UNITEDNATIONSCONFERENCEONTRADEANDDEVELOPMENT

WORKINGPAPER

No.5

UNCTAD/WP/2023/4

AUGUST2023

ClovisFreire

CommoditiesBranch,

DivisiononInternationalTradeandCommodities,UNCTADclovis.freire@

AnjaSlany

PolicyAnalysisandResearchBranch,DivisionforAfrica,LeastDevelopedCountriesandSpecialProgrammes,UNCTAD

anja.slany@

Abstract

Keywords

RealizingproductdiversificationforstructuralchangeinAfricancountries

ExportdiversificationhasbeenamongthemostcitedpolicyrecommendationsforAfricancountriestospurstructuraltransformationandincreaseresilience.However,exportdiversificationthatbenefitsstructuralchangeisnotanautomatedprocessandrequiresananalyticalapproachandcomplexdecision-making.Applyinganadjustedeconomiccomplexityandproductspacemethodologyontradedataof54Africancountriesandtheirtradingpartners,thispaperassessesexportdiversificationopportunitiesthatarefeasibletorealize,associatedwithstructuralchangeandofhighdemandintheworldandontheAfricancontinent.IncreasingcomplementaritiesofAfricanexportsandimportsarecrucialtoyieldhigherbenefitsfromtheAfricanContinentalFreeTradeArea(AfCFTA).Thepaper’sfocusonintra-Africandiversificationopportunitiesallowsforacontinentalmappingofcurrentexportswithexportdiversificationopportunitiesandtheidentificationofnicheareasofindividualcountries.Thepaperfindsthatalmostallcountrieshavesomepotentialforproductdiversificationintolightmanufacturing(machineryandmechanicalappliances;electricalmachinery)andprocessedbasemetalproducts(articlesofironandsteel),thoughindifferentproducts.Thepaper’sfindingscanguidepolicymakersanddevelopmentpartnersinidentifyingindustrializationstrategiesandproductivecapacityneeds.

AfCFTA,exportdiversification,productspace,Africaneconomies

Thefindings,interpretationsandconclusionsexpressedhereinarethoseoftheauthor(s)anddonotnecessarilyreflecttheviewsoftheUnitedNationsoritsofficialsorMemberStates.ThedesignationsemployedandthepresentationofmaterialonanymapinthisworkdonotimplytheexpressionofanyopinionwhatsoeveronthepartoftheUnitedNationsconcerningthelegalstatusofanycountry,territory,city,orareaorofitsauthorities,orconcerningthedelimitationofitsfrontiersandboundaries.Thispaperhasnotbeenformallyedited.

1UNCTADWorkingPaperNo.5

_____________________________________________________________________________________

_____________________________________________________________________________________

Contents

Acknowledgements

1

Introduction

2

1.Theimportanceofproductdiversificationforstructuralchange

3

2.Identifyingproductdiversificationopportunities

5

2.1Data

5

2.2Methodologytoidentifyproductdiversificationopportunities

6

2.3Results

9

3.Leveragingregionalexportdiversificationopportunities

15

4.Realizingdiversificationopportunities:TheroleoftheAfCFTA

1

9

5.Conclusion

27

References

28

Appendix

34

Acknowledgements

ThispaperbenefitedfromcommentsandsuggestionsfromHabibaBenBarka,LisaBorgattiandAnidaYupariAguado,twoanonymousreferees,andparticipantsfromthe‘‘Africa’sregionalandglobalintegration:Lessonsfromthepastandimplicationsforthefuture’’conference,heldinBonnon17-18November2022.

2UNCTADWorkingPaperNo.5

_____________________________________________________________________________________

_____________________________________________________________________________________

Introduction

TheestablishmentoftheAfricanContinentalFreeTradeArea(AfCFTA)givesnewmomentumtopromoteexportdiversificationatthecontinentallevelbyprovidingalargermarketandreducingtradecosts.ThepaperdiscussesthepotentialofincreasingregionaltradecomplementaritiesandthepotentialoftheAfCFTAinredressingthelowlevelsofintra-Africantradeandexportdiversification.MostAfricancountrieslargelydependonexportingafewcommodities,oftencharacterizedbylow-incomeelasticities,withlittlescopeforproductivityincreasesandstrongpricefluctuation.Duetothehighvolatilityofexportearningsanddeteriorationofacountry’stermsoftrade,exportdiversificationawayfromcommoditydependenceiskeytopromotingstructuraltransformationandsustainableeconomicgrowth(e.g.Sarinetal.,2020).

