[*PG371]POVERTY AND SOCIAL ANALYSIS OF TRADE AGREEMENTS: A MORE
COHERENT APPROACH?
Abstract: The collapse of trade negotiations in Canc�n in September 2003 shook confidence in both the Doha Development Agenda and the commitments of industrialized countries and international economic institutions to pursue “coherent” trade and development policies. This Article critically examines the dual commitments of development institutions, especially the World Bank, to trade liberalization and poverty reduction, and the challenges to achieving “policy coherence” through trade “mainstreaming,” “capacity building,” and “impact assessments.” In particular, the Article considers the feasibility of adapting existing tools for poverty and social analysis to assess trade policies and agreements. The Article uses gender as one possible lens through which to analyze the potential impacts of trade policies on vulnerable groups in developing countries. While recognizing their limitations, the Article supports the development of practical tools for poverty and social analysis for use by government trade offices and other ministries, development institutions, research institutes, and civil society groups in developing countries. However, for these tools to be useful, developing countries must have the “policy space” to choose the trade policy options that best support their poverty reduction strategies and broad development goals.
The Fourth Ministerial Conference of the World Trade Organization (WTO) in Doha in November 2001, including the adoption of a Work Programme now referred to as the “Doha Development Agenda” [*PG372](DDA), recast the built-in agenda and new activities of the WTO in terms of developing countries and their concerns in the global trading system. The DDA also extended the principle of “coherence in global economic policy-making,” which is embedded in the Agreement Establishing the WTO, to include human development concerns such as public health. However, the breakdown of trade negotiations at the Fifth Ministerial Conference in Canc�n in September 2003 has shaken confidence in both the DDA and the broader notion of coherence in international policy. In particular, the unwillingness of major industrialized countries to dismantle their agricultural subsidies highlights the incoherence between their trade policies and international development commitments. Not surprisingly, similar tensions can be found in the policy frameworks and activities of international organizations over which these countries have significant influence, particularly the International Monetary Fund (IMF), the World Bank, and regional development banks, as well as the countries’ own aid agencies.
The DDA suggests several related channels through which countries and development institutions can pursue coherent approaches to trade and development. These include support for (1) “mainstreaming” trade into the national development plans and poverty reduction strategies of developing countries, (2) trade-related “capacity building” in these countries, and (3) assessing the development impact of particular trade policies, especially on poor and vulnerable groups. Part I of this Article reviews the DDA in the context of general trends in international development assistance. Part II critically examines the dual commitments of development institutions, especially the World Bank and other multilateral development banks (MDBs), to trade liberalization and poverty reduction, and the challenges to achieving “policy coherence” through trade “mainstreaming,” “capacity building,” and “impact assessments.” Part III considers the feasibility of adapting existing tools for poverty and social analysis to assess trade policies and rules. Finally, Part IV uses gender as one possible lens through which to analyze trade policies and agreements and their possible impacts on vulnerable groups in developing countries.
This Article argues that a comprehensive poverty and social analysis of trade policy should consider social, economic, and legal/regulatory factors. Legal/regulatory analysis can also help to identify areas of tension between international economic law, on the one hand, and other legal frameworks that promote development goals, such as gender equality, on the other. The development of tools for poverty and social analysis of trade policies and rules can improve the quality of development agencies’ advice and assistance to developing [*PG373]countries. More importantly, these tools can be used by trade negotiators, other government officials, research institutes, and civil society groups in developing countries to inform national debates about the proper role of trade in national development strategies.
Development concerns are not new to the WTO or the global trading system. Building on language in the General Agreement on Tariffs and Trade (GATT), the Agreement Establishing the WTO recognizes the shared goals of WTO members in “raising standards of living, ensuring full employment and a large and steadily growing volume of real income and effective demand, and expanding the production of and trade in goods and services, while allowing for the optimal use of the world’s resources in accordance with the objective of sustainable development,” as well as the “need for positive efforts designed to ensure that developing countries, and especially the least developed among them, secure a share in the growth in international trade commensurate with the needs of their economic development.”1 Although the concept of “development” and the ascribed categories of “developing country” and “least developed country” (LDC) remain contested,2 they are nonetheless woven into the fabric [*PG374]of the WTO. The development theme is most evident in provisions of various WTO agreements granting “special and differential” treatment to developing or least developed members,3 and in the operation of the WTO Committee on Trade and Development and Sub-Committee on Least-Developed Countries.
The characterization of the outcome of the Doha Ministerial Conference as a “Development Agenda” was not an afterthought. The failure to launch a new round of trade negotiations in Seattle in 1999 had been attributed in large part to the dissatisfaction of many developing countries with their experience in implementing the Uruguay Round agreements. Moreover, by 2001, the majority of WTO members were developing countries.4 It was apparent to all members and to the WTO Secretariat that further progress would not be made at Doha unless developing country concerns were addressed in some fashion. The dissatisfaction of developing countries with the operation of the WTO system had also found support among prominent economists,5 the World Bank and other development agencies,6 and a broad spectrum of civil society groups in both industrialized and developing nations.7
[*PG375] The Ministerial Declaration of the Doha Conference asserted that developing countries’ “needs and interests [would be] at the heart of the Work Programme” included in that document, and that “the marginalization of least-developed countries in international trade” would be addressed.8 Developing country concerns were reflected most prominently in the separately-adopted Declaration on the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) and Public Health9 and the decision on Implementation-Related Issues and Concerns.10 However, developing country interests were also mentioned in provisions of the main Ministerial Declaration relating to agriculture, services, market access, investment, competition policy, trade facilitation, subsidies and countervailing measures, environment, small economies, debt and finance, technology transfer, technical cooperation and capacity building, LDCs, and special and differential treatment.11
The initial reactions of WTO members to the DDA ranged from jubilant to disappointed, and commentators’ initial assessments were similarly mixed.12 One academic commentator found evidence in the DDA that the WTO might be starting to transform itself into an organization that respects distributive as well as efficiency values.13 An[*PG376]other was less optimistic, noting that the DDA had resolved only two substantive trade law issues of interest to developing countries, and that at Doha “the can was [simply] kicked down the road.”14
Subsequent events confirmed the skeptics’ fears. By fall of 2003, when the Fifth Ministerial Conference was to take place in Canc�n, little progress had been made on the Doha commitments of most interest to developing countries. The only tangible advance—a WTO General Council decision allowing some flexibility in the application of the compulsory licensing provision of the TRIPS Agreement—emerged only in late August 2003, and after months of protracted debate and negotiation.15 The Canc�n Ministerial Conference commenced on September 10, despite wide differences among country groups, especially on agriculture and the so-called “Singapore issues.” The chairperson closed the conference abruptly on September 14, when there appeared to be an impasse over the “Singapore issues,” and wide gaps remained in country positions on agriculture.16
From a developing country perspective, however, Canc�n fundamentally changed the dynamics of multilateral trade relations. For the first time, several negotiating blocks of developing countries emerged—notably, the so-called Group of 20 and the Alliance for Special Products and Special Safeguard Mechanisms—and some of these coalitions continued to act in the informal trade discussions that resumed in late 2003.17 In addition, despite the breakdown of negotiations in Canc�n, there is little question that trade discourse continues to be colored by the development perspective introduced at Doha. Regardless of their underlying interests, trade negotiators and other interested parties are now far more likely to couch their proposals and responses in terms of the likely benefits or harms to developing countries.
The international financial institutions, including the IMF and MDBs, as well as a number of United Nations (U.N.) bodies and agencies and bilateral aid agencies, have been involved in trade promotion or trade liberalization for some time. Indeed, the U.N. Conference on Trade and Development (UNCTAD) was established in 1964 as the principal organ of the U.N. General Assembly dealing with trade, investment, and development issues, and as the U.N. focal point for LDCs.18 The IMF’s trade-related activities are somewhat specialized and relate to both its surveillance of member countries’ exchange rate policies and its extension of balance-of-payments support through credits and loans.19 The trade-related activities of the MDBs, the U.N. Development Programme (UNDP), and bilateral aid agencies are more closely related, although they reflect different modalities of assistance (in particular, the MDBs provide loans as well as grants) and have evolved significantly over time.
