More Favorable and Differential Treatment of Developing Countries Toward a New Approach in the World Trade Organization

Hoekman, Michalopoulos


Introduction and Overview
International trade is important for development and poverty alleviation.It helps raise and sustain growth-a fundamental requirement for reducing poverty-by giving firms and households access to world markets for goods, services and knowledge, lowering prices and increasing the quality and variety of consumption goods, and fostering the specialization of economic activity into areas where countries have a comparative advantage.Trade and trade opportunities are important for generating the investment and positive externalities that are associated with learning through the diffusion and absorption of technology.Policies that shelter economic agents from the world market impede these spillover benefits and dynamic gains (Bhagwati, 1988;Irwin, 2001).This does not imply that 'one size fits all' or to deny that adjustment costs and measures to safeguard the interests of poor households must be considered in the design of policies.
Understanding and addressing the impacts of reform on the poor and vulnerable in society, and taking action to facilitate adjustment is necessary (McCulloch et al., 2001).
The primary determinant of the benefits from trade is a country's own policiesthe principle that 'what you do is what you get'.Determining the appropriate trade policy stance and the associated complementary policies for a country should consequently figure in the design of development and poverty-reduction strategies.In part, however, this policy stance will be affected by what other countries do.A major question confronting many developing country governments is if and how trade policy should take into account that the global trade environment is distorted by a variety of policies pursued by trading partners.Measures that restrict market access for developing countries' exports goods and services and that lower (raise) the prices of their exports (imports) have direct negative effects on investment incentives and the growth potential of their economies.
For example, agricultural support policies-high rates of subsidization and trade barriers-by developed countries increase world price volatility, lock developing countries out of major markets and can lead to import surges that have highly detrimental effects on developing country farmers.The existence of such policies has become a major political barrier to further trade policy reform in developing countries.
The WTO is a forum both to negotiate improved market access and to agree to 'rules of the game' for trade-related policies.Developing countries gain from both dimensions.A rules-based world trading system is beneficial to developing countries as they are mostly small players on world markets with little ability to influence the policies of large countries.The rules of the WTO can also be beneficial by reducing uncertainty regarding the policies that will be applied by governments-thus potentially helping to increase domestic investment and reduce risks.
Much obviously depends, however, on getting the rules 'right'.To a significant extent WTO rules reflect the 'interests' of rich countries: they are less demanding about distortionary policies that are used by these countries and they largely mirror the ("best practice") disciplines that have over time been put in place by them.Thus, the much greater latitude that exists in the WTO for the use of agricultural subsidization, for example, reflects the use of such support policies in many developed countries.The same is true for the permissive approach that has historically been taken towards the use of import quotas on textile products-which in principle was prohibited by GATT rules.
More recently, the inclusion of rules on the protection of intellectual property rights has led to perceptions that the WTO contract continues to be unbalanced.' Ensuring that the rules are supportive of development and are seen to be so by the majority of stakeholders in society is perhaps the most fundamental challenge confronting the WTO from a development point of view.Developing countries have complemented efforts to influence the content of WTO rules with a strategy of seeking 'differential and more favorable treatment' (Hudec, 1987;Finger, 1991;Michalopoulos, 2000).Generally captured in the term special and differential treatment (SDT), many provisions in the WTO call for the granting of preferential access to markets for developing countries, exemptions (transitory and permanent) from certain rules, and development assistance.One approach that has been pursued by developing countries in an effort to increase the net benefits from WTO membership has been to seek to expand the reach of SDT.This paper discusses SDT options and mechanisms that could be considered as part of the WTO's Doha Agenda.2It is based on a longer report that develops the arguments in greater depth and assesses the various proposals that have been submitted to the WTO on SDT (Hoekman et al. 2003).
A premise underlying our approach is that efforts to enhance the development relevance of the WTO need to distinguish the question of more favorable treatment of developing countries from the broader issue of ensuring that WTO rules and disciplines support development.The second dimension is by far more important, but it goes well beyond the specific language that is found in WTO agreements relating to developing country interests.Instead, it revolves around whether a particular WTO rule makes sense for developing countries to implement.SDT has arguably not been an effective instrument to promote development for reasons discussed at greater length below.The key requirements for helping countries to use trade for development include a concerted effort to reduce market access barriers and agricultural trade distortions on a nondiscriminatory basis; strengthening mechanisms to increase the likelihood that WTO rules support development prospects; recognizing that resource constraints in small and low income countries may require temporary exemptions from multilateral rules; and making greater efforts to tie trade-related assistance to national development priorities.Specifically, we suggest that the following be considered by WTO members: * A binding commitment by developed countries to abolish export subsidies and NTBs.(tariff quotas) and to reduce MFN tariffs on labor-intensive products of export interest to developing countries to no more than 5 percent in 2010, and to no more than 10 percent for agricultural products.All tariffs on manufactures should go to zero by 2015, the target date for the achievement of the MDGs.The liberalization should include developing countries, on the basis of a formula approach that reduces the variance in tariffs very substantially and gives credit for past unilateral trade liberalization.* A binding commitment by developed countries on services to expand temporary access for service providers by a specific amount-e.g., equal to an additional one percent of the workforce-and not to restrict cross-border trade (e.g., via telecom channels).* Unilateral action by all developed countries to extend preferential market access for LDCs, and to simplify eligibility criteria, especially rules of origin.* Affirmation by the WTO that core disciplines relating to the use of trade policy should apply equally to all WTO members.* Acceptance of the principle that for small and low income countries, 'one size does not fit all' when it comes to WTO rules on domestic regulation or that require substantial investments to be implemented.
* Recognition that some WTO agreements need to be adapted to make them more supportive of development, and a consequent willingness to change the rules and make them more development oriented.* Expansion of development assistance to bolster trade capacity in poor countries and strengthening the linkages between trade-related technical assistance and the mechanisms through which aid priorities are determined in developing countries.
