creationrelationships across a wide range of countries, we also carried out a broad assessment on thedeterminants of job creation at the firm level. Any analysis on the determinants of a fundamentaleconomicoutcomelikejobcreationfaceshugechallengesofendogeneity.Nevertheless,wewereabletoidentifysomefactorsthatareatleastcorrelatedwithjobcreationoutcomes.First,intermsoffirmtypes,theanalysissupportsmuchoftherecentresearchshowingthatsmalland(especially)youngfirmsareassociatedwithgreaterjobcreation.Morebroadly,byfarthemostimportantdeterminantsofjobcreationatthefirmlevelareeitherfirm-specificcharacteristics(includingdemographics,ownership,andfirmbehaviorslikeuseoftechnologyandtraining)ormacro-economicandstructuralfactors(includinggrowth levels, investment, and labor market responsiveness). Beyond these (and another, arguablystructural, factor – institutional quality), access to formal bank financing, the business regulatoryenvironment, and some limited aspects of labor regulations, appear to be relevant. Indeed, whilebusiness regulatory issues do not appear to be among the most fundamental factors shaping jobcreation outcomes, they may be important at the margin – the evidence presented in this paperidentifiesbusinessregulationasparticularlysignificantinindustrysectorsandfirmsizeclassesthatfacegreater volatility in terms of profits and factor adjustments. Together, these findings point to theimportance of market rigidities that may constrain the ability of firms to respond to growthopportunities. Finally, the analysis of high impact firms suggests the importance of the competitionenvironmentintheemergenceofhighlyproductiveandjob-creatingfirms.

Whiletheanalysisondeterminantsoffirm-leveljobcreationpresentedhereisonlyataveryhighlevel,the above findings may help to frame a potential policy agenda for job creation, including:macro-economic fundamentals, responsive labor markets, access to finance, competition, and a facilitativebusiness regulatoryenvironment.Thesearenotsurprising,butnevertheless frameasetof issues thatcouldbeexploredinfuturepapersinthisseries.

Indeed, bearing in mind the exploratory nature of the paper, the results are promising and suggestfertilegroundforfurtheranalysis.Futurepapersintheseriesmayconsidertwomainavenuestodeepenthisinitialanalysisoneconomywidejobcreation:first,exploringhowsectoralandenterprisestructuresshapepatternsofjobcreation;second,assessinginmoredetailthedeterminantsoffactorintensityandjobcreationatthefirmlevel.

On the former, an immediate follow-on to the findings in this paper may be to probe job creationpatternsinasetofcountrycasestudies.Suchanapproachwouldallowformoredetailedquantitativeassessment, including linking job creation outcomes with specific policy actions. These case studieswouldfocusoncountriesthatappearasoutlierstothepatternsdiscussedinthispaper,orwherethereareparticularlyinterestingepisodesofgrowthanddecline(e.g.surgesandslumps).Thismayallowustoidentifymorespecifically thedynamics jobcreation/ jobsheddingresponses,andthetimingofthoseresponses, in key stages of the business cycle (entering or coming out of a slump) impact longer-runoutcomes.

Anotherpotentiallyvaluablefollow-onfromthefindings inthispaperwouldinvolveassessinginmoredetailhowsectoralspecializationimpactsjobcreationpatterns.Suchanassessmentmightfollowalongthe lines of Loayza and Raddatz (2006), who found that the sectoral composition of growth (and,

41

specifically, relative labor intensity) had important implications for the impact of growth on poverty.Anotherpotentiallyvaluablefollow-up.

Furtherdown the line, future research following the resultsof thispapermay involve thechallengingtaskofincorporatingearningsintotheanalysistoallowforamorenuancedunderstandingofeconomiescharacterizedbyhigh levelsof self-employmentand informality,aswellas identifyingdistortions thatmayleadtoasystematicunderutilizationoflaborinsomeeconomies.

Finally,muchwork remains tounderstand firm-leveldeterminants to job creation.Within this, ahighpriorityinthisseriesofpapersonprivatesectorjobcreationistodeepenourknowledgeonthefactorswhichdeterminewhetherandhowafirmchoosestoemploylabor,particularlywithrespecttocapital,whenfacingaknown(or,converselyanuncertain)futurepatternofgrowth.