The Organization for Economic Co-operation and Development (OECD) has identified four phases or trends in trade-related development assistance:
Not surprisingly, the trends in trade-related development assistance reflect general trends in development assistance. This is particularly evident in the World Bank’s trade-related lending in the 1980s and 1990s, which included not only investment loans to build facilities, such as ports and export corridors, and to support export industries, but also a significant number of structural adjustment loans that included trade liberalization conditions. By its own count, between 1981 and 1994, the World Bank supported trade or foreign exchange policy [*PG378]reforms through 238 loans.21 (This lending preceded the conclusion of the Uruguay Round and resulted in much of the “autonomous liberalization” that developing countries have argued they should be given credit for in current WTO negotiations.) From 1994 to 1999, World Bank investment lending for trade-related activities represented about 26% of its total lending. Between 1990 and 1998, 68% of the World Bank’s adjustment loans included support for trade and exchange rate policy reforms.22
The commitment of WTO members at Doha to prioritize developing country needs and interests in the implementation of the Uruguay Round agreements and new trade negotiations is understandable in light of other recent international commitments and trends in development assistance. In particular, the Millennium Declaration, adopted by the U.N. General Assembly in September 2000, asserted that “the central challenge we face today is to ensure that globalization becomes a positive force for all the world’s people. For while globalization offers great opportunities, at present its benefits are very unevenly shared, while its costs are unevenly distributed.”23 In discussing development, the Millennium Declaration emphasized the need to address “the abject and dehumanizing conditions of extreme poverty, to which more than a billion [people] are currently subjected,” and resolved “to create an environment—at the national and global levels alike—which is conducive to development and to the elimination of poverty.”24 The Millennium Declaration went on to commit all members of the U.N. to several time-bound and continuing goals, now referred to as the Millennium Development Goals (MDGs). These include: reducing by half the proportion of the world’s people living on less than one dollar a day; ensuring universal access to primary school and equal access of girls and boys to all levels of education; reducing maternal mortality by three-quarters and under-five child mortality by two-thirds; and halting and reversing the spread of HIV/AIDS, malaria, and other widespread diseases, all by 2015.25 Continuing goals include:
Among other things, the Millennium Declaration laid the groundwork for the Declaration on the TRIPS Agreement and Public Health adopted by WTO members in Doha.
Less than a year after Doha, the international community convened again for the International Conference on Financing for Development in Monterrey. The Monterrey Consensus adopted at the conference confirmed that the international community would give priority to “[m]obilizing and increasing the effective use of financial resources and achieving the national and international economic conditions needed to fulfill internationally agreed development goals, including those contained in the Millennium Declaration, to eliminate poverty, improve social conditions and raise living standards, and protect our environment.”27 The Monterrey Consensus reaffirmed U.N. members’ commitment to trade liberalization, but noted that developing countries need “appropriate institutions and policies” to benefit fully from trade, and welcomed the emphasis in the DDA on giving priority to developing countries’ needs and interests.28
The Monterrey Consensus committed U.N. members to “improve [their] domestic policy coherence through the continued engagement of [their] ministries of development, finance, trade and foreign affairs, as well as . . . central banks.”29 To complement national devel[*PG380]opment efforts, the Consensus also recognized “the urgent need to enhance coherence, governance, and consistency of the international monetary, financial and trading systems” and encouraged “policy and programme coordination of international institutions and coherence at the operational and international levels to meet the Millennium Declaration development goals of sustained economic growth, poverty eradication and sustainable development.”30 Turning to the role of the multilateral financial institutions, the Consensus stressed the need
in providing policy advice and financial support, to work on the basis of sound, nationally owned paths of reform that take into account the needs of the poor and efforts to reduce poverty, and to pay due regard to the special needs and implementing capacities of developing countries and countries with economies in transition, aiming at economic growth and sustainable development. The advice should take into account social costs of adjustment programmes, which should be designed to minimize negative impact on the vulnerable segments of society.31
Even before the adoption of the Millennium Declaration and MDGs, the World Bank and most regional development banks, as well as a number of bilateral aid agencies, had already redefined their missions in terms of poverty reduction. The World Bank’s reorientation toward poverty reduction coincided with the launch of the Heavily Indebted Poor Countries Initiative (HIPC) by the World Bank and the IMF.32 This initiative came about in response to a well-orchestrated civil society campaign for meaningful external debt relief for highly-indebted countries, particularly in Africa, intended to free up scarce financial resources to address HIV/AIDS and other critical social development needs.33 A key element of HIPC was the requirement that HIPC countries prepare comprehensive Poverty Reduction Strategy Papers (PRSPs) to ensure that the resources freed up through debt [*PG381]relief would be used effectively for poverty reduction.34 The World Bank’s recommitment to poverty reduction also coincided with increased attention to the social impacts of World Bank-financed structural adjustment programs (and macroeconomic stabilization programs financed by the IMF) and greater effort to mitigate negative impacts of these programs on vulnerable groups through the protection of social expenditures and the establishment of social funds.35 This shift in adjustment lending also responded to widespread criticism of the IMF’s and World Bank’s stabilization and adjustment programs from member countries, other international agencies, and civil society in both industrialized and developing countries.36
Among other concerns, external critics of structural adjustment lending have focused on trade liberalization measures included in many of the World Bank’s adjustment loans. For example, the Structural Adjustment Participatory Review International Network (SAPRIN), a global network of civil society groups and researchers, recently completed a four-year, multi-country study of the effects of specific structural adjustment policies on a range of economic sectors and social groups.37 The study was funded by the European Union and several European governments, the UNDP, and others, and it involved collaboration with World Bank staff and government officials, as well as a wide range of non-government stakeholders in each country. Some key conclusions of the study related to trade liberalization were:
Based on these findings, the SAPRIN review recommended that
[t]rade-reform measures should take into account the conditions of domestic producers and should be paced and sequenced in order to create a level playing field that will help to stimulate local production and ensure that domestic enterprises can face competition from foreign goods. . . . Trade reform should be nuanced rather than indiscriminate [and] future reform processes should be designed by governments with the participation of a wide range of sectors and population groups to ensure that policies are consistent with national development aspirations.39
Another recent critique of development bank approaches to trade policy, which focused on the particular challenges facing LDCs, came from UNCTAD in its Least Developed Countries Report 2002. In this case, the critique also extended to the World Bank’s (and IMF’s) cur[*PG383]rent approach to poverty reduction through PRSPs. In summary, the UNCTAD report argued that:
UNCTAD’s main criticism of the World Bank/IMF approach to PRSPs for LDCs was that the poverty reduction strategies prepared thus far have tended to be oriented toward adjustment rather than development. From this perspective, pro-poor public expenditure management techniques have essentially been grafted onto macroeconomic policies and structural reform programs that are substantially similar to the policies the World Bank has promoted since the 1980s through structural adjustment loans. UNCTAD found that these policies generally had not reduced the incidence of poverty in LDCs and concluded that structural adjustment policies are not the appropriate tools for achieving pro-poor growth. Instead, UNCTAD recommended that LDC governments be given leeway to devise more “development-oriented” strategies that focus on improving supply capabilities and productive capacities, establishing a dynamic investment-export nexus, and preventing marginalization of particular groups and regions through a variety of policies.41
[*PG384] Although the SAPRIN review and UNCTAD report used different methodologies and focused on different sets of developing countries, they both highlighted the need for external advisors to take into account a country’s particular constraints and other characteristics in recommending trade policies to promote economic growth and reduce poverty. This has also been the strong recommendation of economists, such as Dani Rodrik, who have stressed the limits of trade policy; the variety of strategies that successful economies have used to open themselves to trade over time; and the need to develop resilient domestic institutions to withstand the external shocks that come with trade openness.42 A recent study on trade and sustainable human development, supported by UNDP and other organizations, and a recent publication by Oxfam International on trade and poverty, echo many of the same themes.43