The elements of this 'package' are both consistent with, and would do much to realize, the objectives laid out in para 2 of the Doha declaration: ".. we shall continue to make positive efforts designed to ensure that developing countries, and especially the leastdeveloped among them, secure a share in the growth of world trade commensurate with the needs of their economic development.In this context, enhanced market access, balanced rules, and well targeted, sustainably financed technical assistance and capacitybuilding programmes have important roles to play" (WT/MIN(01)/DEC/1).In the penultimate section of the paper we discuss briefly how the suggested 'package' can be mapped into the ongoing discussions on SDT.There are currently three major dimensions of SDT in the WTO-preferential access for developing countries to developed country markets; promises by developed countries to provide technical assistance to lower-income economies to help them implement multilateral rules; and exemptions from certain WTO rules.Some of the latter are transitory, e.g. for rules on customs valuation, the abolition of trade-related investment measures (TRIMs), and implementation of stronger protection of intellectual property rights under the Agreement on TRIPS, and some permanent, e.g., limited reciprocity under the Enabling Clause or Article XVIII GATT. 4 A good case exists for all three types of SDT under certain circumstances, but to make SDT a more effective instrument, it must be re-thought and targeted more narrowly towards those that need it the most.

I. Differential and More Favorable Treatment
One rationale for SDT is essentially that very small and/or low income economies lack the institutional development or minimum scale to manage the full panoply of WTO rules or, at least, might find the returns to creating the institutions to apply them effectively outweighed by the costs.This is not to say that the WTO rules are not sound policy prescriptions-although in some cases they are not-simply that some rules become unambiguously beneficial only as countries become richer.Small and/or poor countries may also lack the resources to overcome natural obstacles to trade or to use policies that in principle would be the most efficient in addressing market failures.This could give rise to a case for offering such countries preferential access to markets as well as financial and technical assistance.
Historically, the major focus of efforts to operationalize SDT have centered on preferential access through the Generalized System of Preferences (GSP) and ensuring that the extent of reciprocity in periodic multilateral trade negotiations was limited (Laird, Safadi and Turini, 2002;Finger and Winters, 2002).One option for strengthening 3 For a succinct but comprehensive summary of the post-Doha SDT discussions in the WTO, see the ICTSD/IISD Doha Round Briefing Series, vol. 1, no. 13, February 2003 (www.ictsd.org).
4 Article XVIII allows developing to use trade policies in the pursuit of industrial development objectives and to protect the balance of payments, imposing weaker disciplines than on industrialized countries.There are also many exhortations to developed countries to 'take into account' the interests of the developing countries in the application of WTO rules and disciplines.
SDT that has been actively pursued by developed countries in recent years has been to deepen trade preferences for least developed countries (LDCs) and sub-Saharan Africa.
Examples are the EU Everything But Arns initiative-which grants duty-and quota-free access for all goods exported by LDCs (with delayed implementation for three important products-sugar, rice, and bananas), and the US African Growth and Opportunities Act, which does the same for a large number of African countries.Experience has shown that such schemes can have a significant positive effect on recipients, but that very much depends on their supply side capacity (e.g. a large proportion of the benefits of GSP have accrued to a small number of more advanced developing countries), their ability to put the rents generated to good use, and on the ancillary documentary requirements that are imposed by preference-granting countries.Recent research reveals that liberal rules of origin are critical for a strong trade response in sectors such as textiles and clothing (Inama, 2002;Brenton, 2003;Brenton and Manchin, 2002;Mattoo et al. 2002).
Preferences are by definition discriminatory-to give some countries preferential access implies, and depends for its effects on, not giving such access to others.A major policy question that arises is which countries should be eligible for preferential market access.In practice there is a hierarchy of preferences, with the most preferred countries generally being members of reciprocal free trade agreements (EU, NAFTA, EU-FTAs, etc.), followed by LDCs-which in principle often have free access to major marketsand other developing countries, which generally get GSP preferences.In many jurisdictions, GSP status does not involve duty free treatment, instead being limited to a tariff reduction (Inama, 2003).
From a poverty reduction point of view-and in light of the Millennium Development Goals-a good case can be made that preferences should focus on the poor, wherever they are geographically located, and not on a limited set of countries.In absolute terms, most poor people live in countries that are not LDCs-especially China and India.Limiting preferences to LDCs or concentrating on a specific geographic region such as sub-Saharan Africa-while appropriate in light of limited institutional capacity and infrastructure weaknesses in these countries-ignores the majority of the poor in the world today (Winters, 2001).
One way forward would be to agree on a single preferential tariff rate-zero-for all products currently benefiting from GSP status in developed countries (as is the case presently in the US), thereby removing all partial preferences.Extending preferential duty-free access to large countries such as India and China will be very difficult politically-one reason why duty-free access for much of Africa and the LDCs could be implemented is that these countries account for less than 0.5 percent of world trade.
Given the political reality that developed nations will not grant large and/or higher income developing countries unconditional preferential market access, this will have to occur through MFN liberalization.In turn, this will require a willingness on the part of developing countries to engage in reciprocity.But this is in their own interest, for, as mentioned above, much of the benefit from trade policy reforms is generated by a country's own actions.5Thus, rather than seeking to extend GSP programs for all developing countries-which have proven to be of limited value, with only a few countries benefiting significantly-greater and more sustainable gains can be obtained through MFN liberalization.This is also something that the WTO is designed to deliver.
Thus, our first recommendation: 1. Market access: Give priority to MFN liberalization of trade in goods and services in which developing countries have an actual or potential export interest A strong case can be made that MFN-based market access will have the greatest beneficial impact on development.One reason for this is that it involves an element of 'rebalancing' of the WTO as it implies that elements of 'reverse SDT'-special opt-outs and exemptions that benefit interest groups in industrialized countries at the expense of developing countries-would be removed.Agricultural subsidy programs, textile import quotas, tariff peaks and escalation that imply high rates of effective protection for developed country industries are examples.Such continued protection implies that products produced by poor people are subject to higher tariffs than products produced by the non-poor.Reversing this situation through a MFN-based liberalization program that centers on these sectors would not only be very beneficial to developing countries (and developed country consumers), but also help remove a major political barrier to further trade reforms in developing countries by providing a positive demonstration effect.
