Issues for Negotiation under the Doha Work Programme on RTAs

Spacial articlas

Issues for Negotiation under the Doha Work Programme on RTAs India’s Negotiating Stand

T

RTAs in force as of February 2005 By year of entry into force (left-hand scale) and cumulative (right-hand scale)

Economic and Political WeeklyMay 26, 20071968Chart 1 shows the unabated growth of RTAs notified to theGATT/WTO.Although RTAs may take the form of FTAs, CUs, oragreementsleading to the formation of one or the other, the most commoncategory is the free trade agreement, which accounts for 72 percent of all RTAs. The trend towards the conclusion of free tradeagreements, which require a lesser degree of integration and arefaster to conclude, has intensified in recent years. In a FTA, eachparty maintains its own tariffs vis-à-vis third parties. CUs, whichprovide for the establishment of a common external tariff andharmonisation of trade policy, often take years to negotiate andhave long implementation phases. CU agreements account for19 per cent of all RTAs. 18 of the RTAs identified (17 FTAsand 1 CU) contain commitments on trade in services in additionto tariff concessions on goods. The new generation of RTAs,especially those involving developed countries, tend to go farbeyond traditional tariff-cutting exercises and even beyond therealm of existing multilateral rules. First, while old regionalismwas essentially confined to RTAs between industrial economiesor developing economies, the new regionalism is known for crossalliances between developing and industrial economies. Second,while old regionalism was essentially limited to RTA formationsby contiguous economies, the new regionalism does not seemto be limited to neighbouring economies. In recent times RTAshave been intercontinental. Although there may be coordinationproblems, from the gains-from-trade perspective this could bea healthy development. Third, under the new arrangement RTAsare not exclusive, meaning thereby one country can simulta-neously be a memberof more than one RTA. This may eventuallyturn out to be an aidin promoting multilateralism through RTAs.Fourth, while old regionalismwas limited to shallow integration,the new regionalism is more ambitious. A number of recentagreements aspire for deep integration, with commitments toharmonisation of regulatory measures, freeing factor movements,and other close integrating measures. Such trade agreementsinclude more and more regional rules on investment, competitionand standards, in a few cases;they also contain provisions onenvironment and labour. Many more agreements today containdisciplines limiting the use of quantitative restrictions and sub-sidies among countries forming an RTA.The configuration of RTAs is diverse and becoming increas-ingly more complex with overlapping RTAs. The simplestconfiguration is a bilateral agreement formed between two parties.These account for more than half of all RTAs in force and foralmost 60 per cent of those under negotiation.2 The mostnoteworthy development expected in the next five-years is theemergence of a new category of agreement, namely RTAs whereeach party is a distinct RTA itself.3 Several agreements of thiskind are currently under negotiation.The greatest concentration of RTAs is primarily centered aroundwestern Europe, constituting about 50 per cent of the RTAscurrently in force. Further, the enlargement of the EuropeanUnion (EU) in 2005 and 20074 has increased its strength from15 to 27 with the addition of 12 new countries. The EU is alsonegotiating and concluding second-generation bilateral FTAsbased on a reciprocal exchange of preferences with partners inthe Mediterranean and north Africa, with the objective of a Euro-Med free trade area by 2010. The EU is also looking to forgecloser ties with emerging markets in Latin America. Its FTA withMexico entered into force in mid-2000 and negotiations on FTAswith the Southern Common Market (Mercosur-Argentina, Brazil,Paraguay and Uruguay) and Chile have been taking place.Countries of north Africa are participating in various otherregional trade initiatives both in Africa and the middle east, suchas the recent effort launched by the Arab League to establishan Arab free trade area by 20075 and a broader economic union(including a single market and currency by 2010). For their part,Gulf Cooperation Council countries established a common externaltariff in 2003, and are engaged in discussions with the EuropeanCommission (EC) on the negotiation of a possible RTA.In sub-Saharan Africa, the regional integration process is gainingdepth, although progress is uneven and far from certain due tothe complex web of overlapping RTAs membership. In westernAfrica, there are two advanced levels of RTAs; one is the WestAfrican Economic and Monetary Union6 and the other is CentralAfrican Economic and Monetary Community.7 In eastern andsouthern Africa, members of the Common Market for Easternand Southern Africa8 have achieved their objective of creatinga FTA grouping of 20 countries, and similarly, the parties to theSouth African Development Community (SADC) established aFTA by 2004, despite difficulties mainly due to overlappingmembership with other RTAs, in particular the long-standingSouth African Customs Union (SACU) agreement.9Asia Pacific is a region currently undergoing significant changeswith respect