The outcome documents from the Doha Ministerial Conference suggested three related channels through which development banks and agencies could support the DDA under the general framework of promoting “coherence in global economic policy-making.”44 First, the Ministerial Declaration confirmed that “technical cooperation and capacity building are core elements of the development dimension of the multilateral trading system” and called for “effective coordinated delivery of technical assistance” to developing countries, including relevant international and regional institutions, the OECD Development Assistance Committee, and bilateral donors.45 This could include enhanced use of the Integrated Framework for Trade-Related Technical Assistance to Least-Developed Countries (Integrated Framework), which [*PG385]was established in 1997 and has involved the IMF, International Trade Center, UNDP, UNCTAD, World Bank, and WTO.46 Second, within the provisions on improved technical cooperation and capacity building, the Doha Declaration endorsed efforts to “mainstream[] trade in national plans for economic development and strategies for poverty reduction.”47 This clearly implicates the World Bank in its work with developing countries on PRSPs but is also relevant to all development banks and agencies in terms of their general support to developing countries’ national development plans. Third, the Doha Declaration supported initiatives by WTO members and international organizations to assess the impact of trade policies, for example, from an environmental perspective or as provided in the guidelines for further negotiations under the General Agreement on Trade in Services.48
The concept of policy coherence has generated considerable interest at the international level in recent years. The OECD coined the term “policy coherence for development” in the early 1990s and has defined it as “the systematic promotion of mutually reinforcing policy actions” aimed at “taking account of the needs and interests of developing countries in the evolution of the global economy.”49 Policy coherence was a fundamental concern of trade negotiators during the Uruguay Round and prompted the drafting of a specific ministerial declaration to promote “greater coherence in global economic policymaking.” The declaration noted the importance of achieving harmony between “structural, macroeconomic, trade, financial and development [*PG386]aspects of economic policymaking,” at both national and international levels, and specifically authorized the Director-General of the WTO to pursue cooperation with the heads of the IMF and World Bank.50 At the recent International Conference on Financing for Development in Monterrey, the concept was extended to encompass “coherence and consistency of the international monetary, financial and trading systems in support of development” and specifically to support the MDGs.51 Policy coherence is also the general theme of the eleventh UNCTAD session that will be held in June 2004 in Sao Paolo. The focus of UNCTAD XI is on “enhancing coherence between national devel[*PG387]opment strategies and global economic processes towards economic growth and development, particularly of developing countries.”52
The OECD points out that “[c]oherence has always been . . . a function of competing and conflicting interests and values” in which “stakeholder power and political will” play important parts.53 The likelihood of achieving coherence between different policies can turn on several factors, including the similarity of the policies; the range of interests and stakeholders involved; the level(s) of governance that are engaged (which could include international, regional, national, and community actors); and the types of cooperation or other action involved (which might range from information-sharing and consultations to joint research and co-funding of specific projects). There is a general concern—for example, among development agencies—that efforts to improve policy coherence might result in a homogeneity of approaches.54 Another general concern is that seemingly coherent policies may mask very inconsistent approaches in their implementation.
In the area of trade and development policies, the coherence issues are wide-ranging. For example, some of the substantive concerns about WTO rules from a development perspective relate to the impact of industrialized countries’ agricultural subsidies on farmers in developing countries; the impact of high import tariffs on garments and other products made in developing countries; the tension between the protection of intellectual property and the sharing of technology with developing countries; and the impact of tariff reductions on national budgets, particularly on social expenditures such as public health and education. Other aspects of the international economic system that have also prompted concern include the impact of volatile exchange rates and world commodity prices on developing countries, especially LDCs; the links between high levels of external debt and promotion of exports; the links between trade liberalization and de[*PG388]velopment finance, reflected in trade-related conditions in IMF and development bank loans; and the limited participation of developing country governments and their constituencies in international economic decision-making.55
The international focus on policy coherence has also attracted criticism. Douglas Hellinger, coordinator of the SAPRIN study, has dismissed the notion of “coherence in international monetary, financial and trading systems to support development” as “an oxymoron.”56 He argues instead for “a new coherence, a convergence around a different set of values and principles—namely respect for diversity, equity, self-reliance, self-determination, community and environmental sustainability, democratic economic decisionmaking, transparency and accountability.”57 The economist Joseph Stiglitz and others have also criticized the dominance of the international economic system by the IMF, World Bank, and WTO and have proposed that the U.N. should play a greater role in international financial issues through a new “economic and social security council.”58 Several civil society actors have voiced similar concerns about the role of the IMF as the “gatekeeper” for financial assistance to developing countries and the dominance of the World Bank in country-level development processes such as the formulation of PRSPs, diagnostic trade studies, and trade-related technical assistance.59
[*PG389] The international debate on policy coherence has strong parallels with the academic discourse on “linkages” between international trade law and other international law areas such as the environment, labor, and human rights.60 For example, many of the concerns about the dominance of the IMF, World Bank, and WTO in the international economic system (relative to the U.N. and its agencies) resonate with concerns expressed about the power of international trade rules that are enforceable through trade sanctions (compared with the weak enforcement mechanisms in international labor standards and human rights treaties). Just as the IMF has been criticized for misdiagnosing recent financial crises and prescribing policy measures that actually exacerbated the crises,61 international trade law has been roundly criticized for relying on outdated assumptions about the global economy and for even violating basic principles of trade economics (for example, in the area of subsidies).62
The next section examines the coherence of trade and development policies within one institution, the World Bank.
The goal of policy coherence, even within the same organization, is far easier to articulate than to implement. This is especially true for a large organization such as the World Bank, which has an extremely broad agenda, including multiple policy goals that reflect different analytical and ideological perspectives. Trade liberalization and poverty reduction are clear examples of the “policy tension” that can result. This tension can operate on several levels. Whereas trade liberalization is part of the neoliberal economic policy framework [*PG390]commonly referred to as the “Washington Consensus,”63 which the IMF and MDBs have pursued since the 1980s, current approaches to poverty reduction—which emphasize its multidimensionality, including elements of vulnerability and disempowerment—have their roots in alternative development discourses, such as UNDP’s “human development” framework,64 Amartya Sen’s “capabilities” approach,65 and the participatory methods of Robert Chambers66 and others.67 The two policy goals are also associated with very different methodologies for research and analysis: while trade effects are typically analyzed through economic models using macroeconomic data, poverty analysis relies increasingly on a combination of quantitative and qualitative measures (including participatory assessments) carried out at the community, household, and individual levels.68 In terms of policy development, there are also “cultural” differences: while “trade culture” is commonly viewed as “legalistic and highly centralised,” “development culture” aspires to be “decentralised, demand driven, and based on a country-owned process.”69 Moreover, even among macroeconomists, there are ongoing debates about the relationships among trade [*PG391]openness, economic growth, and poverty reduction that have significant implications for development policy.70
Having adopted poverty reduction as its overall mission, the World Bank has moved poverty reduction to the top of its policy hierarchy. Therefore, one would expect poverty concerns to temper and shape the World Bank’s current approach to trade. The principles of the PRSP process provide one possible pro-poor framework for trade policy advice and support to developing countries.71 These principles [*PG392]suggest that any trade policy recommendations to a developing country should be tailored to the country’s specific context and should directly support its poverty reduction strategy and commitments toward the MDGs. Further, the recommendations should be based on an analysis of various trade policy options and an assessment of the likely impacts—positive and negative—of these options on poor people and other vulnerable groups. Trade policy options should also be evaluated in consultation with a broad range of stakeholders, including non-trade ministries and representatives of civil society and private sector groups. Finally, any trade policy recommendation should allow flexibility to adjust the policy to address any negative impacts that might arise in implementation.
Recent World Bank publications and other documents suggest that there is a framework in place to link trade policy to poverty reduction. In general, however, trade liberalization is still seen as a development goal that is parallel rather than subordinate to poverty reduction. This reflects the general belief of World Bank economists that trade liberalization spurs economic growth, and therefore—indirectly, at least—increased trade should support poverty reduction. The World Bank’s approach to trade policy advice does not adhere consistently to general PRSP principles, although some work is starting to be done to at least analyze the distributional effect of some trade reforms.