Implementing this recommendation has implications for the choice of negotiating modalities on tariffs.A Swiss type formula, as proposed by the Chair of the WTO market access committee, is much preferable to a request-offer or linear cut approach (Francois and Martin, 2003).6Any formula-based MFN liberalization should extend to middleincome countries, which are among the most dynamic markets in the world and where trade barriers are often substantially higher than in developed countries.It would usefully extend to LDCs as well.Reciprocity is the engine of the WTO; it is not engaging in reciprocal exchange of market access concessions that has helped create the current structure of protection confronting developing countries.7 In addition to defining negotiating modalities, WTO members should set a concrete timetable (deadline) and agree on specific benchmarks for product coverage and the maximum tariff that is to be permitted.We suggest that a target be set that at the end of the Doha round implementation period-say 2010-MEN tariffs on manufacturing products of export interest to developing countries, defined to include the set of laborintensive products such as footwear, textiles and apparel, 8 will not exceed 5 percent and those in agriculture 10 percent.Additionally, it could be agreed that by 2015, the target date for the achievement of the Millennium Development Goals all tariffs on developing country exports of manufactures will be eliminated, in effect removing the problem of tariff peaks and associated tariff escalation.Both of these measures will benefit the millions of poor in all developing countries who are employed in these activities.. Reciprocal concessions will be necessary in order to make complete developedcountry liberalization politically feasible but, as we have noted several times, this too will agreements that may not be development priorities.
be beneficial and assist the battle against poverty.Here we believe a formula-based approach to tariff reduction-as has been proposed by many WTO members-is appropriate and useful.This should focus on a commitment by all developing countries to bind all tariffs, to reduce current tariff bindings to come much closer to applied rates, as well as further reduction in applied rates (see Francois and Martin, 2003).Given the resource allocation distortions created by tariff structures that are highly differentiated, one benefit of a formula approach is that it can be used to reduce the variance in tariffs.9 While reciprocal tariff reductions by developing countries need not fully match the proposed liberalization in developed countries proposed above, they must be significant enough to trigger substantial developed country liberalization of products of interest to developing countries.As the variance in tariffs is higher in developed countries than in developing economies, while levels of bindings are much higher in developing countries, a 'Swiss-type' formula that centers on reducing the average levels of bindings can do much to both reduce tariff peaks and escalation in OECD countries and give credit to developing countries for past reforms (Hoekman, 2002).i°S ervices are of great importance to developing countries and there are substantial opportunities both to expand exports and to liberalize further access to developing country markets.While the latter will bring the greatest gains, opening by developed countries of temporary access to service markets for natural service providers-so-called mode 4 of the GATS-and a binding of the current liberal policy set that is applied to cross-border trade (modes 1 and 2 of the GATS)-would both be valuable and assist governments in pursuing domestic reforms.Walmsley and Winters (2002) conclude that an opening of developed country labor markets to allow temporary entry by foreign workers equal to 3 percent of the current workforce would generate welfare (real income) gains that exceed those that could be attained from full merchandise trade liberalization.
In addition, many developing countries have begun to exploit the opportunities offered by the internet and telecommunication networks to provide services through cross-border trade.Currently such trade is largely free of restrictions, and this desirable state of affairs should be locked in through the GATS (Mattoo, 2003).
In the case of services (the GATS), the multilateral rules allow substantial discretion to government to apply policies that discriminate against foreign firms.For developing countries what matters most is market access-as with goods-and we argue that this should imply that negotiating modalities are agreed to focus on the removal of barriers to trade on those services and modes of supply in which developing countries have an export interest.Given that the rapid spread of information technology and telecommunications infrastructure is allowing service firms located in developing countries to contest markets in richer countries, binding commitments to impose no restrictions on such trade would be valuable.Specific commitments to open markets through liberalization of temporary movement of natural service suppliers would also be very valuable, although much more difficult to achieve.
2. Developed countries should continue to extend dutyand guota-free treatment for LDC exports on a unilateral basis, and simplify associated rules of origin In tandem with a commitment to implement deep MFN liberalization by developed countries on goods and services produced by developing countries, they should provide dutyand quota free access on a universal basis to all products originating in LDCs, starting at the conclusion of the Doha negotiations."'Currently the only clearly defined sub-set of developing countries in the 1979 GATT Enabling Clause is the LDC group, and this is the group that has recently been granted deeper preferential access by major trading powers.While it is important to recognize that there are a number of countries that are not formally classified as LDCs but have similar levels of per capita income and equally limited supply capacity, efforts to move beyond the LDC category will be divisive and involve arbitrary distinctions between developing countries.Thus, we suggest that the focus of full dutyand quota-free access continue to center on LDCs.12 " Large, more advanced middle-income countries should also consider granting such treatmnent.
12 In principle, equity considerations suggest that such deep preferences should also be extended to poor countries that are not classified as LDCs, and to very small econornies.However, this runs foul of WTO disciplines (Part IV and the Enabling Clause) that require identical preferential treatment to be extended to all developing countries that are non-LDCs (these provisions allow 'better' treatment of LDCs).
Of great importance here is not only the coverage of duty-free access (which should extend to all products), but also action to apply liberal rules of origin to preferential trade.This applies not only to the exports of LDCs, but also to GSP Programs more generally (Inama, 2002;Brenton, 2003;Mattoo et al. 2002).Efforts to adopt identical liberal rules of origin would do much to reduce transactions costs associated with utilizing preferences.It is noteworthy that this is an area that has never been subject to GATT/WTO rules, despite attempts that date back to the 1960s and 1970s (Hoekman and Kostecki, 2001).Indeed, the lack of progress on simplification and harmonization of preferential rules origin is another reason why we favor a MFN-based approach to market access for developing country exports to developed countries.