42

ANNEX A: OKUN’S LAW

ThebasicbivariatemodelspecificationthatwasusedbyOkun(1962),interalia,byWeber(1995)mayberepresentedbythefollowingsetofequations:

(#,%6 ≡ (#,% − (#,%∗ (1)

"#,%6 ≡ "#,% − "#,%∗ (2)

"#,%6 = '(#,%6 + /#,% (3)where(#,%6 isthelogarithmofcyclicaloutput,theoutputgap;(#,%isthelogarithmofobservedoractualoutput;(#,%∗ is the logarithmofpotentialornaturaloutput;"#,%6 is thecyclicalunemploymentrate, theunemploymentgap;"#,% is theobservedor actualunemployment rate;"#,%∗ is thepotentialornaturalunemploymentrate;and/#,%isastochasticerrorterm.Thesubscriptidenotescountryiinyeart.Okun’scoefficientismeasuredbytheestimatedvalueof',theimpactcoefficient,suchthat' ≤ 0.Obviously,toestimatethiscoefficientweneedtimeseriesontheunobservedcomponents"#,%∗ and(#,%∗ .Thetrendvaluesare foundthrougha linearregressionusingtwo lagsof thecorrespondingvariableasHamilton(2017) recommendsnotusingHodrick-Prescott (HP) filteras itproduces serieswith spuriousdynamicrelationsthathavenobasis intheunderlyingdata-generatingprocessandsuggestusingvariables lagsinsteadasacleanerdetrendingprocess.Wealsouseamovingaverageoftheform(%∗ = ((%E. + (% +(%-.)/3tofindthenaturaloutputandemployment.Sincedeterminingthenaturalrateofgrowthinthisframeworkisnon-trivialforourresultswedecidedtousethedifferencespecificationtomakesurethatthedetrendingoftheserieswasnotthedrivingforceinourresults.However,wemakeuseofthegapspecification to check the robustness of our results. Table 1 shows results using Okun’s coefficientsestimatedbythegapspecificationsusingthetwodetrendingmethodologies.In this Annex, we test some alternative specifications of the Okun model, mainly the idea that therelation between unemployment and output growth does not necessarily need to be linear. AnnexFigure 1 shows the difference between the fitted values estimated by a linear and by a non-linearregressionmodelforthecasesoftheUnitedStates,Ecuador,Tanzania,andVietnam.Weseethatthenon-linearmodeldisplayssomeinterestingpatternsthatarenotaccountedforinthelinearmodel.Forexample, the data for the United States displays a convex shape whereas Ecuador, Tanzania, andVietnam have a concave form. In the case of the United States, the rate at which unemploymentdecreases, slows down as GDP grows at faster rates. On the other hand, the developing countries’concaveform,suggestthatunemploymentactuallyincreaseswithlowlevelofpositivegrowthandonlybecomes negative once a faster rate of growth in achieved. The convex form in the US is probablyrelated to the country reaching itsnatural rateofunemployment, asonce it is reached,moreoutputgrowthwillnot leadto lessunemployment.Eventhoughthesepatternsareextremely interestingandthereismoretobeexplored,theissueisthatwenowhavetwocoefficients,ratherthanone,andhowto we explore the Okun coefficient as an employment output when there is more than one part tocontrol for.However, it is important tohighlight thatboth linearandnon-linearOkuncoefficientsarehighlycorrelated.

