accessing market opportunities: quality and safety standards

�63

��

Accessingmarketopportunities:qualityandsafetystandards*

Crescenzo dell’Aquila and Dario Caccamisi

Introduction

TheimportanceofcompetitivenessofCaribbeanagrifoodsectorsisrelatedtothefactthattradeliberalizationwouldnotresultingrowthinruralareasand increasedfoodsecurityunlessdomesticproducersandtradersareabletotakepartinincreasedtradingopportunities.Qualityandsafetystandardsareamongthemanyfactorsaffectingcompetitivenessinagrifoodtrade,andhavebecomeincreasinglyimportantinthelastdecadeasmajordimensionsofbothtradepolicyandprivatemarketingstrategies.

On one hand, the trade policy of developed countries is increasinglyproviding legal ground for safety and quality standards, which are meantto protect consumer rights to safe food and accurate information aboutthe characteristics of food products. These standards sometimes end upfunctioning as disguised trade barriers, discriminating against foreignproviders, and there is a need for multilateral control over such rules andcustoms practices. On the other hand, the private sector– in particular,major retail chains– is developing and implementing private quality andsafety standards, which can also function as entry barriers that are evenmorerestrictivethantradepolicymeasures.Furthermore,safetyandqualityschemes (based on, respectively, the World Trade Organization (WTO)

* This chapter draws upon work prepared for the regional workshop on Use of produce qualityandfoodsafetyprinciplestoenhancethemarketingofagriculturalandfoodproductswithinandoutsidetheCaribbean(PortofSpain,TrinidadandTobago,9–13October2006),andwascarriedout under the FAO projects “Promoting CARICOM/CARIFORUM Food Security” (GCP/RLA/141/ITA)and“SupporttotheRegionalEconomicOrganizationsfortheImplementationoftheirRegionalProgrammesforFoodSecurity”(GTFS/INT/928/ITA).Sections11.2.2,11.3.2and11.3.3werepreparedbyMrDarioCaccamisi.TherestofthechapterwasdraftedbyMrCrescenzodell’Aquila.

�64

Agricultural trade policy and food security in the Caribbean

Agreement on the Application of Sanitary and Phytosanitary Measures(SPSAgreement)andtheAgreementontrade-relatedaspectsofintellectualproperty rights (TRIPS Agreement) are becoming benchmarks both fornegotiatingstrategiesinthecontextofmultilateraltradenegotiations(WTOandEconomicPartnershipAgreements(EPAs))andforinvestmentstrategiesintechnology,organizationandcapacity-buildingforpublicadministrationsand private operators. These developments, along with the shift in marketpowertowardsretailingstages,callfornewstrategiestointegratemultilevelnegotiations(maintradeplayers,non-governmentalorganizations(NGOs),privatesectorretailchains)andtobuildcapacityatallpointsofthesupplychain.

Thischapterpresentsadescriptionofvariousframeworksforaddressingquality issues, especially as they might function as entry barriers ininternational agrifood trade. It alsoprovides a reference forunderstandingfeatures and implications of the most common food quality and safetyregulationsshieldingdevelopedmarkets.Specifically,thechapter:a)introducesthelinkbetweenqualityandsafetystandardsandsupplymanagementissues;b) discusses current features of the multilateral institutional frameworkproviding technical and legal references for national legislation relevant toquality and safety of agrifood products; c) introduces the major privatequalityassuranceandcertificationschemesanddiscussestheirrelationshipswithmultilateralarrangements;andd)facilitatesawarenessamonginstitutionsandoperatorsofthegrowingrelevanceofqualityandsafetystandardsandprovideessentialreferencesfordealingwiththem.

The next section introduces food quality dimensions and requirements,placing them in the context of an export product supply chain. Section11.2 presents multilateral agrifood regulations and particularly SPS andgeographical indications (GI) requirements, as determined by the WTOagreements and accredited benchmarking organizations. Section 11.3considers themarket-driven sideof the sameprocess,providingdetailsonmajor food quality and safety assurance and certification schemes and therelationship between legal and private standards. The final section drawsconclusionswithreferencetopolicyandinstitutionalsolutionsinsupportofCaribbeanexportingsectors.

��.� Food quality dimensions and the supply chain�4�

The quality of food products is increasingly important to food industries,whether for food safety or other qualitative attributes. On the one hand,nationallegislation,withtheirsetsofpoliciesandinfrastructures,areinplacetoprotectconsumerhealthandprovidethelegalbasesforthedifferentiation

142ThemainsourceofthissectionisCARIRI/INEA(2006).

Accessing market opportunities: quality and safety standards

�65

of products (e.g. extent to which they are authentic, ethical, healthy, safe,etc.).Atthesametime,internationalagreementsandinstitutionstrytomakesanitaryandqualitystandardsobjectiveandpredictableinordernottoharmtrade.Ontheotherhand,privatehealthandqualitystandardsareincreasinglydefining entry barriers to the richer markets of developed countries. Thereality is that an agrifood industry wanting to engage international tradewill have to deal with opportunities and constraints stemming from thosestandards. All in all, both national legislation and private standards spreadresponsibility across everyone in the supply chain, including farmers andgrowers,manufacturersandprocessors,foodhandlersandconsumers.

Publicregulationshavestronglegitimacybecausepeoplehavetherighttoexpectthefoodtheyeattobesafeandsuitableforconsumption.Inaddition,internationalfoodtradeisincreasingand,alongwithitssocialandeconomicbenefits,cancontributetothespreadoffood-borneillnessaroundtheworld.Food-borneillnessandfood-borneinjuryareatbestunpleasantandcanbefatal.Therearealsoeconomicimpactswhenoutbreaksoffood-borneillnessdamage trade and tourism, leading to loss of earnings, unemployment andlitigation. Food spoilage is also wasteful, costly and can adversely affecttradeandconsumerconfidence.Effectivehygienecontrol,therefore,isvitalforavoidingtheadversehumanhealthandeconomicconsequencesoffood-borneillness,food-borneinjuryandfoodspoilage.

Privatestandardsserveagro-industryoperatorswell.Theycanhelpmeethealth or other regulations (such as national SPS, geographic indications,organic or fair trade) and facilitate marketing strategies that emphasizeproductdifferentiationlinkedtohigherorstricterproductattributes.Fromtheprivate sector standpoint, foodqualitycanbeconsideredacomplexofcharacteristicsthatdeterminesitsvalueoracceptabilitytoconsumers,whilefoodsafetyisabasicrequirementoffoodquality.Foodsafetyimpliesabsence,orsafelevels,ofcontaminants,adulterants,naturallyoccurringtoxinsoranyother substance that could make food injurious to health on an acute orchronicbasis.Qualityattributesalsoinclude:nutritionalvalue;organolepticpropertiessuchasappearance,colour,textureandtaste;functionalpropertiesandsymbolicfeatures(FAO,2000).

Both public and private standards based on safety and quality can beconsidered entry barriers (Porter, 1980) when they create the possibilityofhigherrevenues for firmscapableofsellingproductsconsistentlytothegivenstandard;thesefirmscantherebydefineandbenefitfromnewsourcesof competitive advantage. A summary representation of quality and safetystandardscommonlyfacedbyagrofoodsupplychainsforexportmarketsisshowninFigure ��.�.

The emerging relevance of both public and private quality and safetystandards increases pressure on the different actors of the supply chain.Thisisduemainlytoincreasingcostsofcomplyingwithsafetyandquality

�66


standards.Thesecostsarisefromtechnicalandmanagerialrequirementsandadaptationofstrategies–thatis,theneedtocomplywithsystemsofquality control(forthedetectionofdefects)and quality assurance(forthepreventionofdefects),withinwiderquality management systems.

Quality assurance (QA) covers a range of activities related to the life ofthe product: design, development, production, installation, servicing anddocumentation. It includes the regulation of the quality of raw materials,assemblies, products and components; services related to production; andmanagement, production and inspection processes. According to qualitymanagement practice, the main goal of QA is to ensure that the productfulfilsorexceedscustomerexpectations.

Moreover, the adoption of quality standards is increasingly documentedthroughvoluntarycertificationofabusiness.Certificationindicatesthat,intheviewofthecertifyingbodies,thebusinesshasaspecificsetofknowledge,

FIGURE11.1Quality dimensions and the supply chain

Farmers Processors

Retailers

Consumers Traders Service Companies

Production / Pre- Post -harvest stages

Quality assurance/ certification schemes:

• Legal standards (SPS, TBT, Codex, National legislations)

• GAP, GAP based schemes (EurepGAP), GHP, other private standards (product), ISO 22000(2005), traceability

• Geographic Indications

• Organic, Fair trade • SA• 8000

• GAP, GAP based schemes (EurepGAP)

• • Traceability

Supply chain

Quality schemes

Quality dimensions:

Product attributes • Food safety

• Commercial quality

Process attributes

• Environmental issues

• Process organization, information and documentation

Quality assurance/ certification schemes:

• Legal standards ((SPS, TBT, Codex, National legislation)

•GMP/HACCP, GMP based schemes (BRC) GHP, other private standards (packaging/ labelling/ storing/transport), ISO 22000(2005), traceability

• ISO 14001 (2004)

• ISO 9001(2000), traceability

Input

Quality dimensions:

Product attributes • Food safety

• Commercial quality

Process attributes • Produce Origin

• Method of production

• Ethical Issues (worker’s welfare)

• Environmental issues

• Process information and documentation

harvest stages

Source: CARIRI/INEA(2006)


�67

skillsorabilities.Althoughvoluntary,certificationisoftenrequiredbylargeretailchainsoperatinginmoredevelopedmarkets.Certificationneedstoberenewedperiodically(althoughforGIscertificationmaybepermanent–seelatersection).Certificationbodiesarebusinessorganizationsor, lessoften,professional bodies or non-profit organizations. (Sometimes the latterexist primarily to offer a particular certification.) Whatever its nature, thecertifyingbodydeterminesthepoliciesofthecertificationprogramme.Legalandprivatestandards,andtheirrelationshipswiththemultilateralframeworkprovided by the WTO and accredited benchmarking organizations, arediscussedinmoredetailinlatersections.

��.� Multilateral regulation of safety and quality standards

TheCodexAlimentarius143andotherWTO-accreditedorganizations(suchasOIE144,IPPC145andothers146)elaboratebenchmarkingstandardstoguidegovernments inworkingouttheirownnationalstandardsinaharmonizedway,soastofacilitateinternationaltradeinagriculturalandfoodproducts.The Uruguay Round (UR) of General Agreements on Tariffs and Trade(GATT) negotiations (1994), and specifically the WTO Agreement on theApplication of Sanitary and Phytosanitary Measures (SPS Agreement)formally recognized the Codex standards and other recommendations asbenchmarks for international harmonization. The Agreement on TechnicalBarrierstoTrade(TBTAgreement)alsorecognizesCodexstandards.Theseagreements contribute to defining standards relevant for food companiesandalsoserveasthebasictextsforresolutionoftradedisputes(FAO,2000;FAO,2005a).

The SPS Agreement deals directly with trade-related sanitary and

143TheJointFAO/WHOCodexAlimentariusCommissionwassetupin1962toprotectthehealthofconsumersandensure fairpractices in foodtrade. It isan intergovernmentalbodyengaged inpreparing international foodstandardsandotherrelevantrecommendationsthatpromotequalityandsafetyoffood.Codexcanbeattributedwithover200foodstandards;nearly3000maximumresiduelimitsforpesticides,veterinarydrugs,mycotoxinsandenvironmentalcontaminants;codesofhygienicpractices;ageneralstandardforfoodlabelling;acodeofethicsforinternationaltradeinfood;andawiderangeofguidelinesandrecommendationsforgovernmentsandindustry.

144The World organization for animal health (OIE) is an intergovernmental organization createdin 1924. To ensure transparency in the global animal disease situation, each Member Countryundertakestoreporttheanimaldiseasesthatitdetectsinitsterritory.

145IntergovernmentalPanelonClimateChange(IPCC)wasestablishedbytheWorldMeteorologicalOrganization (WMO) and the United Nations Environment Programme (UNEP) to assessscientific, technical and socio-economic information relevant for the understanding of climatechange, itspotential impactsandoptionsforadaptationandmitigation.It isopentoallMemberStatesoftheUnitedNationsandtheWMO.

146Although not mentioned in the Agreement on Agriculture (AoA), organizations such as theInternationalStandardsOrganization(ISO)haveachievedasimilarstatus;seelatersection.

�68


phytosanitary measures for protection of human health. Its principalobjective is to minimize the negative effect on trade from the adoptionandenforcementofSPSmeasures.WTOMemberStatesareencouragedtoadopt internationally recognized standards (if these exist), but are free toapplystricterstandards.Thelatterareallowedconditionalontheprovisionof scientific justification for the measures and the implementation of riskassessmentmechanisms(FAO,2005b;Wilson,2003).

The TBT Agreement addresses “product characteristics or the relatedprocesses and production methods” reflected in technical regulations andrequires that these regulations conform tobasicprinciplesof transparencyand non-discrimination. It seeks to ensure that technical regulations andstandards, including packaging, marking and labelling requirements, andanalytical procedures for assessing conformity with technical regulationsand standards, do not create unnecessary obstacles to trade. Relevantinternational standards developed by bodies such as the InternationalStandards Organization (ISO), if they exist, must be used as the basis fortechnicalregulations,unlessthiswouldbeinappropriatebecauseofclimatic,geographic or technological factors (FAO, 2005b; Wilson, 2003). WTO-accredited standards setbyOIE, IPCC, ISO,etc. arevoluntary,becomingcompulsoryonlywhenrequiredbynationallegislation.

Other WTO-accredited standards are rooted in the UR Agreement ontrade-related aspects of intellectual property rights (TRIPS), which callfor application of some fundamental WTO principles (i.e. most-favourednations (MFN) status) in that field. In the section on TRIPS, entirelydevotedtoprotectionofgeographicalindications(GIs),theaimisexplainedas providing institutional guarantees for the competitive advantages andrevenues that a product derives from its reputation and traditions relatedmainly to geographic origin (De Filippis and Salvatici, 2006; WTO, 2000;WTO,2002).

��.�.� Sanitary and phytosanitary measures (SPS)SPS regulations are an area of increasing importance in national tradepolicies and international efforts for harmonizing trade rules. Regulationof foodsafety,aswellasanimalandplanthealth, isevolvingrapidly inallcountries.While some trends in regulationare consistentwithminimizingtradedistortions,thegeneralorientationtowardsmorestringentregulationofawiderrangeofrisksandqualityattributesraisesnewpotentialbarrierstoagriculturaltrade.Foodsafetyregulationsandstandardsevolvedifferentlyaround the world as countries respond to food safety crises and preparefor perceived exposure to emerging food safety risks. These differences inregulationsandstandardscanleadtointernationaltradeconflictsordisputesandcanultimatelyaffectglobalpatternsoffooddemandandreducetrade.These trends are often entwined with increased consumer demand for


�69

credenceattributesof foodproducts ingeneral,becausequalityandsafetyare often jointly produced (Unnevehr and Roberts, 2003; Wilson, 2003;Buzby,2003).

In addition, non-traditional agricultural exports from less developedcountries – particularly of fresh and minimally processed products – todevelopedcountriesaregrowingrapidly.This tradearises inpart fromthedecreasedrelevanceoftraditionalprotection(tariffsandquotas)forsomeofthese commodities, such as seafood and tropical fruits. But such productsfrequentlyhavehighrisksforcertainSPShazards,whichmaybeexacerbatedbytradeoverlongdistances.Asdevelopingcountriesworktomeethigherandevolvingfoodsafetystandards,theyhaveraisedconcernsaboutwhetherthe increasing standards will impede their participation in world trade(UnnevehrandRoberts,2003;FAO,2004a;Hensonet al.,1999;Athukoralaand Jayasuriya, 2003). Moreover, the private sector is evolving rapidly tomeet demands for process attributes throughout the world, in many casessettingstandardsthatarehigherthanpublicones(Caswellet al.,1998;Lee,2006).Theseefforts frequentlyaffect international trade,especiallyexportsfromdevelopingcountries,exacerbatingtheotherdifficulties.

Taken together, these trends in the international food system posecontinuing challenges to the SPS Agreement, as well as to efforts toreducebarrierstoagriculturaltradeandimprovethetradeperformanceofdevelopingcountries.AlthoughtheWTOasamechanismoflastresortfordisagreementsover such technicalbarriershasmademuchprogress since1995,SPSmeasuresarestillacontentiousfield,duetothewideroomleftbySPSprinciplesandWTOarrangementsforgovernmentstoimposeadhocmeasuresrestrictingmarketaccess.Moreover,duetoincreasingmultilateralconstraints on traditional tariff and non-tariff barriers, the number ofSPS measures is increasing while gains made on traditional trade policymeasuresarereducing.

Forinstance,phytosanitarycontrolsimposedbyimportersarecurrentlylimiting developing country exports of fresh fruit and vegetables. Thesecontrols are particularly stringent in the United States, Australia andJapan.Between1995and2000,nearly270SPSmeasureswere introducedagainst imports of fresh fruit and vegetables worldwide (FAO, 2003a).Thus,amajorhindrancetofreshproducetradeisthelackofharmonizedtechnical standards and treatments for exports. Some countries apply theCodex Alimentarius for maximum residue limits (MRLs) on pesticides,whileothersapplytheirown,oftenstricterMRLsthatmayonlypartiallyconform to the Codex. Another difficulty arises from setting MRLs atthe laboratory limit of determination, as this often makes verification ofcompliancedependentonverycostlymodernanalyticalmethods.

Quarantine regulations are another serious impediment and measuresdesignedtopreventbio-terrorismarelikelytoincreasetheadministrative

�70


and regulatory burden on exporters of fresh fruits and vegetables inparticular. Moreover, developing countries exporting tropical fruit faceserious challenges in meeting the phytosanitary regulations of importingcountriesduetotheunevenphasingoutofmethylbromide.147

Thereareblurredprofilesintheimplementationofallthebasicprinciplesof the SPS agreement, which leaves room for their use as disguised tradebarriers.Theseprinciplesare:1.Harmonizationof rules.Membersshouldusecommoncriteria(international

standards,guidelinesandrecommendations)tosetupSPSmeasuresandin all cases these measures should be justified scientifically. The aboveexamplereferringtowidelyrangingrestrictionsonuseofmethylbromideshowshowcountriescancontinuetoestablishrequirementsthatarenotinlinewithinternationalguidelines.

2. Equivalence of measures. WTO members should recognize anothercountry’s SPS measures as equivalent to their own if those measuresprovideanappropriatelevelofprotection.Inspectionsandaccreditationby public officers from the importing country are often the only wayaroundSPSbarriers,nomatterwhatthelevelofcontrolsanderadicationmeasures carried out by phytosanitary authorities in the exportingcountries. Moreover, costs relating to inspections are usually borne byexporters.

3.Appropriate level of protection.Thisprincipleisoftenchallengedbecauseitisthoughttobeviolatedthroughtheoutrightinterdictiononallimportsofcertainfruitsandvegetablesappliedbymanycountries,rangingfromthe United States, India, China, Japan and Australia, to Mexico andChile, to many others including some Caribbean countries and othersmall island developing states (SIDS). These countries ban all foreignfruits and vegetables from their territories, unless a lengthy and costly“importriskanalysis”or“pestriskanalysis”hasshownthattheimportsdo not constitute a risk to consumer or plant health. This practice istantamounttoareversalofthe“burdenofevidence”.RatherthansettingSPSrequirementsasafunctionoftheriskpresentedbycertainimportsofplantproducts,thesecountriesobligeexporterstodemonstratethattheirproductsaresafe.

4. Non-discrimination. SPS measures must not unjustifiably discriminatebetweencountrieswheresimilarconditionsprevail, and imports shouldbe treated no differently from domestic produce. However, a numberof countries maintain differentiating requirements when dealing with

147Fumigation with methyl bromide greatly affects the quality and shelf life of produce and is stillrequired by many countries for fruits and vegetables imported into their territory. Its use is incontradictionwithinternationalguidelinesestablishedthoughtheMontrealProtocol(1987)signedbyUnitedNationsmembers,whichforeseesthetotaleliminationoftheuseofmethylbromideby2015,inviewofitstoxicityandharmfuleffectsontheozonelayer.


�7�

domesticproduce.Forinstance,theUnitedStatesappliesstrictermaturitystandardsandtolerancelevelstoimportedItalianbloodorangesthanforitsdomesticallyproducedbloodoranges:importedorangesmusthaveanaciditylevelofatleast9°Brix,whereasthestandardfordomesticproduceissetat8°Brix.

5. Transparency. Countries should be required to notify trading partnersof changes in their SPS measures to allow them to adapt to the newmeasures.Delayednotifications,andfrequentandsuddenchangesinSPSrequirementsforimports,arecommon.

6. Regionalization. This principle stipulates that countries should not banimports of plant or animal products from pest- or disease-free areas.However, there are frequent impediments to importing produce frompest-ordisease-freeareaswithincountriesthatarenotentirelypest-ordisease-free.

DespitetheattemptintheUruguayRoundtoprovideadurablemultilateralframework to regulate the use of food safety and quality regulations, aremarkable divergence of views has emerged about this framework in theDoha Round trade talks. Developing country proposals signal frustrationwith the increasingly exigent standards faced by their exports, or the newobligationstojustifytheirownregulatoryregimes,orboth.Thesubstantialcosts facing some developing countries in meeting SPS standards in high-income markets reduces their potential gains from trade and confirms theconcernabouttheirfurthermarginalizationininternationaltrade,regardlessofprogressmade in reducingother tradebarriers (Unnevehr andRoberts,2003;Hensonet al.,1999).

��.�.� Geographical indications (GIs)DefinitionsArticle22.1ofTRIPSdefinesGIsas:“...indicationswhich identifyagoodasoriginatingintheterritoryofamembercountryoraregionorlocalityinthatterritory,whereagivenquality,reputationorothercharacteristicsofthegoodisessentiallyattributabletoitsgeographicalorigin”,148thusrecognizingadirectlinkbetweenqualityoffoodstuffsandtheirorigin.AGIisusuallythenameof a specific location, althoughat times traditionalnamescanbenon-geographical,suchastheGreekcheese“feta”.Article22providesthatGIsforallgoodsmustbeprotectedagainstmisuseandestablishesaminimumstandardofprotectionforallGIs,whateverthenatureofthegoodtowhichitisapplied.ScopeofsuchprotectionislimitedtotheprohibitionoftheuseofGIsbyproducersnotlocatedintheregiondesignatedbytheparticularGI.

Inlinewithmarketingprinciples,fromthestandpointofconsumers,GIsare meant to prevent their being misled about the origin and production

148See,amongothers,WTO(2002)andECRegulations2081/92and2082/92.

�7�


methods (and therefore process attributes) of the product. Consumersare thus assured that they receive the genuine article they pay for. Fromthe standpoint of producers, GIs are a means of branding outputs, whichincreasesproductdiversificationandproducerincome.149

Debate and initiatives on GIsThe WTO Doha meeting in November 2001 agreed to negotiate theestablishmentofamultilateralsystemofnotificationandregistrationofGIsforwinesandspiritsby theFifthWTOMinisterialConference.However,theConferencewasheldinCancuninSeptember2003withoutanyformaldiscussionontheissue.WTOmemberscannotagreeonwhethernegotiationswerealsomandatedforextendingprotectiontoproductsother thanwinesor spirits and on the adoption of a multilateral register of protected GIs.Two proposals underpin the debate on this latter issue: a) a register withlegaleffects(EuropeanUnion(EU)plus17WTOmembers(1998));orb)aregisterforinformationpurposesonly(UnitedStatesplus16WTOmembers(1999))(DeFilippis-Salvatici,2006).