to its stance towards regional integration, with anumber of countries shifting their long-standing policy of MFNtrade liberalisation to actively consider the regional option. Also,the “open regionalism” typically associated with the Asia PacificEconomic Cooperation (APEC) appears to be counteracted bya drive towards preferential trade initiatives. Japan, Singaporeand the republic of Korea have been negotiating and conductingfeasibility studies for the establishment of several RTAs bothamong countries in the region and cross-regionally. Similarly,New Zealand and Australia are exploring the possibility of severalRTAs with regional partners and with countries of the westernhemisphere. Singapore and New Zealand already concluded afar-reaching RTA and discussions are being held for a closereconomic partnership between New Zealand and Hong Kong,China. Japan and Singapore formed their own bilateral agreementin 2002. A notable development in the region has also been theagreement between the members of ASEAN10 and China estab-lishing a FTA in 2005. Japan and the republic of Korea havealso initiated similar negotiations with ASEAN. These develop-ments constitute an enormous change in the east Asia. Prior tothem, Japan and Korea, along with Hong Kong and one othercountry, were the only four members of the more than 140members of the WTO that had not participated in a reciprocalregional trading agreement. See Table 2 for details of RTAs inthe region.IIIEffects of New Regionalism on World Trade If looked at from a global perspective, the regional develop-ments mentioned above clearly point to the emergence of a trendtowards cross-regional RTAs. Most of the major players at theregional level are increasingly looking beyond their regionalborders for partners in selective (often bilateral) preferential tradeagreements.This new regionalism is having significant effects on the worldtrading system. Some of these effects are positive. For example,one motive for a country, which is a member of a multi-countryRTA engaging in bilateralism, might be to force the other memberof the bilateral RTA to make more progress in trade liberalisation

Economic and Political WeeklyMay 26, 20071969and thereby promote deeper integration in the RTA. Rajan et al(2002, chapter 2) give this as one reason behind Singapore’spursuit of bilateralism. They refer to the “convoy problem”whereby the pace of integration is held back by the “least willingmember”. Another benefit is that RTAs can set precedents anddevelop negotiation modalities that can be adopted later inmultilateral negotiations. The Canada-US FTA in particular,developed concepts and modalities in the service trade areas thatwere important in the development of the GATS [Lyold 2002].The negative effects include multiple system of rules, unequalaccess to the world market, undermining the MFN principle, huband spokes strategy etc. RTAs certainly create multiple systemsof rules. This multiplicity may pose a problem for the govern-ments and the traders of a country that is a member of more thanone RTA. In areas such as rules of origin, and sanitary andphytosanitary standards (SPS), export traders may face differentrules depending on the destination of their exports.The precedents set by RTAs may be bad as well. There hasbeen concern over some of the RTA precedents. One exampleis the exclusion of some agricultural products from tradeliberalisation under the Japan-Singapore EPA; specifically, theagreement excludes cut flowers and ornamental fish, Singapore’sprincipal exports of agricultural products to Japan, from the listof products imported under the terms of the agreement into Japan.It has been reported that Japan is pushing for a similar exclusionin negotiations with Mexico, Korea and Australia. Another exampleis the US predilection for side agreements on environment andlabour standards. Having succeeded in embedding these in theNAFTA, the US is now pursuing such agreements in otherbilaterals and in the free trade area of the Americas (FTAA).Since the development of new regionalism, many countries arenow hubs. In the Asia-Pacific area, Singapore, the US, Canada,Mexico, Chile, Peru, Australia, New Zealand and Russia are nowhubs on the basis of RTAs already in force and others such asJapan and Thailand may join them soon. One can crudely measurethis effect by considering the number of spokes for each hub,that is, the number of countries with which one hub country hasseparate bilateral free trade agreements (excluding plurilateralRTAs of which it is a member as these have connections acrossspokes). One might describe the EU as a super-hub because ofthe large number of spokes. Hubs create multilayered preferenceschemes. One consequence is that the spokes have less marketaccess than the hub as the hub enjoys preferential access to allspokes but a spoke has preferential access only to the hub and,for reverse trade, a hub gets unrestricted imports from the spokesbut each spoke gets unrestricted access only from its hub partner.New regionalism has created unequal access to world markets.Most of the bilateral trade agreements are between developedcountries or in a few cases between a developed and a developingcountry; examples of the latter are the agreements Mexico haswith the EC and European free trade agreement (EFTA)11 countries.When the larger