There is a tension between deepening preferences for LDCs and MFN-based liberalization, as the benefit of the former-assuming rules of origin are not too restrictive-is eroded by the latter.Given that research suggests that to date many LDCs have not benefited significantly from preferences-because of limited product coverage, only partial reductions in tariffs, and tight origin requirements as well as local supply-side difficulties-there should be limited concern with the erosion of current preferences that is associated with a MFN approach.We recognize the danger that deepening preferences will increases the scope for future preference erosion and potentially generate more opposition to MFN liberalization from the beneficiaries. 1 3 Provided not too much is invested in extending preferences now and provided that the proposed MFN liberalization is also pursued, this danger can be managed. 4 13 Ozden and Rheinhart (2003a,b) argue that countries with preferential access to developed country markets-even if it is of limited value due to administrative requirements and exceptions-have less of an incentive to pursue trade liberalization.14 Sugar is an example.Current quota allocations in protected markets such as the EU go disproportionately towards a few countries that are relatively high-cost producers-for example, Mauritius has 38% of EU quotas (Mitchell, 2003).Given that a number of LDCs are significant producers of sugar and are lower-cost suppliers than Mauritius, the extension by 2009 of duty-and quota-free access to the EU market for all LDCs will result in preference erosion for Mauritius.This partial unwinding of trade diversion will benefit more cost-effective LDC producers but is still costly in global welfare terms.Moving to free trade in sugar markets would benefit the non-LDC producers, generate higher global welfare gains, and increase both world sugar prices and sugar trade.It should also be recognized that coordinated global liberalization across all products will offset some of the lost preference rents.In the case of sugar, the world sugar price increase would offset about half of the lost quota rents for countries that currently have preferential access.Moreover, the loss in rents would be much less than is commonly expected, because many of the beneficiaries of preferences are high-cost producers, reducing the potential benefit of preferential access to distorted markets.The cost to the EU and US of providing $1 of preferential access has been estimated to exceed $5 (Beghin and Aksoy, 2003).
Another dimension of preferential treatment (SDT) in the WTO relates to commitments by developed countries to apply instruments such as antidumping, countervailing duties and safeguards less vigorously against developing countries through the use of de minimis and similar provisions.We suggest that such provisions be maintained, and be understood to apply to exports from all developing countries as at present.This will help to create an element of desirable differentiation in the reach of SDT, given that full duty-free access is provided only to a subset of developing countries.
3. Technical and financial assistance: actions to help all developing countries improve their trade capacity should be strengthened by linking such activities to the national processes through which development aid is provided at the country level.
It is generally recognized that the major constraint limiting export growth in LDCs and other small and low income countries is not restrictive market access conditions in export markets but a lack of supply capacity and the high-cost environment in which firms must operate.In addition, firms in these developing countries may find it more difficult to deal with regulatory requirements such as health and safety standards that apply in export markets.Development assistance can play an important role in helping to build the institutional and trade capacity needed to benefit from increased trade and better access to markets.This assistance must go beyond the implementation of WTO rules narrowly defined and focus on supply capacity more broadly, as well as addressing adjustment costs associated with reforms.Scarce aid resources should be allocated to priority areas that will help to mobilize growth.Trade-related areas will normally figure among these priorities, but not necessarily.
More funds are needed to address trade-related policy and public investment priorities, to help low-income countries adapt to a reduction in trade preferences following further nondiscriminatory trade liberalization, and to assist poor net importing countries to deal with the potential detrimental effects of a significant increase in world food prices should these materialize.In our view the best approach is to embed the delivery of trade-related technical assistance in the national agenda and priority-setting processes that are used by primarily country-focused.In order to maximize financing for trade-related assistance and to ensure that assistance in this area addresses priority areas for intervention, the trade-related technical assistance and capacity-building agenda must be embedded in a country's national development plan or strategy.In the case of low-income countries the primary example of such an instrument is the PRSP-implying that governments and stakeholders must take action to embed trade in PRSPs in those instances where trade is seen as a priority.
It is important to avoid a situation in which a desire by donor countries to see developing countries implement certain WTO agreements leads them to divert assistance flows towards trade institutions at the expense of recipients' own priorities.For these reasons, we do not support suggestions (e.g., Finger and Schuler, 2000) to make technical assistance a mandatory, binding, requirement and to link implementation of WTO agreements to the provision of such assistance.We recognize that disallowing such linkages may make it more difficult to secure developing country agreement to future WTO rule changes, but would argue that if developing countries are to be 'compensated' for adopting new rules, they should be confident that this is not at the expense of donor support for other objectives.Whatever the national priorities are, over time as income levels and institutional capacity increases, countries will become better able to implement resource-intensive WTO rules.Possible approaches are discussed below.

II. WTO Rules and Economic Development
Much of the SDT debate centers on issues related to making the WTO more development relevant, and the perceived need to both revisit some of the existing disciplines and to take action to ensure that new, future, rules support development.Many WTO rules make sense from a development perspective (Finger, 1991;Hoekman and Kostecki, 2001).Some do not.Several current agreements need to be rebalanced to reflect developing country interests-in particular the Agreements on Agriculture and TRIPS.More generally, looking forward and learning from the experience with Uruguay Round implementation, there is a need to recognize that in this area 'one size does not fit all'.
We make three recommendations to move WTO rules to be more supportive of development: 1. Core trade policv rules: Towards common disciplines We do not support SDT that involves flexibility to pursue trade policies that introduce distortions to trade through protection.Hoekman et al. (2003) survey some of the voluminous empirical literature that has investigated the effects of trade protection and its removal on the performance of firms and industries in developing countries.The overwhelming tendency of this literature is to conclude that the case for using traditional trade policy instruments such as quotas and quota-like policies such as trade-related investment measures to achieve economic development objectives is weak.Government interventions are justified where there are market distortions; but in most circumstances market distortions should be addressed through interventions other than through trade (Bhagwati, 1988;Bora, Pangestu and Lloyd, 2000).