The EU, Switzerland and other WTO members strongly support theclaimthatunauthorizeduseofGIs isharmfultoconsumersandlegitimateproducersandthat increasedmarketaccessneeds togohand inhandwithenhancedGIprotection.EUobjectiveswithintheWTOdebateonGIsare:

• toobtaineffectiveprotectionagainstusurpationofnamesinthefoodandbeveragessector;

• tomakemarketaccesseffective,byensuringthatproductsthathavetherighttouseacertaindenominationarenotpreventedfromusingsuchanameonthemarket;and

• toensureconsumerprotectionandfaircompetitionthroughregulationoflabelling.

ProtectionofGIsby theEU isongoingandbasedon theconclusionofbilateralandregionalagreementsonprotectionofintellectualpropertyrightsforwinesandspiritsandpreventionof fraud in theuseofproductnames.Thoseagreements linkconcessionsonaccess to theEUmarket fromthirdcountrieswiththeprotectionofEUGIsinthosecountries.150

149Aspreviouslysuggested,GIshelpproducersobtainapremiumpricefortheirproducts.AccordingtoarecentstudyoftheFrenchmarket,GIcheesepriceshoveraround+30percentandareupto+230 percent for wines. GIs can also positively affect the position of agricultural producers insharingvalueaddedinthesupplychain:accordingtothesamestudythepriceofmilkforGIsis100percenthigherthanmilkusedforothercheeses.

150Examples of specific agreements on GIs are: EC–Australia (wines, 1994), EC–Mexico (spirits,1997),EC–SouthAfrica(winesandspirits,2002),EC–Canada(winesandspirits,2003).Examplesof general agreements with specific section for GIs are EC–Chile (wines and spirits, 2002) andEC–Mercosur(SouthernCommonMarket),whichforeseesimprovedaccesstoEUmarketversusadequateprotectionofallEUGIsinMercosurcountries(negotiationsstillongoingin2006).


�73

GIs can have several positive effects. They can be an excellent means topromoteruraldevelopment,becausetheyhelpproducersobtainapremiumpricefortheirproducts,allowforabetterdistributionofthevalueaddedtoagriculturalproducers,bringvalue to theregionoforiginandcan increaseproduction and create local jobs, thus preventing rural exodus. GIs canbe an effective marketing tool, as they encourage variety and diversity ofproduction, and allow producers to market differentiated products, withclearly identifiable features. They are a tool to preserve local know-how,natural resources and biodiversity and can play an important role in localculture, contributing to social cohesion (helping local producers worktogether) and raising the profile of local and national identity (makingproducers and consumers proud of their traditions). They can have otherpositiveindirecteffects,suchaspromotingtourism.

GIscanalsohaveshortcomingsandproblemsofimplementation.ProtectingtraditionalproductsthroughGIsiscostlytogovernments,becausetheyrequiremorequalifiedextensionservicesandmoreandbettercontrols;toproducers,becauseofincreasedcostsforinspectionsystems;andtoconsumersbecausethey have more information to gather and process. Moreover, problems ofrecognition of GIs arise on cultural grounds as many names of productshavetravelledwithemigrants,whowouldliketocontinuetomakethesameproductsandusethesametermstoidentifythoseproducts.

��.3 Market-driven agrifood regulations and quality assurance and certification schemes

��.3.� Private and legal standardsWTO agreements and accredited benchmarking organizations provide areferencenotonlyfornationalregulations,butalsoforprivate,commercialstandards. There are many reasons for retail chains and some producersto create or develop their own standards. Retail companies may requireprivate food certification of their national and third-country suppliersto assure consumers that the products they sell are safe and to shield thebusiness from liability in case of unsafe foods sold through their outlets.Specificgoodagriculturalpractices(GAPs)orgoodmanufacturingpractices(GMPs) standards, such as those of the Euro-Retailer Produce WorkingGroup(Eurep)andBritishRetailConsortium(BRC)certificationarecasesin point. Producers can develop standards related to particular productionprocesses, or to raw materials – often linked to local attributes of theproductionprocessor input,or toethicalvaluesorhealthconcerns.Thesequality standards target the willingness of customers to pay a premiumpricefor“authentic”,“traditional”orhigh-qualityfoodproducts.Theyarebased upon international agreements and/or national legislation protectingnames of products belonging to particular regions, obtained by particular

�74


production processes or marketed through contractual arrangements thatprotectsmallproducers.

Private standards are part of commercial agreements between voluntaryparties in a free market, and as such are not subject to state interventionand fall outside the jurisdiction of the WTO. This is the case of the FairTradestandardoftheNGOFairtradeLabellingOrganizationsInternational(FLO-I) and the Eurep and BRC global standards. Whenever a voluntarystandard is taken into consideration for (full or partial) inclusion within acountry’slegislation,151thestandardshouldnotconflictwithSPSandTBTAgreements. In other words, it should not become a disguised barrier totrade(Wilson,2003;FAO,2004b;FAO,2003b).

Private,voluntarystandardscanhaveaverystrongimpactoninternationaltradeasentrybarriers, thisbecauseprivate sectoroftensets standards thatsupersedepublicones (Caswell,Bredahl andHooker, 1998).Theseprivatestandardsfrequentlyaffectexportsfromdevelopingcountries,exacerbatingtheir problems for greater involvement in international trade. Costs ofcompliance with these private standards may be high, and many suppliersin developing countries, especially small farmers, cannot afford the luxuryof private certification. Some developing countries have complained aboutprivate standards constituting de facto SPS barriers to more developedmarkets and ask the authorities of the United States, the EU and othercountries to address this concern (Lee, 2006; Wilson, 2003; Unnevehr andRoberts,2003;Hensonet al.,1999).

Moregenerally, foodcompaniesare finding itdifficult tosimultaneouslymanageoverlappingqualitystandards, suchas theonesdiscussedbelow.152Theyareeitherbecomingtooexpensiveor,asaresultof“simplifying”effortsmadebyso-calledglobalstandards,mayunderminetheefficiencyofthefoodcompanies’qualitystrategiesanddrivethosestrategiesunderthecontroloflarge,multipurposeretailchains.

��.3.� Quality assurance and certification schemesThe hazard analysis and critical control point (HACCP)The HACCP is a system that identifies, evaluates, and controls hazardssignificantforfoodsafety.Itsadoptioniscompulsoryinsomecountriesandvoluntary in others.153 HACCP implementation is meant to be guided byscientificevidenceofriskstohumanhealth.Itidentifiesspecifichazardsandmeasuresforcontrollingthembyfocusingoncriticalcontrolpoints(CCPs)

151SuchasinthecasesoftheEU,theUnitedStatesDepartmentofAgricultureorJapanAgriculturalStandardorganicstandards.

152FrenchandGermanretailershaverecentlydevelopedtheirownqualitystandard,theInternationalFoodStandard(IFS).Itsmanagementmaybeevenmorecomplexthanforthestandardsreviewedbelow.

153FormoredetailsontheHACCPmethod,seeCodexAlimentariusCommission,2003.


�75

alongtheproductionflowfromprimaryproductiontofinalconsumption.154Redesignofanoperationshouldbeconsideredifahazardis identifiedbutno CCPs are found. Any HACCP system is capable of accommodatingchange, such as advances in equipment design, processing procedures ortechnologicaldevelopments.

HACCPisatooltoensurethesafetyoffoodbyfocusingonpreventionrather than relying mainly on end-product testing. HACCP can provideother significant benefits in terms of control and improvement of theproduction flow, improvement of working conditions and reduction ofproduction costs. Moreover HACCP can aid inspection by regulatoryauthorities and promote international trade by increasing confidence infood safety. Successful application and implementation of the HACCPsystem to any stage of the food chain requires the full establishment ofprerequisiteprogrammes,suchasgoodhygienicpracticesaccordingtotheCodexofGeneralPrinciplesofFoodHygiene,andtheappropriateCodexpractice and food safety requirements (Codex Alimentarius Commission,2003).Successful implementation requires trainingandamultidisciplinaryapproach, including expertise as appropriate in agronomy, veterinaryhealth,production,microbiology,medicine,publichealth,foodtechnology,environmental health, chemistry and engineering. The application ofHACCP is compatible with the implementation of quality managementsystemssuchastheISO9000series(seelatersection),andisthesystemofchoiceinthemanagementoffoodsafetywithinsuchsystems.

EurepGAPEurepGAPisaqualitystandardthatbeganin1997asaninitiativeof largeEuropean retailers belonging to Eurep to respond to consumer concernsabout food safety, environmental protection, workers’ health, safety andwelfareandanimalwelfare (madcowdisease,useofpesticides,geneticallymodifiedorganisms(GMOs),etc.).TheobjectivewastodevelopvoluntarystandardsandproceduresfortheglobalcertificationofGAPs.Itworksby:

• encouragingadoptionofcommerciallyviableFarmAssuranceSchemes,whichpromotetheminimizationofagrochemicalinputs;

• developing a GAP framework for benchmarking existing assuranceschemesandstandards,includingtraceability;

• providing guidance for continuous improvement and the developmentandunderstandingofbestpractice;

• establishingasingle,recognizedframeworkforindependentverification;and

• communicating and consulting openly with consumers and partners,includingproducers,exportersandimporters.

154CCPisastepatwhichessentialcontrolscanbeappliedtopreventoreliminateafoodsafetyhazardorreduceittoanacceptablelevel.

�76


Collaboration between retailers and producers resulted in a protocolfor independent, recognized third-party certification of farm productionprocesses, which farmers around the world can use to demonstrate theircompliancewithGAPs.EurepGAPcertificationcoversfruitandvegetables,155flowersandornamentals, integrated farmassurance, aquacultureandgreencoffee.Theschemecoversthewholeagriculturalproductionprocessofthecertifiedproduct, frombeforetheplant is intheground(seedandnurserycontrol points) to non-processed end product (produce handling controlpoints). It also works to establish awareness and responsibility regardingsocial issues and animal welfare criteria for farms. EurepGAP stressesthe importance of residue screening, setting up a standard on MRLs anddeveloping guidance notes to help farmers and growers be better able todemonstratethattheirproducemeetsdestinationMRLrequirements.

BRC global standardsThedevelopmentoftheBRCglobalstandardswasinitiallydrivenbytheneedtomeetthelegislativerequirementsoftheEUGeneralProductSafetyDirectiveandtheUnitedKingdomFoodSafetyAct.Itestablishedastandardforthesupply of food products and acted as evidence for UK retailers and brandowners todemonstrate “duediligence” in the faceofpotentialprosecutionby the enforcement authorities.156 The BRC standard is comprehensive inscope,coveringallareasofproductsafetyandlegality,includingsuchcriticaltopics as the HACCP system, quality management, factory environmentstandards and product and process control. Major business benefits derivefromcustomerconfidencelentbytheBRCcertification.

Each standard is developed under the leadership of the BRC and itsmembers; it is extensively revised to reflect changing EU legislation andcontinuously develops best practice requirements. The use of the BRCstandard is legally voluntary, but strongly recommended for those foodproducers thatarewilling to supply theBritishmultipurpose retail chains.The2005editionincludedchangesinlegislationrelatedto:

• traceabilityoffoodcomponentsthroughthesupplychain;• ensuringthatfoodcomponentsremainuncontaminatedbyotherelements

(importantwhenallergenslabellingisastatutoryrequirement);• foodproductsuppliersbeingabletoadvertisethatfarmedgoodsintheir

productscomefromaparticularsource;• ensuringthatguidelinesgoverningvariousprocessesinthemanufacture

155The normative document for EurepGAP Fruit and Vegetables certification was developed by agroupofEuropeanrepresentativesfromthefruitandvegetablessector,withthesupportofproducerorganizationsoutsidetheEU.Thestandardcoversallfresh,unprocessedagriculturalproductsofplantorigingrownforhumanconsumption.Itdoesnotcoverherbsorplantsexclusivelyusedformedicinalpurposesorfortheiraromaticattributes.SeeEurepGAP,2004.

156ForadditionaldetailsonBRCstandardseeLee(2005)andtheBRCWebsite(www.brc.org.uk).


�77

offoodproductsaresufficientlyrobust;and• whatsupplierscansayintheircommunicationstoinformthebusiness

communityabouttheirBRCcertification.Thestandardhasbecomeglobalandisnowusedbysuppliersfromaround

the world. BRC global standards have been designated for packaging andare being developed for identity-preserved non-genetically modified foodingredients and consumer products. The BRC and Institute of Packaging(IOP) developed the packaging standard, which provides a common basisfor auditing companies supplying packaging for food products to retailersand assists retailers and food manufacturers in the fulfilment of their legalobligations.157

ISO standardsTheInternationalStandardsOrganization(ISO)isanetworkofthenationalstandards institutes of 157 countries, with one member per country anda central Secretariat in Geneva. It occupies a special position between thepublicandprivatesectors:whilemanyofitsmemberinstitutesarepartofthegovernmentalstructureoftheircountries,othermembersarefirmlyrootedintheprivatesector.ThishelpsenableISOtoreachconsensusonsolutionsbridgingtherequirementsofbusinesswiththebroaderneedsofstakeholdergroupslikeconsumersandassociations.

TheISO9000andISO14000familiesareamongitsmostwidelyknownstandards, implemented by 760900 organizations in 154 countries.158 ISO9000helpsorganizationsmeetcustomerqualityrequirementsandapplicableregulatoryrequirements.ISO14000helpsorganizationstominimizeharmfuleffects on the environment caused by their activities and to improve theirenvironmental performance. While most ISO standards are highly specificto a particular product, material, or process, the ISO 9000 and ISO 14000standards are “generic management system standards” because the samestandardscanbeappliedtoavarietyoforganizations.

��.3.3 Food traceability DefinitionsISO defines traceability as the “ability to trace the history, application, orlocationofthatwhichisunderconsideration”.Gellyncket al.(2005)referto“theinformationnecessarytodescribetheproductionhistoryofafoodcropandanysubsequenttransformationorprocessthecropmightundergoonits

157Themain sectionsof the standardare: a) scope;b)organization; c)hazardand riskmanagementsystem; d) technical management system; e) factory standards; f) contamination control; g)personnel;h)riskcategorydetermination;andi)evaluationprotocol.

158MoredetailsontheISOnetworkcanbefoundat:http://www.iso.org/iso/en/aboutiso/introduction/index.html#twoandhttp://www.iso.org/iso/en/iso9000-14000/index.html.

�78


journeyfromthegrowertotheconsumer’splate”.Informationonfoodscanbetracedforwardandbackateachstageofthefoodchain,i.e.production,preparation/processing,distributionandsale.The“traceabilityofaproduct”relatestosourcesofmaterialsandparts,aswellasthehistoryofprocessing,post-shipmentdeliveryandexistenceoftheproduct.159

Thedefinitionoftraceabilityforfoodisnecessarilybroadbecausefoodisacomplexproductandtraceabilityisatoolforachievinganumberofdifferentobjectives. The logistics objectives of traceability (for example, proceduresfor withdrawing food products unfit for the market) can be linked tomarketingobjectivestoallowtargetingspecificmarketsegmentsandtoassureconsumersabouttheoriginandqualityoffood.Foodtraceabilityislinkedtobothconsumersafety issues(foodsafety,bioterrorism,consumer’sright toknow)andtothemarketingandinvestmentbehaviourofproducers.

Traceability can be divided into tracking and tracing. Tracking refers tothelocationofitemsastheymovethroughthesupplychain.Tracingrelatestotherole,compositionandtreatmentofafoodproductduringthevariousstagesofproduction.Thus,“traceabilitycanbedescribedasacombinationoftheflowofsubstancesandofinformation.”160

A traceability system is composed of an organization, a system and aprocess,documentedprocedures,resourcemanagement(personnel,financialresources, machinery, equipment, software, technologies and techniques),rulesandeducation,andtraining.Keyconceptsoftraceabilityare:

• identificationofsupplychainparticipantsandproductsalongthevariousstagesofthesupplychain;

• recordingofrelevantinformationonmanufacturinganddistributionofaproduct;

• identificationofconsistentproductbatches;and• in-factorytracingofrelevantinformationthatisrelatedtotheidentified

productbatches.

PurposesBytrackingand tracing foodand its informationateachstageof the foodchain,traceabilitysystemscanachievethefollowingpurposes:i) Greater reliability of information. Traceability systems can secure the

transparency of distribution routes; the quick provision of informationto consumers, customers and government agencies; and the matchbetweentheproductand its label.Asaresult, thesystemhelpspreventmisidentificationoflabelsandinformationandmakestransactionsfairer.

159MoredetailsontraceabilitysystemscanbefoundinGolan,KrissoffandKuchler(2004).160ISO9000:2000 incorporates theprevious ISO8402:1994standardandprovidesa specific section

coveringtraceabilityandproductidentificationrelatedissues(ISO,n.d.).


�79

In particular, the systems enable consumers to get correct informationaboutfoodanditssuppliers,makegooduseofthisinformationwhentheybuyfoodproductsandtakestepstopreventrisks.Thesystemsalsoenablethecustomerandthecompetentgovernmentagenciestoobtainaccurateinformation for product and risk management purposes and help foodbusinessoperatorsincreasethereliabilityoftheirproducts.

ii) Contribution to food safety.Traceabilitysystemscanhelptracequicklyandeasilythecauseofaccidentsrelatedtofoodsafetyandhelpremoveafoodproductproblempromptlybyzeroinginontheproductandtracingit to itsdestination.Thishelpsminimizebothdamage to the consumerandeconomic lossalong theentire foodchain. Inaddition, thesystemsmake it easier to collect data about unexpected impacts on health andlong-termeffectsandhelpdevelopriskmanagementtechniques.Finally,theyhelpdefinetheresponsibilityoffoodbusinessoperators.

iii) Contribution to achieving higher levels of business efficiency.Traceabilitysystemshelpincreasetheefficiencyofproductmanagement(includinginventory)andqualitycontrolbyusingidentificationnumbersandbystoringandofferinginformationabouttheoriginsandcharactersofproducts.Thiscontributestocost-savingandimprovementinquality.

Inmostcases, thepurposes listedabovearepursuedsimultaneously,buttheirprioritymaybedifferentdependingonproductcharacteristics,stateofthefoodchainorconsumerdemand.Afoodbusinessoperatorwillconsiderthesefactorswhenbuildingatraceabilitysystem.

Costs and limitationsWhiletraceabilitysystemsareeffectivetools,theymayhavelimitationsandproblems. Traceability systems are generally too complex to be complete.Even a hypothetical system for tracking beef – in which consumers scantheirpacketofbeefatthecheckoutcounterandaccesstheanimal’sdateandlocationofbirth,lineage,vaccinationrecordsanduseofmammalianproteinsupplements–couldbeconsidered incompletebecause itdoesnotprovidetraceabilityofbacterialcontrolinthebarn,useofgeneticallymodifiedfeedoranimalwelfareattributessuchashoursinthebarnoratpasture.Therearebothtechnicalandeconomicreasonsforsuchlimitations.

Technical reasons limiting traceability include the differing scope ofapplicationsaccordingtothecharacteroftheproduct,workorsector,aswellasthevariousfactorsdeterminingefficiencylosses.Applicationsareaffectedbythenatureandstateofrawmaterials, lotsize,cargocollection,divisionandtransportationmethod,productionandmanufacturingmethod,packingmethod,numberofstagesfromproductiontoretailingandscaleandnumberoffoodbusinessoperators.Efficiencylossesoccuri)whentheprocesses(e.g.orderplacement and receivingprocedures)differ among the food business

�80


operators concerned; ii) information is unreliable; iii) transmission ofinformationbetweenfoodbusinessoperatorsisdifficultorinterrupted;andiv)lotsarenon-uniform.

Attempts to track or trace food and its information more accuratelymay result inveryhighcosts.Themaincosts involved in introducingandmanagingatraceabilitysystemincludethosefor:

• drafting the basic idea and procedures necessary for construction of atraceabilitysystem;

• purchasing equipment (e.g. measuring apparatuses, informationprocessingequipment);

• managing the system, such as identification, recording, arranging andstoringinformation,educationandtraining;and

• inspectionbythethirdpartytosecurethesystem’sreliability.Food business operators must compare the objectives and effects to be

achievedwith thecosts involvedwhen they seek toestablisha traceabilitysystem.Inparticular,smallenterprisesshoulddeviseeffectivestrategiesforaccessing financial and human resources. They should collect informationabout traceability, define the objectives and scope of their system andconsider cutting costs through joint efforts with other enterprises. Thetraceabilitysystemdoesnotdirectlyperformsafety(sanitation)management,quality control and environmental management in the production process;theserequireseparatesystems.

��.3.4 The EU caseThe relative strength of private standards in relation to public legalrequirements has increased in many parts of the world. In the case of theEU,arecentreportunderlinedthefactthatprivatefoodstandardsaremorestringentthanEUlegalrequirementsonfoodsafety(Lee,2006).

EUlegislationonfoodsafetystipulateslegalrequirementsforsuppliersinthirdcountries.Forfoodofnon-animalorigin,theEUrequires“equivalenceofrisk-outcome”aslaidoutintheSPSandTBTAgreementsoftheWTO.161However, with the exception of MRLs for some specific products, it doesnotspecifyhowtomeetthoselegalstandardsanddoesnotrequirespecificcertification.OnlyinthecaseoforganicproductsmayimportsintotheEUbe facilitated by an initial certification obtained in the country of origin.Basically, as long as the final imported products pass official controls inmembercountries,theEUdoesnotlookintotheprocessbywhichproductsareproducedorprocessedinthirdcountries.

161Regulation (EC) 882/2004 of the European Parliament and of the Council of 29 April 2004, onofficialcontrolsperformedtoensuretheverificationofcompliancewithfeedandfoodlaw,animalhealthandanimalwelfarerules.


�8�

Competent authorities in third countries are relied on to carry outinspectionsonfarms.However,theinspectionbodiesoftheimportcountriescannot oblige competent authorities in third countries to bring theirsystems in line with EU ones. There is no way to verify effective officialcontrols of competent authority in third countries, with the exception oforganic production. Also, many developing countries do not have soundnationalfoodsafetysystems,andsomeofthemdonotevenhavecompetentfunctioning authorities. Therefore EU business operators must resort toprivate certification to show due diligence and to protect themselves fromlegalclaims.

Private certification based upon standards such as EurepGAP and BRCrequire“equivalenceofsystems”,settingoutspecificmeasureswithreferenceto EU legislation and that of member countries, to ensure that productsimportedintotheEUarelegallycompliant.Assuch,theydonotnecessarilyset higher standards on the safety of final products. Rather, they requiretight controls over the process through which products are produced orprocessed.Asameanstoanend,itisclaimedthatprivaterequirementsonproduction processes assist developing-country suppliers to comply withlegal requirements, which otherwise would involve a complicated processof aligning with both EU legislation on food safety and those of membercountries.

Thetrendtowardsstrengtheningprocesscontrolisalsodemonstratedbythegrowingbodyoflegislationinthefield,162whichimpliesaninstitutionalshift towards sector-oriented quality assurance schemes and away fromenterprise-levelqualitymanagementapproaches(Schiefer,2004).Thistrendalsofuelsnetworkingorotherformsofhorizontalandverticalcoordinationof the supply chains, which become a relevant source of competitiveadvantage(HanfandHanf,2005;Gellyncket al.,2005).