size of the markets in developed countries andespecially the US and EU is taken into account, there is no doubtthat the increase in market access resulting from bilaterals hasgone overwhelmingly to developed countries and not todevelopingcountries. The one significant exception among the developingcountries appears to be Mexico, which has secured mostly freeaccess to its major markets in both north America and Europe.In the APECarea, the countries that have gained improved marketaccess from the bilaterals are all the higher income countries ofthe region, again, with the exception of Mexico.IVIndia’s Stand on RTA IssuesIndia is a marginal player in the global trade scenario. Its sharein global trade is below 1 per cent. India is also not a part ofany RTA that has substantial influence on world trade. As a partof the integration process with the world, the signing of theBangkok agreement (India, Bangladesh, Laos, Philippines, SouthKorea and Thailand) in 1975 was the first initiative. The agree-ment failed to go a long way in achieving its objective of tradeexpansion. Recent developments like proposals of accession ofChina to the Bangkok agreement have given rise to new expec-tations. India’s second initiative on this front is the SAARC freetrade agreement (SAFTA) with Bangladesh, Nepal, Bhutan,Maldives and Pakistan which came into full effect in 2006. Dueto political tension between India and Pakistan and also forreasons like Bangladesh, Maldives and Nepal have a very limitedexport basket to offer to comparatively larger economies likeIndia, Sri Lanka and Pakistan, India has not achieved much fromthis regional arrangement. India is also a part of the Bay of BengalInitiative for Multi Sectoral Trade and Economic Cooperation(BIMSTEC) with Bangladesh, Myanmar, Sri Lanka and Thai-land. India also entered into a bilateral free trade agreement withSri Lanka in 2000 and more recently with Thailand in 2004 andhas negotiated a comprehensive economic cooperation agree-ment (CECA) with Singapore and ASEAN in 2005.12 TotalSAFTA13 and BIMSTEC trade constitutes about 1.5 per centand 2-3 per cent of total world trade approximately. This impliesthat about 96 per cent of India’s trade is outside the preferentialzone. More than half of this 96 per cent is with countries thatare part of one or more RTAs. For instance, NAFTA and theEU constitute 50 per cent of India’s exports, 10 per cent goesto ASEAN and another 10 per cent to Japan and south Asia.Table 2: Preferential Trade Agreements in the Asia andPacificRegion, June 2006Already in ForceRegional Trading AgreementsBilateral Trade AgreementsAFTA (ASEAN Free Trade Area),1992,1993ASEAN-China Free Trade Agreement,2004,2005Bangkok Agreement, 1975,1976Pacific Island Countries Trade Agreement(PICTA), 2001SAARC Preferential Trading Agreement(SAPTA), 1993,1995South Asian Free Trade Agreement(SAFTA), 2002, 2006Trans-Pacific Strategic EconomicPartnership Agreement (TPSEPA),2005 2/*Note:*(2/) Not yet in force, in both RTAs and bilateral trade agreements, thefirst year refers to the year of signing of the agreement and the secondyear is the year it came into force.Source:Tumbarello (2006).Australia-New Zealand (CER, closerEconomic Relation, 1983,1983)Australia-Singapore, 2003, 2003Australia-Thailand, 2004,2005Australia-United States 2004,2005China-Hong Kong, SAR, 2003, 2004China-Macao SAR, 2003, 2004China-Pakistan, 2005, 2005China-Thailand, 2003, 2003India-Sri Lanka, 1999, 2001India-Thailand, 2003, 2004Japan-Mexico, 2004, 2005Japan-Singapore, 2002, 2002Korea-Chile, 2003, 2004Korea-Singapore, 2002, 2002Lao PDR-United States, 2003, 2005Lao PDR-Thailand, 1991, 2001New-Zealand-Singapore, 2000, 2001New-Zealand-Thailand, 2005, 2005Singapore-European Free TradeAssociation, 2002, 2003Singapore Jordan, 2004, 2005Singapore-United States, 2003, 2004Sri Lanka-Pakistan, 2005, 2005Sri Lanka – Pakistan, 2005, 2005Vietnam-United States, 2000, 2001

Economic and Political WeeklyMay 26, 20071970Therefore on the whole, 70 per cent of India’s trade is withcountries that are part of strong and well established RTAs. So,with India being a part of only SAFTA, Bangkok agreementBIMSTEC and an FTA with Sri Lanka and Thailand in recentyears, the country needs to take a strong view and seek tighterdisciplines in the WTO on RTAs.Within the GATT and WTO, the examination of specific RTAshas been plagued by disagreement about the interpretation ofcertain elements of the rules relating to RTAs as well as by certainprocedural aspects. In practice, the CRTA of the WTO has alsonot been able to resolve many of the systemic issues. The WTOsecretariat has prepared a synoptic paper on procedural andsystemic issues (document TN/RL/W/8/Rev 1), which summarisesthe discussion that has already taken place on RTAs and high-lights the issues on a factual basis.On the goods side, the single most important issue probablyrelates to the interpretation of the term “substantially all thetrade”, which relates to the requirement that “duties and otherrestrictive regulations of commerce…are eliminated on substan-tially all the trade between the constituent territories” as definedin GATT Article XXIV: 8. This is particularly relevant for thoseagreements where the coverage of agriculture is presently limited,for example, many of the RTAs formed by European countries.Afew RTAs have eliminated all duties