Thestrandofliteratureoneconomiccomplexity(e.g.Hidalgoetal.,2007;HidalgoandHausmann,2009)hasbeenusedtoidentifypotentialnewexportsthatrequireproductivecapabilitiessimilartotheonesalreadyexistentinthecountry,whichcouldfacilitateeconomicdiversificationandserveasaninputtoindustrialpolicies.

1

Yet,forAfricancountries,thereremainslimitedpolicyrecommendations,basedontradedataandeconomiccomplexityanalysis,onwhichsectors/product-groupshavethegreatestpotentialforexportdiversificationduetohighfeasibilitytoproduce,aswellasfavorabledemandconditions.Successfulexportdiversificationrequirestheobservationofmarketincentives(growthopportunities)andstateinterventiontoprovidesectorinputsandfacilitatemarketentrance(Freire,2017).

Fewattemptshavebeenmadetoguidedevelopingcountriesonexportdiversificationopportunities,andnoattemptshavebeenmadetodatefortheAfricancontinent.ThispaperprovidesevidenceonfeasibleexportdiversificationopportunitieswithbenefitsforstructuralchangeforallAfricancountries.Theanalysisconsidersthedemandforthesefeasibleproductsbothgloballyandonthecontinent.Althoughintra-Africanexportsasashareoftotalexportsarecurrentlylow,around18percent,regionalintegrationpromisesanimportantinstrumenttopromoteexportdiversification.Intra-Africanexportsarelessconcentratedonasmallnumberofproductsthanexportstotherestoftheworld,andcomprisealargershareofmanufacturedgoods.

AliteraturereviewinSection1givestherationaltoassessproductdiversificationopportunitiesthathavethepotentialtobenefitstructuralchange.Section2brieflydescribesthedataandmethodology,andpresentstheresultsofoverallexportdiversificationopportunities.Section3focusesonproductdiversificationopportunitieswithdemandonthecontinent.Section4providesatoolkitofnecessarypolicyconsiderations.Section5concludes.

1.Theimportanceofproductdiversification

forstructuralchange

Successfullypromotingdiversificationwithinandacrosssectorscanhelpreducevolatilityandcontributetolong-termstructuraltransformationandsustainabledevelopment(e.g.G?zg?randCan,2017).InAfrica,where45ofthe54countriesarecommodity-dependent(UNCTAD,2022),exportdiversificationcanbeasignificantdriverofproductivityandproductiongrowth(BenHammoudaetal.,2010).

1Thisliteratureassumesthatproductsrequireaspecificsetofproductivecapacitiestobeproduced.Countrieshavesomeofthesecapacitiesandwillproducetheproductsforwhichtheyhavethecapabilities.Countriesthatareverydiversifiedandproducemanyproductsthatareproducedbyonlyafewcountries(i.e.notubiquitous)wouldthenhavealargesetofproductivecapacitiesandareconsideredmorecomplexeconomies.Ifaproductisproducedbyfewcountries(i.e.notubiquitous)thatarealsoverydiversified,theseproductsareconsideredtobemorecomplex(Freire,2021b).

3UNCTADWorkingPaperNo.5

_____________________________________________________________________________________

_____________________________________________________________________________________

Diversificationresultsfromincreasingthenumberofgoodsandservicesproduced/exported(horizontaldiversification),measuredbythenumberofactiveexportlinesorqualityupgradingwithinalreadyproducedgoodsandservices(verticaldiversification),measuredbytheirunitvalue.WhileImbsandWacziarg(2003)showedevidenceforaslowinghorizontaldiversificationwithincreasingGDPpercapitalevels,qualityupgradingwithinproductsandsectorsincreasedwithrisingincomelevels(Mau,2016).Thevarietyofexportsisstronglypath-dependent,asacountry’scurrentproductioncapabilities(i.e.technologies,productionfactors,institutions,resourceendowments)influencewhatcanbeproducedinthefuture(e.g.Hidalgoetal.,2007).Spilloversoftechnologyandskillsthroughdiversificationtriggerlong-termstructuralchange(Hausmannetal.,2007).Coniglioetal.(2021)explorepathdependencyinthediversificationprocessforasampleof221countriesandfindthatmostcountriesthatarelesslikelytodiversifyawayfromtheircomparativeadvantagearenaturalresourcedependent.