This does not imply that developing countries should give up their rights to use all kinds of trade policies-countries have the right under the WTO to impose tariffs and quotas as well as export taxes if they desire to do so, under certain well specified circumstances.' 5 Nor does it deny the distortions that are created by industrialized country trade and subsidy policies, or the need to recognize that there are adjustment costs associated with trade reforms.What it does imply is that there is a clear ranking of policy instruments, with quotas and quota-like instruments being particularly costly, with both a significant body of theory and evidence suggesting these are not efficient tools to promote industrial development (Noland and Pack, 2003;Hausmann and Rodrik, 2002).
Not using such instruments will benefit consumers and enhance welfare in developing countries.Similarly, there are benefits associated with binding tariffs-including that this is a major negotiating coin in the WTO.Moreover, we argue that WTO rules relating to transparency of trade policy and to the criteria that should be applied when taking actions against imports that are deemed to injure a domestic industry are also beneficial.
We recognize, however, that there are cases where weaknesses in institutional capacity and severe market imperfections combined with lack of financial resources may require that developing countries be permitted to pursue second best trade policies in certain well defined circumstances.Indeed, as part of this a good case can be made that some existing WTO disciplines and provisions discriminate against developing countries-for example, the special safeguard provisions of the Agreement on Agriculture-or are not in their interests.These issues are important and are discussed at length in Hoekman et al (2003).In the next sub-section we synthesize the conclusions that emerge on some of the major WTO agreements.They are best regarded not as questions of SDT but as a call for revision of the relevant WTO rules (see below).

Rule-making and Implementation of Agreements
A major feature of the 1979 Enabling Clause is that it calls on industrialized countries not to seek reciprocal concessions from developing countries that are 'inconsistent with their individual development, financial and trade needs'.No one supports the making of inappropriate concessions, but the overuse of the 'nonreciprocity' clause has, in the past, excluded developing countries from the major source of gains from trade liberalizationnamely the reform of their own policies.Non-reciprocity is also a major reason why tariff peaks today are largely on goods produced in developing countries.Reciprocity is the engine of the WTO negotiating process.By not engaging in reciprocity countries lose both a mechanism that can be used to support the pursuit of beneficial trade policy reforms, but also remove an instrument that can generate better access to export markets.
The reciprocal exchange of trade liberalization commitments benefits all the countries that engage in the process-this is one reason why we believe the WTO trade policy rules should apply to all members.
The value of reciprocity when applied to regulatory policies is much more doubtful, given analysts' uncertainty about the appropriateness of particular policy instruments, and the differences between countries' domestic priorities (Hoekman, 2002).
A good case can be made that when it comes to regulatory policies that affect trade only indirectly (if at all), one quickly gets into a situation where apples are traded for oranges, with significant potential for a negative net outcome for low-income countries.
This suggests that attention should focus on redefining the principle of calibrated reciprocity in the Enabling Clause to apply only to policy disciplines that do not directly concern trade policies or that are resource-intensive to implement.In the case of small and poor countries, there is at least a prima facie case that such agreements may not be development priorities.Consider for example the so-called Singapore issues.It is very difficult to conceive of rules in these areas that do not differentiate across countries.In the case of competition law, for example, most developing countries that have such laws have not had them for long, often have not been enforcing them and generally need to develop much more experience to determine what works and what does not.To some extent differentiation across countries can be achieved through the choice of negotiating modalities.Thus, in the case of investment or trade facilitation, the adoption of a GATStype positive list approach to the sectoral coverage or depth of policy disciplines can ensure that inappropriate commitments are not made by low-income countries.But such an approach will be much more difficult to apply to issues such as procurement or competition policy if the multilateral disciplines are to have meaning.In these cases there is a need to recognize that implementation may be resource-intensive and not a priority for low-income countries.
The ability to implement and to benefit from implementation of WTO disciplines will vary from country to country, depending not only on size and income but also on factors such as the skills of the workforce and institutional capacity.These observations suggest there is a need for 'differentiation' between developing countries in determining the reach of resource-intensive WTO rules.The basic rationale for differentiation is that certain agreements may simply not be development priorities or they may require many other preconditions to be satisfied before implementation will be beneficial.These preconditions can be proxied by the attainment of a minimum level of per capita income, institutional capacity and economic scale.Some WTO disciplines may not be appropriate for very small countries in that the regulatory institutions that are required may be unduly costly-i.e., countries may lack the scale needed for benefits to exceed implementation costs.
Several options have been proposed to take into account and operationalize country differences in WTO agreements.Such "rule-related SDT" could involve: * Total flexibility for developing countries as long as other WTO members are not harmed (Stevens, 2002); * An agreement-specific approach involving country-based criteria that are applied on an agreement-by-agreement basis to determine whether (when) agreements should be implemented.This could be linked to the provision of technical assistance and development of a national action plan for ultimately assuming the WTO obligations concerned (Wang and Winters, 2000); * A country-based approach that places trade reforms priorities in the context of national development plans such as the PRSP, and would employ multilateral surveillance and monitoring to establish a cooperative framework under which countries are assisted in gradually adopting WTO norms as part of a more general program of trade-related reforms (Prowse, 2002;Hoekman, 2002).
Of these options, the first has implicitly underpinned the approach taken in the proposals made by many countries to the CTD in 2002.Both the second and third approaches would allow the issue of defining general eligibility (country differentiation) to be avoided.They are also likely to result in more attention being devoted to the economic costs and benefits of implementation of WTO rules.However, they have downsides as well.An agreement-specific approach entails 'implementation audits' and associated negotiations and makes the WTO a focal point for trade-related development assistance.Reasonable people may disagree about the magnitude of assessed costs and benefits or the specific criteria that are suggested for agreement-specific SDT.
Determining criteria that could be used in the implementation context will require input from stakeholders, government agencies and development institutions.While this could help to strengthen the coherence of policy at both the national and international levels, it would also make the WTO negotiation and enforcement process much more complex.