on agricultural goods butgeneral agricultural trade, even on a preferential basis, remainssubject to exceptions. Average agricultural preferential tariffsremain high and concessions on agricultural products grantedbyRTA partners tend to be parsimonious in nature. The debate on“substantially all the trade” has centered on two possible interpreta-tions, which are not mutually exclusive. The first, a quantitativeapproach, favours the definition of a statistical benchmark, suchas a certain percentage of trade between the parties. The second,a qualitative approach, would require that no sector (or at leastno major sector) be excluded from intra-RTA trade liberalisation.For India too, agriculture is a sensitive sector. But whether Indiawould like to grant concessions on agricultural products willdepend much on the partner country’s export basket and India’sexport competitiveness. But perhaps, it is logical for India to gowith the second option, i e, a qualitative approach and wherenecessary use of the positive list approach in granting concessionson agricultural products, as is done in the majority of RTAs.Another issue deals with “the general incidence of duties”,which are not on the whole to be higher or more restrictive againstthird parties upon the formation of a CUs (Article XXIV: 5).This issue has been largely clarified with the 1994 understanding,which specifies that the evaluation of the general incidence ofduties and other regulations of commerce applicable before andafter the formation of the CU shall be based on an overallassessment of weighted average tariff rates, for which the appliedrates shall be used. If it were desired to ensure that even statictrade diversion were avoided, this could also be achieved byrequiring that the MFN rates also be reduced to a level whichwould prevent or minimise trade diversion. In relation to “otherregulations of commerce (ORCs)” (Articles XXIV: 5) and “otherrestrictive regulations of commerce (ORRCs)” (Article XXIV:8),the systemic debate also runs up against the issue of the definitionand measurement of non-tariff barriers. The only exceptionsconcern quantitative restrictions that satisfy GATT provisions(eg, agriculture, balance of payments, and health or safety con-siderations). “Regulations of commerce” is an expression whichhas been used in the GATT legal texts only in connection withRTAs. No definition of the term is provided. Some members haveargued that what is important is not whether some specificmeasures fall under the umbrella of ORRCs but rather if theirapplication among RTA parties leads to a restriction on the tradeof third parties. The question that concerns India is that whethersafeguards and anti-dumping measures are considered as ORCCs.Moreover, should consideration of ORCs and ORCCs be dif-ferent in fully implemented RTAs and interim agreements? SAFTAhas encouraged tariff concessions, but significant non-tariff barriersremain in place. Moreover, anti-dumping investigations continueto be a major barrier to trade in the south Asian sub-region. Also,with respect to other trade measures in agriculture which alsoconstitute barriers, it needs to be clarified, for example, whatmethodology should be used to aggregate commitments ondomestic supports and export subsidies.There are no explicit WTO disciplines on the use of preferentialrules of origin. The rules of origin can multiply distortions asoverlapping FTAs have begun to form. Origin rules should bejustified to prevent products from non-parties to an RTA gainingpreferential access to the market through the party, which maintainsthe lowest external import restriction (ie, to avoid “tradedeflection”).There are different interpretations so as to whether RTA originrules constitute “other regulations of commerce” or not, whichhavebeen debated upon. However, it is believed that RTA rules oforigin definitely constitute an ORC. Most countries in the worldagree that a case-by-case examination of the preferential rulesoforigin in RTAs is needed. That would clearly indicate whethertheserules had restrictive effects on the trade vis-à-vis third parties.The meaning of certain aspects of article V of the GATS hasalsobeen raised. The basic provision is that an “economic integrationagreement”, the term used in the GATS for an RTA coveringtradein services, should have “substantial sectoral coverage”, under-stood in terms of the number of defined sectors used in GATSschedules of commitments, volume of trade affected, and modesof supply. This coverage is to be achieved through the eliminationof existing discriminatory measures and the prohibition of newormore discriminatory measures. For the purposes of evaluation,account may be taken of the contribution of such an RTA to awider process of economic integration or trade liberalisationamong the members. Some flexibility is allowed for such agree-ments involving developing countries. Many countries, includingIndia, agree that the unavailability of detailedtrade data in servicesmakes it difficult to arrive at a percentage-type test for quan-titatively measuring “substantially all discrimination”.One important issue deals with RTAs established under theenabling clause, ie, RTAs in the area of trade in goods betweendeveloping countries. As Laird (1999) says, that an RTA formedunder the enabling clause need not substantially cover