The Global Poverty Report 2001, which was prepared by the IMF, World Bank, and four regional development banks for the G8 Summit in July 2001, set out an analytical framework to guide the IMF’s and MDBs’ work related to trade and poverty.72 The report traced several channels through which increased trade openness could affect an economy, particularly including its poor people. These channels include: (1) effects on the prices of goods and services that the poor con[*PG393]sume and produce; (2) effects on the demand for and returns to factors of production that the poor can offer, such as labor; (3) effects on government revenue and resources available for anti-poverty programs; (4) the potential for economic growth; and (5) transition costs, including the possibility of increased volatility of growth, resulting in the need for social protection mechanisms. The report concluded that “comprehensive trade reform can help reduce poverty when it is part of a set of reforms that improve the domestic macroeconomic and investment climate, enhance infrastructure and technology, and contribute to the provision of knowledge and skills.”73 However, the report cautioned that “these effects vary significantly across countries, regions, and groups within countries, which makes it difficult to generalize about the effects of trade liberalization on poverty.”74
In terms of recommendations, the Global Poverty Report 2001 called on developed countries to remove protective barriers on imports of agricultural and manufactured goods, reduce subsidies, and increase market access for goods from developing countries. The report encouraged developing countries to integrate trade reform in their broader development agendas, together with taking measures to increase the asset base of poor people (including education, health, and land), to make markets work better for poor people, and to reduce social barriers that disadvantage women and certain ethnic or racial groups. The report also called for “a careful sequence and mix of reductions of border barriers and domestic reforms so as to maximize the growth potential from trade liberalization while minimizing its adverse impact on poverty.”75
The Global Poverty Report 2001 did not address all of the criticisms of past World Bank support for trade liberalization—for example, UNCTAD’s concern about the structural problems of commodity-exporting LDCs—but it nevertheless provided a basic framework for examining the possible impacts of trade policy changes on poor people. However, subsequent World Bank research reports have been much more upbeat about the benefits of increased trade for developing countries, although they have criticized some trade agreements and practices and have shown increasing concern about the implementation costs of trade agreements for developing countries.
[*PG394] In Global Economic Prospects 2004, for example, the World Bank’s research economists marshaled statistics and arguments in support of a “pro-poor Doha outcome,” including further reductions in tariffs and agricultural subsidies and further liberalization of services (including labor services). The report provides compelling statistics on the magnitude and negative development impact of agricultural subsidies in industrialized countries, as well as statistics on the generally higher tariffs these countries impose on imports from developing countries (compared with imports from other industrialized countries). In addition, the report acknowledges the industrialized-country bias in WTO agreements and considers the possibility of renegotiating some of the agreements—for example, the Agreement on Agriculture and the TRIPS Agreement—to address the resource and capacity constraints of developing countries. The report also suggests that “more attention and resources should be devoted to costing out the implementation requirements of proposed rules and to calculating their costs and benefits.”76 The last recommendation echoes earlier findings by World Bank economists on the substantial implementation costs of some Uruguay Round agreements.77 The report therefore questions the wisdom of introducing new uniform trade rules on “behind-the-border” issues that would require substantial changes in domestic regulatory schemes. In the area of tariff reform, however, the report argues that both “[d]eveloped and developing countries alike could affirm their commitment to poverty reduction by accepting an ambitious program of liberalization.”78
[*PG395] Other recent World Bank publications promote a similarly optimistic view of the benefits of trade liberalization for developing countries, in which adjustment costs are relatively small and short-term and can be addressed through compensatory programs carried out under the country’s poverty reduction or general development strategies. This optimism flows through the trade chapter of the World Bank’s Poverty Reduction Strategy Sourcebook,79 which one commentator characterizes as reflecting “a one-size-fits-all mentality” that is “based more on faith than evidence.”80 The optimism is also reflected in the handbook that the World Bank recently published to help developing countries participate in the trade negotiations under the DDA.81 Notably, only one of the fifty-five chapters directly addresses the relation between trade policies and poverty reduction. The implication is that, rather than being one of several policy tools that developing country governments may use to reduce poverty, trade liberalization is a goal to be pursued in any case, and any social costs associated with it can simply be mitigated through supplemental, “poverty-reduction” schemes (instead of providing a basis for reexamining the trade policy itself).
The World Bank’s pro-trade agenda is most evident in the 2003 Progress Report of its newly established Trade Department. While this report mentions the need for “greater attention towards the distributional impact of trade and trade reforms,” the main thrust of the agenda outlined in the report is to (1) support the Doha process, (2) support regional integration initiatives, and (3) “promote trade integration as a core element of country development strategies.”82 The implications of this agenda for the PRSP process and other country-level activities are considered in the following subsections.
Although “mainstreaming” is a familiar concept in development practice, “trade mainstreaming” is a relatively new variation. As used by the WTO, World Bank, and other core agencies in the Integrated Framework, it refers to the integration of trade into country development plans and poverty reduction strategies.83 Trade mainstreaming was first adopted as an outcome of a review of the Integrated Framework, but it now has broader relevance under the DDA.84 In principle, trade mainstreaming should further the goal of policy coherence by incorporating trade policy in a country’s overall development framework and ensuring that it complements the country’s other economic and social priorities.
There is a risk, however, that trade policy analysis and recommendations could simply be inserted in a national development plan or poverty reduction strategy without any effort made to ensure that the trade policy furthers the country’s poverty reduction goals. This possibility has concerned some international agencies and civil society groups who fear that “the Bank’s lead role in mainstreaming trade into PRSPs . . . offers the opportunity to lock the trade liberalization agenda into national development plans of the most vulnerable countries.”85 The concern was heightened by the proposal to add “trade integration strategy chapters” to PRSPs and to have these chapters be based on “diagnostic trade integration studies,” such as those that have been under[*PG397]taken on a pilot basis under the Integrated Framework.86 The primary purpose of these studies is to identify opportunities for trade liberalization and export promotion, not to identify opportunities to reduce poverty through trade policy. The studies are technical exercises carried out by consultant teams rather than as part of the consultative processes carried out for the rest of the PRSPs. Therefore, they are not appropriate in their current form simply to “fold into” the PRSPs. What is needed is for the key PRSP principles—including country specificity and ownership, comprehensiveness, and broad stakeholder participation—and a thorough poverty impact analysis to be “mainstreamed” into the development of any new trade policy.
The World Bank’s Trade Progress Report acknowledges that relatively little attention has been paid to trade issues in PRSPs to date but predicts that upcoming PRSPs will include more discussion of trade issues as part of “an increasing focus on growth.”87 However, the report also notes that “mainstreaming trade into PRSPs . . . of the countries that are now not actively interested in trade is likely to take time. Furthermore, the trade agenda has to compete for policy makers’ attention with other immediate concerns” such as HIV/AIDS, excessive debt, and fiscal crises.88 This raises serious questions about country ownership of the trade agenda, as well as the general PRSP process, if World Bank teams are pressing developing country governments to address trade issues in their PRSPs, even though the countries themselves have other, more urgent development priorities. Notably, recent World Bank documents now refer to the “embedding” of trade policies in PRSPs and national development strategies.89 This terminology further reinforces the impression that trade policies can be developed in isolation from other development priorities, poverty and social considerations, and consultative processes, and can then simply be inserted in national policy documents.