Widening the set of actors involved in implementation of a new approach towards SDT may reduce the risk of inducing countries to adopt and pursue a program of trade and regulatory reform that may not be appropriate, but care will be needed to ensure that this would not lead to cross-conditionality.' 6 16 Many countries were concerned in the Uruguay Round about avoiding possible "cross-conditionality" between WTO and international financial institutions; this led to a ministerial declaration on "coherence" to call for "avoiding the imposition on govermnents of cross-conditionality or additional conditions" resulting from cooperation between the WTO and the international financial institutions.See Declaration on the Contribution of the World Trade Organization to Achieving Greater Coherence in Global Economic Policymaking, December 15, 1993.
A country-specific approach has the advantage of not centering on WTO requirements only.The focus would be to support developing countries in managing their trade reform agenda.Although eminently sensible from a purely development perspective, it is difficult to make consistent with the binding nature of the WTO negotiating process.The presumption would be that national considerations would take precedence over implementation of WTO obligations.In effect it implies a shift towards development institutions-national and international-taking the lead in regulatory areas, with the multilateral negotiations focusing on the market access agenda, an area there they have proved they can do well.This could certainly help to avoid unconstructive outcomes caused by taking up domestic policy issues in multilateral negotiations in the absence of clearly defined developing country priorities and constituent interests.
However, it is likely to be too far-reaching to be acceptable to many WTO members.' 7 Another alternative is to adopt a rule of thumb -based on variables such as size and income per head -that would allow the bulk of the identified difficulties to be tackled at low (or zero) negotiating cost.This could be done by defining country eligibility for rule-related SDT provisions more narrowly.Such SDT could comprise an opt-out for countries that satisfy the criteria and would be broadly applicable across those disciplines where it has been agreed there are substantial implementation issues.' 8This approach implies redefining the three-fold country group classifications currently used in the WTO-the LDCs, all other developing countries, and the developed country group.
This would seem sensible given that many countries that define themselves as developing have per capita incomes that are many multiples of those in the poorest countries.Ghana, Nigeria and Saint Lucia are very different from Argentina and Korea and their institutional capacity significantly weaker.' 9 Their needs for assistance are much closer to those of the LDCs than middle and higher income countries which still call themselves 17 Finger (2002) argues that in the behind-the-border areas the development banks must lead.Their processes include technical, cost-benefit analysis, their instruments include country-project specific legal commitments.These give the development banks comparative advantage over multilateral negotiations to address behind-the-border issues.In this view, SDT on such issues can never be more than identifying what the multilateral negotiations cannot do, not about what they can do to support development.
18 Once countries have increased per capita incomes or exceeded the criteria thresholds that are established, the various disciplines would become applicable.The objective is not to create permanent exclusions.
developing.In our view low income and small economies should be included in SDT.
Others will not.Although this has been a politically sensitive issue in the WTO, much of the discussion on greater country differentiation has (implicitly, if not explicitly) been driven by market access preferences, where country classification is inherently arbitrary.
In the case of implementation of resource-intensive WTO agreements, a more general approach based on objective criteria should be feasible.To allow for flexibility, such a rule of thumb approach could be supplemented by a fairly demanding appeals procedure for countries that feel they have been particularly hurt as a result of not satisfying the 20 criteria.
What are the "right" criteria is ultimately something that WTO members must determine for themselves, and determining which agreements are resource-intensive in the sense used here is an important part of this.Arguably, what matters most at this point is that WTO members recognize that capacities and priorities differ hugely across the membership and consider alternative approaches along the lines sketched out above.
Given the steady expansion of the WTO into regulatory areas, this would help make 'development relevance' more than a slogan.A new approach towards SDT that is anchored much more solidly on economic analysis and a national process of identification of development priorities could do much to enhance the 'ownership' of the institution in developing countries.Whatever the specific approach that is chosen, agreement on a methodology through which to determine ex ante the costs and benefits of implementation will help provide a basis for identifying the rules and disciplines that small or poor countries should not be expected to implement.It must be recognized that some types of rules will fail a cost/benefit test at the level of individual WTO members. 21 In the interim, extension of transition periods and postponing implementation of some 20 Arguments have also been made in favor of a less centralized approach to categorization by negotiating different country categories for different WTO agreements, on the basis of the presumption that systemwide categories may not always be the right identifier across different agreements.Such an approach is likely to give rise to similar discretion and process-related negotiating costs as a case-by-case approach.We believe that when it comes to resource-intensive agreements there will not be that much variation in the ability to implement.That is, if a country confronts institutional/resource constraints in implementing TRIPS it is also likely to have problems implementing the Agreement on Customs Valuation.
21 An example of an agreement that may require both significant investments for implementation and give rise to potentially large net resource outflows is the Agreement on TRIPS.
agreements (especially TRIPS) for LDCs and other low income and small economies should be adopted.
Mechanisms also need to be strengthened to allow for regular monitoring of the implementation of SDT.This should extend to the provision of information on traderelated development funding and investments in trade capacity enhancement, as well as to regular reporting to the WTO on the importance of trade-related priorities identified in national development plans or strategies.A first step in this direction has already been taken by the DAC and WTO, building on a database of bilateral and multilateral development projects.More attention is needed at the national level to identify priorities, where trade issues rank in this overall set of priorities, and within the trade agenda, what will generate the highest social rate of return.The Integrated Framework for Traderelated Technical Assistance is one important vehicle that supports the achievement of this objective (Tsikata, 2003); the WTO Trade Policy Reviews provide another mechanism that can be used for this purpose.

Rebalancing Existing Agreements
As mentioned, much of the SDT debate revolves around specific WTO agreements.
Getting agreements 'right' requires an agreement-specific approach, and may require differentiation, not only by type of country, but by an explicit focus on what makes sense from the point of view of safeguarding the interests of vulnerable groups in poor societies.This paper cannot consider all WTO agreements.Instead, we focus briefly on two agreement that are of great importance for developing countries-agriculture and TRIPS.