all thetrade; does not require duty elimination; has no fixed timetableforimplementation; and is not subject to periodic reporting require-ments. The only main obligations of parties to such an RTA areto notify the agreement or its modification to the WTO Committeeon Trade and Development, to furnish information deemedappropriate, and afford the opportunity for prompt consultationswith respect to any difficulty or matter that may arise. India isunlikely to oppose this point and would not like to touch theenabling clause while flagging other issues in the negotiating group.Another point that would be of concern to many developingeconomies including India is the unclear issues related to tran-sition periods. Could RTA parties apply for a longer (more than10 years) transition period for some products? When it is saidthat any interim agreement shall include a plan for the formationof a customs union or an FTA, it is not clarified so as to

Economic and Political WeeklyMay 26, 20071971whatshould constitute a “reasonable length of time”. Whenshould interim agreements fulfil the requirements spelled out inparagraphs 5 and 8 of Article XXIV? As noted by Laird (1999),developed countries tend to have wider trade coverage andgenerally apply their commitments over a stricter time frame thantheir partners. There is no explicit provision for such asymmetri-cal application of WTO rules, although this would seem con-sistent with the principle of special and differential treatment fordeveloping countries.Developing countries have also voiced their support for othercountries on the issue of elaborating disciplines aimed at ensuringthat third parties are compensated for the possible negative effectsbrought by the creation or enlargement of an RTA. For instance,there was large-scale trade diversion for India in the textiles sectorbecause of Mexico in NAFTA14 and therefore, it would beobvious for India to seek compensation from the NAFTA members.Finally, regarding the notification requirements (paragraph 7),it had been observed by India that clarification with respect tocontents of notification is required. Since notification require-ments were very ambiguous, it does not obligate members toprovide substantial information. Keeping this in view, the GeneralCouncil, on December 14, 2006, established on a provisionalbasis a new transparency mechanism for all regional tradingagreements. The new transparency mechanism – negotiated inthe negotiating group on rules-provides for the early announce-ment of any RTA and notification to the WTO. Members willconsider the notified RTAs on the basis of factual presentationby the WTO secretariat.15 The transparency mechanism isimplemented on a provisional basis. Members are to review, andif necessary modify, the decision, and replace it by a permanentmechanism adopted as paper of the overall results of the Doharound. The transparency mechanism with emphasis on sufficientinformation to be provided by the RTA members will facilitatea build-up of a strong database to help all members.Since India is not a member of any RTA that has a stronginfluence on world trade, she will stand to lose because of tradediverting effects of any RTAs and the new formations where itis not involved. The Indian textile sector, for instance, has beenbadly affected since the US gives preferential treatment and dutyfree access to textile products from Mexico under NAFTA. Thequestion of what should be India’s general stand on RTAs isdifficult to arrive at. For this, India will have to look at whetherRTAs promote global welfare or not, ie, it has to analyse theextent of trade diversion due to RTAs and its impact on Indianexports. However, all its present agreements with regional partnershave opened markets for Indian goods in the countries concerned.All these agreements constitute unilateral tariff reduction exceptthe India-Sri Lanka FTA. India’s overall trade balance withSAFTA is positive. It is evident that the share of India’s exportsin south Asian countries has increased from 2.73 to 6.2 per centover the period 1990 to 2006.16 Hence, its existing and recentinitiatives in regional/bilateral trade liberalisation may help divertsome trade of the countries concerned from their other tradingpartners in favour of India given their supply constraints, andtherefore may be beneficial for India.Other Problem Issues in RTAsSome of the other problematic issues with respect to RTAsare (i) the use of tariff peaks17 by developed countries,(ii)equivalence notion under the SPS agreement not beingextendedon an MFN basis, (iii) rules of origin problem; and(iv)non-reciprocal tariff concessions. Some of these issues arealso quite important as far as the trade diversion aspect ofRTAsisconcerned.The issues are dealt here briefly one by one.