Duetothesepath-dependenciesandpotentialmarketimperfections(e.g.highdiscoverycostsandlackofmarketinformationkeepcountriesfromproducingagoodthatcouldbeproducedwithacomparativeadvantage),detailedinformationondiversificationopportunitiesandstrategicgovernmentinterventionisnecessary.Moreover,exportdiversificationmustbetargetedtowardsproductswithahigherbenefitforstructuralchangeduetohighercomplexity.

2

Inouranalysis,structuralchangeisdefinedbythemovementfromlesstohighercomplexitygoods.Inlessdiversifiedcountries,diversificationpoliciesthattarget‘related’productsaremorelikelytosucceedanddevelopcomparativeadvantagesinproductsthatrequiresimilarproductioncapabilities.

3

Alargeportionofliteraturehasemphasizedtheimportanceofregionaltradeasasteppingstonetowardgreaterintegrationintoworldtrade.Countrieswithsimilarendowmentsandincomelevelshaveagreaterexportvarietyintheirbilateraltradebasket(e.g.Parteka,2020),andscaleeconomies(domesticandforeignmarketsize)positivelydriveexportdiversification(ParketaandTamberi,2013).Also,reducingtradecostssignificantlyincreasestheexportdiversificationlevelsofdevelopingcountries,whiletradeagreementspositivelyimpactexportdiversificationforAfricancountries(Vogel,2022).Regolo(2017)confirmsthatregionalmarketsaremoreaccessiblefornewlyexportedproductsthankstolowertradecosts,includingtariffs.Whileintra-Africantradecostsarestilloftenhigherthanextra-Africantradecosts(UNCTAD,2021a),theAfCFTAplaysaroleinreducingtariffsandnon-tariffbarriers,andisthereforeexpectedtopromotetheintroductionofnewexportproducts.

Currently,intra-Africantradeislow.Intra-Africanexportsaccountfor17.73percentoftotalexports(comparedto52.68percentintheAmericas,58.47percentinAsia,68.10percentinEuropeand6.41percentinOceania(UNCTADStatistics)).Themainobstaclestointra-Africantradeincludelowtradecomplementarityduetoloweconomicdiversificationandlimitedproductivecapacities,highnon-tariffbarriers,andlackofinfrastructure.TheAfCFTAisexpectedtoaddressthesetradefrictionsthroughvariousinitiatives(UNCTAD,2021a).Thepotentialoffacilitatingintra-Africantradeforstructuralchangestemsfromtheobservationthatintra-Africanexportsaremoresophisticated(i.e.highershareofmanufacturedgoods)andmorediversifiedthanexportstotherestoftheworld(UNCTAD,2021a;IITCandUNCTAD,2021).

Moreover,duringatimeofglobaltensions,suchasthepandemic,thewarinUkraine,andadeclineinglobalization,South-Southtradecanprovidedevelopingcountrieswiththeuntappedpotentialforexportdiversification(OECDandWTO,2019).Countrieswithshorterbilateraldistancetrademorewitheachother

4

2AsshowninGalaetal.(2018),asthecomplexityoftheexportbasketofdevelopingcountriesincreases,thereisagreaterlikelihoodthattheirincomewillconvergewiththoseofhigh-incomecountries.

3Pinheiroetal.(2022)examinestheroleofrelatedandunrelateddiversificationintheprocessofeconomicdiversificationandfindsthatrelateddiversificationismorecommonamonglessdevelopedcountries,whileunrelateddiversification(i.e.largejumps)becomesmoreimportantascountriesclimbthecomplexityladder.UNIDO(2023)examinescountries’diversificationexperiencesoverthepasttwodecadesandshowsthat61%of“jumps”observedwereshort(highrelatedness).Theseresultshighlighttheimportanceofbothformsofdiversificationinpromotingeconomicgrowthandavoidinglock-in,whileemphasizingtherelevanceofrelateddiversificationforlessdevelopedcountries.

4Empiricalevidenceintheliteratureisbasedonthetheoreticalfoundationsofthegravitymodel(seeBergstrand,1985;BaierandBergstrand,2007;Egger,2004).