Thedebatecontinuesonwhetherthesestandardsbasedonequivalenceofsystemsareamarketopportunityforsuppliersordisguisedtradebarriers.163Ontheonehand,theiradoptionmaygiveoperatorsbetterandeasieraccessto developed markets. The retail industry, food manufacturers, importers,

162TheintroductionoftheGeneralFoodLawintheEUmadeamovetowardsprocess-basedcontrolsfor primary production, such as HACCP and traceability systems, to be implemented from 1January2005ineachfoodcompanyintheEU.

163InthecaseoftheEU,severalDirectoratesGeneral(DG)havebeeninvolvedinqueriesrelatedtothis issue. The DG for Agriculture and Rural Development and the DG Joint Research Centrehaverunapilotprojectonprivatefoodschemes.TheEuropeanAidandCooperationOfficewaspresentedwithrequestsfortechnicalassistanceinmeetingprivatefoodstandards,i.e.EurepGAP,fromanumberofcountriesinAsia.TheDGforExternalRelationswasposedwithenquiriesfromthird countries about private food standards. The DG for Health and Consumer Protection isconcernedwithreportedconfusionbetweenofficialEUstandardsandprivateones,inparticularindevelopingcountries.TheDGforTradehasreceivedcomplaintsfromdevelopingcountriesintheWTOaboutprivatefoodschemesconstitutingSPSbarrierstomarketaccess.

�8�


caterers, ingredient suppliers and the food service industry can all benefitgreatly from global standards, such as BRC. On the other hand, there isno doubt that BRC adoption may become a supplier selection criteria inthehandsofdominantmarketplayers (namely,multipurposeretailchains)and thus an entry barrier with significant trade impact. The independenceof accreditedcertificationbodiesbecomesparticularly important toensureproducerssufficientlyfairaccesstoBRC-orientedmarkets.

ItisintheEU’sinteresttobecomeinvolvedinthedynamicsofprivatefoodschemes.Firstly, theEuropeanCommission (EC) shouldbe theonly legalbodytosetprotectionlevels,andlegalrequirementsshouldbeadequatetoguarantee foodsafety.Secondly,asamemberof theWTO,theEUshouldensure that SPS measures do not constitute barriers to trade. Finally, theEU should be aware that the trend of European retail chains to be moredemanding on the safety and quality features of their developing countrysuppliers could end up seriously harming both the EU commitments oninternational development and its efforts to strengthen the capacity ofdeveloping-countryfarmerstoexport.

EC Regulation 882/2004 does open the possibility for public–privatecollaborationoncontrolsoffoodsafety.AlthoughitisnotadvisablefortheEUtorecognizeanyprivatefoodschemes(unlesstheEUiswillingtoassumeliability for what the private sector is doing), it could be appropriate tomaintainacontinuousdialoguewithprivatestandard-settingorganizationsandretailers,ifonlytosensitizethemtothespecificconcernsofdeveloping-countrysuppliers.TheEUshouldalsopayattentiontothefunctioningofthemarketforcertificationtoavoidpricehikesthatwouldunderminedonors’efforts to assist developing countries. In terms of technical assistance, theEUmaywishtomakeuseofEurepGAPspecifications tohelpdevelopingcountriesupgradingtheirsystemsandmeetingEUstandardsonfoodsafety(evenwithoutreferringspecificallytoEurepGAP).

��.4 Conclusions in a Caribbean perspective

Future challenges for Caribbean firms competing in foreign markets willinvolve fewer traditional trade policy barriers (tariffs, quotas) and morenon-tariff barriers based on quality, safety and technology. SPS measurescouldplayaprominentrole in these.Additionalchallengeswillarise fromthesegmentationofmoredemandingmarketswhereentrybarriersrelatedtoprivatesafetyandqualitystandardsmaybehigher.

Ifnotmanagedeffectively,orleftunattended,nationalregulatorysystemsof safety and quality standards can be impediments to maintaining andexpandingtrade,especiallyfordevelopingcountries.Managedsuccessfully,they can be a stimulus to trade and enhance the opportunity to exploitcomparativeadvantagetothemutualbenefitofall.


�83

In the multilateral arena, where the harmonization, transparency andappropriateness of safety and quality standards must be established, it isbecoming increasingly complex and difficult for institutions to implementandregulate thequalityandsafetystandardsset.Oneof themost strikinginfractions against the SPS Agreement is the outright ban on imports offruitsandvegetables imposedbymanycountries.Thesecountriesshift the“burden of evidence”: rather than setting SPS requirements in function ofthe risk presented by imports of certain plant products, these countriesobligeexporters todemonstrate that theirproductsaresafe.Moreover, thecapability of the current structures to deal with emerging issues (such asGMOs) or with the structural disadvantages of least developed countries(LDCs)isratherlimited.

Thus,exportersfromdevelopingcountriesarefacingacomplexsetofpublicandprivaterules,oftenconsideredbythemtobemanagedinanunfriendlymanner, which can pose organizational and technological challenges thatcanputthematacompetitivedisadvantage.Ontheonehand,nationalSPSregulations and related technical requirements are often an obscure andarbitrarydeviceforselectingproviders,usuallyinfavourofnationalproducersas opposed to foreign companies. On the other hand, while market powershiftstowardsbigretailchains,thecomplicatedterrainofoverlappingprivatequality standards suchasHACCP,BRCandISO isbecoming increasinglydifficultandcostlytomanage.Ithasreachedthepointthatmanysuppliersin developing countries – and especially small farmers– cannot afford theluxuryofprivatecertificationandhaveraisedthe issueofprivatestandardsconstitutingdefacto sanitaryandphytosanitarybarrierstotrade.

We have seen that private, voluntary standards can have a very strongimpactoninternationaltradeasentrybarriers,astheysometimessupersedepublic standards. These private standards frequently affect exports fromlessdevelopedcountries,exacerbatingtheproblemsfordevelopingcountryinvolvement in international trade. The issue of restrictiveness of public/private safety regulations in some cases could boil down to whether itis necessary to go further than “equivalence of risk outcome” to require“equivalence of systems” from third countries. One proposal is to allowcontrolsonriskoutcomestobesufficienttoensurefoodsafety;toestablishpublic authorities as the only legal body entitled to set health protectionlevels,withlegalrequirementsadequateenoughtoguaranteefoodsafety.

It is necessary to recognize private schemes as a part of a commercialcontract between suppliers in developing countries and retailers in moredevelopedcountries,andassucharenot imposedondevelopingcountries;their acceptance is a voluntary business decision. Business operators oftenarguethattheyrequireprivatecertificationtoensurefoodsafetyandreflectconsumer concerns. The SPS Agreement is binding only for its memberStates,andnotforbusinessoperators.Thus,forgovernmentstointervenein

�84


thefreemarkettoensurethatsafetyandqualityrequirementsdonotbecomeimpediments to developing country exports they will have to establishdialoguebetweenstates,NGOsandoperators.

Although from the perspective of developing countries – and above allLDCs–thecurrentsystemoffoodregulationandmultilateralrulescannotbeconsideredsatisfactory,therulesarenotbadforthemper se.Onthecontrary,therulescanactasacatalystforchange,andbydoingsoincreasedevelopingcountries’ competitive advantage and contribute to more sustainable andprofitable trade in the long term. In the consumer-driven, media-drivenworldoftoday–aworldoffoodscares,single-issuecampaignsandintensepublicscrutinyofissuesaffectinghumanhealthandfoodquality–rulesarefactsoflifeandwillnotgoaway.Itislegitimateforconsumerstoinsistontheirentitlementtobuyproductsthatmeetcertainlevelsofsanitary,healthandqualityrequirements,butcountriesshouldnotallowtheirstandardstobebasedonprejudiceortobeestablishedinresponsetopressuregroups.Inthe longrun,all countriesmustgain fromcloser internationalcooperationon these issues. This is of interest to Caribbean countries themselves andrequiresregionalcooperation,useofinternationalassistanceforinstitutionalbuildingintheseareasandcollaborationwithinternationalstandard-settingorganizations.

Caribbean countries should consider quality and safety issues in theframework of regional strategies, comprising different components thatare consistent with the overall objective of improving the quality of theregion’s agricultural supplies, improving recognition of regional qualityproducts, increasing intra-regional trade and expanding appropriate nichesinhigher-incomemarkets.Theframeworkshouldalsoconsideradoptionofenvironmentalqualitystandardstofacilitateenvironmentalmanagementandcertificationofterritories.

Onestartingpointforacomprehensiveregionalapproachtoqualityissuescouldbetobroadenandstrengthenregionalagenciesrelatedtoqualityandsafetyofproductsandservices,suchastheCaribbeanRegionalOrganisationforStandardsandQuality(CROSQ).Thereareseveralreasonstorecommendaregionalapproach.Firstly,exportersoftenfinditverydifficulttoconvincetheirnational administrationsof the importanceof resolvingSPSdisputes.This is because agrifood exports are often fragmented across a wide rangeof destination countries, and even if they are concentrated, the value oftheseexportsistoolowtomakeitworththeeffortofnationalgovernmentsenteringintonegotiationswithoneimportingcountry.Nationalgovernmentsin developing countries generally do not have the financial and humanresourcestoengageinlengthyandcostlynegotiationsofoftenverycomplexand technical SPS matters with importing countries. Secondly, as the setofqualitydimensions relevant to international agrifood trade expands, the


�85

costsofdealingwiththosedimensionsincreases.Aregionalapproachwouldoptimizetheuseofhumanandfinancialresourcesonaregionalscale.

By gathering representatives of regional institutes (e.g. the CaribbeanAgriculturalResearchandDevelopmentInstitute(CARDI)andtheCaribbeanFoodandNutritionInstitute (CFNI));national institutions (e.g.ministriesof agriculture and health, universities and research centres); parastatalorganizations (e.g. the National Agricultural Marketing and DevelopmentCompany (NAMDEVCO) and the Barbados Agricultural Developmentand Marketing Corporation (BADMC)); and private organizations(Agroempresarial, National Flour Mills Ltd, Guyana Manufacturers &ServicesAssociation,etc.), theoperationalizationofa regionalagencymaycontribute to addressing the many aspects of quality and safety issues.The establishment and operationalization of the Caribbean AgriculturalHealth and Food Safety Agency (CAHFSA) could provide the Caribbeanregion with a wide-ranging agricultural health and food safety agency todealwithphytosanitaryissues,policy-makingregardingplanthealthissues,programmeplanningandimplementationandobligationsunderthevariousinternationalagreements.Itcouldalsoassistinthedevelopmentofcommonpositions on plant health issues for Caribbean Community and CommonMarket(CARICOM)MemberStatestopresentatinternationalfora.

On the export side, the regional quality agencies should define, in closecooperation with national governments, the quality policy and practicesin the region, and shouldharmonize andcoordinate effortsonSPSexportdossiers.Thecriticalareasthattheregionalqualityagencyshouldfocusonare:developmentofsupplychainpractices,startingfromnegotiationswithprivate global quality standard owners; detection of traditional productssuitable for recognition of GIs; setting up of regional and sub-regionalinitiativesaimedatlinkingcertifiedproductstotheoperationofothersectors(trade, tourism); identification and launching of specific quality controlprogrammes;anddevelopmentof informationtechnologytools forqualitymanagementintheregion.

Improving the recognition of quality products can be important forCaribbean countries, especially as market access widens. In the contextof both market strategies and negotiations this should emphasize the richvarietyofCaribbeanfoodproductsbasedontraditionalknow-how,orwhichhave clear features attributable to their geographical origin. This approachhas considerable potential for building market reputation and increasingrevenues. Some examples of products that have acquired recognition anda reputation worthy of protection on external markets are: bananas fromGrenadaandStLucia,peppersfromJamaicaandBelize,coffeefromJamaicaandseaislandcottonfromseveraloftheCaribbeanislands.

CaribbeancountriescouldhavemuchtogainfromtheEU’sstronginterestinsupportingtheextensionofprotectionofGIsataWTOlevel,aswellas

�86


theEU’sneedtobuildalliancesinthisfield.BydemandinganextensiontoallgoodsoftheprotectioncurrentlyawardedundertheTRIPSAgreementonly to wines and spirits, and establishment of a related binding registerof GI names, Caribbean countries could boost the commercial value oftheir traditional products, and could also build a negotiating ground withtheEU,bothforalliancesintheWTOnegotiationsandexchangesinEPAnegotiations.

Cooperationbetweendevelopedcountries,donorcountriesanddevelopingcountries can help shape a better trade environment through a number ofavenues:

• Developed countries that genuinely pursue trade liberalization shouldadjusttheircooperationschemestohelpdevelopingcountries improvetheir capacity to meet SPS rules and requirements This is crucial ifdevelopingcountriesaretobeproperlyandprogressivelyintegratedintotheglobaltradingsystem.Specificprovisionsfortrade-relatedtechnicalassistanceinthefieldofSPSshouldbeincludedinaidprogrammes(forexample, cultivation or breeding programmes, food-chain integrationprogrammesforslaughterhouses,etc.).

• Developed countries should help developing countries identify andfocus on products that can be more easily exported to higher incomemarkets. For instance, the sensitivities of EU consumers are highestwithsomeproductssuchasmeat,wheredevelopingcountries face themostchallengesinmeetinghygieneandotherrequirements.Incontrast,sensitivities are lower when it comes to plants and vegetables. Thereshould also be greater efforts to increase transparency of EU, UnitedStatesandotherexportbuyers’regulatorysystems.Oneexamplewouldbe toaccelerate theprocessofharmonization in theapplicationof theEU’sborderinspectioncontrols.

• Developed countries should put more effort into standard-setting atan international level and ensure effective participation of developingcountriesintheformulationofthesestandards.Thedefinitionofsafetyandqualityregulations forhigher-incomemarketswouldthentakeonboardthespecificneedsofdevelopingcountriesfromthebeginningandallowthemtoagreeonspecificcarve-outsandtransitionperiodswhereappropriate.

• Finally, the functioning of international organizations also matters.Improving coordination between international aid donors, as well asincreasingcoherencebetweenWTOandotherinternationalorganizationssuch as the International Monetary Fund (IMF) and the World Bank,wouldmakeaidinthefieldofsafetyandqualitystandardsmoreeffective.The issueof resources for international standard-settingorganizations,suchastheCodexAlimentarius,isrelevantinthisregard:theyarenotsufficientlyequipped,giventheimportanceoftheirtask.Althoughthe


�87

CodexhasmadehugeeffortstosetupaTrustFundtohelpitsmembersparticipate in Codex standards, there is still a great need to continuethe push for harmonization of SPS product and process requirementsthroughtheestablishmentofmoreandbetterinternationalrules.

References

Athukorala, P.-C. & Jayasuriya, S.2003.Foodsafetyissues,tradeandWTOrules:adevelopingcountryperspective.The World Economy,26(9):1395–1416.

Buzby, J.C., ed. 2003.International trade and food safety: economic theory and case studies.USDAAgriculturalEconomicReportNo.828.Washington,DC,UnitedStatesDepartmentofAgriculture.

CARIRI/INEA. 2006. Presentations and other training materials for the regionalworkshoponUseofproducequalityandfoodsafetyprinciples toenhance themarketingof agricultural and foodproductswithin andoutside theCaribbean,Port of Spain, Trinidad and Tobago, 9–13 October 2006. Caribbean IndustrialResearch Institute/Istituto Nazionale di Economia Agraria [accessed throughFAOandINEAlibraries,Rome].

Caswell, J.A., Bredahl, M. & Hooker, N. 1998.Howqualitymanagementmetasystemsare affecting the food industry. Review of Agricultural Economics, 20(20): 547–557.

Codex Alimentarius Commission. 2003.Recommended international code of practice-general principles of food hygiene. CAC/RCP 1–1969, Rev.4–2003 (available athttp://www.codexalimentarius.net/download/standards/23/cxp_001e.pdf)

De Filippis, F. & Salvatici, L., eds. 2006.Un percorso difficile. Il negoziato agricolo nel Doha round del WTO.Rome,ProcomEdizioni.

EurepGAP. 2004. General regulations: fruit and vegetables. Version 2.1 (October2004)(availableathttp://www.eurepgap.org/fruit/Languages/English/documents.html).

FAO. 2000. The importance of food quality and safety for developing countries(availableathttp://www.fao.org/trade/docs/LDC-foodqual_en.htm).Rome.

FAO.2003a.Important commodities in agricultural trade: fruits and vegetables.FAOFactSheets:InputfortheWTOMinisterialMeetinginCancún.No.13(availableathttp://www.fao.org//DOCREP/005/Y4852E/y4852e13.htm).

FAO.2003b.Environmental and social standards, certification and labelling for cash crops.FAOCommoditiesandTradeTechnicalPaperno.2.Rome.

FAO. 2004a.The market for non-traditional agricultural export.FAOCommoditiesandTradeTechnicalPaperno.3.Rome.

FAO. 2004b.Voluntary standards and certification for environmentally and socially responsible agricultural production and trade. FAO Commodities and TradeTechnicalPaperno.5.Rome.

FAO. 2005a. Codex and the international food trade (available at http://www.fao.org/docrep/w9114e/W9114e06.htm).

�88


FAO.2005b.Perspectives and guidelines on food legislation, with a new model food law.DevelopmentLawServiceStudy87.Rome,FAOLegalOffice.

Gellynck, X., Januszewska, R., Verbeke, W. & Viaene, J. 2005. Firm’s costs of traceability confronted with consumer requirements(availableathttp://www.eaae.uni-goettingen.de/Startseite/EAAE-Vortraege/Gellynck%20et%20al..pdf).

Golan E., Krissoff, B. & Kuchler, F.2004.Foodtraceability:oneingredientinasafeandefficientfoodsupply.Amber Waves,2(2):14–21.EconomicResearchService,USDA(on-linedocument,availableatwww.ers.usda.gov/Amberwaves/April04/Features/FoodTraceability.htm,2004).

Hanf, J. & Hanf, C.-H. 2005.Does food quality management create a competitive advantage?Paperpreparedforthe92ndEAAEseminaronQualitymanagementandqualityassuranceinfoodchains,2–4March2005.Göttingen,Germany.

Henson, S., Loader, R., Swinbank, A. & Bredahl, M.1999.The impact of sanitary and phytosanitary measures an developing country exports of agricultural and food products.PaperpreparedfortheWorldBankandpresentedattheConferenceonAgricultureandtheNewTradeAgendaintheWTO2000Negotiations,Geneva,1–2October1999.

International Standards Organization (ISO).(n.d.)ISO 9000:2000 series(availablethroughhttp://www.iso.org).

Lee, G. C-H.2006.Private food standards and their impacts on developing countries.European Commission, DG Trade Unit G2 (available at http://trade.ec.europa.eu/doclib/docs/2006/november/tradoc_127969.pdf).

Porter, M.E. 1980.Competitive strategy.NewYork,FreePress.Schiefer, G.2004.Traceabilityandcertificationinfoodqualityproduction:acritical

view. In A.G.J. Velthuis, L.J. Unnevehr, H. Hogeveen & R.B.M. Huirne, eds.New approaches to food-safety economics. Norwell, MA, Kluwer AcademicPublishers.

Unnevehr, L. & Roberts, D. 2003. Food safety and quality: regulations, trade, and the WTO.InvitedpaperpresentedattheinternationalconferenceonAgriculturalpolicy reform and the WTO: Where are we heading? Capri, Italy, 23–26June2003.

Wilson, J.S. 2002. Standards, regulations and trade. WTO rules and developingcountriesconcerns.InB.Hoekman,A.Mattoo&P.English,eds.Development, trade and the WTO: a handbook,pp.428–39.Washington,DC,WorldBank.

World Trade Organization (WTO). 2000. Agreement on trade-related aspects of intellectual property rights (TRIPS) (available at http://www.wto.org/english/tratop_e/trips_e/t_agm3b_e.htm).

WTO. 2002. The relationship between the TRIPS agreement and the Convention on Biological Diversity and the Protection of Traditional Knowledge. Councilfortrade-relatedaspectsofintellectualpropertyrights. 24June2002(availableathttp://commerce.nic.in/wto_sub/TRIPS/sub_Trips-ipcw356.htm).

�89

��

Tradeandfoodsecuritypolicyanalysis:apracticalguide*

Crescenzo dell’Aquila, Piero Conforti, J.R. Deep Ford and Hansdeep Khaira

Introduction

The consequences of policy decisions are becoming more complex andfar-reaching every day, mostly as a result of the deepening of economicinteractionsamongagents,activitiesandpolicies.Theseinteractionsaretakingplace within an increasingly wider trading environment, characterized bydiverse technologies, infrastructures, resourceendowmentsandconsumer’spreferences.Asimpledecision, liketheestablishmentofatarifforsubsidyregime in a specific sector,or evena change in the implementation ruleofone particular regime, may imply consequences that go well beyond thesector itself, and well beyond the trading parties more directly involvedin that regime. This means that understanding the likely effect of a policydecisiontendstorequiretheconceptualizationofcomplexlinkagesamongalargenumberofvariables,whichiscreatinganincreasingdemandforpolicyanalysis.

The interest of policy-makers is usually multifaceted. Often, changes intradepoliciesareassessedintermsoftheirlikelyconsequencesonthedegreeofexposureoftheindustryinvolvedtoforeigncompetition,ontherelatedeffectsintermsofemploymentoronthebalanceofpayments.Increasingly,the impacts on poverty and food security levels are investigated. Policy-makers’attentionisprimarilyattractedbytheshorttermimpactsofreforms,

* This chapter partly draws upon the work done by the authors for the regional training course“TradePolicyAnalysisandAgriculturalTradeAgreements”(Parmaribo,Suriname,31January–18February 2005), carried out in the framework of the FAO project “Promoting CARICOM/CARIFORUMFoodSecurity”(GCP/RLA/141/ITA).

�90


however there are considerable longer term effects that also need to beevaluated.Policyanalysisisusefulinalltheseareas–forexample,inassessingthecurrentdegreeofcompetitivenessofan industry,or thepossibilitiesofmaintaining competitive advantages in a wider environment through time.Inthosecontexts inwhichpovertyandfoodsecurityare important issues,the need to understand its linkages with policies and the farther-reachingconsequencesofreformsrequiresmorecomplexanalyticaltools.

Most policy matters boil down to establishing the extent to whichconveniently computed benefits outweigh costs. This calls for numericalestimatesoftheconsequencesofpolicychanges,computedonthebasisofasetofexplicitlypostulatedrelationshipsreflectingtheinteractionsinvolved.

In the specific area of global agricultural trade, a number of differentquantitative models have proven to be potentially useful, particularly fordemonstrating how specific reform packages might impact on differentcountries and commodities, and for helping to settle controversial issuessuchastradedisputesbetweencountries.Aninterestingrecentdevelopmentin this area is the considerable degree of networking now undertaken byresearchers and analysts around the world. Based on the potential offeredby the growing power of computers and the Internet, increasingly oftenanalysts share their conceptual approaches, analytical frameworks andtools, lowering significantly the start-up costs of the analyses in terms ofdata collection, organization and even model development. In turn, this isenhancingthedegreeoftransparencyandreplicabilityoftheresultsonkeyquestions,andiswideningtheareasofanalysisandthepublicinvolvementin the investigations, with beneficial feedback effects on the quality of theresultsthemselves.