2 2 In the case of agriculture, the so-called 'Green box' of permitted subsidies does not cover the types of market imperfections that are likely to be found in developing countries.Developing countries may also need to pursue 'second best' policies insofar as their realities dictate that in the foreseeable future they will not have the institutional capacity to pursue first best policies in reducing poverty.An example here is a need to allow for special safeguards in agriculture for low-income countries that do not have the capacity to implement an adequate safety net.The Agreement on Agriculture needs to focus on a set of measures to enhance food security and stimulate agricultural production of the rural poor in developing countries on a permanent basis.These measures should not be seen as 'exceptions' or SDT, but rather as a rebalancing of the rules.Allowance should be made for: * Direct and indirect investment and input subsidies or other supports to households below the national poverty line in order to encourage agricultural and rural development.Such supports could be product specific as well as general-what is needed is that they are effectively targeted to the rural poor.* Programs that support product diversification in small, low income developing countries currently dependent on a very small number of commodities for their exports, including programs involving government assistance for risk management.* Foodstuffs at subsidized prices in targeted programs aimed at meeting food requirements of the poor, whether urban or rural, as part of an overall effort to enhance food security. 23* Transportation subsidies for agricultural products and farm inputs to poor remote areas; * Programs involving government assistance for the establishment of agricultural cooperatives or other institutions that promote marketing, quality control or otherwise strengthen the competitiveness of poor farmers.* A new Special Safeguard provision, available only to developing countries, to provide rapid but time limited protection against import surges that hurt poor producers.
Many of the above provisions were included in the so called 'Harbinson' draft that suggests approaches for future liberalization commitments in the current WTO negotiations on Agriculture. 2 4 The draft also contains a large number of provisions permitting developing countries greater leeway in protecting agriculture through border measures, such as tariffs and tariff quotas, than would be the case for developed countries.Such SDT could be justified because low income developing countries do not have the fiscal capacity to support agriculture through less trade distorting direct income supports.On the other hand, this additional flexibility, over 23 Consumption subsidies are already available under the WTO, but providing them via producer subsidies for goods that are barely traded is generally not permitted because developing countries registered no subsidies during the Uruguay Round and are bound by a commitment not to increase subsidies above historical levels.The delivery of such subsidies via producers may be desirable for reasons of administrative simplicity. 24Curiously, a recommendation to include programs in support of product diversification has not been included.See WTO, Negotiations on Agriculture, 'First Draft of Modalities for the Further Comnmnitments' (Revised), TN/AG/W/l.Rev. 1, March 18, 2003.time, could also lead to the same kind of inefficiencies in agriculture as have undermined competitiveness of many developing country industries nurtured behind high protective barriers.
Agreement will be needed on thresholds or criteria to determine to which countries the above type of provisions will apply.In our view it is important that criteria include administrative capacity and an indicator of poverty.As noted above, a simple measure that could be used is a combination of national GDP per capita, size and human development.One option would be to use a rule-of-thumb approach, as proposed above for resource-intensive disciplines. 2 5 As far as the TRIPS agreement is concerned, the most appropriate level of intellectual property protection varies by income level (Grossman and Lai, 2002;Lai and Qiu, 2003).Thus, low-income countries that are less likely to benefit from domestic innovation will benefit from staging the implementation of the agreement.Clearly a key issue is to resolve the difference in views among members regarding the ability of countries without domestic supply capacity to license firms abroad to provide drugs if this is deemed necessary to attain public health objectives.In addition, the TRIPS agreement affects the supply of international collective goods in several areas and is weak in reflecting developing country (and global) interests.Possible remedies to this situation include: * Ensuring that TRIPS is interpreted and implemented in a manner so as to ensure that governments' rights to protect public health and ensure access to medicines is not impeded; * Providing for the development of suitable contracts that balance private interests and public objectives in the area of extracting biogenetic resources from developing countries; * Providing for the establishment of new forms of IPRs over collective and traditional knowledge; and * Permitting the reconciliation of possible conflicts between a global IPR system enforced by the TRIPS agreement and public interest in resource conservation and biodiversity.
In these cases, and more generally, more information and analysis of the costs and benefits of alternative rules, and the distribution of these costs and benefits, is also necessary.There is often substantial uncertainty regarding both dimensions.Given substantially weaker social safety nets and insurance mechanisms, and the high rates of poverty in developing countries that entail much greater vulnerability to negative shocks, there is a clear need for more attention and resources to be devoted to a costing out of implementation requirements and calculation of cost-benefit ratios.The same applies to monitoring of outcomes, to allow for policies to be adjusted if necessary.
An important issue that needs to determined is how much can and should be done up front, as a 'pre-condition', and how much of these types of changes will need to be negotiated.Clearly some of what is proposed can and should be implemented without reciprocal concessions-e.g., ensuring that TRIPS is interpreted so as to allow public health concerns to be addressed by governments, and developing effective monitoring of actions by developed countries to implement provisions relating to technical assistance, technology transfer, de minimis thresholds, etc.However, most substantive changes to rules will need to be pursued in the context of negotiations.

III. Moving Forward: Beyond the SDT Review
The SDT discussion in the WTO has focused on specific proposals relating to existing agreements and provisions.There has also been a discussion of a possible new 'Framework Agreement' and a Monitoring Mechanism to track delivery of SDT and its effectiveness.to address balance of payments difficulties which are incompatible with good development practice.We also feel that some of the proposals in Category 3-e.g., the proposal to exempt LDCs from the TRIMS agreement, or the suggestion that LDCs, "notwithstanding any provision of any WTO Agreement, shall not be required to implement or comply with obligations that are prejudicial to their individual development needs..."28 are not likely to be beneficial to developing countries.Nonetheless, we agree with the overall approach to address specific SDT proposals as part of ongoing negotiations on these subjects-this is consistent with our suggestion to adopt a renegotiation approach to rule-related issues.It can be argued that developing countries should not be expected to pay again for improved rules; and indeed in some cases it may be possible to develop interpretations of the rules which are favorable to development without formal renegotiation of the agreement.An obvious example is what is being considered in the context of TRIPS and health.While this approach should be used as much as possible, we suspect that without active engagement in defining and defending specific rule changes, the objective of making the rules more development-friendly will not be realized.We therefore support the suggestion to (re-)negotiate when it comes to the rules of individual agreements.