4UNCTADWorkingPaperNo.5

_____________________________________________________________________________________

_____________________________________________________________________________________

andalsohavemorediversetradebaskets(Baharetal.,2014;BallandandBoschma,2021).Empiricalevidencehasestablishedapositivelinkbetweenexportdiversificationandcommonlanguage,aswellascommonborder(forareviewofempiricalliterature,seeVogel,2022)whichhighlightstheimportanceofSouth-Southtradefordiversification.Inaddition,marketaccessintheformofpreferentialtradeagreementsalsomattersfordiversification.Currently,Africancountrieslargelyexportnaturalresourceswithlittlevalueaddition.Infact,thevaluecreationinhigh-incomecountriesdependsontheannualimportingofresourcesfromlower-incomeregions,whichcausesanunequalecologicalexchange(Doringeretal.,2021).Hence,are-orientationofAfricantradetowardsthecontinentwouldnotonlykeepmorevalueaddedonthecontinent,butcouldalsopromotemoreequaltraderelationships.

Amajoraimofdiversificationispromotingstructuralchangebyincreasingproductcomplexity.However,increasingthenumberofproductsexporteddoesnotnecessarilymeanthattherehasbeenadiversificationintoproductswithhighervalue.Literatureoneconomiccomplexity(e.g.Hausmannetal.,2007;HidalgoandHausmann,2009)showsthatexportdiversificationshouldbeassessedaccordingtothelevelofthecurrentcomplexityoftheexportbasketasitstronglyinfluencesacountry’scapabilitiestoexportcertainproducts,atleastintheshort-tomedium-term.Linkingtheconceptofproductcomplexitytotheproductspacemethod,proposedbyHidalgoetal.(2007),facilitatestheidentificationofpathwaysforfuturediversificationintomorecomplexproducts.Particularly,bydisaggregatingtradedatareportedattheHarmonizedSystem(HS)6-digitlevel(furtherbyquantityunitcodeandunitvaluerange),ourapproachinthispapercontributestotheliteraturebyaccountingforproductdifferentiation.

Availableliteraturethatappliestheproductspacemethodtoidentifyexportdiversificationremainslimited,butisemerging.FortheexampleofTanzania,Estmannetal.(2022)usetheproductspacemethodtoexplorediversificationopportunitiesthatarehighlyrelatedtocurrentexportsandrequirerelativelylarge-jumpstonewproducts.Theirassessmentshowsthatwhilelow-hangingfruitproductsarealreadytargetedintheFiveYearDevelopmentPlanofTanzania,theproductclusterswithhighexportdemandanddiversificationopportunities(suchasplasticsandmachineryproduction)currentlyreceivelittleattention.

Similarly,forRwanda,HausmanandChauvin(2015)calculatefeasiblestrategicproductsfordiversificationwithhigher-than-averagecomplexityandahighdemandinboththeworldandregion,basedonHS4-digitleveltradedata.Threeclustersemergeintheiranalysis:i)machineryandelectronics;ii)constructionmaterials,andmetalandwoodproducts;andiii)chemicalproducts.Forinstance,withinthecategoryofmachinery,theauthorsfindthattheopportunitieslayinsupposedlysimpleragriculturework,andfoodprocessingmachineryandparts.HausmanandChauvin(2015)acknowledgetheambitioninsuchadiversificationstrategy,butthatgreatercoordinationandinvestmentwouldalsogeneratespilloversandfacilitatefuturediversificationanddevelopment(HausmanandChovin,2015).ApplyingtheGrowthIdentificationandFacilitationFrameworktoUganda,LinandXu(2016)identifiedarticlesofapparelandclothing,footwear,leather,electricalmachineryandequipment,agro-processingbusiness,andironandsteelasfeasibletoproduceandhavingexportgrowthopportunities.

AttheAfricanregionallevel,anapplicationoftheproductspacemethodisprovidedbySiTou(2021)fortheEastAfricanCommunity,showingahighdegreeofcomplementarityamongnationalproductivestructuresofsomeofitspartnerstates(Burundi,Kenya,Rwanda,TanzaniaandUganda).Theresultssuggestthatregionalvaluechains(RVC)cancontributetoproducingalargershareofgoodsonthecontinentthatareotherwiseimportedandareameanstopromoteindustrialdevelopment.DespitethewidelycitedpotentialofRVCdevelopment,concreteRVCupgradingprogrammesarelacking.Moreover,thereisaresearchgapinidentifyingdiversificationopportunitiestoinformregionallycoordinatedsectorpromotionstrategiesandindustrialpolicymaking.Thispaperclosesthatgap.