Thischapterpresentselementsofcommonquantitativetoolsused intheinvestigation of the consequences of trade and agricultural policy changes.ItisintendedtobeapracticalintroductionforagriculturalandtradesectorpolicyanalystsintheCaribbean.Emphasisisplacedontradepolicy,andanattempt is made to show how the linkages with food security, agriculturaldevelopment and rural development can be addressed. Particularly, thechapter aims to: a) show the potentials of quantitative analysis whilehighlightingtheassociatedchallengesandlimitations;b)introducedifferentapproaches and analytical frameworks; and c) facilitate awareness of theavailabilityofdatabasesandcomputerbasedtoolsthatcanbeusedasstartingpoints.

The following section presents an overview of the major approaches,consideringthetwowidecategoriesofex postandex anteevaluationsandintroducing modelling approaches and relatedpolicy representation issues.Section2deepensex postapproachesandexplainsandimplementssomeofthe most common descriptive indicators used for food security, trade andtradepolicyanalysis.Section3deepensex anteapproachesby introducing

Trade and food security policy analysis: a practical guide

�9�

partialandgeneralequilibriumframeworksandincludingreferencestosomeofthemodelsanddatabasesthatcanbeaccessedmoreeasilytostartusinganalyticaltools.Theappendicestothechapterpresentaglossaryandmorespecificreferencestodatasourcesandotherresourcesforpolicyanalysis.

��.� Approaches to quantitative trade policy analysis and main models’ characteristics

Amongthenumberofwaysofclassifyingtheapproachesemployedintheanalysisofpolicychanges,abroadandconvenientdistinctioncanbemadebetweenex anteandex postmethods.Theformerincludethosestudiesaimedatansweringa“what-if”typeofquestion,thatis,atprovidingcomparativeinformationonacounterfactualscenariobuiltbymakingassumptionsonthevalueofapolicyvariable.Forinstance,giventhatthetariffonsugarintheEuropeanUnion(EU)is€400/metricton(tonne),onemaywanttoanalysehowtradeandpriceswouldlooklikeinthesugarmarketshouldthetariffbe€200/tonne.Thisrequiresacrediblerepresentationofthesugarmarketasitis,withthetariffat€400/tonne,makingitpossibletoanalysecomparativelythe effecton trade andpricesunder a scenario inwhich the tariff is €200/tonne.

Bycontrast,ex post studiesarebasedon theanalysisandcomparisonofcurrent and past data with the aim of assessing the effects of trade policymeasuresontrade,welfare, foodsecurityandotherdimensionsof interest,whichoccurredfollowingimplementationofthegivenpolicies.Suchstudiescanbebasedeitheroneconometrictechniques,oroncomputationofsetsofdescriptiveindicators.

Theresultsofeconometricexercisescanrelyuponstatisticaltests,indicatingtheexistenceofastatisticallysignificantrelationbetweenachangeinapolicyvariableandthechangeofagivenindicator.Forinstance,onemaywishtoanalyse the extent to which the implementation of a free trade agreementbetweentwocountrieshasbroughtaboutanincreaseinthevolumeoftrade.Aneconometrictestwouldthenberunondataencompassingboththeperiodpriortotheimplementationoftheagreementandtheperiodafterit.Thiscanindicate theextent towhich thevolumeof trade is related to the loweringofthetariffs,and,undergivenassumptions,alsoprovideindicationsonthecausaldirection.

If the same exercise is conducted on the basis of descriptive indicators,instead, there are no measures of the statistical reliability of the evidenceproposed. This approach has the advantage of simplicity and is far lessdemanding in terms of data and technicalities; but it also involves a costin termsofmore limited analytical content.However,oneof themethodsintroducedinthechapter,thePAM,showsaconsiderableanalyticalcontentdespitebeingbasedmostlyonasetofdescriptiveindicators.

�9�


��.�.� Models and their features Bothex anteandex postapproachestoquantitativeanalysisarebasedonsetsoffunctionalrelation,whicharecommonlyreferredtoas“models”.Modelsaresetsofequationsaimedatrepresentinginastylizedwaythebehaviourofeconomicagentsandtheir interaction,andcanbeclassifiedinanumberofways;usefulsummaryreviewsareavailable invanTongerenandvanMeijl(1999),andFAO(2006).Forthepurposesofthischapter,thecriteriatolookatare:

• theextentoftherepresentationoftheeconomy;• thepresence/absenceofatimedimension;• thenatureandoriginoftheparameters,andtheavailabilityofmeasures

ofthestatisticalreliabilityoftheresults;• thetypeoffunctionalform;and• thetypeofclosurerule.One possible model classification divides partial equilibrium (PE) and

generalequilibrium(GE)models,dependingonthefirstofthecriterialistedabove. PEs are those models in which the analysis excludes at least somemarkets, assuming that they will not be affected by what happens in themarket analysed. In contrast, GE models include the entire economy bydefinition.

Thesecondcriterionyieldsanothermodelclassificationbydividingstaticfromdynamicmodels.Staticmodelsarethoseinwhichthetimedimensionisabsent;theycomparetwoalternativestatesoftheworldwithoutobservingthe adjustment path between them. In contrast, dynamic models include atimedimension,sothatthevaluesofthevariablescandependuponpastandfuturevalues;theyadjusttochangesthroughmorethanoneperiod.

Depending on the origin of the parameters, models can be classified as“econometric”. Econometric models are those in which parameters areestimated,allowingastatisticalvalidationoftheresults.Parametersarethenumbersthatshapethebehaviouralrelations.Inex postanalysesparameterscan constitute an output of the analysis, since the objective of deriving astatistically-controlled measure of the relationship between two or morevariables is achieved through one or more tests upon the “soundness” oftheparameterscomputedbymeansofavailabledata.Inotherwords,intheaboveexample,theexistenceofarelationbetweenthelevelofthetariffsandthevolumeof tradebetween twocountries is analysedon thebasisof thestatisticalreliabilityoftheparameterthatcanbecomputedtorepresentthisrelation.

Incontrast,inex anteanalyses,parameterscanconstituteoneoftheinputs:whatever their origin (ad hoc estimation, literature, or calibration164) they

164Calibrationimpliesfittingtheunknownparameterstothevaluesthatreproducethedatainthebaseperiod.


�93

embedassumptionsaboutthebehaviourofthevariablesemployedtodescribetheexistingenvironment,basedonwhichthecounterfactualexperimentwillberun.Returningtotheaboveexample,acrediblerepresentationofthesugarmarketimpliesanassumptionontheelasticityofdemandandsupplyintheEU,ifwewanttoconsidercomparativelytheeffectsonthepricesandvolumeoftradeofreducingthetariff from€400/tonneto€200/tonne.Underbothscenariosweshallassumethatthebehaviourofproducersandconsumersisunchangedwithrespecttoprices,incomesandinputsinproduction.

Inex ante analyses,however,theavailabilityofameasureofthestatisticalreliabilityoftheparametersconstitutesanimportantadvantage,asitallowsforcheckingthereliabilityoftheresults.Othertypesoftestscanalsoberuntochecktheperformanceofex ante simulationmodels,suchascomputationof data from past periods for which the observed value of the variables isknown.Suchtestscanmeasurethestabilityoftheresults,whichallowsforverifyingthereliabilityofdynamicmodels.

Themathematicalfunctionalformoftheequationsisanotherkeyfeatureofthemodels,whichembedsassumptionsaboutthebehaviourofeconomicagents.Modelscanbebasedeitheronreducedformequations(inwhichitis implied that optimizing behaviour is modelled through the restrictionsontheparameters,suchasthoseofadding-up,homogeneityorsymmetry)oronstructuralformequations(inwhichoptimizationisexplicit).Modelsemployedinex anteanalysistendtobebasedonrelativelysimplefunctionalforms,suchasthelinearorlog-linear,theconstantelasticity,thelog-log,ortheCobbDouglas.Incontrast,moresophisticatedandtheoreticallysoundfunctionalformsarefoundineconometricexercises,whicharemoresuitableforex post analyses.

Finally, the closure rule is an important characteristic for equilibriummodels. This differentiates variables into exogenous and endogenous,thereforedeterminingthecriteriausedtosolvethemodel.Aswillbeshownin Section 3, the closure rule is particularly important because it impliesassumptionsaboutthefunctioningofthemarketrepresented.

To conclude this section, it may be useful to recall six basic “rules ofthumb” valid for model-based analyses, which however commonsensicaltheymayappear,arestillveryimportantandoftenoverlooked.

Firstly,nomodel issuitableforanalysingall typesofproblems,andit ismoreusualthatamodelissuitedforonlyoneparticularproblem.Adaptinga model built for one purpose to a totally different problem is seldom asuccessfulstrategy.Secondly,nomodelcanbebetterthanthedataonwhichitisbased.Dataarealwaysoneofthemostimportantpartsofanyanalysisand often a major source of the limitations of the results; they should becarefully considered and extensively discussed. Thirdly, the credibility oftheassumptionsisimportant,butnotalwaysnornecessarilyavalueper se;modelscansometimescaptureessentialandpertinentaspectsofrealityeven

�94


bystartingfromveryunrealisticassumptions.Fourthly,generatingnumbersis important, but understanding how they are generated, and under whichassumptionsandlimitations,ismoreimportant.Thereforetimeneedstobespentonunderstandingthenumericalresultsthataregenerated.Fifthly,theresultsofmodelsusuallyindicateatbestasignandanorderofmagnitude.Finally, results seldom speak for themselves, and more often requireinterpretationbasedondeepknowledgeoftheproblemanalysed.

In all cases, results of models are the mere outcome of the interactionamong exogenous assumptions, behavioural hypotheses and parameters,andshouldbestrictlyconsideredassuch.However,theycanassistpolicy-makersinbuilding“menus”ofpoliciesandtheirconsequences,tosupportagivenchoiceandtocommunicatewithotherpolicy-makers,especiallyinthecontextoftradenegotiations.

��.�.� Policy representationPoliciesaredifficulttorepresent inamodel, forat leasttwomainreasons.Firstly,theyarespecifiedatafarhigherlevelofdetailthanisnormallyusedinamodel.Forinstance,policynormallyspecifiesatariffatafarmoredetailedproductlevelthanit ispossibletoadoptinamodel,givendataavailabilityand the need to avoid making the model unmanageable. Representingpolicyinamodelrequiresaggregations,whichimpliesmakinganumberofassumptions.

Secondly, it is often difficult if not impossible to represent policiesexplicitly.Anexplicitrepresentationisoneinwhichthemodelincludesasanexogenousvariable(thatis,avariablethatcanbeshocked)thesamevariableoperated by the policy-maker (Anania, 2001). In the practice of policyanalysis,thisisseldompossible.Moreoftenaone–to–onerepresentationofpolicy measures is not possible, due both to the difficulty associated withcapturingthedetailsofthedecision-makingprocess,andtothepresenceofpoliciesthatimplysimilarandcumulativeeffectsthatcannotbeseparatedinastylizedsetting.

Asanexample,considertherepresentationofafixedtariff inamodelinwhichtariffsaredefinedinpercentageterms.Theanalystwillneedtoconvertthefixedtariffintoanad valorem tariff,onthebasisofsomerelevantprice.Suppose, further, that the tariff is coupled with a domestic price supportmechanism that operates in conjunction with the tariff. The analyst willneedtofindsomemeasurescapableofcapturingtheeffectsofbothpolicies,avoidingdouble-countingandwithoutgivingupthepossibilityofsimulatingchangesinonlyoneofthetwopolicies.

These problems call for the calculation of some kind of “equivalent”measure, capable of aggregating over policies implemented on differentproducts, and aggregating over different types of measures operated onthe same (group of) product. This equivalent measure would be capable


�95

of capturing those policies that are in fact put in place and changed.For tariffs, a standard approach employed in trade policy analysis is thecomputation of some “tariff equivalent”, based on the difference betweenaworldmarketpriceandthecomparabledomesticprice.165Problemswiththistypeofmeasuresarisebothintheaggregationacrosstariffsdefinedfordifferentspecificgoods(wheretheoreticalfoundationsareabsentincommoncalculations)andinaggregationacrosstypesoftariffs,duetothedistortionofworldmarketprices,whichcomplicatestheidentificationofaconvenientpricetobeadoptedintheconversion.

Thelevelofcomplicationincreaseswhenmorearticulatedpolicymeasuresare taken into account. For instance, quantitative restrictions cannot bemeaningfullyrepresentedthroughatariffequivalent,noritisitpossibletorepresentothercommonnon-tariffmeasuresintheseterms,suchastariff-ratequotasormultipleandvariabletariffs.WiderdiscussionsonthesetopicscanbefoundinLaird(1997),Anania(2001)andCipollinaandSalvatici(2006).

��.�.3 Linkages with poverty Authoritativeeffortshavebeenmaderecentlytoshedlightonthekeymatteroflinkagesbetweenchangesintradepolicyandpovertyoutcomes.166Ithasbeen observed that the analytical tools employed for this purpose need toaddressanumberoflinkagesinvolvingthefollowingvariables:167

• priceandavailabilityofgoods;• factorprices,incomeandemployment;• taxes, subsidies and the availability of public resources for financing

them;• investmentandinnovationforlong-termgrowth;• externalshocks,particularlyfromprice;and• short-runadjustmentcosts.Apartialrepresentationoftheeconomyissufficienttoassessthefirstof

suchlinkages,whileamoresatisfactoryrepresentationisneededtoanalysethesecondandthirdones.Thefourthandsixthareusuallystudiedinadhocframeworks(suchasnon-structuralcross-countryanalysisoraggregatedGEapproaches)whilethefifthcanbeanalysedwithinseveralapproaches.

In many cases, a key starting point on the linkage between changes intradepolicyandpovertyoutcomesisthemodellingofthelabourmarket.Ashighlighted recentlybyAckerman (2005),oneof themajorgaps incurrent

165AcommonmethodtoderivetariffequivalentsistouseProducerSupportEstimates(PSE).Thesecomprise price distortions or market price supports (transfer from consumers to producers) aswell as transfers fromgovernment toproducers.A similar concept exists forConsumerSubsidyEquivalents(CSE).SeealsothePolicyAnalysisMatrixapproachinChapter10.

166Oneexample,coveringavarietyofapproaches,isabookeditedbyHertelandWinters(2006)andpublishedbytheWorldBank.

167HertelandReimer,2005.

�96


tradepolicyanalysisliteratureisthatrepresentationofthelabourmarkettendstobepoor.Thisisoneimportantdrawbackoftheresultsofseveralmodels,and it may significantly affect the conclusions of many trade liberalizationstudies,particularlythoseondevelopingcountries.Thislackclasheswiththeimportancethatpolicy-makersrightlyattachtothelabourmarket.

��.� Ex post trade policy analysis: the Policy Analysis Matrix and the descriptive indicators

Ex postevaluationsassesstheeffectsoftradepolicymeasuresontrade,welfare,foodsecurityandotherdimensionsofinteresttopolicy-makers,whichoccurfollowingimplementationofthegivenpolicies.Theseapproachesusevariouskindsofstatisticalandeconometrictools,rangingfromdescriptiveanalysesoftrendstoeconometricmodels.

Econometricmodelsarethestandardwaytocontrolforandanalysefactorsaffectingthetradepattern.Theseapproachestypicallyestimateworldtradeflowsbydefiningasimplifiedexplanatoryhypothesis,andascertainwhethertheestimatedrelationshipschangeasaconsequenceofimplementingacertainpolicy. The main strength of these models is the possibility of statisticallyvalidating hypotheses about various variables affecting trade. Weaknessesinclude the lack of details in the definition of variables relevant for policyanalysis, theneed for agreatdealofdata and the impossibilityof treatingthestructuralbreakdeterminedinthemodelwhenlargepolicychangestakeplace(Taylor,2004;Lucas,1976).

Ontheotherhand,non-parametricapproaches(suchastheindicesdiscussedinthissection)avoidtheproblemofdefiningamodelfortradeflows.Theycanprovidefirst-glance ex postassessmentsoftheimpactofbothtradeandfoodsecuritypolicymeasures,aswellaspreliminarypicturesofrealitiestobemodelled.Byusingdescriptiveindicators,analyticalproceduresarefasterandlessdemandingintermsofdata.Thismakesthemparticularlysuitableforprovidingquickanswers topolicyquestions,especially forphenomenaoccurring on a world scale and involving a large number of commodities.Themainweaknessof this approach is that itsdescriptivenaturedoesnotallowfordetailedanalysisofvariousfactors(notnecessarilypolicyfactors)affectingthetradepattern,noroftheleveloffoodsecurity;nordoesitallowstatistical validation of hypotheses (Hoekman, English and Matoo, 2003;DrysdaleandGarnaut,1982).

The third framework introduced is the policy analysis matrix (PAM)(MonkeandPearson,1989).ThePAMisatoolthatconstructstwoenterprisebudgets,onevaluedatmarketpricesandtheotherateconomic/socialprices;theimpactofpolicyisthenassessedasthedivergencebetweenthemarketandeconomicvalues.ThePAM,onceassembled,providesaconvenientmethodofmeasuringpolicyeffects,competitivenessandcomparativeadvantage.


�97

This sectionprovidesdefinitions andexamplesof indicators relevant fortradepolicy,foodsecurityandvulnerabilityanalysis,withspecialreferenceto:thecontributionoftradetofoodsecurity,theanalysisoftradeflows,theopenness anddependenceof the trade regimeand the exposureof exportsandimportsinthinly-tradedmarkets.168ItthenintroducesthePAMtool.

��.�.� Indicators of dependency, vulnerability and food security related to tradeTherearenumerousgeneralindicatorsforeachofthesethreeconcepts,butmostarenotnecessarilyrelatedtotradeandthereforenotmentionedhere.The relationship focusedonhere canbe illustrated throughan exampleofhow the three concepts are linked: dependency on food imports is linkedtoforeignexchangegeneration,whichcanmakeacountryveryvulnerabletopricedeclinesinexportmarkets(especiallyinthosemarketswhereitisapricetaker).Dependingontheparticularnationalcontexttheanalystwouldidentifyoneindicatorasbeingmorerelevantthananother.

Cereal supply indicator (SI)usescerealsasrepresentativeoffoodneededand is measured by dividing the total supply for domestic utilization(production + imports – exports + changes in stocks) by the population.Anexample (Table ��.�) of this indicator (cereal supply/kgper capita) forthree Caribbean countries over three years indicates three very differentoutcomes:

168Other policy indicators of agricultural support and trade competitiveness are defined within thePolicyAnalysisMatrixapproach(seesection12.2.3)andappliedtoCARICOMcountries/productsinChapter10.

100*

∆+−+=

r

rrrrr

pop

xmySI

where:yr =cerealproductionmr =cerealimportsxr =cerealexportΔr =changesinstocksofcerealspopr =countryr’spopulation

TABLE12.1Cereal supply indicator

�995 �000 �003

SaintVincent/Grenadines 97.6 113.3 118.9

Grenada 100.8 90.2 88.6

Jamaica 104.4 98.4 103.7

Source:FAOSTAT,2006

�98


At a very general level SI provides a snapshot of how much food hasbeen available domestically on average. One word of caution is to payattentiontohowsupplyisdefined(totalsupplyisproductionplusimportsminus decreases in stock), another is to remember that the supply is notevenlydistributed,andstill another is tonotice, forexample, that it isnotclear (without going back to the base data) whether Saint Vincent and theGrenadines’supplyisincreasingbecauseofdomesticproductionorduetoagreaterrelianceonimportedfood.Conclusionsneedtobemadewithcare.

Food import capacity indicator(ICI)istheratioofthefoodimportvaluetothetotalexportvalue(excludingservices):

100*=r

rr

X

MICI

Thiscouldalsobemeasuredmoredirectlyandeasilyusingcerealimportsasaproxyforfoodimportsanditisalsooftencomparedtototalagriculturalexport earnings as opposed to total merchandise trade. The table below(Table ��.�)showsfoodimportvaluesovertotalagriculturalexportearningsandagainrevealsthreeverydifferentsituations.

Belize’s(stableandconsiderable)agriculturalexportcapacity,whichallowsittopurchasefoodimports,canbecontrastedwiththevulnerabilityofHaiti,bothintermsofforeignexchangeearningsfromtheagriculturalsectorandofnaturaldisastersandtheirimpactsonagriculturalproductionandexports.(Haitiwashitbyseverehurricanesin2003and2004.)Changesinthetypeof foodconsumed (forexamplemoreprocessed foodorhigher-value foodimports)canleadtoanincreaseintheindicator,butisnotnecessarilyasignofincreasingfoodinsecurity.Inaddition,thecapacitytoimportbasedonanexpandingserviceindustrymaynotbereflected,dependingonthevariablesusedintheindicator.

where:Mr =valueofthecountryrtotalimport(excludingservices)Xr =valueofthecountryrtotalexport(excludingservices)

TABLE12.2

Food import capacity indicator

�995 �000 �004

Barbados 1.59 1.66 1.52

Belize 0.32 0.38 0.36

Haiti 10.33 10.83 22.10Source:FAOSTAT,2006


�99

Food import coverage indicator (FIC) compares the foreign exchangereserve balances of the country (at end of year) with the food import billvalue(annual)andindicateshowvulnerablethecountry’sfoodsecuritycouldbe to severe shocks that might disrupt either its domestic supply (whichwould have to be replaced) or lead to a loss of foreign exchange earningcapacitythroughpriceorexportsupplyshocks.

100*=r

rr

FibFIC

Fb

Table ��.3 shows four very different situations; it reflects effects ofpositive and negative economic (international price) or political shocks onthechangingcapacityofacountrytocovertheimportsofitsfoodfromitsforeignexchangereservesinthecaseofacrisis.

Guyana’sforeignexchangereservecoverageoffoodimportsincreasedfivetimesbetweenthe1990 level,where therewasaratioof .86 (less thanoneyear’s food import coverage), and 1996 (potentially a five-year coverage).In Guyana, 1990 was the beginning of a period of significant economicgrowth that also coincided with major political change. Rice and sugarexports increased substantially in the first half of the 1990s and sufferedfrom declining growth rates and prices thereafter. Suriname was similarlyaffectedbydecliningriceprices.ThehealthyTrinidadandTobagosituationreflects clearly an expansion of oil revenues while Haiti’s consistentlyhigh vulnerability position worsened. Several factors in addition to globalcommodity prices affect this ratio, such as changing food import levels,performanceoftheeconomyasawholeand,inmanyCaribbeancountries,theperformanceofthetourismsector.Inseveralcountries,tourismreceiptshave been a major factor in improving results when this indicator is themeasuringrod.

TABLE12.3Food import coverage indicator

�995 �000 �004

Guyana 5.18 4.23 3.41

Suriname 2.93 0.87 1.86

TrinidadandTobago 1.62 5.38 10.03

Haiti 0.65 0.68 0.31Source:FAOSTAT,2006andIMF,2006

where:Fbr =foreignexchangereservebalancesofcountryrFibr = food import bill value (annual) ofcountryr

300


��.3.� Trade shares and indicators of trade structure and performanceCoverage ratio (XC)indicateshowmuchofthevalueofimportsisfinancedbyexport.Itisgivenbythepercentageratioofexportoverimport:

Theindexvariesbetween0(importfullycoveredbyentriesofthebalanceof payments other than export) to +∞, with a value of 100 indicating thatexportisfullycapabletocoverimport.Overtimetheindexcanmonitorthedevelopmentof sectoral surplusesordeficitsofexportover import. In thecaseoftheCaribbeancountries,thisindextendtoshowanexcessofexportoverimportwiththerestoftheworldforbananasandsugar,determinedbythepreferentialtradingtieswiththeEU15.