“Trade capacity building” presents similar opportunities and risks. Unlike trade mainstreaming, which is a relatively new and untested concept, developing countries have had substantial experience as recipients of trade-related technical assistance (or technical coop[*PG398]eration).90 Unfortunately, the experience has been quite mixed. A proliferation of international, regional, and bilateral agencies has been involved in trade-related assistance to developing countries,91 leading inevitably to inconsistency and duplication. There has also been a lack of clarity about the meaning of trade-related capacity building or technical assistance/cooperation. Since the 1970s, international, regional, and bilateral agencies have provided several types of trade-related support to developing countries in the areas of export marketing, trade liberalization, trade facilitation, and “trade capacity development,” which the OECD defines as “building capacity by facilitating a country-driven participatory trade policy process as part of a comprehensive approach to overall development goals and poverty reduction strategies.”92
Trade-related technical assistance within the WTO initially focused on assisting developing country officials to understand the nature of their countries’ multilateral trade commitments and how to comply with them. However, one developing country commentator has noted that the short courses provided by the WTO were quite general and did not take account of developing countries’ constraints in terms of specialized staff and resources.93 Other international and bilateral agencies have financed studies related to export promotion, but their recommendations did not take account of “supply con[*PG399]straints” within developing countries or the effects of nontariff barriers maintained by industrialized countries.94 There has also been a general perception that trade-related technical assistance, particularly if provided by other countries, is not “value-free” but is in fact “deeply political” and tends to further primarily the commercial and other interests of the donors.95
The Integrated Framework, established in 1997, was intended to improve the effectiveness of trade-related technical assistance to LDCs through the coordinated efforts of the six agencies involved. However, a review completed in 2000 found that the Integrated Framework had not functioned well due to (1) different perceptions of LDCs and donors about its objectives; (2) absence of linkages to overall development assistance processes; (3) assistance that was not sufficiently demand-driven; (4) weak administration; (5) insufficient coordination among agencies, LDCs, and donors; and (6) lack of adequate funding.96 Based on the review, the six core agencies approved a number of changes in the Integrated Framework to address these problems, including support for “the integration of trade, trade-related technical assistance, and capacity-building into the national development strategies and plans of LDCs,” to be led by the World Bank.97
Following the Fourth Ministerial Conference in Doha, the WTO adopted a new general strategy for technical cooperation that tracks both the Integrated Framework and the DDA in emphasizing technical support for “trade mainstreaming.”98 This mandate resulted in the creation of a new technical assistance fund—the Doha Development Agenda Global Trust Fund—and expanded commitments by international and bilateral agencies under the Integrated Framework and other technical assistance programs.99 However, developing countries and civil society groups expressed concern about the WTO’s initial plan for the new technical assistance fund, which prioritized areas that did not necessarily reflect developing country interests, including controversial “Singapore issues” such as investment, competition, and [*PG400]government procurement.100 Similar criticism was leveled at UNCTAD’s post-Doha program for technical assistance, which reportedly allocated 60% of funds to “Singapore issues,” including investment, competition, and trade facilitation.101
Since the World Bank has been designated to lead efforts to “mainstream” trade into PRSPs and other national development planning exercises, its trade capacity building activities under the DDA are likely to focus on integrating trade in the PRSP process. However, the type of technical support now contemplated—the drafting of “trade integration strategy” chapters for PRSPs based on “diagnostic trade integration studies”—is at odds with the principles and objectives of the PRSP process.102 Instead, the World Bank and other external agencies should support capacity building of government agencies, research institutes, and civil society groups to evaluate independently trade policy options and to assess their likely impacts—positive and negative—on poverty reduction, other national development goals, and development budgets.103
As noted earlier, the World Bank’s new trade agenda recognizes the need to analyze the distributional impact of trade and trade policies.104 This responds to recent calls from developing countries and civil society groups for systematic assessments of the development impact of new trade proposals within the WTO.105 It also reflects the [*PG401]general work already being done by the World Bank, IMF, and some bilateral donors to develop tools to assess the poverty and social impact of major economic policy changes (discussed in more detail in Part III below).
The World Bank’s recent Trade Progress Report indicates that the World Bank’s main vehicle for analyzing the potential impact of trade reforms on vulnerable groups will be diagnostic trade integration studies (DTIS) similar to the pilot studies originally carried out in selected countries under the Integrated Framework.106 As noted earlier, however, the general objective of these studies—which are typically carried out by trade consultants—is to identify opportunities for trade liberalization and export promotion. Assuming that a DTIS team has the expertise to do poverty and social analysis, the main task of the team will continue to be identification of trade reforms and trading opportunities in a particular country. Therefore, a DTIS team will have little time or incentive to thoroughly analyze the possible poverty and social impacts of the trade options being considered. In addition, based on past practice, DTIS teams are likely to consist mainly of foreign consultants who are generally disconnected from the PRSP process and who will have little time to consult widely with government ministries, parliamentarians, civil society groups, and other stakeholders.
As noted above, there is another process under way within the World Bank to develop tools to analyze the distributional effects of various economic reforms. Part III considers whether these tools would be more helpful than the DTIS approach to analyze possible impacts of trade policy changes on poor and vulnerable groups.
Development banks and agencies originally developed tools for poverty and social analysis to improve the design of traditional development projects. As they expanded their work on the social dimensions of projects—for example, to include greater consultation with project beneficiaries and other project-affected groups and to address the needs of women, indigenous people, and ethnic minorities—they developed additional tools for social analysis, borrowing heavily from the [*PG402]field research techniques of anthropologists and sociologists.107 The recent focus on poverty reduction has prompted the development of additional tools to carry out participatory poverty assessments and design appropriate interventions.108
For some time, civil society groups and some governments have been urging the IMF and the World Bank to incorporate poverty and social assessments in the design of stabilization and adjustment loans in order to assess the potential impacts of proposed reforms on marginalized and vulnerable groups.109 Similar recommendations have been made relating to their loans in support of PRSPs. For programs supported by its Poverty Reduction and Growth Facility (PRGF), the IMF now documents any poverty or social analysis that has been done in the relevant country on the measures included in the program. However, there has been little experience to date in undertaking this sort of analysis specifically for PRGF programs.110 At the same time, the World Bank has begun to support governments in carrying out poverty and social impact analysis of major policy changes included in PRSPs,111 and has also proposed options for carrying out poverty and social assessments in connection with its adjustment loans.112 Very few PRSPs to date have included analysis of distributional impacts. However, most recent PRSPs mention plans to carry out this type of analy[*PG403]sis for policy changes that are likely to have large distributional impacts, including trade reforms.113
As part of a larger work program supported by the United Kingdom’s Department for International Development (DFID) and other donors, the World Bank has developed a User’s Guide to Poverty And Social Impact Analysis (PSIA) and a related Toolkit for use by policy-makers and analysts in developing countries, as well as by other development agencies and civil society groups.114 Tools related to particular policy areas—including trade policy—will continue to be developed and refined. The same work program also included several pilot studies that have used PSIA tools to analyze a variety of economic policies and programs in developing countries, including tariff changes and export promotion programs. Initial reactions to the User’s Guide have been mixed.115 Nevertheless, the User’s Guide provides a starting point for developing approaches to analyze the PSIA of trade policies and rules. Other relevant resources include a handbook on trade liberalization and poverty recently developed by DFID.116
PSIA is defined in the User’s Guide as “analysis of the distributional impact of policy reforms on the well-being or welfare of different stakeholder groups, with particular focus on the poor and vulnerable.”117 The User’s Guide suggests a framework for PSIA built around seven key questions:
The choice of welfare measures to be used in PSIA has become more complicated, as development agencies and governments now recognize poverty as multidimensional, including both quantifiable and qualitative aspects. Until recently, poverty analysis was typically based on income or consumption measures. However, as participatory poverty assessments have become more common, a variety of participatory and qualitative measures have been developed.
Determining which groups to analyze also requires reflection. Some form of stakeholder analysis is needed to identify and disaggregate the groups and subgroups most likely to be adversely affected by the particular policy change. Analysis based on broad categories such as “the poor” is unlikely to produce meaningful information. In the area of trade policy, for example, impacts may be experienced differently by groups in particular sectors or regions, by producers and consumers, and on the basis of class, gender, ethnicity, age, and other factors.
Identifying likely transmission channels for a particular policy change is also critical. The User’s Guide suggests five main transmission channels for most policy reforms: (1) employment; (2) prices (including production, consumption, and wages); (3) access to goods and services (including public goods and services); (4) various assets;119 and (5) transfers (including gifts, remittances, and subsidies) and taxes.120 For trade policy changes, resources such as the Global Poverty Report 2001121 and DFID’s handbook on trade liberalization and poverty122 suggest additional transmission channels that are likely to be relevant. These include government revenue (e.g., related to tariff changes) and adjustment costs (e.g., the social costs associated with [*PG405]the displacement of workers from adversely affected industries, the impact on unpaid household work, and the costs of complying with particular trade agreements).123 Finally, for trade policies and trade agreements, this Article argues that laws and regulations, particularly those that affect the welfare of poor and vulnerable groups, are other important transmission channels.