(i) Use of tariff peaks by developed countries: After the imple-mentation of the Uruguay round, and the consequent tarifficationof non-tariff protection in agriculture, dispersion in tariff ratesdid not fall substantially, and even increased in some instances.In the case of agriculture, protection was lowered mostly on theitems already characterised by relatively low barriers, while thetariffication procedures did little to reduce protection on highlyprotected goods such as dairy, meat, sugar and so on. Overall,the phenomenon of tariff peaks seems to have been aggravated.Although the average tariff rates in industrialised countries arelow, they have high peak tariffs for certain products, some ofwhich are of export interest to many developing countries. Incertain Quad markets (EU and Japan, especially) MFN tariffpeaks in some processed agriculture and food categories can behigh enough completely displace exports from developing coun-tries in the absence of any preferential regime. As can be seenfrom Table 2, in Canada and the US, tariff peaks are concentratedin textiles and clothing; in the EU and Japan, in agriculture andfood products.Preferences granted by the Quad are of a cascading nature:Countries with FTAs get the best treatment, followed by LDCsand other developing countries. The US grants preferences tothe members of the Andean pact, Caribbean community and toMexico under NAFTA. Two different groups of LDC countriesare constructed in the EU case: LDCs that are not ACP members,and LDCs that are. For Canada, the developing countries aregrouped into several categories: those benefiting from LDC,generalised system of preferences (GSP), and Mexico and Chile,which benefit from FTA status. Finally for Japan, developingcountries are divided into GSP beneficiaries and LDC bene-ficiaries. On an average, the preferential schemes are quitegenerous but are much less generous for tariff peak products.Preference margins on tariff peak items for GSP beneficiariesare only 9 per cent in Canada, 18 per cent in Japan, and 23 percent in the US [Hoekman and Olarreaga (2002). For LDCs, themargins fall to 25 per cent in Canada, and 30 per cent in theUS and Japan. The EU, by contrast, has a 50 per cent marginfor GSPbeneficiaries and 70 per cent margin for LDCs in tariffpeak items. It is said that tariff peaks in Japan affect about $500million in LDC exports to the world and that in EU affect about$800 million.Indian exports such as textiles and garments, and agriculturalcommodities can be greatly affected by the prevalence of tariffpeaks. Market access for these products could be facilitated byour ability to secure reduction in these tariffs in the industrialisedcountries through future tariff negotiations in the WTO frame-work. The phasing out by all countries of tariff peaks andescalation (tariffs rising with the degree of processing of imports)is critical to the development dimension of the current round ofmultilateral trade negotiations, and could best be achieved throughformula approaches that ensure deep across-the-board reduc-tions. Hence, the issue of tariff peaks and tariff escalation shouldbe addressed very carefully, since this holds back export-ledgrowth and leads to greater trade diversification in countries thatare not members of any significant RTA and the developingcountries in general. Moreover, higher the MFN rates of a developedcountry, the greater the leverage strength it will enjoy in termsof asking for special privilege from the developing countries,particularly in any of the North-South RTA formation.

Economic and Political WeeklyMay 26, 20071972(ii) Equivalence clause under SPS agreement not being extendedon an MFN basis: The SPS agreement encourages the use ofequivalence and mutual recognition agreements in Article 4.According to Article 4 as well as the decision by the SPScommittee, two SPS measures are said to be equivalent to oneanother when they are not identical but they yield the same levelof sanitary and phytosanitary protection.Developed country representatives noted that equivalence,although a useful principle in theory was in practice difficult todeal with even for large developed countries.18 Formal equiva-lence agreements are rare even between developed countries. Thereason for this is that negotiations are very demanding in termsof resources and time. At the same time, ad hoc acceptance ofthe equivalence of specific products, or of the equivalence ofcertain technical aspects related to SPS measures, are common.The acceptance often takes place without any formal agreements.The agreement between the EC and the US on sanitary measuresis aimed at facilitating trade in live animals and animal productsbetween the two countries, by establishing a mechanism for therecognition of equivalence of sanitary measures. The proceduretoreach recognition of equivalency is, however, rather complicatedand consists of several steps. Basically, the importing countryhasto explain the objective of the sanitary measure for which recog-nition of equivalency is sought and identify its appropriate levelof sanitary protection. The exporting country has to demonstratethat its sanitary measure achieves the importing country’s appro-priate level of sanitary protection. On the basis of the evidenceprovided by the exporting country, the importing country decideswhether the foreign measure achieves its appropriate level ofsanitary protection and, therefore, can be regarded as equivalent.The recognition of the equivalence is not easy to achieve andusually implies the fulfilment of several requirements. However,for developing countries, this option is worth pursuing since itwould greatly facilitate market access for their products.While the WTO agreement on technical barriers to trade (TBT)and the agreement on SPS recognise that countries have the rightto introduce measures necessary to protect human, animal andplant life, these