Normalized trade balance (NB)isthenettradeindicatorusedmostoften.Itistheratioofthetradebalanceofacountry/industryoveradimensionalmeasureoftheflows(i.e.thetotalvalueoftrade,measuredasasumofimportandexport):

Theindexvariesintherange-100≤NBi≤100.Negativevaluesmeanthatthecountry/industryiisanetimporter,totheextremevalueof-100,whichsignals that only import takes place in the country/industry considered.Positivevalueshavetheoppositemeanings(netexportpositions).

Trade balances give a synthetic measure of the degree of disequilibriumoftradeflows,whiletheirnormalizationismeanttomakethemsuitableforcomparisons. The improvement, over time, of the NB suggests improvedtradeperformanceofthesectorevenwhenthetradebalanceworsens.Thiscanhappen,forexample,whenwestartfromasizeabletradedeficit,andtheexport growth is higher, in percentage terms, than that of imports. In thissense,NBscanshowmoreaccuratelythansimpletradebalancesthechangesthatoccurredintradeperformance.

Moreover,indisaggregatedanalysis,thenormalizedtradebalanceisofteninterpretedasanindicatoroftradespecialization.HighandpositiveNBsarerecordedforcommoditiesinwhicheithermarketorpolicydeterminants,orboth, make national production competitive in both foreign and domesticmarkets.Therefore,theNBmaybeconsideredanex postsyntheticindicatorofthecompetitivesuccessofnationalproducts.

100*=r

rr

M

XXC

where:Xr =valueofthecountryrtotalexportMr =valueofthecountryrtotalimport

where:Xi =valueofexportsofcountry/industryiMi =valueofimportofcountry/industryi

( )

( )100*

ii

iii

MX

MX

+

–=NB


30�

The NB for Caricom’s agrifood trade performance is expected to begenerallynegative,andstronglynegativeforagriculturalproducts.

Trade/GDP ratio (O)isaconventionalmeasureofopenness,givenbythepercentageshareofGDPtraded:

Evaluatingthe“degreeofopenness”totradeofeconomiesorsectorsisarathercommonapproachforassessingtheimpactoftradepolicies.Althougheconometric attempts to estimate measures of openness have often beeninconclusive, the idea that the share of GDP (or consumption) traded candetectchangesinopennessofthecountry/sectorhavemadethoseratiostheconventionalmeasuresofopenness.

Due to its shortcomings,169 the trade–to–GDP ratio is used to describebroadchangesinopennessoveralongtimeor,attimes,tocomparedegreesof openness before and after implementation of trade agreements. In fact,this indexgathersmoreinformationthanitshouldinordertobearefinedmeasureofopenness. It isnotable todifferentiatehistorical,geographical,economicandpoliticalfactorsaffectingtheshareofagriculturalGDPtraded;thereforeitisaffectedbymanyfactorsnotdirectlyinvolvingtradepoliciesandpolicy-determinedopenness.Also,theindexisnegativelycorrelatedwiththesizeofaneconomy,becauselargecountries,withlargerandmorediversestocksofresources,arebetterabletomatchdemandandsupplydomestically,and transportation costs are likely to favour domestic producers for arangeofproductsthatwidensasthesizeofthecountryincreases(Perkins-Syrquin,1989).Thisimpliesthattheratioisnotcomparableamongdifferentcountries.

Trade shares (S) are useful indicators of the structure of trade, tradeperformanceanditsevolutionovertime.Mostoftheindicatorsintroducedinthissectionaretradeshares,oracombinationinvolvingthem.

Trade shares canbecalculated in severalways, according to thepurposeof the analysis. For instance, the share in world export of a country (or asingleindustryofthecountry)istheratioofacountry’s(orcountry’ssingleindustry)exporttotheworldoverworldexport(orworld’ssingleindustryexport):

169Adiscussionoflimitationsofthetrade–to–GDPratioandanattemptofaneconometricestimateofthismeasureofopennessareavailableinLeamer(1988).

where:Xr =valueofthecountryrtotalexportMr =valueofthecountryrtotalimportGDPr =grossdomesticproductofcountryr

100*+

=r

rrr

GDP

MXO

30�


Theindexrangesbetween0≤S≤1(or0≤S≤100,ifweprefertheindextobeexpressedinpercentagetermsandthereforemultiplyitby100),wherehighervalues indicategreater importanceof the country inworld exports.Onthedemandside,forshareinworldimports,thesamemeaningmaybeobtainedbyswitchingexportwithimportinthepreviousdefinition.170

In the case of CARICOM, shares of world market are expected to benegligible,howeverimportanttheproductsareforthelocaleconomy.Thisindicates that CARICOM countries are not expected to be able to affectworldprices.However,sharescanbehigherinaspecifictargetmarketand/orproduct.Forinstance,ifwecomputethesharefortheEU,expressedas

This will turn out to be significant, especially for certain products, suchas sugar, rice or bananas, while still not significant for the whole agrifoodsectorasawhole.Ingeneral,CARICOMagrifoodtradesharesareexpectedto be relatively higher for beverages, sugar, prepared cereals and fruit andvegetables.

Tradesharescanalsobecomputedtoevaluatetherelevanceofaspecificsectoronthetotalexportorimportofacountry.Forexample,eiistheexport(import)ofthei sectorofcountryj,whileEisthetotalexport(import)ofthesamecountry:

Or,toevaluatetherelevanceofaspecificsectorontheexportto(importfrom)agivencountry,eiistheexport(import)oftheisectorofcountryjtothepartnerk,whileEiistheexport(import)ofsectoriofcountryj:

170Itmustbenotedthatexportstoapartnercountryaregenerallyexpressedinfreeonboard(f.o.b.)price,whereas imports usually include costs of insurance and freight (c.i.f. price).Therefore, thevalueofexportsfromcountryAtowardscountryBdiffersfromthevalueofimportsintoBfromA(thevalueofimportsexpressedinc.i.f.pricewillbegreaterthanthevalueofexportsexpressedinf.o.b.price).Caremustbeappliedwhengatheringimportandexportflowsinthesameindicatorsinordertoavoidinconsistencyinthedata.

where:xi =exportsofcountryiX =worldexport

X

xS i

i =

where:mEU,i=EUimportfromcountryiMEU=EUimportfromtheworld

ii

M

mS

,=

EU

EU

where:ei=exportofproducti bycountryjE=totalexportofcountryjE

eS i

i =


303

In percentage terms, in the first case, the closer the indicator is to 100thehighertherelevanceofthesectorinthestructureofexport(import)ofcountry j. In the second case, the closer the indicator is to 100 the higherthe relevance of partner country k in the structure of export (import) ofcountryj.

Revealed comparative advantage. Balassa (1965) suggested measuringcomparativeadvantagesastheyarerevealedbytradedatausingaspecializationindicator,oftencalled the indexof revealedcomparativeadvantage (RCA).RCA is the ratio between the share of product j in country i’s export(numeratorofRCA)andtheshareofproductjinworldexport(denominatorof RCA). In practice, it detects the relative specialization of country i inexportingproductjonthebasisofj’simportanceinworldtrade.

RCAijisalwayspositive(≥0)and,bybeingexpressedinpercentageterms,RCA>100signalsarevealedcomparativeadvantageofcountryiinproductj.Theindexcanbeadjustedforexaminingcomparativeadvantagesinreferenceareasotherthantheworld(i.e.asingleCARICOMcountrywithreferencetototalCARICOMexport).

RCAdealswith thedifficultiesofmeasuringcomparative advantagesbyobservingtherelativespecializationinexportofproductj.Sincepricescannotbeobservedinconditionsofautarky,measuringcomparativeadvantagesforthe purpose of defining a country’s position in the international divisionof labour becomes rather arduous171. Even if one expects, following thetraditionalneoclassicalapproach,thatacountry’sinternationalspecializationis determined by its relative endowment of production factors, significantproblemsarisethathamperthequantitativeevaluationofsuchendowments.

FortheCaribbeanitisexpectedthattheRCAsaresignificantforproductssuch as bananas, rice and cane sugar, given that preferential trade policiesmeantthattheareaisengagedsignificantlymorethantheworldaverageintheseproducts.

171SeealsoChapter10.

where:ei,k=exportofproductifromjtokEi=totalexportofproductibycountryi

kiki

E

eS

,, =

where:xij =countryi’sexportofproductjXit =countryi’stotalexportxwj =worldexportofproductjXwt =worldtotalexport

100*/

/i j

Xx

XxRCA =

ij it

wj wt

304


Concentration index (HX). This indicator measures the degree ofconcentration of the export structure of a country. This feature makes itinterestingformanydevelopingcountries,whosestructureofexportisoftenhighlydependentonrelativelyfewprimarycommodities.TheHXindexisbased on the Hirschman Index, calculated using the shares of the variousproductsintheexportstructureofagivencountry,

Thesmaller thevalueof the index, the lessconcentratedthestructureofcountryj’sexport.Inthisversiontheindexrangesfrom1to∞,whichmakesitdifficulttocompareamongcountries.However,thisindexcanbenormalized(i.e.forcedtoassumevaluesbetween0and1)bydividingitbythenumberofdifferentproductsthat,theoretically,couldbeexported(n).Thisimpliesthatthevaluestheindexwillassumearedependentonthenomenclatureandthedigitleveladoptedfortheanalysis(i.e.ifthecalculationisperformedforagrifoodsectorbyusing2-digitlevelHSnomenclature,thentherewouldbe24productsconsidered,andn=24).

InthisnewversiontheHXindexiscomparableamongcountriesandtakesthevalue1formaximumconcentration(oneproductcoversallexports).

Indicators of trade similarity.Sometradespecializationindicatorsmeasurethemerchandisesimilaritybetweentheexportflowsoftwocountriesinthesamereferencemarket.Theexportstructuresimilarityindex(ES)comparestherelativedimensionoftheexportsharesforagivenmerchandiseaggregatebetween twocountries towardsa specific targetmarket.Foreach item theshareof total agrifoodexports is considered for eachof the twocountriescompared.172

Inmath:

172Starting from the original export structure similarity index (ES), other related indicators weredeveloped,suchastheproduct similarity index(PSI)andthequality similarity index(QSI)(GrubelandLloyd,1975;FingerandKreinin,1979).InthischapterwegointodetailontheESonly.

where:Si =shareofproductiintheexportofcountryjn=numberofproductexported

2

= ∑

→niisHX

−

−

⇒

−

−

=

∑→

n

ns

n

nHX

HXni

i

norm 11

1

11

12


305

The use of these indices as an analytical instrument for evaluatingcompetitionbetweenagrifoodexportstowardsaspecificmarketisbasedonthe ideathat themoresimilar thestructuresandfeaturesof twocountries’exportsaretowardacommonreferencemarket,thestrongerthecompetitionbetweenthosetwocountriesinrelationtothosegoods.

Theindexvariesbetween0and100:inthefirstcasethesimilarityisnull,whileintheoppositecasethestructuresofthetwoflows(intermsoftradeshares) are identical. The results, however, depend heavily on the level ofdisaggregationadopted173.

��.�.3 The policy analysis matrix (PAM)Asmentioned,thePAMisananalyticalframeworkaimedatexaminingtheimpact of policies based on two enterprise budgets: one valued at marketprices,andtheotherateconomicorsocialprices.Thedivergencebetweenthemarketandeconomicvaluesindicatesthestaticimpactofthepolicysetting,andconstitutesaconvenientwaytoshedlightonthecompetitivenessoftheeconomicsector(s),andtheircomparativeadvantage.

The enterprise budgets used to construct the PAM comprise revenueand cost data for the production and marketing of a specific commodityorganized into two accounting identities. One calculates profit as thedifferencebetweenrevenuesandcost.Theothercalculatesthevalueofthedivergence(distortion)inducedbypolicyasthedifferencebetweeneconomicandmarketvalues.

ThestructureofthePAMmatrixispresentedinTable ��.4.Itallowsforviewing the two accounting identities and readily calculating the profitsand divergences. The first column displays data on revenue. The next twocolumnsseparatethecostitemsintotradableandnon-tradablecomponents,with“value” asa sumofquantityandprice. (Intermediate inputs suchasseeds,fertilizers,pesticidesandtransportationareseparatedintotradableandnon-tradablecomponents.)The finalcolumn iscalculatedasprofits = total revenue – total costs.

ThefirsttworowsofthePAMvaluetherevenuesandcosts(andtherebythe profits) using different valuation methods. The first row uses marketprices,whichcaptures theeffectsofpolicies (distortions).The secondrowuseseconomicprices,whicharetheefficiencyprices,devoidofdistortions.

173Understandably,thehigherthelevelofdisaggregation,thehighertheaccuracyofthecomparisonbetweentradestructures,butalsothelowertheprobabilityofhaving“similar”tradeshares.

( )[ ]∑=i

iBiA xxES 100*,min

wherexiA andxiB arethesharesoftotalagri-industrialexportsofcountryAandcountryB(respectively),regardingitemi.

306


The final row reinforces the second identity, using distortion = market price – economic price;itcapturesthedistortion(ordivergence)inrevenues,costs,andprofits.

TABLE12.4The Policy Analysis Matrix (PAM)

Benefits Costs Net profit

Gross revenue Tradable inputs Domestic factors

Budgetatmarketprices A=Pid* Qi B=Pjd* Qj C=Pnd*Qn D

Budgetateconomicprices E =Pib* Qi F=Pjb*Qj G=Pns*Qn H

Divergences I J K L

Where:Pid = domestic market price of output iPjd = domestic market price of tradable input jPib = economic price of output iPjb = economic (parity) price of tradable input j Pnd = domestic market price of non-tradable input nPns = domestic economic (shadow) price of non-tradable input nQi = quantity of output iQj = quantity of tradable input jQn = quantity of non-tradable input n

ThePAMprovidesavisuallyappealingwayofcapturingandpresentingthedataondivergencesandprofits,whichmaybelabelledasfollows:

privateprofits: D=(A–B–C)socialprofits: H=(E–F–G)outputtransfers: I=A–Einputtransfers: J=B–Ffactortransfers: K=C–Gnettransfers: L=D–H;orL=I–J–KThe PAM allows for calculation of the indicators of policy effects,

competitiveness and comparative advantage. Indicators of the effects ofpolicies on the farm system include the nominal protection coefficient(NPC),theeffectiveprotectioncoefficient(EPC)andtheproducersubsidyequivalent(PSE).TheNPCmeasurestheimpactofpoliciesonproductionprocess and output prices. The EPC measures the effects of policies onvaluedadded(revenuelessvalueoftradedinputs).ThePSEmeasuresthenetcontributionofpolicies tofarmrevenues, that is, thenetvalueof transfersasapercentoffarmrevenuesvaluedinprivateprices.Theprivateprofit isameasureofinternationalcompetitiveness.Ineffectcompetitivenessmeansthattheproductionunitshaveprofitableproductionwiththepolicysupportprovided.


307

TheDomesticResourceCost(DRC)isameasureofcomparativeadvantageandimpliesthattheproductionunitscanhaveprofitableproductionevenintheabsenceofpolicysupport.Theindicatorsofpolicyeffects,competitivenessandcomparativeadvantagemaybecalculatedfromthePAMasfollows:

nominalprotectioncoefficient(NPC) = A/Eeffectiveprotectioncoefficient(EPC) = (A–B)/(E–F)producersubsidyequivalent(PSE) = (L/A)privateprofits = D=(A–B–C)socialprofits = H=(E–F–G)domesticresourcecostratio(DRC) = G/(E–F)The major limitation of the indicators (NPC, EPC, PSE, DRC) and the

PAMis that theytypicallyuse fixed input–outputcoefficients.Asaresult,it is not possible to use them directly to indicate producer or consumerresponsestopolicychangesthatreducedistortions.

��.3 Ex ante policy analysis and equilibrium models

As pointed out above, ex ante policy analysis constitutes an attempt toanticipate the likely result of a policy change, through the building of acounter-factualscenariothatiscomparedwithastatusquoscenariocapableofansweringa“what-if”question.Theanalysisinvolvesthecomparisonoftwodifferentstatesoftheworldonthebasisofsomevariablesof interest.Onestaterepresentsrealityduringagivenperiodoftime,usuallycalledthebaseline (orbasecaseorbenchmark),andtheotherstaterepresentsrealityunderadifferentpolicyoption,usuallyreferredtoasthecounter-factualorpolicyscenario.

Models considered here are economic equilibrium models, in which thesolutioncorrespondsto(at leastsome)marketclearingconditions.Amongthe classes of models available, computable partial equilibrium (PE) andgeneralequilibrium(GE)modelsarethosebasedontheinteractionamongendogenous variables, which is absent in the analyses based on simplestatistical indicators. The large number of microeconomic details involvedmakesthesemodelssuitableforpredictingchangesinproduction,demand,trade,prices,incomesandwelfare.

Thevariablesutilizedinthesetypesofmodelsareclassifiedasexogenousandendogenous.Exogenousarethosewhosevalueisdeterminedoutsidethemodel, while endogenous are those whose value is determined by solvingthe model. Examples of typical exogenous variables in policy analysismodels are the population, the rate of technical change and the policyvariables themselves. Examples of typical endogenous variables includeprices,consumption,productionandtrade.

Dynamic models can be classified as “recursive dynamic”, when thesolutionisbasedontheforecastedvalueoftheexogenousvariablesandon

308


thevaluesof theendogenousvariables in thepreviousperiod. In thiskindofmodels,agents’behaviourisoptimalwithineachperiod,butnotthroughtime174. On the contrary, fully dynamic models can be based on dynamicoptimization, in which solutions are provided in the form of an optimalbehaviouralpath,andbehaviourisoptimizedthroughtime.

Onelastremarkisinorderonthenatureoftheresultsofmodelsusedinex anteanalyses.Theonlyadditionalinformationthatsuchexercisescanconveyis the outcome of a policy experiment, indicating how the world wouldlookwith,forinstance,adifferenttarifforadifferentsubsidy.Bynomeansshould thisbe confusedwith aprovisional exercise,which tellshow somephenomenonmayevolveinthefuture.Theconfusionarisesparticularlywiththeresultsofdynamicmodels,whichoftenneed toutilize theoutcomeofsome provisional exercises – typically for exogenous background variablessuchasthepopulation,GDP,productivityfactors–inordertobuildacrediblerepresentation of the world through time. Provisional exercises, however,pertain to the building of a baseline against which the policy experimentwill be run. Therefore it would be misplaced to judge the “soundness” ofthepolicyanalysisonthebasisof itscapacity toproducecorrect forecaststhatarenormallynotaproductoftheanalysis,butarerathertheproductofdifferentexercisesemployedasastartingpointinpolicyanalysis.

General and specific limitations of equilibrium models (from modellingassumptions,todataquality,topolicyrepresentation)werediscussedabove.ThissectionprovidesanintroductiontoPEandGEapproachesasapplicabletotradepolicyandfoodsecurityanalysis.

��.3.� An introduction to partial equilibrium (PE) modelsOur focushere ison largeglobalPEmodels that includemultiple regionsand countries, and multiple product markets, and that are employed inmacro-level analysis of agricultural policy (particularly agricultural tradepolicy). Usually, these tools include the main agricultural markets only,whilenofactormarketsareconsidered.(Theirfeaturesaresupposedlytakenintoaccountbythevalueoftheparameters.)Demand,supplyandtradeforagricultural commodities are generated simultaneously with equilibriumprices,givenanumberofexogenousmacroeconomicassumptions–suchastheGDP,theexchangerate,theconsumerpriceindexandtechnicalchange–andthelevelofpolicyvariables.Therestoftheeconomyisassumednottobeaffectedby,andnottoaffect.whathappensinagriculture.AbasictextbookreferenceforthistypeofanalysisisFrancoisandReinert(1997).

174Itshouldalsobenotedthatinrecursivedynamicmodels,whichareconsiderablymorefrequentinagriculturaltradepolicyanalysis,theendogenousvariablesofperiodtbehaveinfactasexogenousvariables with respect to the solution of period t+1. They are sometimes referred to as “pre-determined” variables, since they are endogenously computed in one period, but employed asexogenousinthefollowingone.


309

Inthepast,goodswereoftenassumedtobeperfectlyhomogeneous;thishaschangedinsomeofthemorerecentcontributions,severalofwhichassumesomedegreeofdifferentiationacrossthesamegoodindifferentmarkets.Ingeneral,thepartialequilibriumapproachhasbeenemployedextensivelyforcommoditymarkets,undertheassumptionthatallmarketsrepresentedarelinkedby(atleastsome)degreeoftransmissionbetweenprices.

Thistypeofmodelassumesthepresenceofarepresentativeagent,whileitdoesnotprovideindicationsaboutpossibleunderlyingdistributionaleffects.Agent behaviour is assumed to be maximizing, if restrictions are imposedon the parameters, and models are calibrated on a base year, in order torun counter-factual scenarios (if comparative, static) or paths (if recursive,dynamic).

Figure12.1offersaschematicillustrationofthefunctioningofaPEmodelforoneindividualproduct.

A typical partial equilibrium model175 consists of a set of behaviouralequations, a set of equilibrium relations between supply and demand, andasetofidentitiesthataggregatevariables.Equationscanbegroupedintoasupplycomponent,ademandorutilizationcomponent,andaforeigntradecomponent; this pattern is repeated for each region and product included.Inaddition, thereareprice transmissionequations, linkingworldprices to

175Thefollowingcanbereferredtoanumberofmodels,suchastheCosimo-AGLINKmodeloftheOrganisation for Economic Co-operation and Development (OECD) and FAO, or the FAPRImodeloftheCenterforAgriculturalandRuralDevelopment(CARD),ortheEuropeansimulationmodel(ESIM)builtbytheUniversityofBonn.AreviewofsomeoftheseexercisescanbefoundinvanTongerenandvanMeijl(1999).

FIGURE12.1A partial equilibrium net trade model

Pw0 Pw1

Pw0

A Bw

Pw2

Pw0 Pw1

Pw0

A B

Pw2

CountryAisanexporter,whilecountryBisanimporter.Thegraphinthemiddleshowsthetransactionintheworldmarket.IfcountryBimposesanad valoremtariffraisingthepricefromPw0toPw1theworldpricewillbereducedfromPw0toPw2,andtradewillbereducedaccordingly.

3�0


BOX ��.�The structure of a standard partial equilibrium agricultural model

Crop products Livestock products supply(1) si,n = s(pv,i,n, pv,j,n, Pols) (8) ci,n = c(pz,i,n, pz,j,n, Polc)(2) rv,i,n =r(pv,i,n, PR) (9) AL = al(pv,i,n, pv,j,n)(3) Qov,i,n = si,n rv,i,n (10) rz,i,n = r(pz,i,n, AL, PR) (11) Qoz,i,n = ci,n rz,i,n

demand(4) Cuv,i,n = cu(pv,i,n, Yn, POPn) (12) Qd z,i,n = qd(pz,i,n, Yn, POPn)(5) AAv,i,n = aa(Qoz,i,n) (6) SEv,i,n = se(sv,i,n) (7) Qdv,i,n = Cuv,i,n + AAv,i,n + SEv,i,n

price transmission (13) pi,n = p(pi,w, tc, Polp)

trade (14) (E i,n - Ii,n) = Qoi,n - Qd i,n

closure (15) Σ (E i,n - Ii,n) = 0

where:i,j=products E=exportsv=crops I=importsz =livestock tc=exchangeraten=country PR=yieldtrendand Y=GDPs=land(hectares) POP=populationc =heads(number) pn=priceincountrynAL =indexoffeedcost pw=worldpricer=yield(perhectareorperhead) AA=demandforfeedPolp=policiesdirectlyaffectingprices SE=demandforseedsPols=policiesbasedonland Qd=totaldemandPolc=policiesbasedonlivestockheads Qo=supplyCu=demandforhumanconsumption


3��

domestic prices, and world market equilibrium conditions that closes themodel.