Legal/regulatory analysis of changes in trade policies or trade rules has become more relevant as the scope of WTO and other trade agreements has widened to include services, intellectual property, health and food safety regulations, and any other regulations that may be deemed “technical barriers to trade.” Careful examination of the text of a trade agreement is needed to assess its likely impact, including any ambiguities in wording that might affect its implementation by parties to the agreement,124 and any new laws, regulations, institutions, and procedures that may be required to implement it.125 Analyzing a trade agreement’s text can also identify areas of possible tension or conflict between the agreement and a country’s existing obligations under other international instruments, such as human rights or environmental treaties, or between the trade agreement and national laws or regulations already in place. Although related, the legal/regulatory analysis suggested here can be distinguished from institutional analysis, which the User’s Guide also recommends. Institutional analysis is broader in scope and could include a review of the [*PG406]functioning of government agencies, markets, community organizations, and informal networks, as well as legal or judicial institutions. Part IV of this Article describes in more detail a possible framework for legal/regulatory analysis of trade agreements, with gender as the main category of analysis.
Another important element of any PSIA—which the User’s Guide discusses in detail—is the gathering of data and the selection of appropriate tools for analysis. In this area, the User’s Guide cautions analysts to be aware of constraints in the availability of data, the difficulty of disentangling trade from other effects and tracing different impacts over time, the capacity of government agencies, academia, and civil society organizations to use complex tools and methods, and the time available for data collection and analysis.126 The User’s Guide reviews various data collection methods, both quantitative and qualitative, and recommends the use of mixed methods to assess more fully the impacts of policy changes on poor and vulnerable groups.127 A wide variety of tools for data analysis is reviewed, including tools for
Each of these tools has advantages and disadvantages, and a number of tools may not be practicable because of data and time constraints. In addition, the tools may not adequately capture the different impacts of a policy change on particular social groups, especially qualitative impacts. For example, computable general equilibrium (CGE) models, which attempt to simulate key aspects of an entire economy or region, are extremely complicated to build, rely entirely on quantitative data, and simplify the economy in a number of ways. These models have also been criticized for not taking into account gender and other social differences. Over the past few years, however, several researchers and research teams have been working on alterna[*PG407]tive CGE models that better simulate the impact of particular policy changes on different social groups, including differentiations by gender, rural/urban location, education, income level, occupation, and other factors.129 There are also research networks, such as Micro Impacts of Macro and Adjustment Policies (MIMAP), that are helping to build the economic modeling capacity of researchers and research institutes in developing countries.130 However, there is much more work to be done in this area.
Another general problem is that the household survey data on which many of these tools rely are collected at the household—rather than individual—level. Because data are not readily available on the allocation of income, consumption, and assets within households, a simplifying assumption is made that all household members are treated equally. Thus, the tools may fail to capture different impacts on household members related to gender, age, or other factors.131
To assess the impact of trade policy changes, such as tariff reductions, a variety of tools can be used, including consumer assessments, demand/supply analysis, and more elaborate economic models. However, the usefulness of these tools will depend on the availability of data, time limitations, and the types of impacts that seem to be most relevant. Moreover, since trade policy changes can extend to a wide variety of trade-related areas—including services, agriculture, investment, and various regulations—other tools and approaches may be needed. A general analytical challenge for the PSIA of any trade policy or agreement is to disentangle its likely impact from the effects of other eco[*PG408]nomic policies, other trade agreements, and general changes in the local and global economies.
The pilot PSIA studies supported by DFID and the World Bank provide an early indication of the tools and approaches that can be used to analyze trade policy changes, as well as the constraints and complexities involved. To date, most of the studies have relied on partial equilibrium or CGE models. However, a pilot study of Uganda’s Strategic Export Initiative found that the quantitative data available—for example, from poverty measurement surveys—were inadequate for economic modeling purposes and relied instead on case studies of changes in farming and fishing households. The use of qualitative studies permitted the researchers to examine intra-household dynamics, which led the team to make several recommendations to address underlying gender issues in the coffee and fishing sectors.132
Based on the experiences in early PSIA studies, researchers and development agencies involved have concluded that PSIA tools can be very helpful in estimating the distributional effects of major policy changes. However, in light of the limited quantitative data available in most developing countries; the conceptual and technical difficulties in conducting ex ante analysis of policy reforms; and the limited resources and capacity of research institutes in developing countries; they stress the need to keep expectations realistic about future PSIA work. They strongly recommend the use of several data sources (including qualitative as well as quantitative data) and several tools (such as economic, social, and institutional analysis). They also emphasize the importance of integrating PSIA in existing processes (such as the development of PRSPs and national development plans), ensuring that the PSIA is country-led, and involving a broad array of stakeholders.133
[*PG409] These findings echo a number of issues and concerns that some development agencies and civil society groups had raised in comments on the initial draft of the User’s Guide.134 These included:
In the case of PSIA of trade issues, it is also important to link the analysis and outcomes to a country’s various ongoing trade initiatives, which are likely to include not only WTO negotiations but also negotiations at regional and bilateral levels. (For example, bilateral trade agreements may include more aggressive trade liberalization measures in certain sectors, which should be considered along with the country’s more general commitments under WTO and regional agreements.) PSIA could also be extremely useful to any countries involved in the protracted process of joining the WTO. This is especially the case for LDCs, which are increasingly being pressured by powerful WTO members to agree to “WTO-plus” terms—for example, on liberalization of their agriculture sectors and broadening the coverage of their intellectual property rules—as conditions for WTO accession.135
There is still much work to develop high-quality, usable tools for PSIA of economic policy options, including analysis of changes in trade policies and rules. It is particularly important to ensure that these tools are of practical use to developing country governments, research institutes, and civil society groups, and that governments can use them to assess a variety of policy options in light of national conditions, constraints, and development priorities. Increased funding of [*PG411]research institutes and networks in developing countries is needed to support independent policy research and local debate on important national policy issues. Given the multidimensional nature of poverty and social exclusion, multidisciplinary approaches to PSIA should also be strongly encouraged. This is particularly important for PSIA studies in the trade area, which also need to factor in the messy tangle of multilateral, regional, and bilateral trade agreements to which most countries—including developing countries—are parties. Part IV provides an example of one multidisciplinary project that employed social, economic, and legal/regulatory approaches to analyze possible impacts of trade agreements from a gender perspective.
For over two decades, economists and other social scientists have been investigating the relationships between gender differences and trading patterns, focusing primarily on employment associated with export-promotion policies.136 Through formal economic modeling, empirical work, and policy analysis, they have demonstrated that trade policies do not necessarily have the same impact on men and women and that gender relations and differences themselves can influence trade policy outcomes.137 At the same time, feminist legal scholars [*PG412]have drawn attention to the various ways in which international legal frameworks—including international economic law—reflect and extend typically “masculine” assumptions and norms and reinforce the dominant position of major industrialized countries in world politics and the world economy, resulting in the exclusion or marginalization of alternative perspectives and interests.138 Their work has also explored the tensions between international economic law and human rights law and the implications for women and societies in general.
Reflecting growing concern about the potential adverse impact of trade liberalization on women, particularly in developing countries, the Beijing Platform for Action, which was adopted unanimously at the Fourth World Conference on Women in 1995, called on all governments to “ensure that national policies related to international and regional trade agreements do not have an adverse impact on women’s new and traditional economic activities.”139 This call has been taken up by several international and regional bodies, including UNCTAD, Asia-Pacific Economic Cooperation (APEC), the Caribbean Community (CARICOM), the Southern Common Market (MERCOSUR), and the Southern African Development Community (SADC), [*PG413]which have established advisory groups or offices to ensure that gender concerns are reflected in their activities related to trade and foreign investment. In 2003, a U.N. interagency task force on gender and trade was created, headed by UNCTAD, to coordinate research, capacity building, and advocacy activities within the U.N. system. Gender and trade has also been identified as a cross-cutting issue to be considered at UNCTAD XI, taking place in Sao Paolo in June 2004.140
Building on these developments, the Coalition for Women’s Economic Development and Global Equality (Women’s EDGE Coalition), which advocates for improvements in women’s and girls’ economic opportunities, rights, and well-being in developing countries, recently commissioned a study by two economists and the author of this Article to develop a possible framework for carrying out gender assessments of trade and investment agreements.141 The study supported Women’s EDGE Coalition’s efforts to persuade the Office of the U.S. Trade Representative (USTR) to carry out gender assessments of new trade and investment agreements being negotiated between the United States and other countries, with a particular emphasis on impacts in developing countries. Women’s EDGE Coalition based its argument for gender assessments in part on the precedent of U.S. Executive Order 13141, which requires the U.S. government, [*PG414]through the Council on Environmental Quality and the USTR, to conduct environmental reviews of proposed trade agreements.142
The gender assessment study was conceived as multidisciplinary and combined social, economic, and legal/regulatory analyses within a human development and human rights framework. The study included a review of relevant social science, trade policy, and legal literature; the development of socioeconomic and legal/regulatory frameworks for analysis and several preliminary assessment tools; and policy recommendations for conducting gender assessments, including the participation of a wide range of stakeholders in the process.