measures are not to be applied in a manner whichwould constitute a means of arbitrary or unjustifiable discrimi-nation among WTO members. Developing countries have fre-quently complained about the lack of implementation of Article4regarding equivalence, particularly on an MFN basis. They statethat importing countries often require “sameness” instead of“equivalence”, the former implying that the measures must beidentical not only in outcome but in formulation too.19 Moreover,an RTA may enact a regime of positive integration through rulesof mutual recognition. The NAFTA treaty for instance, providesfor the mutual recognition of SPS measures if the exportingcountry’s regulations achieve the importing country’s appropri-ate level of protection. The burden of proof is on the exporter.The final decision about equivalency stays with the authoritiesof the importing country who take decisions on a case-by-casebasis. It is with the help of such mutual recognition agreementsthat regional agreements effectively increase barriers to importsfrom third countries, thereby leading to trade diversion.Whether the SPS agreement, including the international stan-dards and guidelines is a sufficiently strong instrument to assurethat developing countries can, in practice, derive benefits fromthe use of the principle of equivalence, as well as it promotesmultilateralism, remains a question to be answered in future. Thecore of the problem is the lack of trust developed countries havein the capacities of the food safety systems of developing coun-tries. More work by international organisations on clear guide-lines on the establishment of equivalence agreements could bevery helpful and could help distinguish between “equivalence”(as defined by the SPS agreement) and “sameness”. Developingcountries must take part in the international setting of guidelineson how to achieve equivalence in these areas so as not to beleft out. In fact, a number of developing countries have requestedmore information about how and under what circumstancesequivalence can and should be implemented through mutualrecognition agreements. India has asked the developed countriesTable 3: Distribution of Tariff Peaks by Product Groups in Quad CountriesProduct GroupTotalNo of ItemsNo of PeaksShare in Total12-1922-2932-99100-299>=300Total(Per Cent)Per CentPer CentPer CentPer CentPer CentEuropean UnionAgricultural and fishery products (1-24)2779544313312122197.7Mineral products, fuels (25-27)2570000000Leather, textiles, clothing (41-43, 50-64)15656000060.5Industrial products (28-96)7771277800423.3All products (1-96)108075713383413121263100JapanAgricultural and fishery products1897204299111816576085.1Mineral products, fuels1940000000Leather, textiles, clothing241042391528713114.7Industrial products688044391528713314.9All products897124833812610972893100USAAgricultural and fishery products17791387099151133336.6Mineral products, fuels1830000000Leather, textiles, clothing1814374110400052457.4Industrial products8123407127450057963.4All products100855451971441511912100CanadaAgricultural and fishery products142965101668015927.4Mineral products, fuels1875000050.9Leather, textiles, clothing12093202700034760.1Industrial products67913743900041371.6All products84074444916680577100Note:HS chapters are given in parenthesis.Source:Nagesh Kumar (2001).

Economic and Political WeeklyMay 26, 20071973to notify the WTO of any equivalence agreements they enter intobetween themselves so that developing countries can study themand negotiate similar agreements with the developed countries.(iii) Rules of origin problem: Preferential rules of origin (ROO)are of crucial importance in the functioning of any FTA inadministrating a number of trade issues and in avoiding tradedeflection. ROO, in general, serve as important elements in theadministration of a range of other trade regulations, including:duty drawback provisions; antidumping (AD) provisions;countervailing duty and safeguard proceedings; quantitativerestrictions; prohibited imports, etc. Each of these trade regu-lations involve distinguishing domestic from foreign goods, ordistinguishing among foreign goods. A study by Bhattacharyaand Dasgupta (2000) on triangular trade between China, Indiaand Nepal clearly showed that in the absence of any value additionnorm in respect of manufacturing products according to the Indo-Nepal trade and transit treaty, there was large-scale trade diver-sion via Nepal from China to India. However, despite theimportance of ROO, their design can result in their developmentas a trade protectionist tool hence leading to trade diversion.Krishna and Kruger (1995) demonstrate how rules of origin canact as “hidden protectionism” and induce a switch in demandin free trade partners from low-cost external inputs to higher-cost partner inputs to ensure that final products actually receiveduty free access.ROO is also a major problem in an RTA and acts as a barrierto trade particularly when developed and developing countriesare involved. In the EU, the ROO criterion is very complex anddiscriminatory. Also, satisfying the certification requirements ofthe ROO of the EU is often too costly. The benefit conferredby the preferential schemes in certain cases becomes marginalin comparison with the administrative workload and cost to planthe product mix to comply with the preferential