A simplified representation of the standard structure of the modelsconsideredisshowninBox ��.�.

Thesupplycomponentconsistsofequationsforcropsandfor livestock;supplyisobtainedastheproductofayieldperhectareoflandorperhead,timesthenumberofhectaresemployedortheherdsize.Yieldsdependonatrendvariable(whichisusedtorepresenttechnicalchange)onoutputpricesandonfeedcostsforlivestock.Theseareincludedinanaggregatefeedpriceindex.Landandheadsallocationdependsonrelativeoutputprices,andonthepoliciesdirectlyaffectingtheirallocation.

Thistypeofmodelingissimplifiedinseveralrespects.First,productionisentirelydeterministic:nouncertaintyfactors,suchasclimaticvariability,areaccountedfor.Noassumptionsaremadeconcerningfarmers’attitudetowardrisk,unless theyare included in theparameters. Inputdemand is includedonlyforland,herdsandwhereprimaryproductsareemployedasinputsintheproductionofother (processed)goods included in themodel, as is thecasewithfeedcrops,oilseed(whereseedsareinputsformealcakesandoils)andindairyproduction,wheremilkistheinputofbutter,cheesecasein,etc.The demand for non-agricultural inputs, such as fertilizers, pesticides andmachineryisnotincluded.Landuseandherdsizedependsolelyonthepriceobtainedforagriculturalproducts,ratherthanonthepricesoflandandheadsthemselves.

The demand component for crops consists of an aggregation, by meansofanidentity,oftheamountusedforhumanconsumption,forfeedandforseeds.Forlivestock,onlyfeedisincluded,alongwiththepricesofproducts,thedemandforhumanconsumptionincludesthepricesofafewmoredirectsubstitutes, together with the GDP level and the population as exogenousshifters.Thedemandforfeedisdirectlyrelatedtothenumberoflivestock,through technical coefficients. By the same token, the demand for seed isdirectlyrelatedtothenumberofcultivatedhectares.

Thetypicalpartialequilibriummodelconsideredhereiscomparativestatic,and does not include stock formation. This choice is usually justified byconsideringthatstockscannotbeincreasedordepletedafteragivenpoint,and thus, their variation must add up to zero. Nonetheless, the absenceof stocks from the model can be a problem, especially in modeling thosemarketswheretheymayassumeastructuralcharacterandmaysignificantlyaffectthebehaviourofeconomicagents.

Inthemorestandardapplicationsthetradecomponentismadeupofexcesssupplyequations.Goodsproducedindifferentcountriesareassumedtobeperfectlyhomogeneous,andworldmarketsaretreatedasasinglearbitragemechanismofexcess supplies.Allmarkets influenceprices throughout the

3��


model,thatis,pricechangesoccurringinonemarketarealwaystransmittedto all the others. The closure rule is defined by the sum of the excesssuppliesinallmarkets,whichhavetoadduptozero.Thesolutiongeneratescountries’nettradepositions,butitdoesnotincludeinformationonbilateraltradeflows.

Apopularalternative to thisapproach in the tradecomponent is theso-calledArmingtonassumption,whichisbasedontheideathatsubstitutabilitybetweendomesticandforeignproducts ineachmarket is less thanperfect.FrancoisandHall(1997)offerasimpletreatmentofthisapproachwithinaPE setting. This assumption allows the endogenous generation of bilateraltradeflows,sothatthemarketclearsthroughthesumoftotalexportsandtotalimportsinthemodel.

Thepossibilityofgeneratingendogenouslybilateraltradeflowsisindeeda very important feature of the model in the exercises aimed at analysingdiscriminatory trade policies, such as preferential trading schemes, or anyotherprovisionwhichdoesnotapplymultilaterally.HencethepopularityoftheArmingtonapproach,whatever its limitations.Anextensivediscussionof the limitation of this approach and of the models capable of handlingsatisfactorily bilateral trade and discriminatory policies may be found inAnania(2001).

��.3.� An introduction to general equilibrium (GE) modelsThe basic structure of a GE model can be described through blocks ofrelations dealing with production, consumption, factors market, savings/investmentandthebalanceofpayments.Basicreferencesforthismodellingapproachare inGinsburghandKeyzer (1997)andDevarajanet al. (1997).De Muro and Salvatici (2001) offer a useful review of the different typesofmodelswithin thisapproach,highlightingrelevantmatters related to itspotentialandactualusesintheanalysisofagriculturalandtradepolicy.

GE models are characterized by the fact that they take into account allactivitiesandinstitutionalentities(suchashouseholds,governmentandfirms)without assuming absence of feedback effects for any of them within theeconomy.Inprinciple,thischaracteristicistotallyindependentfromthedetailsincludedinthemodel:evenifitisformulatedforonesinglegood,oneproducerandoneconsumer,thesewillberepresentativeoftheentireeconomy.

Averybasicrepresentationcanbeprovidedwiththegraphshownbelow,with reference to one consumer who is also a producer. Given an initialfactorendowmentwhichdefinestheproductionpossibilityfrontierofgoodsX1 and X2, and the utility function parameters which define the structureof the indifference map, an autarky equilibrium is found in Q, where theindifferencecurveandtheproductionpossibilityfrontieraretangenttotheisocost line. If we allow for trade (within a small, open economy setting),therepresentedcountrywillhaveaccesstoaconsumptionlevelofC,while


3�3

producinginP,givenexogenously-definedtermsoftrade,andwillbetradingthedifferencebetweenthesetwopointsdefinedintermsofthetwoproducedgoodsX1andX2.

In more realistic settings, GE models include three sets of relations,determining real flows, expenditure and income, respectively. Moreover,equilibrium conditions make it possible to “close” the model, by definingendogenousandexogenousvariables.Asetofidentitiesensuresthatincomedoesnotexceedexpenditureandequalsthatofthefactorsofproduction.

A very simple example, with only initial factor endowments and utilityfunctionparameterbeingexogenous (whileall the rest is calculatedby themodel),isgiveninBox ��.�(adaptedfromMagnaniandPerali,2002).

In their more standard setting, GE models are solved by imposing theequilibriumconditioninallmarkets,followingtheso-calledWalraslaw.Itisimportanttohighlightthattheclosureruleacquiresafundamentalmeaning,whichisthatofdefiningthebeliefsoftheanalystontheultimatemechanismthatregulatestheeconomy.A“neo-classical”closurewillthereforeattributeapropulsiverole tosavings,andassumethat the levelof investmentvariestoensureequivalencebetweenthetwo.A“Keynesian”closure,instead,willpermit theexistenceofunemployment,andensureequilibriumthrough itspresencewithanendogenouslabourdemand.Inotherexercises,inwhichadecisiveroleisattributedtoinvestment,thesewillbeadjustedtosavings,andconsumption will be determined by sales. In other exercises, it is assumedthat factors of production are not paid for according to their marginalproductivity,andequilibriumisachievedthrougharedistributionofincome,which influences the savings rate. By representing the entire economy,therefore,GEmodelsmakeallassumptionsmoreexplicit.

FIGURE12.2A general equilibrium model

Q

P

Q

X1

X1

X1

X1

C

Q

P

Q

X1

X1

X1

X1

C

3�4


BOX ��.�The basic structure of a general equilibrium model

ProductionGoods – sectors Production functionagriculture XSagr=f(L,K)textiles XStex=f(L,K)

ConsumptionAgents Utility functionrural Urur=f(Xagr,Xtex)urban Uurr=f(Xagr,Xtex)

Factors of productionEndowments

labour L=Lcapital K=K

Variablesendogenous exogenouspi priceofgoodi LS

h Labourendowmentofconsumerhw wage KS

h Capitalendowmentofconsumerhr Returntocapital ahi UtilityfunctionparameterXS

i Supplyofsectori

hXDi demandfori ofconsumerh

DLi Labourdemandofsectori

DKi CapitaldemandofsectoriYh Incomeofconsumerh

Thetradecomponentcanbespecifiedeitherasaresidualofthedomesticmarket–whichisespeciallythecaseinthoseexercisesinvolvingonecountrywhere the small,openeconomyassumption isoftenadopted–or throughanArmingtonstructure,asforPEframeworks.TheArmingtonhasgainedpopularityinGEmodellingalso.

In order to become “computable”, general equilibrium models (thecommonacronymisCGE)require:

• adatabasedescribingtheflowsofresourcesintheeconomy,atthelevelofaggregationconsideredinthemodel;and

• asetofparametersforthebehaviouralrelationsofthemodel.


3�5

The database on which a CGE is based is known as Social AccountingMatrix (SAM): it is a consistent set of accounts describing resource flowsbetweenconsumers,producers,thegovernmentandforeigneconomies.

Parameterscanbeobtainedthroughcalibrationorestimation.Inthelattercase, once the database to be employed as a benchmark equilibrium forthe economy has been constructed, the model will be solved in “reverse”mode,sothatthesolutionwilldeterminethevaluesoftheparametersthatare compatible with the known values of the exogenous variables, and ofthe endogenous variables of the benchmark equilibrium. Given that thebenchmarkperiod isnormallyrepresentedbyoneobservation–either the

BOX ��.�Continued

Equations

real flows expenditure

XSi=f(DLi,DKi) supplyofgoodi hXD

i=ahi(Yh/pi) demandfori

w=( ∂ XSi/ ∂ XiDLi)pi

labourdemand

r=( ∂ XSi/ ∂ XiDKi)pi

capitaldemand equilibrium conditions

XSi= ∑h hXD

i

demandequalssupply

income flowsYh=wLS

h+rKSh ∑i DLi= ∑h

LSh

demandforlabourequalslabourendowment

∑i DKi= ∑hKS

h

demandforcapitalequalscapitalendowment

identities

PiXSi=DLiw+DKir

Yh= ∑i hXDipi

AdaptedfromMagnaniandPerali(2002).

3�6


referenceyearorsomeaverageofafewyears–itisclearthatthecalibrationprocedure generally does not make it possible to assess the statisticalreliabilityoftheparametersobtained.

Toavoidthisproblemonesolutionistheestimationofparameters,whichinvolvescalculatingthemthrougheconometrictechniques.Whiledesirable,this isusuallyunfeasible.Firstly, theaverage sizeofaCGEmodel impliesthe need to estimate a high number of parameters, which increases withthenumberofsectorsandhouseholdsconsidered.Inturn,thisimpliesthata large number of consistent observations must be available, especially ifparameters are to be estimated simultaneously. Separate estimations formodelblocks–suchasoneforproduction,onefordemand,oroneforeachproduct – still would not take into account all the equilibrium conditionsconsideredinthemodel.

��.3.3 Which model is better?Compared to the PE approach, the GE approach removes one majorsimplifying assumption. In fact, when some activities are excluded fromtheanalysis,itisassumedthatwhathappensinoneactivitydoesnotaffectdemandandsupplyinthesectorsthatareconsidered.Thisalsoappliestothefactormarket,which is seldom included in thePEmodelsemployed inex anteanalysisofpoliciesforagriculturalproducts.TheextenttowhichthisisanacceptableassumptiondefinestheextenttowhichaPEanalysiscanbesuitableforaparticularproblemathand.Thepossibilityofincludingahigherlevelofdetail,whichhasfrequentlybeenconsideredasadriverofthechoicein favourofPEmodels– tends tobe an increasinglymisplacedargument,sincethepowerofcomputationaltoolsseemsnottopreventthespecificationofrelativelylarge-sizemodels.

Instead,itisaquestionofthefocusoftheanalysis.Ingeneralterms,PEandGEmodelsfarebetteratrepresentingredeploymentofresourcesthanatcapturingproductivityandgrowth. If theaim is tounderstandchangesin agricultural supply and demand, a PE framework can provide usefulanswers,especiallyiftheanalysisincludesmanypolicydetails.Ifthefocusis more general, and answers have to be provided in terms of changes inincome, factor allocation or distributional consequences, then the modelmustaddressthelinkagesbetweentradeandtheseaspects,andaGEappearsmoreappropriate.Inthisrespect,aGEapproachmaybeamoreeffectivechoice when analysing issues in which it is important to highlight theexistenceand theeffectsofageneral“budgetconstraint” in theeconomy,sothatchangesintheresourceallocationimplysignificantfeedbackeffectstobetakenintoaccount;andwhenconsidering thesecond-roundeffectsofpolicychanges.

DealingwiththeseaspectswithinaGEframeworkrequires:• moredatatobeassembledandmadecoherent;


3�7

• moreparameters tobeeitherestimatedorderivedthroughcalibration;and

• moreexplicithypothesesonthefunctioningoftheeconomyandonallmarketsrepresented.

Moreover,giventhatanaccuraterepresentationofalltheseaspectsusuallyrequires an increase in the number of non-linear relations included in themodels, a more complex representation may involve more difficulties insolvingthemodelusingstandardalgorithms.

In summary, “it may be difficult to justify devoting otherwise scarceresources to more complex and less transparent models, when they mayyield only marginal extensions of the basic insights taken from simplerapproaches” (Francois and Hall, 1997, p. 122). In fact, among the modelsemployed in theanalysisof agriculturalpoliciesGEapproacheshavebeenusedmorefrequentlyinthosecasesinwhichagricultureformsalargeshareoftheeconomy;andinthosecasesinwhichthefocusismoreonintersectoraleffectsratherthanonthepeculiaritiesofsingleproducts.However,inrecentyears, given the increased power of computers and the easier exchangeof information among analysts, the use of GE models has become morecommon.

��.3.4 Where to begin? As mentioned earlier, policy analysts around the world are benefitingincreasingly from networking, and a number of initiatives have beenundertakenaimedatsharingdata,modellingcodesandotherresourcesthatcancontributetoloweringsignificantlythecostsinvolvedinstartingquantitativetrade policy analysis. Here we present some key networking experiences,withtheaimofprovidingthereaderwithpracticalstartingpoints.ForPEmodels, reference will be made to a number of networking experiences,including those related to the Agricultural Trade Policy Simulation Model(ATPSM), jointly built by the United Nations Conference on Trade andDevelopment(UNCTAD)andFAO;andtotheCosimo-AGLINKmodel,jointly developed by the Organisation for Economic Co-operation andDevelopment(OECD)andFAO.ConcerningGEmodels,referencewillbemadetotheexperienceoftheGlobalTradeAnalysisProject(GTAP),whichisprobablythecaseinwhichnetworkinghasdevelopedthemost.

ATPSMTheAgriculturalTradePolicySimulationModel(http://r0.unctad.org/ditc/tab/atpsm.shtm) is a deterministic, comparative, static, partial equilibriummodel of world agricultural markets, built by UNCTAD and FAO. Themodelanddatabasearepubliclyavailable.Themodelisintendedtoserveasatoolforquantifyingtheeconomiceffectsattheglobalandregionallevelsofrecentchangesinnationaltradepolicies,andtoanalysepotentialchanges

3�8


that would result from future policy reforms in individual countries. Itprovidesestimatesofchangesintradevolumes,pricesandwelfareindicatorsassociatedwithchangesintradepolicies.

Domestic supply and demand equations are specified as functions offarm and wholesale prices (respectively) in proportional terms, allowingfor cross-effects in production and substitution in consumption. Exportsupplyisproportionaltoproduction,whileimportsarederivedasaresidualof thedomesticmarket.Thetradecomponentof themodel isessentiallyaresidualofthedomesticmarkets,andthereforethemodelcanbeemployedtocomputechangesinthenettradepositionsofthecountries.Concerningpolicies, themodel includesad valorem tariff equivalents, export subsidiesand the domestic subsidy component which exceeds trade protection, inordertoavoiddoublecountingthetrade-distortingeffect.Wheretariff-ratequotasareimplemented,thedomesticpriceiscomputedonthebasisoftheout-of-quotatariff.

Giventhenatureofthetradecomponent,themodelhasbeenusedmainlyto study the impact of multilateral trade agreements; a recent applicationis in Poonith and Sharma (2004). The results are easily accessible, and theassociated software isparticularly simple and intuitive (anExcelversion isalsoavailable),whichalsofacilitatesitsuseincapacity-building.

AGLINK and Cosimo-AGLINKAGLINK is a partial equilibrium dynamic model of world agriculturebuilt by the OECD Secretariat in cooperation with its member countriesandanumberof independentconsultants.Results,whicharegeneratedonthe basis of member-country responses to questionnaires, are employedin the preparation of the OECD Medium-term Outlook, the periodicalreporting the medium-term forecasts on the market development formain agricultural commodities in OECD countries and their main tradingpartners.ThemodelwasalsousedforseveralpolicyexperimentsrunbytheOECD Secretariat.The model assumesperfect competition in all markets,and perfect homogeneity for products from different countries. For mostproducts and countries, trade is the residual of the domestic market, andthereforethemodeldoesnotgeneratebilateralflows.AGLINKisveryrichintherepresentationofpolicies;itexplicitlytakesintoaccounttariffs,exportsubsidies, domestic subsidies and taxes, and complicated mechanisms likefloorpricesandtariff-ratequotas.ThemodelisavailabletotheOECDandanetworkofauthorizedco-operators.

Cosimo is a partial equilibrium dynamic agricultural model, built as acomplementtotheAGLINKmodeloftheOECD176.Thetwomodelscanbesolvedsimultaneously,andCosimocontainsboththecountriesandregionsincluded inAGLINK,plus thedetails for countrieswhichwereoriginally

176Seehttp://www.fao.org/docrep/meeting/009/J4756e.htm.


3�9

included in the “Rest of the World” region of AGLINK. Like AGLINK,Cosimo is aimed at generating medium-run market outlooks (jointlypublishedbyFAOandtheOECD)andatconductingpolicysimulations.

Cosimowasdevelopedbyconsidering a slightlydifferentproduct spacethan AGLINK, reflecting the production and consumption mixes of thecountries involved. The product structure of the model is flexible, basedon a number of aggregated products for which market clearing conditionsarespecified,whicharemadeupofdifferentindividualproductsspecifictoeachcountrymodule.Therefore,forinstance,theaggregate“coarsegrains”mightbemadeupofmaize,barleyandsorghuminonecountry,whileitmayincludemilletandoatsinanothercountry.

Concerning policies, the model includes both bound and applied ad valorem tariffsinthepricetransmissionequations.Otherpoliciesconsideredare some of the more important tariff rate quotas (TRQs), throughconditional statements; intervention prices, which are also introduced asconditionalstatements;anddirectpayments,whicharemodelledassubsidiesaffectingthereturnsperhectareinthelandallocationsystem.

Parameters are derived from a number of sources, including availableestimatesfromtheliterature(particularlyfromtheWorldFoodModel,theFood and Agricultural Policy Research Institute (FAPRI) and the USDAmodel),calibrationthroughtheconstraints imposedonthesystem,adhocestimationandmodelvalidationwithhistoricalsimulation.Allarecheckedandvalidatedbyspecializedcommodityanalysts.

Cosimoiscurrentlyemployedtoproducethemedium-termoutlooksonkeyagriculturalmarkets,withinajointexercisewiththeOECDSecretariat,and is not publicly available. However, databases on which the model isbasedareavailablefromtheOECDwebsite,andtheworkinggroupoftheCommodities and Trade Division of FAO can be contacted to verify thepossibilityofextending/detailingthemodelforsomeregions,andofrunningparticularpolicysimulationexperiments.Beingessentiallyanet-trademodel,Cosimo is more suited for analysing phenomena that involve the globalmarkets and non-discriminatory policies, such as MFN tariffs reductions,and less suitable for analysing discriminatory policies, such as preferentialtradeschemes.

The Global Trade Analysis Project (GTAP)TheGTAP(https://www.gtap.agecon.purdue.edu/)waslaunchedinthe1990swiththeideaofbuildingaglobalgeneralequilibriummodelanddatabaseforanalysing trade policies. The initiative gradually evolved into a worldwidenetworkofpayingusers,sharingacommonstartingpointinglobalgeneralequilibrium analysis. The database and the associated standard models areavailable fora fee,whichvariesaccording to thedegreeofparticipation intheproject.

3�0


The GTAP was initiated by Purdue University in the United States,in cooperation with other research institutions around the world, whichformedaconsortium.Theprojecthasdevelopedconsiderablyovertheyears,due inpart to theactiveparticipationofa largepoolof institutionswhichincludes, amongothers, theWorldBank, theWTO,UNCTAD,FAO, theInternational Food Policy Research Institute (IFPRI), the MassachusettsInstituteofTechnology,andalargenumberofnationalagencies.Severalveryinfluentialanalyseshavebeencarriedoutonthebasisofthedatabaseandtheassociatedmodel.

TheGTAPstandardmodelisaperfectlycompetitive,comparative,static,general equilibrium computable framework (Hertel, 1997). A standarddynamicversionhasbeenmadeavailablerecently.Thestructureofdemandand supply, which is homogeneous across regions and products, is builtupon the social accounting matrices of individual countries and regions,whileparametersaredrawnmostlyfromtheliteratureandcalibratedonthereferencedatabaseperiod.Themodelassumesthepresenceofrepresentativeconsumers and producers together with a government sector, and allincomesareassumedtoaccruetoasingle“regional”household.Therefore,alldistributionalaspectsareoverlooked,andallconsumersareassumedtopurchaseallgoods.Bythesametoken,governmentcostsandrevenuesdonotneedtobalance,asitisassumedthatanydiscrepancyaccruesdirectlytothehouseholds(i.e.thesingle“regional”household).Government’sconsumptionbehaviourisendogenous,whilepoliciesareexogenous(Hertel,1997).

Substitutabilityamongprimaryfactorsandwithintermediateconsumptionismodelledthroughasetofnestedconstantelasticityofsubstitutionsystems,whiletheproductionoffinalgoodsisaggregatedthroughafixedcoefficientfunction of the Leontiev type. On the demand side the representativeagent allocates his or her income among savings, government and privateconsumption through a Cobb-Douglas utility function, while allocationwithin different private goods is modelled through a constant differenceof elasticity demand system. Bilateral trade flows are modelled throughproductdifferentiationonthedemandside,withtheassumptionofimperfectsubstitutability between similar goods produced in different countries andregions.Transactioncostsarealsoaccountedforinthemodel,astransportservicesareexplicitlyconsideredamong theactivities in theeconomy.Thestandard model adopts the Walrasian closure rule, by which investment atthegloballevelisadjustedtoglobalsavings,andthebalanceofpaymentsisendogenousinindividualcountriesandregions.

The most recent publicly available database version (Version 6) includesdata on up to 92 regions and countries, 57 industries and 5 endowments,andreferstoyear2001asabaseperiod.Ingeneral,therearetwogroupsofdata which are of particular relevance for global models: those on borderprotection and those on bilateral trade flows. The GTAP database is built


3��

from the COMTRADE data, supplied by the United Nations StatisticalOffice, through an ad hoc reconciliation procedure based on a reliabilityindicator of the information supplied by each importing and exportingcountry.TradepolicydataareretrievedfromtheMacMapsdatabase(Bouëtet al.,2001),whiledataondomesticsupport inagriculture isbasedontheOECDandUSDAproducersupportestimates.Exportsubsidiesaredirectlyderivedfromcountries’notificationstotheWTO.