The analytical frameworks developed in the study were based on several stylized facts that help to explain why trade liberalization may have gender-differentiated impacts and why gender differences in turn may affect the outcome of trade policies.143 These include the observation that, in most countries, women and girls typically have less access to productive resources such as land, credit, education, and occupational training. In addition, in most countries the labor market is strongly segregated along gender lines, with men holding relatively better-paying and more secure jobs than women. Furthermore, women bear most of the burden of reproduction, and social norms reinforce the asymmetric division of labor in the household, with the result that [*PG415]women are generally responsible for maintenance of the household, including care of children and elders. This can lead in turn to different consumption and time use patterns among women and men. Social norms in certain societies also restrict women’s mobility and participation in economic and civic activities outside the home. These characteristics tend to overlap and reinforce each other. The net effect, however, is that men and women face different opportunities and constraints that can affect their ability to benefit from trade liberalization or foreign investment, or to adjust to changes resulting from these trends.144
The frameworks were also based on an understanding that categories such as “women” and “men” are generally inadequate for purposes of social analysis and need to be further disaggregated to reflect differences based on age, race or ethnicity, class or caste, household status, citizenship status, location, and other factors. In the area of trade liberalization and foreign investment, this disaggregation of social categories is particularly important, since the most vulnerable social groups are likely to be located at the intersections of multiple social categories (such as undocumented migrant women or indigenous women who are also heads of households).
The socioeconomic framework developed in the gender assessment study is built on a gender critique of mainstream economic analysis of trade. Mainstream trade theory, which is based on several unrealistic assumptions about economies (such as perfect competition and full employment), identifies four principal channels through which trade liberalization will affect the economy and individual actors: (1) the prices of goods and services; (2) employment; (3) wages; and (4) consumption of goods and services.145 The framework identifies ways in which market imperfections—including market power, unemployment, restrictions on labor mobility, and gender inequality—can cause these “transmission channels” to operate differently from what trade theory [*PG416]would otherwise predict. Thus, while trade theory suggests that a sector with a comparative advantage in low-cost labor would expand production for export under trade liberalization—resulting in increased demand for labor and increased wages for workers in that sector—a number of factors can operate to depress or freeze wages in an export sector, including unemployment, capital mobility, a shift to subcontracting, and gender segregation and discrimination. For example, it has been argued that gender segregation and discrimination in labor markets can explain the findings of persistent gender wage gaps in newly industrialized, export-oriented countries where the workers employed in export industries are predominantly women.146
The framework extends the traditional analysis of the four transmission channels to include the effects of trade liberalization on conditions of work (as well as the availability of jobs) and the allocation of consumption within households, and it also considers effects on access to public goods and services such as health care, education, and water supply. Finally, the framework attempts to take account of trade impacts on the informal sector and the reproductive sector, neither of which is captured in traditional economic analysis. In terms of the tools available for socioeconomic analysis, the gender study considers a variety of economic models and ways that they could be modified to reflect better the gender differences in economies, as well as areas for empirical research. For purposes of both analysis and monitoring, the study notes the need to collect gender-disaggregated data, including intra-household data, and identifies a number of indicators that could be relevant for monitoring impacts of trade policies and agreements. The study also stresses the need to anticipate implementation costs and implementation issues related to the trade policy or agreement that is under review.
The legal/regulatory framework developed under the gender assessment study provides for analysis of a particular trade or investment agreement on three levels: (1) the text of the agreement itself; (2) related implementation and enforcement mechanisms; and (3) the agreement’s possible interactions with the various laws and norms that influence gender roles and women’s opportunities and con[*PG417]straints in a particular country.147 Under this approach, a trade or investment agreement would first be examined to identify any overtly “gender-biased” provisions or apparently “gender-neutral” provisions that might have a disparate effect on women and men (or on groups of women and men) in light of their different socioeconomic circumstances. For example, U.S. free trade agreements with Chile, Jordan, and Singapore commit each party to ensure that its labor laws are consistent with internationally recognized labor rights and to enforce these laws effectively.148 However, the agreements fail to include the rights to nondiscrimination in employment and equal pay in their definitions of “internationally recognized labor rights,” even though these two rights are clearly included in the Fundamental Principles and Rights at Work of the International Labour Organization (ILO)149 and are particularly important to women and other historically disadvantaged groups. In contrast, the ILO conventions on nondiscrimination and equal pay are specifically included in the special [*PG418]incentive arrangements for the protection of labor rights under the European Community’s Generalized System of Preferences.150
In terms of implementation and enforcement mechanisms, the Uruguay Round Agreement on the Application of Sanitary and Phytosanitary Measures (SPS Agreement) generally requires WTO members to base their sanitary and phytosanitary regulations on international standards promulgated by organizations such as the Codex Alimentarius Commission.151 However, these organizations have longstanding ties to industry members and associations and have not historically been open to consumer and other citizens’ groups.152 It is therefore possible that the standards set by these bodies may not reflect the concerns of women, particularly from developing countries. The requirement of the SPS Agreement that sanitary and phytosanitary regulations must be “based on scientific principles and . . . not maintained without sufficient scientific evidence”153 also has gender implications. Until recently, less medical research was devoted to women’s health issues, and therefore less data may be available on the long-term effects on women of exposure to certain pollutants and other harmful substances.154 Thus, should a country choose to regulate a particular substance more stringently than the applicable international standard in order to address health risks to women, it could be more difficult to establish a sufficient scientific basis for the regulation under the SPS Agreement. Moreover, under the WTO’s dispute settlement procedures, dispute panels also have considerable flexibility to engage experts to advise on scientific and other technical matters. However, there is no assurance that the panel members— who are primarily trade law experts—or their expert advisors will be [*PG419]sensitive to women’s health concerns in their review of health-based regulations.155
Turning to the possible interactions between trade and investment agreements, on the one hand, and other legal frameworks relevant to women and other social groups, on the other, a few general points should be considered. First, there is an array of formal and informal laws and norms that shapes the status and rights of groups in a society. These may include international treaty commitments a country has made—for example, through ratification of the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW) and various ILO conventions—as well as provisions of the national constitution, statutes, regulations, administrative directives, and judicial rulings. In many countries, the picture is complicated further by the existence of religious, traditional, or customary laws, which may be codified in whole or in part and interpreted and enforced by a separate system of courts. Second, the situation of women, even within the same country, may vary greatly depending on other social factors such as religion, ethnicity, class, and geographic location. Third, there are both institutional and practical factors that can influence the enforcement of laws and the access to justice for women and certain other social groups.156
With these points in mind, the gender assessment study developed a tentative “checklist” of laws and norms relevant to the situation of women and other social groups, against which a trade or investment agreement could be evaluated. The checklist includes the following general categories:
—Non-discrimination and equal treatment laws;
[*PG420] —Affirmative action and other laws providing special treatment, such as small business assistance and government procurement preferences; and
—Gender neutral laws, such as those pertaining to fair wages, food labeling, and health and safety;
Not all of these categories would be relevant in each case, and the scope of the analysis would also depend on the time and resources available. The last two categories were included in the checklist to ensure that provisions of a trade or investment agreement do not reinforce the inequitable treatment of women that may result from gaps in the enforcement of beneficial laws or from the application of certain religious, traditional, or customary laws or norms.