What the current approach to SDT, including the procedural proposal by the Chair of the General Council, does not do is go beyond Doha Ministerial mandate on SDT.
While this is understandable, in our view it is very important that a clear decision be taken in Cancun to re-think the framework for SDT in the WTO, especially when it comes to determining the reach of the rules.To a large extent this is as much a forwardlooking issue as it is a backward-looking one (dealing with existing rules).The suggestion by the Chair to address most of the substantive SDT proposals in specific negotiating groups, if adopted, makes sense in terms of allowing countries to re-negotiate existing rules, but could also result in a defacto choice to pursue an agreement-byagreement approach to SDT.We argued previously that there are downsides to making eligibility for SDT within agreements negotiable, as this could give rise to substantial transactions costs and uncertainty.It is important that transparency and predictability is preserved, to avoid wasteful strategic behavior and to target SDT to those countries that are most in need of it.In determining SDT eligibility, non-negotiability once a deal has been reached is therefore of great importance.A major contribution of the WTO to world welfare is that it promotes the transparency and predictability.These are essential for both producers and users of internationally traded goods and underpin the investments that ultimately help to raise living standards.Thus, our preference is that the WTO specify broad criteria for access to SDT.If these prove unduly burdensome in exceptional cases-as may be the case for very small economies-these can be handled by an appeal to the Committee on Trade and Development and ultimately requests for waivers (based on existing WTO rules). 29  These types of issues are best addressed in the context of designing a new framework agreement for SDT-the cross-cutting approach that is not the focus of current deliberations.What the precise approach should be requires considerable thought, consultation and debate. 30Realism suggests that it will be very difficult to come to closure on the complex issues associated with differentiation before the next WTO ministerial meeting in September 2003.What can be sought, however, is agreement in principle to move in the directions advocated above, not least because that would likely facilitate progress on new subjects (e.g., the Singapore issues).A first step could be to establish a broad-based, high-level group operating under the auspices of the WTO General Council to explore different options, possible mechanisms and details of an alternative approach, including establishing criteria to determine which rules are resource-intensive in implementation, with recommendations to be made before the end of the Doha Round.The terms of reference of such a working group should be relatively broad.We would encourage the membership of any group that is established to pursue such a work program to go beyond the community of trade officials and include both 29 that this approach is very conservative in tenrs of implied changes in terms of groups of countries.To a large extent the approach would be consistent with existing groups identified within the WTO -LDCs for deep preferential access, current recipients of GSP, which is voluntary and defined by developed countries individually, and technical assistance and self-declared developing countries for advantageous treatment under safeguards etc.Indeed, only one new category is called for in the WTO: a set of countries that should have the right not to implement resource-intensive agreements (and that should also benefit from proposed changes in the Agreement on Agriculture, TRIPS, etc.), comprising poor and disadvantaged countries (LDCs and similarly poor states) supplemented by very small countries where implementation of agreements requiring substantial up front costs and scale economies is likely to be excessively costly and that are generally not diversified.30 See Stevens (2002) and Page (2002) for discussions of options and issues in this connection.national economic policymakers and representatives of the international development community.

IV. Concluding Remarks
The traditional approach to SDT in the GATT/WTO has not been a success in promoting development.Indeed, a good case can be made that the approach is fundamentally flawed in that it helped create incentives for developing countries not to engage in the process of reciprocal liberalization of trade barriers and the rule-making process.There is a need to go beyond the traditional SDT debate if Doha is to make progress on helping developing countries to use trade for development.
First and foremost is to improve access to markets-both developed and developing.This can do much to help achieve the income MDG target.In our view market access should be pursued on a MFN basis and rely on the mechanics of reciprocity, with the major element of SDT comprising an acceptance on the part of the major WTO members (large markets) to eliminate tariffs and other trade barriers on the goods and services in which developing countries have a comparative advantage, and negotiating modalities that give 'credit' for past autonomous trade reforms.
Second, there is clearly a need to "get the rules right" from a development perspective, which will require the re-opening of certain existing agreements.The heart of the SDT issue revolves around the need to recognize that one size does not fit all when it comes to regulatory disciplines and the 'behind the border' policy agenda that is increasingly being pursued in the WTO.Here there is a clear need for differentiation, both in terms of negotiating mechanics-should reciprocity extend to trading 'apples for oranges', e.g., market access for goods in return for rules on domestic policies?-and the reach of disciplines across countries.Country differentiation requires agreement on the criteria used to define eligibility for SDT.This has for a long.timebeen a non-starter in the WTO, with the result that SDT provisions have not been very effective.Indeed, the experience to date suggests that the depth of the differential treatment granted will be inversely related to the number of eligible countries.Thus, eligibility for SDT should be restricted to fewer WTO member countries than is currently the case under the selfdeclaration approach that is used to identify developing countries.
In practice, SDT and 'development relevance of WTO rules' are often conflated in discussions, with calls for specific types of SDT essentially being motivated by a perception that a certain rule is 'anti-development' and that developing countries should be exempted from it.In our view the appropriate solution to such problems is to renegotiate the rules rather than seeking an opt-out.Indeed, defining the negotiating set is an urgent matter that should be decided by Ministers.
In order to assist developing countries to benefit from market access opportunities a significant increase is needed in technical and financial assistance to expand supply capacity and improve the investment climate in low income countries.The need for this is acute in absolute terms, but is made even stronger as the trading system moves in the direction of lower MFN trade barriers and the consequent erosion of preferences for those countries that currently benefit from effective preferential access.What is required is a de-linking of development assistance from trade policy-a shift from the current strategy of permitting a small subset of countries to benefit from the large distortions created by developed countries on their markets, to one that puts the emphasis on direct support to expand trade capacity and improve performance.
governments and the donor community.The development community made commitments to this effect at the International Conference on Financing for Development in Monterrey in March 2002-what is needed is an equally clear articulation of trade-related demands by developing countries.Development assistance is (and should be)