5UNCTADWorkingPaperNo.5

_____________________________________________________________________________________

_____________________________________________________________________________________

2.Identifyingproductdiversification

opportunities

2.1.Data

TradedataisfromtheUnitedNationsCOMTRADEdatabaseusingHSclassification19926-digitlevel,covering243economies,includingall54Africancountries.Onlybilateralimportdata(countrytocountry)atthe6-digitlevelwasconsideredfortheanalysis.Duetotheusualbetterqualityofthereportingofimportsfordutiespurposes,importdata,ratherthanexportdata,isused.Theassessmentofpotentialnewproductsisbasedontheyears2018/2019(average).Furthermore,themethoddifferentiatesproductsaccordingtodifferencesinunitvalues.Empiricalliteraturehasshownthatwhileexportingthesameproducttoonecountry,thehigher-incomecountrywillexportthehigherunitvalueproduct,whilethelower-incomecountrywillsystematicallyexportthelowerunitvalueproduct(Schott,2004;Fontagnéetal.,2008).Therefore,intheanalysis,wehaveconsideredproductsofdifferentpricesasdifferentproducts,asthecapabilitiesavailableinthesecountriestoproducetheproductswillbeconsideredinherentlydifferent.Analytically,tradedataattheHS6-digitlevelisfurtherdisaggregatedbyquantityunitcodeandunitvaluerange.ThemethodfollowsFreire(2017,2021a).UnitvalueswerecalculatedbasedonquantityandvaluedatafromCOMTRADE

5

.Themethodstatisticallycomputestheinterquartilerangesoftheunitvaluedistributionforeachproductandconsiderstheproductswithunitvaluewithindifferentinterquartileasdifferentproducts.Forexample,itconsidersawomen'sdressmadeofcottonwithintheunitvaluerangefromUS$0.10to$13.50asadifferentproductthanawomen'sdressmadeofcottonwithintheunitvaluerangefromUS$172to$223.

6

Aspreviouslymentioned,theanalysisusesdatareportedbyimporterstocalculatetheexportofcountries.Thus,dataonexportsfromcountriesinAfricaareofthesamequalityasdatafromothercountriesbecausetheyarefromreportersthatimportfromthosecountries,andtheleadingimportersaredevelopednations.Toreiterate,thisapproachmakesanimportantcontributiontotheliteraturebyaccountingforproductdifferentiation(differentunitvalueswithinproductgroups/qualityupgradingwithproducts).TheadvantageofthismethodisthatitallowsfortheassessmentofdifferentcapabilitiesofcountriesevenwhentheyexportthesameproductattheHS6-digitlevel,iftheproductsareofdifferentunitvalueranges.ThedisadvantageisthatitrequirestheavailabilityofquantityandvaluedataintheCOMTRADEdatasettocalculatetheunitvalues,andquantitydataisnotreportedaswellasvaluedata.Freire(2017)estimatesthatfrom2005to2013,around10percentoftradeflowsbycountriesinCOMTRADEweremissingquantityinformation.Thisresultsinlowerestimatesforthecomplexityofcountriesthanwouldbepossibleifallquantityinformationwasavailable.

2.2.Methodologytoidentifyproductdiversificationopportunities

AshighlightedinSection1,thereisaneedtoensurethateffortstodiversifyexportsareinlinewithtargetsofstructuralchangeandsustainabledevelopment.Thispaperfocusesontheprospectsforstructuralchange,proxiedbyproductsthatareofhighercomplexitythanacountry’saverageproductbasket.Futureresearchandcountrystrategiesmustalsotakeintoaccountenvironmentalconsiderations,aswellastheparticipationofwomenandyouthinthetargetedsectors.Thisexerciseisbeyondthescopeofthispaper(butthecasestudycanprovideanentry-pointfordiscussions).

Thepaperfocusesonthediscussionofthemostfeasiblenewproductswithahighercomplexity,aswellastheopportunitytoreplaceimports(whichwouldrelaxthebalance-of-paymentsconstraintofexportpromotion).Theanalyticalelementsaredescribedasfollows:

5Countriesreportthedatabasedoncriteriathathavebeenestablishedtoensurethecomparabilityofthedata.