ROO. This oftenleads to instances where firms, although meeting the necessaryconditions for origin, decide that it is simpler and cheaper to paythe MFN tariff rates. The different rates of value added criteriareflect the discretionary protectionist interests in formulating theROO in the EU. This has a negative impact on many developingcountries, which are engaged with the EU in preferential RTAs,and their exports bundle is concentrated in such sensitive productsas textiles and apparel. For example, a study found that about75 per cent of EC-EFTA trade benefited from the preferentialtrade regime, while the use of tariff preferences under otherschemes such as the GSP was low, mainly as a result of therestrictive ROO. According to Brenton et al (2002), informationon the implementation of the EU’s preferential scheme of accessfor developing countries, shows that only one-third of EU importsfrom developing countries which were eligible for preferencesactually entered the EU market with reduced duties. This pri-marily reflects the treatment of textiles and clothing products,which accounted for over 70 per cent of EU imports fromcountries covered by the GSP but where the utilisation rate (theratio of imports receiving preferences to eligible imports) wasonly 31 per cent.In some RTAs, ROO can be subject to a “cumulation” pro-cedure. According to this procedure, ROO are broader in theirgeographical coverage required for a certain product to conformorigin. In other words, ROO of a certain product in a givenexporting country confer to the required ROO set by the RTAif they are partly allocated in that exporting country onconditionthat the rest of the requirements be done in othercountries that are agreed upon from the members of the RTA.This procedure relaxes the restrictiveness of the ROO and reducestheir negative impact on production distortions and trade andinvestment diversion. However, some experts argue that cumu-lation is likely to have negative consequences on the develop-mental efforts of developing countries engaged with developedpartners in RTAs.In NAFTA, the change in tariff heading (CTH)20 supplementedby value added criteria are the main rules applied in determiningROO. Compared to the European Economic Area (EEA) agree-ment, the NAFTA places a greater reliance on CTH and less useon value added criteria, which is applied more intensively in thecase of sensitive sectors such as textiles and automobiles. Regardingautomobiles, there is an initial regional value added requirementof about 55 per cent. In the textiles sector, the value addedrequired is so high that it results in a very restrictive ROO inthis area. Here, the ROOs for textiles and apparel are “tripletransformation”, implying that all the materials used in producingthe goods are made in North America. In other cases, ROOs arebased on the material inputs used. The reason why NAFTA doesnot follow the EEA system is that the fear of the US from thelow cost Mexican automobile assembly industry which couldresult in the import of components from a non-NAFTA countryfor further processing in Mexico to produce a final product of“Mexican” origin.ASEAN rules of origin allow up to 60 per cent of non-members’originating input21incorporated in the ASEAN products entitledto the common effective preferential tariff (CEPT)22 underASEAN Free Trade Area (AFTA). Also ASEAN applies acumulative ASEAN original input for CEPT products, so thatin practice the net cumulative regional content may be lower thanthe 40 per cent and the eligibility for ASEAN-origin is stillvalid.23In SAFTA, the ROO distinguish between goods that are “whollyproduced or obtained” and goods that are not “wholly producedor obtained” in an exporting SAPTA country. In order to en-courage regional value addition, the agreement also includes acumulative ROO, which initially said that goods processed inmore than one member country can be eligible for concessionsprovided that the value added in SAPTA countries was at least60 per cent of the fob value. The ROO local content provisionshave been a contentious issue and were subject to continuousscrutiny by members who realised that the effectiveness ofSAPTA was limited, in part due to low value addition in manyof their most competitive exports. After much resistance, par-ticularly from India, the local content requirement was reducedto 40 per cent for the non-LDC members (India, Pakistan, SriLanka) and to 30 per cent for the four LDC members (Bangladesh,Nepal, Bhutan, Maldives), and the cumulative ROO was reducedto 50 per cent.Thus, given the prevalence of PTAs, the question of rules oforigin is of particular policy relevance. It is essential that clear-cut principles and criteria for determining the origin of goodsbe adopted. ROO will vary among the FTAs depending on theunderlying sensitivity to intra-regional competition and on membercountries strategic goals and the level of development.(iv) Non-reciprocal tariff concessions: One of the importantproblem areas is the non-reciprocal tariff concessions exchangedin an RTA. For instance, SAFTA distinguishes between membersaccording to their level of development. The agreement provides“special and favourable treatment” for the LDCs by the non-LDCs, including deeper and wider tariff preferences. Also,tariffand other concessions accorded by a non-LDC to another

!Ii