43

Annex1-Figure1Non-LinearOkun

44

Annex2-Figure2DistributionofOkun’scoefficientbyincomegroup

0.5

11.5

22.5

Density

-1 -.8 -.6 -.4 -.2 0okun_b

02

46

Density

-.3 -.2 -.1 0 .1okun_b

0.5

11.5

22.5

Density

-.5 0 .5 1okun_b

02

46

810

Density

-.2 -.1 0 .1 .2okun_b

45

Annex3-Table1HighImpactFirmsbyCountry

country year all firms himpcf_15 himpcf_20 himpcf_50 jobs himpcf_15 himpcf_20 himpcf_50Angola 2006 834 3.8 3.1 1.0 5,006 13.8 11.2 3.4Angola 2010 618 11.9 8.9 3.9 5,418 47.7 42.2 32.3Albania 2007 2,126 9.0 6.4 0.9 12,672 40.7 27.5 10.4Albania 2013 736 2.6 2.3 0.5 3,010 38.5 35.2 23.2Armenia 2009 1,098 14.1 9.1 3.5 19,905 58.9 52.6 41.6Armenia 2013 1,439 4.3 4.1 1.6 7,695 58.1 57.3 28.7Burundi 2014 661 5.2 3.0 0.4 6,543 25.8 15.9 11.8Benin 2009 1,577 8.4 6.1 3.1 4,575 44.4 38.8 24.2Burkina Faso 2009 1,655 3.2 2.7 1.2 10,116 37.7 36.3 28.5Bangladesh 2007 26,829 5.8 4.3 2.7 399,710 69.9 62.6 51.6Bangladesh 2013 7,319 11.0 10.2 7.1 344,918 86.5 85.6 79.5Belize 2010 759 0.3 0.0 0.0 1,475 8.8 0.0 0.0Bolivia 2006 6,086 10.3 9.2 2.9 41,750 44.9 42.3 18.6Bolivia 2010 4,063 4.5 4.4 1.0 27,514 37.4 36.9 19.8Brazil 2009 292,663 12.2 11.0 6.5 4,655,269 78.8 76.6 69.5Botswana 2006 1,166 6.6 5.6 1.0 10,565 16.8 15.7 6.4Botswana 2010 675 9.2 7.9 3.7 9,275 79.1 77.0 66.5Côte d'Ivoire 2009 8,998 4.0 2.6 1.2 39,355 21.9 17.3 13.6Cameroon 2009 1,329 3.4 3.0 2.2 9,245 62.9 59.2 49.9Congo, Rep. 2009 1,423 6.1 1.7 1.0 8,775 31.1 23.9 17.3Ecuador 2006 6,185 7.6 6.6 3.7 60,147 54.3 51.5 40.9Ecuador 2010 7,830 2.8 2.5 1.0 64,564 36.7 34.5 22.8Egypt, Arab Rep. 2013 114,226 2.6 2.2 1.8 818,282 42.0 40.4 37.9Georgia 2008 3,122 18.9 14.2 9.6 108,718 91.0 89.3 81.8Georgia 2013 7,497 11.3 10.7 7.2 117,155 68.2 67.9 64.6Ghana 2007 6,386 6.2 4.8 2.1 50,816 52.3 47.9 39.0Ghana 2013 1,371 4.9 3.3 1.3 7,971 40.6 31.7 18.1Guatemala 2006 14,234 7.4 3.7 1.4 101,025 53.5 40.8 28.6Guatemala 2010 9,629 2.7 2.5 1.6 105,572 65.8 65.5 58.4Guyana 2010 321 7.1 4.9 2.5 3,324 45.1 40.2 26.8Honduras 2006 7,653 13.2 12.7 8.1 145,346 80.0 78.9 71.5Honduras 2010 5,122 3.4 3.4 0.5 27,545 66.7 66.5 26.4Croatia 2013 11,944 3.8 3.0 2.3 54,601 44.4 42.9 35.5Jordan 2013 8,396 3.3 2.7 0.6 55,407 47.5 45.3 36.3Kyrgyz Republic 2009 1,002 9.9 9.0 1.8 11,652 41.6 40.2 18.7Kyrgyz Republic 2013 939 3.8 1.2 0.8 8,131 50.4 38.5 34.4Lebanon 2013 6,348 6.2 5.2 2.4 32,671 51.2 47.3 29.9Madagascar 2009 1,891 5.5 4.3 2.8 23,571 44.4 42.9 39.7Madagascar 2013 2,434 3.0 3.0 2.2 29,154 55.8 55.8 52.7Mexico 2006 76,686 4.3 3.5 1.4 515,381 53.3 50.7 36.7Mexico 2010 68,465 4.4 2.7 1.3 652,430 63.7 62.1 56.4Macedonia, FYR 2009 4,935 