6Assumingthatthoseunitvaluerangesresultedfromthestatisticalanalysisofinterquartilerangesofunitvaluedistributionfortheproductwomen'sdressmadeofcotton.AdetailedexplanationofthemethodologyisavailableinFreire(2017).

6UNCTADWorkingPaperNo.5

_____________________________________________________________________________________

_____________________________________________________________________________________

i)ProximityintheProductspace

TheProductspacemethod(Hidalgoetal.,2007)–whichmapsthedistancebetweenacountry’scurrentexportsbasketandanewproduct–isbasedonhowoftencountriesexporttheseproductssimultaneously.NewproductsshouldbenearbyintheProductspaceinordertohaveahigherlikelihoodofsuccess.Thereisclearlyatrade-offbetweendiversifyingintomorecomplexproductsandthepossibilityofasuccessfuldiversificationpath(moredistantproductsarealsomoredifficulttodevelop)(SiTou,2021).Ouranalysisshouldnotbeunderstoodasanupperlimitofdiversificationpotential.Ifcountriesrealizelargerjumpsintheproductspacethroughtargetedinvestments,additionalexportopportunitiescanarise.Althoughthecommoditiesavailableinthecountryplayanimportantroleindiversificationthroughvalueadditionandasinputstodownstreamsectors,insomecases,largerjumpsintheproductspacecanbenecessarytopromotestructuralchange.

ThemeasureofproximitybetweenproductsAandB(ΦA(chǔ)B)intheProductspaceiscalculatedusingamethodsimilartothatproposedbyHidalgoetal.(2007),astheminimumvaluebetweentheconditionalprobabilityP(A|B)ofacountryproducingAgiventhatitproducesBandtheconditionalprobabilityP(B|A)ofacountryproducingBgiventhatitproducesA:

ΦA(chǔ)B=ΦBA=min(P(A|B),P(B|A))(1)

Therefore,theproximitybetweentwoproductsrangesfrom0percent,inthecaseofnocountryproducesbothproducts,to100percent,inthecaseofallcountriesthatproduceonegoodalsoproducestheother.Inthispaper,weassumethat“feasibleproducts”arecloseproductswithan80percentprobabilitythatthecountryhassimilartechnologicalcapabilitiesandknowledgetoproducethoseproducts.

ii)Productcomplexityindex

Adesirableoutcomeofexportdiversificationshouldbeanincreaseineconomiccomplexitytobenefitstructuralchange(seeSection1).Therefore,inthenextstep,onlyproductsofhigher-than-averagecomplexityareconsideredtopromotediversification.

TheproductcomplexityindexiscalculatedfollowingFreire(2017)asarevisedversionofthemethodofreflectionsproposedbyHidalgoandHausmann(2009).Themodificationisthatthemethodconsidersallexports,notonlytheoneswithacertainrevealedcomparativeadvantage(RCA),becauseanRCAisavolatilemeasureforthosecountrieswithlowlevelsofdiversificationandthatrelyonfewcommoditiesforexports.Themethodofreflectionsiterativelycalculatesmeasuresofdiversification(kc,N)andubiquity(kp,N),andproduces,foreachproductp,anorderedlistofNrealnumbers(kp,0,kp,1,kc,p,…,kc,N),whereNisthenumberofiterationsofthemethod.Thenumberofcountriesthatexportproductpisrepresentedaskp,0.Theproductcomplexityindex(PCOMP)iscalculatedas:

PC0MP=

kp,5?-,

u

(2)

Where-,isthemeanandGisthestandarddeviationofthedistributionofkp,5.

iii)Importdemand(exportopportunity)

Theremustbeanactualglobaldemand,hence,favorabledemandconditions,fortheseproducts.Theimportdemandin2018/2019isusedasanindicatorofmarketopportunitywhichis,byassumption,thesameforeachcountry.

Toestimatetheproduct’sexportpotential,thispaperusesamonetizedtypeofoverlapindexdesignedtomeasurethedegreetowhichthepotentialnewexportsofonecountrymatchtheexpandingimportmarketsofanother(Freire,2017).Themethodassumesthattherearegoodopportunitiesfortradeexpansiontowardstheseproductsbasedonthepastgrowthrateoftheirimportmarkets.

7