References

Ackerman, F. 2005. The shrinking gains from trade: a critical assessment of DohaRound projections. Global Development and Environment Institute Working Paper05-01.MedfordMA,TuftsUniversity.

Anania, G.2001.Modelingagricultural trade liberalizationand its implications fortheEuropeanUnion.Osservatorio sulle politiche agricole dell’UEWorking Paper no. 12.Rome,IstitutoNazionalediEconomiaAgraria(INEA).

Balassa, B. 1965. Trade liberalisation and “revealed” comparative advantage. The Manchester School,33:99–123.

Bouët A., Fontagné L., Mimouni M. & Pichot, X. 2001. Market access maps: Abilateralanddisaggregatedmeasureofmarketaccess.Cepii Document 18.Paris,Centred’EtudesProspectivesetd’InformationInternationale.

Cipollina M.P. & Salvatici, L. 2006. Measuring protection: mission impossible?TRADEAG Working Paper 2006-0� (availableathttp://lotus5.vitamib.com/hnb/tradeag/tradeag.nsf/Web/Frame?openform).

De Muro P. and Salvatici L. 2001TheCommonAgriculturalPolicyinMultisectorialModels, Working Paper n 11, INEA, Osservatorio sulle Politiche Agricoledell’UnioneEuropea,http://www.inea.it/opaue/WP11.pdf

Devarajan S., Go, D., Lewis J., Robinson S. & Hall, H. 1997. Simple generalequilibriummodelling.InJ.F.Francois&K.A.Reinert,eds.Applied methods for trade policy analysis. A handbook. Cambridge,UK,CambridgeUniversityPress.

Drysdale, P. & Garnaut, R. 1982.Tradeintensitiesandtheanalysisofbilateraltradeflowsinamany-countryworld.HitotsubashiJournalofEconomics,22:62–84.

FAO. 2006.Tradepolicysimulationmodels:estimatingglobalimpactsofagriculturaltradepolicyreformintheDohaRound.FAO trade policy technical notes no. 13.Rome.

Finger, J. & Kreinin, M. 1979.Ameasureof`exportsimilarity’anditspossibleuses.The Economic Journal,89(356):905–912.

Francois, J. & Hall, H. 1997.Partialequilibriummodelling.InJ.F.FrancoisandK.A.Reinert,eds.Applied methods for trade policy analysis. A handbook. Cambridge,UK,CambridgeUniversityPress.

Francois, J. & Reinert, K., eds. Applied methods for trade policy analysis. A handbook. Cambridge,UK,CambridgeUniversityPress.

3��


Ginsburgh, V. & Keyzer, M. 1997. The structure of applied general equilibrium models.Cambridge,MA,MITPress.

Grubel, H. & Lloyd, P.1975.Intra-industry trade: the theory and measurement of international trade in differentiated products.London,Macmillan.

Hertel, T. 1997.Global trade analysis: modelling and applications.Cambridge,UK,CambridgeUniversityPress.

Hertel, T. & Reimer, J.2005.Predictingthepovertyimpactsoftradereform.Journal of Int. Trade and Econ. Dev.,2005,14(4):377–405.

Hertel, T. & Winters, A., eds. 2006. Poverty and the WTO. Impacts of the Doha Development agenda. Washington, DC, World Bank and Basingstoke, UK,PalgraveMacmillanUK.

Hoekman, B., English, P. &. Matoo, A., eds. 2003. Development, trade and the WTO. A handbook.Washington,DC,WorldBank.

Laird, S. 1997. Quantifying commercial policies. In J. Francois & K. Reinert,eds. Applied methods for trade policy analysis. A handbook. Cambridge, UK,CambridgeUniversityPress.

Leamer, E.E.1988.“MeasuresofOpenness”,inBaldwin,R.E.,TradePolicyIssuesandEmpiricalAnalysis,UniversityofChicagoPress,Chicago.

Lucas, R. 1976. Econometric Policy Evaluation: a critique. Carnegie-Rochester Conference Series on Public Policy,1:19–46.Amsterdam,North-Holland.

Magnani, R. & Perali, F. 2002. Laboratorio di Politica Economica. Benessere,EquilibrioGenerale,eSviluppoEconomico,UniversitàdiVerona.

Monke, E. & Pearson, S. 1989.The policy analysis matrix for agricultural development.Ithaca,NY,CornellUniversityPress.

Perkins, D. & Syrquin, M.1989.Largecountries:theinfluenceofsize.InH.Chenery&T.Srinivasan,eds.Handbook of development economics,Vol.II.Amsterdam,North-Holland.

Poonith D. & Sharma, R. 2004. The impact of WTO negotiating modalities ondevelopingcountries.InG.Anania,M.Bohman,C.Carter&A.McCalla,eds.Agricultural policy reform and the WTO: Where are we heading? Cheltenham,UKandNorthampton,MA,EdwardElgarPublishingLtd.

Taylor, L. 2004.Reconstructing macroeconomics: structuralist proposals and critiques of the mainstream.Cambridge,MA,HarvardUniversityPress.

Van Tongeren, F. & van Meijl, H., eds.1999.Reviewofappliedmodelsofinternationaltradeinagricultureandrelatedresourceandenvironmentalmodelling.Agricultural Economics Research Institute Report 5.99.11.TheHague,LEI.


3�3

Appendix12.1

Glossaryoftradeterminology�77

AD VALOREM TARIFF Atariffcalculatedonthevalueofthedutiableitemandexpressedasapercentageofthevalueofgoods;forexample,10percentad valoremmeans10percentofthevalueoftheenteredmerchandise.

AGREEMENT ON AGRICULTURE (AoA)

AWTOagreementestablishingrulesandcommitmentstoensureafairandmarket-orientedsystemfortradeinagriculturalgoodsandproducts.TheAgreementonAgricultureconsistsofrule-basedcommitmentstoreduceprotectionandsupportofagriculturalgoodsandproductsoveraspecifiedimplementationperiod.TheUruguayRoundofAgreementonAgriculturesignedin1994wasthefirstmajorinternationalagreementonagriculture.

AGREEMENT ON RULES OF ORIGIN

AWTOagreementaddressingtherulesthatdeterminethecountryoforiginofanimportedproduct.UsuallyapplicableamongmembersinaFreeTradeAgreement.Adecisionbyacustomsauthorityonorigincandeterminewhetherashipmentfallswithinaquotalimitation,qualifiesforatariffpreferenceorisaffectedbyananti-dumpingduty.

AGREEMENT ON SUBSIDIES AND COUNTERVAILING MEASURES

Theagreementpermitssignatoriestoimposespecificdutiesonimportstooffset–or“countervail”–thebenefitsofsubsidiestoproducersorexportersprovidedbythegovernmentoftheexportingcountry.

177Thisglossarydrawssomeofitsdefinitionsfromthefollowingsources:FAO,UNCTAD,USAIDandWorldBank.

3�4


APPLIED TARIFF Thetariffactuallyappliedbyacountryatitsborder.WhenthecountrybelongstoWTO,appliedtariffsrespecttheceilingdefinedbythebound tariffsagreedupon(MFNrates).ManycountriesactuallyapplytariffslowerthanMFNrates.AlargersetofcountriesappliestariffslowerthanMFNunderpreferentialagreements(freetradeagreements,systemofgeneralizedpreference,preferentialaccessforcertaincountriesorregions,specialagreementwithdevelopingcountries,etc.).Appliedtariffsalsoincludethelower-than-MFNtariffsagreeduponintheWTOframeworkthatareappliedwithintariffratequotas(called,ingeneral,the“in-quotatariffs”).

ARBITRATION Anarrangementthroughwhichtwopartiestoadisputeagreetotheappointmentofanimpartialchairpersonoragroupofcompetentpersonstodecidethedisputedissueandagreeinadvancetoabidebythedecisionrendered.

BALANCE OF PAYMENTS

Thedifferencebetweenthefundsreceivedbyacountryandthosepaidbyacountryforallinternationaltransactions.

BALANCE OF TRADE (BOT)

Thevalueofacountry’sexportsminusthevalueofitsimports.

Base tariffs Thebasetariffswerethe1995MFNtariffs,whichhadtobedecreasedovertheimplementationperiodoftheUruguayRoundagreement.Anumberofdevelopingcountrieswerefreetodecidethebasetariffonwhichthereductioncommitmentswereapplied.Theresultingtariffiscalledthebound tariff,i.e.aceilingtariffattheendoftheimplementationperiod(2005).

Binding overhang Expressionusedwhenacountryhassetabound tariffatalevelhigherthanthetariffappliedinpractice(often,inordertomaintainamarginforapossibleincreaseinapplied tariffuptotheboundtariff).


3�5

Border price CanbebasedonFOB(freeonboard)orCIF (cost,insuranceandfreight)prices.FOBareusuallyadoptedforexportvalues,whileCIFvaluesareusuallyadoptedforimports.

BOUND TARIFF (BINDING)

MaximumtariffratesresultingfromGATTnegotiationsthatareincorporatedintoacountry’sscheduleofconcessionsandenforceableasanintegralelementoftheWTOregime.Bindingisaprovisioninatradeagreementthatnotariffratehigherthantheratespecifiedintheagreementwillbeimposedduringthelifeoftheagreement.

CAIRNS GROUP Agroupofagricultural-exportingnationsestablishedtodevelopacommonnegotiatingpositionfortheUruguayRound.ItcomprisesAustralia,Argentina,Brazil,Canada,Chile,Colombia,Fiji,Hungary,Indonesia,Malaysia,NewZealand,thePhilippines,ThailandandUruguay.

CARIBBEAN BASIN INITIATIVE (CBI)

Apreferentialtradingarrangementthatcameintoeffecton1 January 1984andprovidedseveraltariffandtradebenefitstomanyCentralAmericanandCaribbeancountriesexportingintotheUnitedStatesmarket.

CIF Acommercialtermmeaningthatthesellingpriceincludesall“costs,insuranceandfreight”foranygoodssold.Thesellerarrangesandpaysforallrelevantexpensesinvolvedinshippinggoodsfromtheirpointofexportationtoagivenpointofimportation.

COMMON AGRICULTURAL POLICY (CAP)

AsystemofEUagriculturalsubsidiesandlevies.Thesesubsidiesworkbyguaranteeingaminimumpricetoproducersandbydirectpaymentofasubsidyforparticularcropsplanted.

COMMON EXTERNAL TARIFF (CET)

Atariffrateuniformlyappliedbymembercountriesofacommonmarketorcustomsunion,suchastheEuropeanCommunity,toimportsfromcountriesoutsidetheunion.

3�6


COMPETITION POLICY

Legislationandregulationsdesignedtoprotectandstimulatecompetitioninmarketsbyoutlawinganti-competitivebusinesspracticessuchascartels,marketsharingorpricefixing.

COMPOUND TARIFF Acombinationofanad valoremtariffplusaspecifictariff.Alsocalleda“mixedtariff”.

Current access The1994Marrakech(UruguayRound)AgreementonAgriculturespecifiedthataftertariffication,currentmarketaccess(i.e.thelevelofimportsthatexistedduringthereferenceperiod)hadtobemaintainedorincreased.Forsomecountries,thiswasachievedbytheopeningoftariffratequotas,called“currentaccess”quotas(asopposedtoquotasopenunderminimum access).

CUSTOMS UNION Agroupofcountriesthatadoptfreetrade(zerotariffsandnootherrestrictionsontrade)ontradeamongthemselves,andthatalsoagreetolevythesametariff(onagivenproduct)onimportsfromoutsidethegroup.

De Minimis IntheWTOUruguayRoundAgreementonAgriculturethisreferstotherulespermittingexemptionofnotificationofassistancefromAggregateMeasurementofSupport(AMS),relatedtodomesticsubsidies-ifthatsupportisbelowacertainthreshold-thetotalmustmakeupnomorethan5%ofthetotalvalueofagriculturalproductionofthatproductorwhereitisnotproductspecific,itshouldnotbemorethan5%ofthevalueofagriculturalproduction-fordevelopedcountries.Thevaluesare10%fordevelopingcountries.

Dirty tariffication Tarifficationistheconversionofnon-tariffbarriersthatexistedforagriculturalproductsbeforetheUruguayRoundintotariffsmeanttobringanequivalentlevelofprotection.Somecountries,however,aresaidtohavesetbasetariffsatalevelhigherthantheoneactuallyprovidedbythemeasurestheyreplaced.Thispracticeiscalled“dirtytariffication”.


3�7

EFFECTIVE TARIFF RATE

Theconceptthat“effectivetariff”protectionforaproductdependsontariffandothernon-tariffbarriersonbothitsinputsandoutputs.

ENABLING CLAUSE EnablesWTOmemberstoaccord“specialanddifferentialtreatment”todevelopingcountries,withoutaccordingsuchtreatmenttoothercontractingparties.

EXPORT SUBSIDY Theincentivespaidbythegovernmenttoanexporterbasedonthequantityofcommodityexported.

Fill rate Theproportionofimportsduringagivenyear,relativetothecommitmentintermsofimportquantityasdefinedbythetariff rate quota (TRQ).

FOB The“freeonboard”priceofaproduct,thatis,afterloadingontoashipbutbeforeshipping,thusnotincludingtransportation,insuranceandothercostsneededtogettheproductfromonecountrytoanother.

FREE TRADE AREA AGREEMENT

Anagreementbetweentwoormorecountriestoeliminatetariffandnon-tariffbarriersontradeamongthemselves,whileeachparticipatingcountryappliesitsownindependentscheduleoftariffstoimportsfromcountriesthatarenotmembersoftheagreement.

GENERALIZED SYSTEM OF PREFERENCES (GSP)

TheGSPisasystemthroughwhichindustrializedhigh-incomecountriesgrantpreferentialaccess(mostlylowertariffs)totheirmarketstodevelopingcountries.

HARMONIZED SYSTEM (HS)

AcompleteproductclassificationsystemdevelopedbytheInternationalCustomsOrganizationthatisorganizedinaparticularframeworkandthatemploysanumberingorcodingsystemconsistentwithitsorganizationalarrangement.ForexampleHS2002has97codesforallmerchandiseproductsatthemostaggregatedlevelofproductgrouping.

3�8


IMPORT-SENSITIVE PRODUCTS

Seesensitive products

IMPORT QUOTAS Importquotasarequantitativerestrictionsthatcontroltheamountorvolumeofvariouscommoditiesthatcanbeimportedintoacountryduringaspecifiedperiodoftime.

INTERNATIONAL/WORLD PRICE

Representswhatthecommoditycanearnasanexportorwhatitcoststotheeconomyasanimport.Itisthe(foreign)opportunitycostforacountryforaparticularcommodity.

LINEAR REDUCTION OF TARIFFS

Areductionbyagivenpercentageinalltariffsmaintainedbycountriesparticipatinginaroundoftradenegotiations.

MARGIN OF PREFERENCE

Thedifferencebetweenthedutypayableunderagivensystemoftariffpreferencesandthedutythatwouldbeassessedintheabsenceofpreferences.

MARKET ACCESS Theconditionsthatgoverntheentryofforeigngoodsintoadomesticmarket.Theextenttowhichtheforeignmarketisaccessiblegenerallydependsontheexistenceandextentoftradebarriers,includingtariffandnon-tariffbarriers.

MINIMUM ACCESS TheWTOMarrakechAgreementspecifiedthat,fordevelopedcountries,startingin2001,accesstodomesticmarketshadtobeopentoimportsforupto5percentofthedomesticconsumptionovertheperiod1986–1988.Forcountriesthatstillmaintainedhightariffs,thiswasachievedbytheopeningoftariffratequotas.Thesequotasarecalled“minimumaccessquotas”.

MOST-FAVOURED NATION (MFN) TREATMENT

Thepolicyofnon-discriminationthatappliestoallWTOmembers,providingallWTOtradingpartnerswiththebestcustomsandtarifftreatmentgiventoanyotherpartner.

MULTILATERAL AGREEMENT

Aninternationalcompactinvolvingthreeormoreparties.


3�9

MULTILATERAL TRADE NEGOTIATIONS (MTN)

NegotiationsheldundertheauspicesoftheGATTfrom1947to1994andthereafteroftheWTO,aimedatmutuallybeneficialagreementsforreducingbarrierstoworldtrade.

NOMINAL TARIFF RATE

Therateofdutychargedonthegrossvalueofagivenproduct,ratherthanonthevalueofitscomponents(i.e.inputsandoutputs).Contrastswitheffective tariff rate.

NON-TARIFF BARRIERS (NTBs)

Measuresotherthantariffsthatrestrictimportsorthathavethepotentialforrestrictinginternationaltrade.Theseincludequotas,licensingandvoluntaryexportrestraints.

NOTIFICATIONS GATTrulesspecifythat,undertheobligationsoftransparency,membercountriesmustnotifyastothewaytheyfilltheirobligationsandimplementtheircommitmentsunderthemarketaccessprovisions.AsetofdocumentsissubmittedtotheWTOonaregularbasis.ItincludesmodificationsintheSchedules,thewaytariffratequotasarefilledandadministered,etc.

PREFERENCES Specialadvantagesextendedbyimportingcountriestoexportsfromparticulartradingpartners,usuallybyadmittingtheirgoodsattariffratesbelowthoseimposedonimportsfromothersupplyingcountries.

PROGRESSIVE TARIFF Seetariff escalation

PROTECTION Governmentmeasuresincludingtariffandnon-tariffbarriersthatraisethecostofimportedgoodsorotherwiserestricttheirentryintoamarketandthusstrengthenthecompetitivepositionofdomesticgoods.

PROTECTIONISM Thepolicyofrestrictingimportsthroughmeasuressuchastariffs,quotas,etc.inordertoprotectthedomesticproducersoftheproduct.

QUOTA FILL RATE Describestheproportionofimports,duringagivenyear,underacommittedandnotifiedquotaamount.

330


RECIPROCITY Thepracticebywhichgovernmentsextendsimilarconcessionstoeachother,aswhenonegovernmentlowersitstariffsorotherbarriersinexchangeforequivalentconcessionsfromatradingpartneronbarriersaffectingitsexports(a“balanceofconcessions”).

RETALIATION Thesuspensionofconcessionsorotherobligationsunderatradeagreement,ortheimpositionofotherbarrierstotrade,byagovernmentinresponsetotheviolationofatradeagreementortheimpositionofotherunfairtradebarriersbyanothergovernment.

RULES OF ORIGIN SeeAgreementonRulesofOrigin.

Safeguards TheMarrakechAgreementonAgricultureallowsforspecialtemporarysafeguardmechanismsforproductssubjecttotariffication.Theyareimposedifincreaseinvolumeordropinimportpricesexceedcertaintriggerlevels.

SANITARY AND PHYTOSANITARY MEASURES (SPS)

Measuresappliedtoensurefoodsafetyandprotectionofhumanoranimalhealth.

SCHEDULES TheofficialtariffcommitmentsforWTOmembersarespecifiedintheSchedules,whicharelegallybindingdocumentsdefiningthebound tariffs(MFN)foralistofcommodities.

SENSITIVE PRODUCTS

Intradenegotiationsandagreements,countriesoftenidentifylistsofparticularsensitiveproductsthattheyregardasespeciallyvulnerabletoimportcompetitionandthattheywishtoexemptfromtradeliberalization.

SPECIAL AND DIFFERENTIAL TREATMENT (SDT)

Theprinciplethatdevelopingcountriesshouldbegivenfavourabletreatmentsuchaspreferentialaccesstomarketsofdevelopedcountriesandthatdevelopingcountriesparticipatingintradenegotiationsneednotfullyreciprocateconcessionstheyreceive.


33�

SPECIFIC TARIFF Acustomsdutyassessedasastatedmonetaryamountperunitofphysicalquantity,suchasUS$1000oneachimportedvehicleorUS$50oneachmetricton(tonne)ofwheat.

STANDARD INTERNATIONAL TRADE CLASSIFICATION (SITC)

Aclassificationofgoodstoenablecomparisonbetweencountriesandforreportingtrade.ItwasestablishedbytheInternationalCustomsOrganizationandissimilartotheHSnomenclature(seeharmonizedsystem).

TARIFF Customsdutiesonmerchandiseimports.Tariffscanbeleviedeitheronanad valorembasis(percentageofvalue)oronaspecificbasis(e.g.US$10per100kg),oronbothformssimultaneouslyforthesametariffline.

TARIFF CUT DILUTION

TheMarrakechAgreementspecifiedthattariffshadtobereducedbyagivenaverage(36percentfordevelopedcountries)overtheimplementationperiod.Theterm“tariffcutdilution”referstothefactthatmanycountrieshavereachedthisobjectivebyhigherpercentagecutsonlesspoliticallysensitivetariffs,andminorcuts(often15percent)onmoresensitiveproducts.

TARIFF ESCALATION Tariffsincreasingwiththedegreeofprocessing.Itoccurswhentariffsonprocessedformsofacommodityarehigherthanthetariffsontheprimaryformofthecommodity.

TARIFF PEAKS Veryhigh(oftenprohibitive)tarifflines,significantlyhigherthantheaverage.

TARIFF-RATE QUOTA or TARIFF QUOTA (TRQ)

Acombinationofanimporttariffandanimportquotainwhichimportsbelowaspecifiedquantityenteratalow(orzero)tariffandimportsabovethatquantityenteratahighertariff.

33�


TARIFFICATION Tarifficationistheconversionofnon-tariffbarriersthatexistedforagriculturalproductsbeforetheUruguayRoundintotariffsmeanttobringanequivalentlevelofprotection.Duringthetarifficationprocess,developedcountriesusedcurrentboundrateforproductsthatwerepreviouslybound.Inthecaseofdevelopingcountries,ifthetariffwaspreviouslyunbound,thecountrycouldofferaceilingbinding.

TECHNICAL BARRIERS TO TRADE (TBT)

Technicalregulationsorstandardssuchastestingrequirements,labellingrequirements,packagingrequirements,marketingstandards,certificationrequirements,originmarkingrequirements,healthandsafetyregulations,andsanitaryandphytosanitaryregulationsthatrestricttradeflows.

Terms of trade Usuallyreferstotherelationshipbetweentheaveragepriceofacountry’sexportsandtheaveragepriceofitsimports.Itindicatestherelativeprofitabilityofexportsvis-à-visexports.

WATER IN THE TARIFF

Whenusedingenericterms,referstoasituationwhenacutinthetariffwillnotleadtoaneffectiveincreaseofmarketaccessandcoversthecasesofbindingoverhang,largepreferentialmarginsandprohibitivetariffs.Inmorerestrictivetermstheexpressionimpliesadifferenceintariffratelevelsbetweenappliedandboundtariffs.(Forexample,ifappliedtariffonaproductis20percentandtheboundrateonthesameproductis100percent,thewaterinthetariffis80percent.)


333

Appendix12.2

Tradedatabases

FAOSTAT TradeStat

FAOSTATisanagriculturalinformationdatabasemaintainedbytheFoodand Agriculture Organization of the United Nations (FAO). It providesaccesstoover3milliontime-seriesandcrosssectionaldatarelatingtofoodandagriculture.FAOSTATcontainsdatafor200countriesandmorethan200primaryproductsandinputitems.ThecorethematicareasaroundwhichtheFAOSTATdatabaseisorganizedareagriculturalproduction,consumption,trade,pricesandresources.