12.4 9.6 6.7 43,169 42.3 40.4 32.7Macedonia, FYR 2013 3,056 3.0 2.5 1.2 11,789 44.2 40.1 25.8Mali 2010 970 1.1 0.8 0.2 1,729 18.6 17.2 8.4Montenegro 2013 1,965 0.0 0.0 0.0 2,422 0.0 0.0 0.0Mongolia 2009 2,567 18.5 14.6 5.5 30,445 66.7 61.7 40.0Mongolia 2013 4,769 3.8 0.4 0.1 28,673 36.7 10.5 6.8Mauritius 2009 4,238 8.6 6.8 3.5 60,949 79.3 77.8 66.1Malawi 2009 1,778 11.7 7.4 3.9 26,585 70.6 67.4 58.9Malawi 2014 2,924 3.5 3.3 1.3 71,763 86.0 85.7 81.5Namibia 2014 7,119 3.1 2.2 0.7 18,026 23.7 22.0 15.8Niger 2009 921 5.1 2.9 0.5 4,010 30.4 23.5 15.0Nicaragua 2010 1,878 3.8 2.3 0.7 11,505 49.1 46.3 27.0Nepal 2009 11,930 2.7 2.5 0.2 36,213 19.4 18.2 6.3Nepal 2013 9,877 2.8 2.8 0.1 22,300 34.7 33.6 5.9Pakistan 2007 25,114 2.2 1.8 0.5 104,299 52.3 49.7 36.9Pakistan 2013 12,549 3.7 3.4 1.9 196,698 64.1 63.0 53.5Panama 2006 3,437 4.1 2.7 1.6 18,463 50.2 40.6 31.2Panama 2010 5,359 7.7 5.8 4.5 37,763 51.8 46.4 38.5Peru 2006 6,076 18.9 14.5 12.3 159,257 80.0 76.0 71.7Peru 2010 13,712 9.5 8.9 1.7 160,409 71.8 70.0 41.4Philippines 2009 33,366 6.3 5.8 2.5 379,664 81.6 80.0 65.5Paraguay 2010 2,144 10.7 10.2 5.3 29,250 68.2 67.6 56.3Russian Federation 2009 84,647 14.9 14.1 8.5 3,940,826 88.8 88.0 83.8Russian Federation 2012 83,316 6.5 5.3 2.6 663,197 48.0 45.9 34.3Rwanda 2006 598 8.0 6.4 2.4 5,776 27.1 25.8 13.8Senegal 2007 1,714 3.2 2.2 1.0 8,616 24.2 22.1 10.6Senegal 2014 2,606 1.6 1.4 0.3 11,287 18.6 17.5 11.7Sierra Leone 2009 870 4.4 2.5 0.6 2,704 38.7 35.0 22.4El Salvador 2006 10,287 9.0 6.6 2.0 78,474 59.7 53.1 37.0El Salvador 2010 4,809 3.1 1.8 1.3 23,478 34.6 23.5 16.7Serbia 2013 14,762 0.7 0.7 0.3 51,145 19.3 17.0 10.1Suriname 2010 274 1.7 1.1 1.1 1,022 43.4 39.1 39.1Chad 2009 647 7.0 3.9 0.9 3,943 25.8 16.2 7.2Togo 2009 1,382 2.0 2.0 0.0 4,710 6.8 6.8 0.0Tajikistan 2013 1,240 6.3 4.1 2.7 11,581 42.4 36.7 32.5Trinidad and Tobago 2010 5,852 1.1 0.5 0.4 15,711 12.6 4.7 3.9Turkey 2008 43,680 15.4 12.5 6.9 1,220,251 87.0 85.5 79.0Turkey 2013 103,852 7.5 7.1 3.2 597,295 56.6 54.7 30.2Tanzania 2006 7,293 6.9 5.0 1.8 59,191 49.1 45.1 30.6Uganda 2006 3,969 6.5 4.4 1.1 24,237 39.8 34.4 16.0Uganda 2013 6,752 2.5 2.5 0.3 17,119 35.8 35.4 16.0Ukraine 2008 48,119 7.8 6.3 2.4 445,132 48.0 44.3 33.4Uruguay 2006 5,492 7.5 6.1 2.9 37,881 42.0 38.0 23.7Uruguay 2010 3,385 2.2 1.9 0.3 17,246 39.8 36.9 18.8Venezuela, RB 2006 18,552 7.3 5.8 2.1 162,155 37.6 35.3 24.4Venezuela, RB 2010 10,910 3.5 3.5 2.1 56,074 75.3 75.3 69.0Yemen, Rep. 2010 5,934 3.2 2.8 0.5 17,279 44.0 41.7 23.8Yemen, Rep. 2013 5,273 1.7 1.3 0.6 10,849 52.5 48.2 36.1South Africa 2007 120,831 7.0 4.4 1.7 1,288,628 31.2 26.5 13.7