Two examples of areas where this comparative approach could be useful are small business assistance and government procurement preferences for women and minority-owned businesses. In the case of small business assistance, the so-called “green light” subsidies originally allowed under Article 8 of the Uruguay Round Agreement on Subsidies and Countervailing Measures (SCM Agreement) included government assistance to “disadvantaged regions,” but not to economically disadvantaged groups such as women or ethnic minorities.158 Thus, a national or local government program to provide low-cost loans, tax credits, training, or other support to women or minority entrepreneurs in an export manufacturing industry might conceivably be challenged as an actionable subsidy under the SCM Agreement.159 Similarly, pro[*PG421]grams to assist women and minority-owned businesses in obtaining government contracts could run afoul of the Uruguay Round Agreement on Government Procurement (AGP), depending on the type of assistance program and specific commitments that the state party has made under the agreement.160
One example—the agriculture sector—will illustrate the types of issues that can be identified through collaborative analysis that considers social, economic, and legal/regulatory factors.
A majority of the female labor force in developing countries is engaged in agriculture—62% in 1990.161 Although women are involved in production of both cash and subsistence crops, research indicates that they are primarily responsible for household food production in several regions—70 to 80% in Sub-Saharan Africa and 65% in Asia.162 However, extensive literature documents the constraints that women farmers face due to limited access to and rights over land, limited access to credit and other inputs, and neglect by agricultural extension services.163
Market liberalization and promotion of export crops have had ambiguous employment effects on women farmers. For example, in some regions, women’s representation in the agricultural labor force has increased as more men migrate to jobs in manufacturing or other [*PG422]off-farm work.164 However, expansion of export crops has also displaced women farmers from land they had used for subsistence farming, and many of these displaced women have taken seasonal jobs as agricultural workers. In Africa and Latin America, women make up the large majority of the workforce in nontraditional agricultural exports such as flowers and specialty vegetables. While these jobs provide cash income, they are also characterized by low wages, health hazards, and seasonality.165 Market liberalization has also created new opportunities for women involved in food processing and trading, although they tend to be concentrated in small-scale operations. In general, the liberalization of agriculture markets has benefited mainly medium and large-scale commercial producers and traders. Women farmers and traders, who generally operate on a smaller scale, have been found less likely to benefit. It is especially difficult for small farmers, including women, to compete with heavily subsidized agricultural imports.166
Case studies from Africa have shown that, where women are primarily responsible for household food production but also help with cash crop production as unpaid family laborers, they are likely to realize little or no direct benefit from programs to increase production of cash crops for export. In many cases, they may be required by male family members to give up part of their subsistence farmland for cash crop production, which further weakens their property rights and may also threaten their family’s food security if they are no longer able to grow subsistence crops. Women farmers may therefore resist or undermine male family members’ efforts to expand production of export crops. This analysis provides one possible explanation for the weak “supply response” that has been observed by economists where the prices of cash crops have been liberalized, particularly in Africa.167
[*PG423] On its face, the Uruguay Round Agreement on Agriculture does not exhibit any overt gender bias. The agreement extends “special and differential treatment” to developing countries in several respects. Moreover, a number of its “green box” exemptions from aggregate limits on domestic supports—for example, for extension services, stockholding for food security, domestic food aid, and disaster relief—would seem to be beneficial to women farmers in developing countries.168 However, the general goal of the agreement is to facilitate agricultural exports. As noted above, women farmers in many developing countries are primarily responsible for household food production; therefore, household food security and sufficiency are high priorities.169 The Agreement on Agriculture includes a domestic support exemption for “food security” stockpiles. However, the agreement treats food security in terms of government (or international) support rather than household self-sufficiency.170
At the Fourth Ministerial Conference in Doha, one of the proposals put forward by developing countries, with strong support from civil society groups, was for the creation of a “development box.”171 Among other things, this proposal allows developing countries to exclude specified staple crops from their tariff reduction commitments and also allows governments to give greater assistance to poor farmers. Similar proposals were floated at the Fifth Ministerial Conference in Canc�n [*PG424]for a “special products” category and a “special safeguard mechanism” to address the food security and rural development needs of developing countries.172 These proposals address the general bias of the Agreement on Agriculture toward agricultural exports. However, they would need to be complemented by gender-sensitive agriculture policies and targeted resources in order to benefit women farmers, who have historically been excluded from agricultural extension services, credit programs, and other support.173
The gender assessment study described here also recognized the importance of ensuring that women’s interests and concerns are represented in the processes for formulating trade policy. Historically, women and other segments of civil society have not been included in these processes; in contrast, private sector interests have been actively engaged in trade policy formulation, lobbying of legislators on trade issues, and applications for trade remedies.174 However, the growing [*PG425]concerns among civil society groups in many countries about the benefits of further trade liberalization have caused at least some trade and commerce ministries to open up the trade policy process through some form of public consultation.175
The gender assessment study included a number of policy recommendations related to the process for conducting gender or other social assessments of trade policy changes and new trade commitments. In particular, the study recommended that a wide range of stakeholders be involved in the assessment, including non-trade government departments for labor, education, health, and international development, state and local officials, women’s organizations, environmental organizations, and representatives of indigenous groups, farmers’ unions, and community-based organizations. These stakeholders should also have input in the development of trade policy and in monitoring the impacts of trade policies and commitments. The study called for improvements in the collection of gender-related data—for example, on women’s and men’s employment in export and import-competing industries, and on consumption patterns, time use, and control of productive assets within households. The study also called for the provision of technical support to developing countries to build their own capacities to carry out gender and other social assessments of trade commitments.176
The gender assessment study for the Women’s EDGE Coalition highlighted the need for further research in a number of areas to determine the precise impact of specific trade and investment policies on women in particular countries, sectors, and circumstances. As discussed in Part III above, the study also noted the need for further work on practical tools for conducting gender and other social assessments of trade agreements, including a mix of quantitative and qualitative/participatory approaches.
In the area of legal/regulatory analysis, the study identified cases in which trade agreements had been negotiated “gender-blind” or with gender concerns as an afterthought. For example, the potential conflict identified between the SCM Agreement and export-promotion programs for women-owned businesses could easily have been avoided if trade negotiators had included an appropriate exception in the agreement for “assistance to historically disadvantaged groups” along the lines of the “green light” subsidy for “assistance to disadvantaged regions.” At the same time, the study suggested that current proposals to address developing country concerns about the operation of the international trading regime—such as the proposal for a “development box” to be added to the Agreement on Agriculture—might need to be further refined to ensure that women farmers and other marginalized groups share in the benefits of the DDA.
The gender assessment study recommended that industrialized countries such as the United States, in formulating their own trade policies and commitments, consider the potential impact on women and other marginalized groups in developing countries. However, each country would still have the primary responsibility for assessing the impact of trade and other economic policies within its borders. The study also recommended that developing countries should have access to trade-related technical assistance funds established under the Integrated Framework to finance poverty and social/gender assessments of proposed trade commitments and to monitor the impacts of trade agreements that are already being implemented.
In terms of timing, the gender assessment study recommended that a social/gender assessment be undertaken as early as possible in the trade negotiation process and that it be updated periodically as [*PG427]the negotiation progresses. Part of any assessment should also involve the selection of indicators to monitor future impacts. Given their resource and capacity constraints, developing country governments and civil society groups will need to be selective in the trade policy options they choose to assess. (As discussed in Part III above, several countries are already on this path, having identified key trade policy reforms to examine using PSIA tools.) By drawing on expertise from national research institutes and universities rather than international consultants, developing countries can conserve scarce resources while also ensuring greater ownership of their trade assessments.
Following the collapse of trade negotiations in Canc�n, uncertainties remain about the direction of the Doha Development Agenda and the type of support that development banks and agencies will provide to developing countries to engage effectively in this process. In light of international and national commitments to meaningful poverty reduction, it is increasingly important for any new trade commitment to be examined carefully from a poverty and social perspective, including considerations of gender and other key social categories. It is also important to encourage the development of a variety of diagnostic tools for this type of analysis for use not only by government trade offices and development institutions, but also by other government ministries, research institutes, universities, and civil society groups. The use of these tools by various stakeholders can lead to a fuller and more balanced understanding of the trade commitments under consideration and may inspire greater public confidence in the international trading system, particularly for developing countries. However, for these tools to be useful, developing countries must also have the flexibility to choose the trade policy options that best support their poverty reduction strategies and broad development goals.