Non-traded Gains From Trade - Selection in the Non-Traded Sector: Evidence from Brazil

Non-tradedGainsfromTrade SelectionintheNon-tradedSector:EvidencefromBrazil RafaelMachadoParenteandRowanShiWP/23/265 IMFWorkingPapersdescriberesearchinprogressbytheauthor(s)andarepublishedtoelicitcommentsandtoencouragedebate. TheviewsexpressedinIMFWorkingPapersarethoseoftheauthor(s)anddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement. 2023 DEC ©2023InternationalMonetaryFundWP/23/265 IMFWorkingPaper WesternHemisphereDepartment Non-tradedGainsfromTrade SelectionintheNon-tradedSector:EvidencefromBrazilPreparedbyRafaelMachadoParenteandRowanShi AuthorizedfordistributionbyGustavoAdler December2023 IMFWorkingPapersdescriberesearchinprogressbytheauthor(s)andarepublishedtoelicitcommentsandtoencouragedebate.TheviewsexpressedinIMFWorkingPapersarethoseoftheauthor(s)anddonotnecessarilyrepresenttheviewsoftheIMF,itsExecutiveBoard,orIMFmanagement. ABSTRACT:Weinvestigatehowtradeshocksaffecttheallocationoflaboracrossplantsatthelocallabormarketlevel.UsingBrazil’simportliberalizationasaquasi-naturalexperiment,weuncoveranewmarginforthegainsfromtrade:thereallocationoflaborfromsmallertolargerproducersinthenon-tradedsector.Wefindthatinresponsetoliberalization,largernon-tradedproducersself-selectintoimporting,expandingastheygainaccesstoinputsfromabroad.Wethendevelopaparsimoniousmodelofheterogeneousproducersincorporatingthismechanism.Thetheoryisconsistentwiththeempiricalfindingsandshowthatreallocationamongnon-tradedproducersiswelfare-enhancing.Incontrast,thisreallocationeffectdisappearswhenallnon-tradedproducersmakethesameimportingdecision. RECOMMENDEDCITATION: JELClassificationNumbers: F12;F13;F14;F15;F16;O24 Keywords: Gainsfromtrade;tradeliberalization;non-tradedsector;firmSelection Author’sE-MailAddress: rmachadoparente@imf.organdrowanxshi@torontomu.ca WORKINGPAPERS Non-tradedGainsfromTrade Selectioninthenon-tradedsector:EvidencefromBrazil PreparedbyRafaelMachadoParenteandRowanShi1 1TheauthorsthankSteveReddingandseminarparticipantsatTorontoMetropolitanUniversity,PrincetonUniversity,andtheFederalReserveBankofSt.Louis.NardeenAbdulkareemprovidedexcellentresearchassistance.BothauthorsgratefullyacknowledgefinancialsupportfromtheInternationalEconomicsSectionatPrincetonUniversity.Allerrorsareourown. Contents 1. Introduction 1 2. Empiricalanalysis 4 2.1.Policyenvironmentanddata................................ 5 2.2.Measuringlocalexposuretothetradeliberalization................... 6 2.3.Reallocationinlocallabormarkets............................. 7 2.4.Establishment-levelemploymentpatterns......................... 9 2.5.Importingbehaviorandnewimporters.......................... 12 3. Atheoreticalmodelofimporting 15 3.1.Framework........................................... 16 3.2.Importliberalizationpolicy................................. 19 3.3.Theroleofselection...................................... 22 4. Conclusion 23 A. Additionaltablesandfigures 27 B. Mathematicalappendix 31 C. Structuralshockanalysis 37 ListofFigures 1.Aggregateimportsversusaverageimporttariffs.2 2.DescriptivestatisticsofimportinginBrazil.12 3.Determinationofefficiencycutoffsinpartialequilibrium,beforeandafterliberal-ization.19 4.Extensivemarginconsequencesofanimportliberalization.20 A.1.Distributionoftheeffectivetariffreduction.27 A.2.Plantsizeacrossregions,conditionalonregionalemploymentquartile.28 A.3.Extensivemarginreallocationoflaborinresponsetothetradeliberalization,byindustry29 A.4.Intensivemarginreallocationoflaborinresponsetothetradeliberalization,byindustry30 ListofTables 1.Theeffectofthetradeliberalizationonjobflowsbyindustrytradability8 2.Totaljobsdestroyedorcreatedattributabletodifferencesinexposure.9 3.Thedifferentialeffectsofthetradeliberalizationonplantsofdifferentsizebysector.11 4.Establishmentcharacteristicsversusimportingstatus.13 5.Plantexit,growth,andimportingbehaviorinthenon-tradedsector14 6.Theeffectsofthetradeliberalizationonplantsimportingdecision.16 A.1.Thedifferentialeffectsofthetradeliberalizationonplantsofdifferentsize:astatis- ticalcomparisonwiththesmallestplants.28 C.1.Thedifferentialeffectsofthetradeliberalizationonplantsofdifferentsize:amodel-basedmeasureofexposure39 C.2.Theeffectsofthetradeliberalizationonplantsimportingdecision:amodel-basedmeasureforexposure.39 1.Introduction Tradepolicyandtradeshockscanhavewidespreadandvariedimpactsontheaffectedeconomies.Inrecentyears,thelocallabormarketasachanneloftradeshocktransmissionhasreceivedconsid-erableinterest.Thekeyinsightisthat,giventheiroriginalcompositionofeconomicactivity,locallabormarketsmaybemoreorlessexposedtoatradeshockcomparedtoothers.Thisdifferentialexposuremaytranslatetostrongerorweakertradepolicyeffectsfortheregion’seconomicagents,regardlessoftheagent’sownindustryofactivity.Differentialeffectsespeciallymayoccuriflocallabormarketsarenotperfectlyintegratedwitheachother,aswithmobilityfrictionsortradecostsamongthem.Inturn,acountry’sinitialpatternofeconomicactivityacrossspacemayultimatelyhaveconsequencesfortheeffectsofnational-leveloftradepolicy,par