share of firms share of positive jobs

46

BIBLIOGRAPHY

Acs,ZoltanJ.,WilliamParsons,andSpencerTracy.2008.“High-ImpactFirms:GazellesRevisited.”unpublishedmanuscriptpreparedfortheUnitedStatesSmallBusinessAdministration.

Ahmed,Masood,DominiqueGuillaume,andDavideFurceri.2012.“YouthUnemploymentintheMENARegion:DeterminantsandChallenges.”inAddressingthe100MillionYouthChallenge:PerspectiveonYouthEmploymentintheArabWorldin2012.Geneva:WorldEconomicForum.

Ball,Laurence,DanielLeigh,andPrakashLoungani.2016.“Okun’sLaw:Fitat50?”RevisedversionofNBERWorkingPaperNo.18668.

Botero,J.,S.Djankov,R.PortaandF.C.Lopez-De-Silanes.2004.“TheRegulationofLabor.”TheQuarterlyJournalofEconomics.119(4):1339–82.

Caballero,RicardoJ.,KevinN.Cowan,EduardoM.R.A.Engel,andAlejandroMicco.2004.“EffectiveLaborRegulationandMicroeconomicFlexibility.”DiscussionPaper1480.CowlesFoundation.

Davis,StevenJ.andJohnC.Haltiwanger.1999.“OntheDrivingForcesBehindCyclicalMovementsinEmploymentandJobReallocation.”AmericanEconomicReview,89(5):1234-1258.

Davis,StevenJ.,JohnC.Haltiwanger,andScottSchuh.1996.JobCreationandDestruction.TheMITPress:Cambridge,Massachusetts.

Decker,Ryan,JohnHaltiwanger,RonJarmin,andJavierMiranda.2014."TheRoleofEntrepreneurshipinUSJobCreationandEconomicDynamism."JournalofEconomicPerspectives.28(3):3-24.

Fort,Teresa,JohnHaltiwanger,RonS.Jarmin,andJavierMiranda.2013.“HowFirmsRespondtoBusinessCycles:TheRoleoftheFirmAgeandFirmSize.”IMFEconomicReview.61:520-559

Fox,Louise,andA.M.Oviedo.2008.“AreSkillsRewardedinSub-SaharanAfrica?DeterminantsofWagesandProductivityintheManufacturingSector.”WorldBankPolicyResearchWorkingPaperWPS4688.Washington,D.C.:WorldBank.

Freund,CarolineandBobRijkers.2014.“EpisodesofUnemploymentReductioninRich,Middle-Income,andTransitionEconomies.”PolicyResearchWorkingPaperNo.6891.Washington,D.C.:WorldBank.

Geroski,PaulA.1995.“WhatdoWeKnowaboutEntry?”InternationalJournalofIndustrialOrganization.13:421-440.

Gertler,MarkandSimonGilchrist.1994.“MonetaryPolicy,BusinessCycles,andtheBehaviorofSmallManufacturingFirms.”TheQuarterlyJournalofEconomics.109(2):309-340.