FAOSTAT has several different modules, including TradeStat, PriceStat,ProdStat, ResourceStat, AquaStat, FIGIS, Food Security Stat andCountryStat.

ThenewFAOSTATTradeSTATmodulecontainsagricultural tradedatafrom 1986 to 2005. The agricultural trade data are detailed official data,providedelectronically(CD-ROMs,etc.)byover100countries/territorieson an annual basis. The national commodity classification (usually theHarmonizedSystem) isconvertedtotheFAOcommodityclassificationtocoverover600 foodandagriculturecommodities.All tradedatadisplayedis converted (standardized) from detailed trade (including transformedcommodities) intoprimaryequivalents.TradeStatcanbeaccessedfromthemain FAOSTAT Web page or directly at http://faostat.fao.org/site/534/default.aspx.

ThemainusesoftheFAOSTATTradeStatdatabasearetodownloaddataseries on primary equivalent food and agriculture exports and imports intermsofquantity,unitprice,value,agriculturaltradesharesandnettradeforcomparativeanalysisacrosscountriesandtime,andimportantlyforusewithotheranalyticalapproaches/tools.

WORLD INTEGRATED TRADE SOLUTION (WITS)

WITSisawidelyused“hub”thatcontainsanumberofimportanttradeandtradepolicydatabasesandanalyticaltools.Itsnotablefeaturesincludewidegeographic(numberofcountries),period(timesseriesdata)andproduct(allmerchandiseproducts)coverage.WITS isa free software.However, accessto its databases can be fee-charging depending on user status. For moreinformation,seehttp://wits.worldbank.org/witsweb/Faq/default.aspx.

334


ThefollowingarethedatabasesandanalyticaltoolsavailableinWITS:

COMTRADE:Globalmerchandisetradeflowsincludingagriculturalflowsare contained in this database. It has information on exports, imports andre-exports (quantity and value) for more than 140 countries. Useful forseekingdataon,forexample,mainexportedcommodities,mainsuppliersofthese commodities and main importers of these commodities. Informationcanbeobtainedforanumberofinternationallyrecognizedtradeandtariffclassifications such as SITC, ISIC, MTN and HS. The HS (HarmonizedSystem) classification is the most common and data for products can beobtainedatthe2-,4-and6-digitlevel(2digitsisthemostaggregatedlevelofaproductgroupwhile6digitsisahighlydisaggregatedlevel).Thetimeseriesdataavailabilitygoesbackasfaras1962forsomecountries.

TRAINS: Trade Analysis Information System (TRAINS) containsinformationonimports,appliedtariffs,para-tariffsandnon-tariffmeasuresfor 119 countries. The data on applied tariffs, para-tariffs and non-tariffmeasuresareavailableatthemostdetailedcommoditylevelofthenationaltariffs (i.e. at the tariff line level). The data are recorded according tothree internationally recognized trade and tariff classifications. Optionalinformationincludesad valoremequivalentsofspecific,mixedandcompounddutiesandpreferentialduties.

IDB and CTS: The Integrated Data Base (IDB) contains imports bycommodity and partner country and MFN applied tariffs for over 80countries at the most detailed commodity level of the national tariffs andthe Consolidated Tariff Schedule (CTS) data base contains chiefly WTOboundtariffs.TheCTSistheofficialsourceforboundtariffs,whicharetheconcessionsmadebycountriesduringanegotiation(e.g.theUruguayRoundofmultilateral tradenegotiations).Thedataare recordedaccording to twointernationallyrecognizedtradeandtariffclassifications.

AMAD: The Agricultural Market Access Database (AMAD) containsinformation on tariffs (bound and applied) and tariff rate quotas (TRQs)(scheduledquantities, countryallocations,out-of-quotaand in-quota tariffrates).Italsocontainssupplementaryinformationonimports(volumesandvalue),supplyutilization,worldunitvaluesandexchangerates.Thecoverageof countries in it is lower than the WTO CTS and TRAINS databases.However,itsmostimportantcontributionistheinformationonTRQs.Itisfreelyavailableatwww.amad.org.

SMART:TheSystemofMarketAnalysisandRestrictionsonTrade(SMART)isoneoftheanalyticaltoolsinWITSforsimulationpurposes.SMARTisa


335

simulationmodelcontaining in-builtanalyticalmodules that support tradepolicy analysis such as effects of multilateral tariff cuts, preferential tradeliberalizationandadhoctariffchanges.Theunderlyingtheorybehindthisanalyticaltoolisthestandardpartialequilibriumframeworkthatconsidersdynamic effects constant. Like any partial equilibrium model, it has thestrong assumptions allowing the trade policy analysis to be undertaken acountryatatime.WITS/SMARTcanhelpestimatetradecreation,diversion,welfareandrevenueeffects.

MARKET ACCESS MAP (MacMap)

Market Access Map is an interactive database of tariffs and market accessbarriers. It contains the market access conditions applied at the bilaterallevel by over 170 importing countries to the products exported by over200countriesandterritories.MarketAccessMap’sstrengthlies in itswidegeographical coverage; its taking into account of almost all multilateral,regional and bilateral trade agreements; the integration of ad valoremequivalents of specific tariffs; as well as certificates and rules of origin.MarketAccessMapallowsuserstoanalysetheprotectionofanygeographicgroupingandsectoralaggregation.Italsooffersthepossibilityofsimulatingtariffreductionsusingvariousnegotiationformulae.DevelopedbyITCincollaborationwithCEPII,UNCTADandWTO,MarketAccessMapaimstoenhancemarkettransparency,supportinternationaltradepromotionandfacilitate the analysis of related trade policy issues. Market Access Map isavailable online at www.macmap.org. The ITC software is available to thepublicbutonlyagainstacontributionthatisusedtofundtheongoingdataandsoftwaredevelopmentwork.

ECONOMIC RESEARCH SERVICE (USDA)

Ausefulsourceforinformationontradepoliciesviz.,domesticsupportandexport competition is theEconomicResearchService (ERS)of theUnitedStatesDepartmentofAgriculture(USDA).ThepubliclyavailablefreeWebsite(http://www.ers.usda.gov/db/wto/) containsinformationondomesticsupport (expenditure on aggregate measurement of support (AMS), GreenBox,de minimis,etc.)andexportsubsidies (quantityofsubsidizedexportsandexpenditureonexportsubsidies)asnotifiedbyWTOmembers.

CARIBTRADE

CARIBTRADE is a merchandise trade and transportation databasemaintained by the Economic Commission for Latin America and theCaribbean (ECLAC) Subregional Headquarters for the Caribbean. Apart

336


fromansweringqueriesondirectionoftrade,thedatabaseprovidesanalysisoptions through indicators listed on the site that enable the evaluation ofrecenttrendsintradeandintheperformanceofitemstraded.Accesstothedatabase at the Web site (http://celade.eclac.cl/redatam/CARIBTRADE/index.html) has been designed at two levels. The first level of accessaccommodatesthequeriesofawidevarietyofusersandisprovideduptothethirddigitoftheSITSRev.3andHSclassifications.Anotherlevelofaccessis accorded to a limited number of personnel at national level. The chiefstatisticiansofthecontributingcountrieshaveaccesstotheirdataatthemostdisaggregatedlevelofdatasupplied.Researcherswishingtheuseofdataatalowerlevelofdisaggregationthan3digitsmaycontactthechiefstatisticiansofthecountriesforthatlevelofdata.Thedatabasecontainsexternaltradeandtransportationstatisticsfor16Caribbeancountries.Thedataseriesbeginsin1995andextendedto2003(attheendof2005).Thecountrieswhosedataareincludedinthepresentdatabaseare:Anguilla,AntiguaandBarbuda,Aruba,Barbados, Belize, British Virgin Islands, Dominica, Grenada, Guyana,Jamaica,Montserrat, theNetherlandsAntilles,SaintKittsandNevis,SaintLucia,SaintVincentandtheGrenadines,TrinidadandTobago.

FREE TRADE OF THE AMERICAS

HEMISPHERIC TRADE AND TARIFF DATABASE and TariffTheinformationintheHemisphericTradeandTariffDataBaseforMarketAccessconsistsof nationalcustomstariffs basedontheHarmonizedSystem(HS) at the most detailed tariff line level, with corresponding productdescription. For each tariff line, the following information is available, asapplicable:MFNappliedtariffrates;preferentialtariffratesandthecountriesto which they apply; tariff lines for which agricultural tariff rate quotas(TRQs)mayapply;andagriculturalexportsforwhichexportsubsidiesmayapply;importandexportstatisticsbypartnercountry,invalueandvolume,atthemostdetailedlevelofthenationalcustomtariff.ThedatainthedatabasearecompiledbytheInter-AmericanDevelopmentBankfromtheofficialsubmissionsbycountriesparticipatingintheFTAAinitiative.Thedatabaseisupdatedonanannualbasiswithtariffsavailableinthesecondquarterandtradeflowsinthefourthquarterofeachyear.Thedatabasecanbeaccessedathttp://www.ftaa-alca.org/NGROUPS/NGMADB_e.asp.

EUROPEAN COMMISSION HELP DESK FOR DEVELOPING COUNTRIES

The Export Helpdesk is an online service, provided by the EuropeanCommission, to facilitate market access for developing countries to theEuropeanUnion.Itcanbefoundathttp://exporthelp.europa.eu.


337

TheWebsitecoversthefollowingcategories:• The requirements and taxes section enables users to get access to

detailedinformationonEUandmembercountries’importrequirementsaswellasinternaltaxesapplicabletoproducts.

• The import tariffs section enables users to get access to detailedinformationconcerningimporttariffsandothermeasuresthatapplytoaparticularproduct.

• The customs documents section provides information concerningthe documents to be produced by an exporter in order to qualify forpreferentialdutytreatmentunderthedifferenttraderegimesavailablefordevelopingcountries.

• Therules of originsectionprovidesinformationontheconditionsthatneed to be met for goods to qualify for advantageous tariff treatmentundertheGSPandACPsystems.

• The trade statistics section provides detailed trade statistics coveringimports and exports for the 25 EU member countries collectively andindividually,andintra-EUtrade.

338


Appendix12.3

ProductnomenclaturesandWITSutilities

Since 1988, OECD member countries provide data according to theHarmonizedSystem(HSormoredetailedclassification).TheInternational Trade by Commodity databaseofOECDstoresdata inHS.Dataare thenconvertedtodifferentnomenclaturesusingacorrelationtable.

The harmonized system (HS) is an international 6-digit commodityclassification developed under the auspices of the International CustomsCooperation Council. Some countries have extended it to 10 digits forcustomspurposes,andto8digitsforexportpurposes.

Intheharmonizedsystemgoodsareclassifiedbywhattheyare,andnotaccordingtotheirstageoffabrication,useororigin.TheHSnomenclatureis logically structured by economic activity or component material. Forexample,animalsandanimalproductsarefoundinonesection;machineryand mechanical appliances (grouped by function) are found in another.The nomenclature is divided into 21 sections, while additional sections(i.e. Section22 and23) areused for country-specific specialpurposes.Forexample,SouthAfricausesSection22foritemssuchaspostalarticles,shipstores and platinum and Section 23 for Original Equipment Manufacturermotorvehiclecomponents.

Eachofthesesectionsgrouptogethergoodsproducedinthesamesectoroftheeconomy.Eachsectionconsistsofoneormorechapters,withtheentirenomenclaturebeingcomposedof99chapters.Somechaptersarereservedforspecialpurposesorfutureuse.ChaptersofsectionsItoXV(exceptsectionXII)aregroupedbybiologicalsimilarityorbythecomponentmaterialsfromwhicharticlesaremade.Forthosechaptersinwhichgoodsaregroupedbyrawmaterial,averticalstructureisusedinwhicharticlesareoftenclassifiedaccording to their degree of processing. For example, Chapter 44 containsitemssuchasroughwood,woodroughlysquaredandsomewoodenfinishedproductssuchaswoodentableware.Articlesmayalsobeclassifiedaccordingto the use or function. This classification (by function) mainly occurs insectionXIIandsectionsXVI toXXI.Forexample, sectionXVIIcontainschaptersonmotorvehicles(87),aircraft(88)andships(89).

The Standard International Trade Classification (SITC)hasbeendevelopedby the United Nations with the purpose of classifying traded productsnot only on the basis of their materials and physical properties, but also


339

according to their stage of processing and economic functions in order tofacilitateeconomicanalysis.

As SITC has been developed principally for statistical purposes, it hasto maintain a correlation with the tariff nomenclature, given that customsdeclarationsaretheprincipalsourceoftradedata.Forthisreason,SITChasundergonethreerevisions,toalignitselfwiththedevelopmentofthetariffnomenclatures.

ThelatestrevisionofSITC(Rev.3)establishesacorrelationwiththeHS,whilethepreviousrevisionswererelatedtoBTN(SITCRev.1)andCCCN(SITCRev.2),respectively.

SITC Rev. 3 was adopted in 1988 and maintains the basic 10-sectionstructureofthepreviouseditions;thesectionsaresubdividedinto67two-digit divisions, 261 three-digit groups, 1033 four-digit groups, and 3118five-digitheadings.

A useful characteristic of WITS is that it makes it possible to work withdifferentproductclassifications.Productclassifications,or“nomenclatures”,arewaysofaggregatingdataforaspecificpurpose.(Forexample,theGTAPnomenclatureaggregatestradedatauptoalevelcompatiblewiththeinput–output tablesused tobuild themodel.)Thepossibilityof linkingdifferentnomenclatures isusefulparticularly in thosecases inwhichmorethanonedatabase,usingdifferentclassifications,isrequiredtoobtainacompletedataset.

WITS also allows for checking the composition of commodity classeswhen working with different nomenclatures. For instance, when workingwithaGTAPdatabase,whereaggregateddataarebasedontheoriginalHSnomenclature,wecanchecktheWITStool“Nomenclaturesconcordances”(accessiblethrough“HelpandInformation”)inordertofindoutwhichHSgoodsareincludedintheGTAPcategoryof“paddyrice”.ThetoolhelpsfindtheconcordancesbetweentheGTAPcodeandtheHS2002nomenclature.Under“paddyrice”wefindtwoHSlines,paddyrice(100610)andhuskedrice(100620).

Appendix12.3

ProductnomenclaturesandWITSutilities

Since 1988, OECD member countries provide data according to theHarmonizedSystem(HSormoredetailedclassification).TheInternational Trade by Commodity databaseofOECDstoresdata inHS.Dataare thenconvertedtodifferentnomenclaturesusingacorrelationtable.

The harmonized system (HS) is an international 6-digit commodityclassification developed under the auspices of the International CustomsCooperation Council. Some countries have extended it to 10 digits forcustomspurposes,andto8digitsforexportpurposes.

Intheharmonizedsystemgoodsareclassifiedbywhattheyare,andnotaccordingtotheirstageoffabrication,useororigin.TheHSnomenclatureis logically structured by economic activity or component material. Forexample,animalsandanimalproductsarefoundinonesection;machineryand mechanical appliances (grouped by function) are found in another.The nomenclature is divided into 21 sections, while additional sections(i.e. Section22 and23) areused for country-specific specialpurposes. Forexample,SouthAfricausesSection22foritemssuchaspostalarticles,shipstores and platinum and Section 23 for Original Equipment Manufacturermotorvehiclecomponents.

Eachofthesesectionsgrouptogethergoodsproducedinthesamesectoroftheeconomy.Eachsectionconsistsofoneormorechapters,withtheentirenomenclaturebeingcomposedof99chapters.Somechaptersarereservedforspecialpurposesorfutureuse.ChaptersofsectionsItoXV(exceptsectionXII)aregroupedbybiologicalsimilarityorbythecomponentmaterialsfromwhicharticlesaremade.Forthosechaptersinwhichgoodsaregroupedbyrawmaterial,averticalstructureisusedinwhicharticlesareoftenclassifiedaccording to their degree of processing. For example, Chapter 44 containsitemssuchasroughwood,woodroughlysquaredandsomewoodenfinishedproductssuchaswoodentableware.Articlesmayalsobeclassifiedaccordingto the use or function. This classification (by function) mainly occurs insectionXIIandsectionsXVI toXXI.Forexample, sectionXVIIcontainschaptersonmotorvehicles(87),aircraft(88)andships(89).

The Standard International Trade Classification (SITC)hasbeendevelopedby the United Nations with the purpose of classifying traded productsnot only on the basis of their materials and physical properties, but also

34�

Index

A ACPpreferences110,113,115,117,

118,182,224,241ACPtariffproposal98AGLINK309,317,318-319Agriculturalexports14,112,114,269,

336Agriculturaltariffs26,75-79,114Agriculturaltariffprofiles79Agriculturaltradepolicy10,37,308AidforTrade67,119,121,216,218,

219Appliedtariffs160,324,334ATPSM317

B Bananas110,113–118,158,191–219,

285,300,302,303Barbados117,118,159,160,177–180,

184–186,285,336Belize112,114,116,117,118,133–139,

177,178,179,185,196,256,259,260,285,298,336

Borderpolicies23–25Boundtariffs82,160,161,163,171,

173,324,330,332,334Businesscompetitivenessindex250

C CARICOM7,263,285,289,297,301,

302,303CaribbeanRegionalNegotiating

Machinery30Cerealsupply20,47,297Commoditydependence47,49CommonAgriculturalPolicy(CAP)

109,177,181,325

Commonexternaltariff26-29,81,188,325

CommonMarketOrganisationforBananas208,210

CommonwealthSecretariat72Competitiveadvantage251,263,268,

281,284,290Comparativeadvantage246,250–253,

296,303,305,306,307Competitiveness53,120,179,186,192,

214,245–248,263,290,296,297,305,306,307

COSIMO309,317–319CSME25

D De minimis63,326,335Differentiation249,264,265,309,320Diversification108,120,179,272DohaRound(DohaDevelopment

Agenda)108,115,118,157,161,166,167,173

Dollarbananas209,210DomesticSupport62,163,178,321,

335Domesticresourcecost255,307Dominica114,196,197,198,199,202,

203,205,206,209,214,215,216,256,336

DominicanRepublic112,116,

E EBA114,116,120,178,181–185,187,

240EconomicPartnershipAgreements58,

119,120,121,188,264Enablingclause107,327

34�


EUACPriceagreement237–238EUricetradepolicy231Exportcompetition63,335Exportshares14,304Externalshocks157,158,173,295

F Factormarkets31,308Foodaid64FoodImports21,22,23,297–299Foodsecurity9,10,12,129,158,163,

191,192,194,196,203–206,208,218,263,289–291,296–300,308

Foodsecurityindicators128,297–304Flexibility60

G G-2036,G3336,161,162,167,171,172,173G-9036GATS193,210GeneralizedSystemofPreferences107,

119,120,181,188,327Generalequilibriummodels111,

312–317GlobalCompetitivenessIndex248GreenBox335Grenada196,197,208,216,260,285,

336GSP240,327,337GTAP309,319–321,339,Guyana227–230,259,262,285,299,

336

H Harmonization267,270,283,286,287HongKongministerial213

I (economic)importanceofpreferences

116,119Investmentenvironment34

Investment110,119,121,179,188,193,216,245–248,264,278,295,312,313,320

importsurges157–161,164,170,172,173

instability109,157,172

J Jagdeoinitiative25,38Jamaica112,113,115,118,158,177,

178,179,180,186,188,196,256,260,262,285,336

L Livelihoodsecurity129,193,218,LeastDevelopedCountries107,178,

181,183,184,187,241,283Lomé774

M Marketaccess116,191,192,194,207,

208,210–213,269,272,281,285,326,328,329,332,334,335,336

MDG1

N NegotiatingGroups36–37

O OCTexports239Overhang324,332OrganisationofEasternCaribbean

States(OECS)191,197

P Partialequilibriummodels308–312Preferences66,107–121,182,186,200,

218,289Preferenceerosion108,114,115,119,

182,203,327,328,329Preferencevalue110,116,117,120Preferential(trade)agreements107,109

Index

343

Policyanalysismatrix253–255,295,296,297,305–307

Productmarkets31,308Productdifferentiation188,265,320

R Realeffectiveexchangerate254Referenceprices163,164,173RegionalSpecialProducts139–148RegionalTransformationProgramme24Revealedcomparativeadvantage252,

303Ricetrade223–226Ruraldevelopment129,130,191,196,

218,219,273,281,290,309

S SaintKittsandNevis112,114,117,

118,178,180,256.336SaintLucia114,115,118,196,197,198,

199,200,201,202,203,204,205,206,214,215,216,260,336

SaintVincent336Smallandvulnerableeconomies41–45Specialanddifferentialtreatment42,

60,64,108,119,327,330Specialproducts61,123,173,216,218Specialproductsidentification126Statetradingenterprises64Sugarpreferences177–189Specialsafeguard62,157,158,160,161,

162,171,173,218Sugar110,113–116,119,120,158,

164–167,169,177–188,191,201,299,300,302,303

SugarProtocol115,180,182,184Suriname112,180,196,226,259,260,

262,299

T Tariffconcentration82Tariffcuttingproposals84–86,88,96

Tariffequivalent295,318Tariffformulae96Tariffpeaks331Tariffpolicy75Tariffprofiles79Tariffrevenue50,52TariffRateQuota(TRQ)181,212,327Tradeagreements107,301,318,324,

335Tradecosts116,183,184,Tradedependence15,45Tradeliberalization107,118,119,120,

263,286,296,330,335Tradeopenness15,45Tariffoverhang89,92Tradepolicy10,119,191,192,195,207,

208,214,263,269,282,290–295,308-317,333,335

Tradepolicyanalysis3,6,33,124,291–321

Tradereformobjectives87,90Traderemedy157,159,161,162,169,

172,173Transportcosts53,217TrinidadandTobago112,117,118,177,

178,179,257,260,262,263,299,336

U Undernourishment18,19

V Valueofpreferences110,116,117,120Vulnerability168,297–299

W WindwardIslands191,196,197,214,

217WITS333-335,338-339WTO,smalleconomies57WTO,UruguayRoundExperience125

Agricultural trade policy and food securityin the Caribbean

Structural issues, multilateral negotiations and competitiveness

COOPERAZIONEI T A L I A N AI N E A

Promoting CARICOM/CARIFORUM Food Security

FAO Trust Fund for Food Security and Food Safety – Italian Contribution

This book is largely a result of work implemented under the trade

policy component of the project “Promoting CARICOM/CARIFORUM

Food Security”. The financial contribution of the Italian Directorate

for International Cooperation to the FAO Trust Fund for Food Secu-

rity and Food Safety funded the project. Italy’s National Institute of

Agricultural Economics (INEA) was actively involved in capacity-

building activities under the trade policy component, particularly in

organizing and carrying out training in the areas of trade policy

analysis and negotiations and quality and safety requirements in

international trade and marketing. The book examines various

dimensions of trade policy and related issues of relevance to the

countries in the CARICOM/CARIFORUM region and presents policy

instruments to address trade and food security and rural develop-

ment linkages. It will serve as a useful guide and reference docu-

ment for agricultural trade policy analysts, trade negotiators,

policy-makers and planners in both the public and private sectors.


Structural issues, multilateral negotiations and com

petitivenessFA

O

Agricultural trade policy and food securityin the Caribbean

TC/M/A1146E/1/06.07/1500

ISBN 978-92-5-105747-6

9 7 8 9 2 5 1 0 5 7 4 7 6

accessing market opportunities: quality and safety standards

Documents