González,AlvaroS.,LeonardoIacovone,andHariSubhash.2013“RussianVolatility:ObstacletoFirmSurvivalandDiversification.”WorldBankPolicyResearchWorkingPaperNo.S6605.Washington,D.C.:WorldBank.

Gordon,RobertJ.1984.‘‘UnemploymentandPotentialOutputinthe1980’s.’’BrookingsPapersonEconomicActivity.537–564.

47

Haltiwanger,J.,S.ScarpettaandH.Schweiger.2006.“AssessingJobFlowsacrossCountries:TheRoleofIndustry,FirmSize,andRegulations.”WorldBankPolicyResearchWorkingPaperNo.4070.Washington,D.C.:WorldBank.

Hanusch,Marek.2012.“JoblessGrowth?Okun’sLawinEastAsia.”WorldBankPolicyResearchWorkingPaperNo.6156.Washington,D.C.:WorldBank.

Heckman,JamesJ.andCarmenPages(editors).2004.LawandEmployment:LessonsfromLatinAmericaandtheCarribean.NBERConferenceSeriesReport.UniversityofChicagoPress,ChicagoandLondon.

Hendry,DavidF.,andGrayhamE.Mizon.1978.‘‘SerialCorrelationasaConvenientSimplification,NotaNuisance:ACommentonaStudyoftheDemandforMoneybytheBankofEngland.’’EconomicJournal.88:549–563.

Kaufman,RogerT.1988.‘‘AnInternationalComparisonofOkun’sLaw.’’JournalofComparativeEconomics.12:182–203.

Knoester,Anthonie.1986.‘‘Okun’sLawRevisited,”WeltwirtschaftlichesArchivReviewofWorldEconomics.122(4):657–666.

Knoteck,EdwardS.,II.2007.“HowUsefulisOkun’sLaw?”FederalReserveBankofKansasCity.

Loayza,NormanV.andClaudioRaddatz.2006."Thecompositionofgrowthmattersforpovertyalleviation."WorldBankPolicyResearchWorkingPaperWPS4077.Washington,D.C.:WorldBank

LaPorta,R.,F.Lopez-De-Silanes,A.ShleiferandR.Vishny.1999.“TheQualityofGovernment.”JournalofLaw,EconomicsandOrganization.15(1):222–79.

Micco,AlejandroandCarmenPages.2006.“TheEconomicEffectsofEmploymentProtection:EvidencefromInternationalIndustry-LevelData.”DiscussionPaper2433.IZA.

Mizon,GrayhamE.1995.“ASimpleMessageforAutocorrelationCorrectors:Don’t."JournalofEconometrics.69:267–288.

Moosa,ImadA.1997.“ACross-CountryComparisonofOkun’sCoefficient.”JournalofComparativeEconomics.24(3):335–56.

Okun,ArthurM.(1962).“PotentialGNP:ItsMeasurementandSignificance,”AmericanStatisticalAssociation,ProceedingsoftheBusinessandEconomicsStatisticsSection,pp.98-104

Pakes,Ariel,andRichardEricson.1998.“EmpiricalImplicationsofAlternativeModelsofFirmDynamics.”JournalofEconomicTheory.79(1):1-45.

Prachowny,MartinF.J.1993‘‘Okun’sLaw:TheoreticalFoundationsandRevisedEstimates.’’ReviewofEconomicsandStatistics.75:331–336.

Rajan,RaghuramG.andLuigiZingales.1998.“FinancialDependenceandGrowth.”AmericanEconomicReview.88(3):559-586.

Smith,Gary.1975.‘‘Okun’sLawRevisited.’’TheQuarterlyReviewofEconomicsandBusiness.37–54.Winter1975.

48

Sutton,John.1997.“Gibrat'sLegacy.”JournalofEconomicLiterature.35(1):40-59.

Weber,ChristianE.1995.‘‘CyclicalOutput,CyclicalUnemployment,andOkun’sCoefficient:ANewApproach.’’JournalofAppliedEconometrics.10:433–445.

WorldBank.2013.WorldDevelopmentReport2013:Jobs.WashingtonDC:WorldBank.