labor standards, developing countries, conditions, industrialized countries, empirical evidence, Labour Standards, University of Michigan, labor rights, international labor standards, provisions, submissions, trade agreements, working conditions, United States, child labor, European Communities, protectionist, public-interest groups, the United States, International Trade, ILO, Robert E. Hudec, Robert E. Baldwin, L. Alan Winters, Alan V. Deardorff, Richard B. Freeman, OECD, International Labour Organization, Jagdish N. Bhagwati, Core Labour Standards, labor side agreement, Robert Lawrence, Robert M. Stern, labor issues, trade sanctions, labor opposition, consumers, public goods, interest groups, private goods, labor issue, European Union, International Labor Organization, Elliot Berg, Gerald R. Ford School of Public Policy, The Canadian NAO, employment conditions
RESEARCH SEMINAR IN International economic
S Gerald R. Ford School of public policy
The University of Michigan
Ann Arbor, Michigan 48109-1220 Discussion Paper No. 496 Labor Standards and Trade Agreements Robert M. Stern University of Michigan August 18, 2003 Recent RSIE Discussion Papers are available on the World Wide Web at: http://www.spp.umich.edu/rsie/workingpapers/wp.html
2 LABOR STANDARDS AND TRADE AGREEMENTS Robert M. Stern University of Michigan Abstract There is a wide disparity of views on issues of international labor standards. Labor and social activists are concerned about the increased imports from countries in which labor standards are ostensibly not enforced at a sufficiently high level. They fear that these imports will be detrimental to wages and employment conditions in the industrialized importing countries and that workers in the developing countries
will be exploited, their wages suppressed, and that they will be subjected to abusive work conditions. This paper explores these different views and the available options for addressing the issues involved. The paper begins with the definition and scope of labor standards and then turns to theoretical aspects of the economic effects of labor standards and a summarizes the empirical evidence on the effects on wages, trade, and foreign direct investment, and the role of interest groups. Global, regional, national/unilateral, and other arrangements for the monitoring and enforcement of labor standards are discussed and implications for policy presented. To be published in a Special Issue of Revue D'Economie Du Developpement, "Analyser les Relations Nord-Sud" (Analyzing North-South Relations), Essais en l'honneur d'Elliot Berg. Revised August 18, 2003 Address Correspondence To: Robert M. Stern Gerald R. Ford School of Public Policy University of Michigan Ann Arbor, MI 48109-1220 Tel. 734-764-2373 FAX 810-277-4102 E-mail [email protected]
LABOR STANDARDS AND TRADE AGREEMENTS* Robert M. Stern University of Michigan I. Introduction The interaction of labor standards and International Trade
has become a key issue in the relations between the advanced industrialized and developing countries in the past decade. Labor interests and social activists in the United States and other industrialized countries have argued that "unfair" labor practices and conditions exist in developing country trading partners and need to be offset by appropriate trade policy measures in order to "level the Playing Field
." The concern is that the increased imports from countries in which labor standards are ostensibly not enforced at a sufficiently high level will be detrimental to wages and working conditions
in the industrialized importing countries. It is also argued that workers in developing countries are subject to exploitative and abusive working conditions, and that their wages are suppressed. As will be noted in the discussion to follow, there is a wide disparity of views on issues of international labor standards. In this paper, I explore these different views and the available options for addressing the issues involved. The paper is structured as follows. Section II deals with the definition and scope of labor standards. Theoretical aspects of the economic effects of labor standards are considered in Section III, while Section IV summarizes the empirical evidence. Global, regional, national/unilateral, and other arrangements for the monitoring and enforcement of labor standards are discussed in Section V. Conclusions and implications for policy are presented in Section VI. *One of my greatest pleasures in life was my close association with Elliot Berg for nearly four decades, as his academic colleague at the University of Michigan and during the subsequent years that he spent in Washington, D.C. Elliot was always generous with his time, and we shared many ideas over the years in each other's research and writings. I particularly valued his comments and suggestions on my previous work dealing issues of trade and labor standards. In writing this paper, I have received helpful comments from Lindsay Benstead and Andrew Brown.
2 II. Definition and Scope of Labor Standards Labor standards are multi-faceted and may vary from country to country depending on the stage of development, per capita income, and political, social, and cultural conditions and institutions. It may be difficult therefore to distinguish unambiguously between universally agreed labor standards and those standards that depend on national circumstances. Nonetheless, efforts have been made to identify and achieve consensus on a group of so-called core labor standards that ideally should apply universally. For example, according to OECD (1996, p. 26), core labor standards include: (1) prohibition of forced labor; (2) freedom of association; (3) the right to organize and bargain collectively; (4) elimination of Child Labor
exploitation; and (5) nondiscrimination in employment. Agreement on the universality of these core labor standards derives ostensibly from adoption of the United Nations Universal Declaration of Human Rights
in 1948, acceptance (though not necessarily ratification) of the pertinent Conventions of the International Labor Organization (ILO) that deal with human rights and labor standards, and the ILO Declaration on Fundamental Principles and Rights at Work in 1998.1 Besides the aforementioned core standards, there are other labor standards that are less universally accepted, and that relate to "acceptable conditions of work," which include: a minimum wage; limitations on hours of work; and occupational safety and health in the workplace.2 Some of the difficulties that may arise in interpreting and implementing core standards and distinguishing between core and other standards can be illustrated in the attempt by Fields (1995, p. 13) to articulate what he considers to be: ...a set of basic labour rights for workers throughout the world: i) No person has the right to enslave another or to cause another to enter into indentured servitude, and every person has the right to freedom from such conditions. 1 According to OECD (2000, p. 20), there are eight fundamental ILO Conventions that form the basis of consensus among the ILO's constituents. These include: prohibition of forced labour (No. 29 and 105); freedom of association and protection of the right to organize and collective bargaining (No. 87 and 98); equal remuneration for men and women for work of equal value (No. 100); nondiscrimination in employment and occupation (No. 111); and minimum age of employment of children and abolition of the worst forms of child labor (No. 138 and 182). 2 See Brown, Deardorff, and Stern (1996, Appendix Table 1) for the definitions and principles of the core and other labor standards that are articulated in U.S. trade law, based on Lyle (1991, pp. 20-31).
3 ii) No person has the right to expose another to unsafe or unhealthy working conditions without the fullest possible information. iii) Children have the right not to work long hours whenever their families' financial circumstances allow. iv) Every person has the right to freedom of association in the workplace and the right to organise and bargain collectively with employers. It would be no easy matter to make operational Fields' proposed basic worker rights. For example, it is unclear how to interpret what is meant by "the fullest possible information" about working conditions or "families' financial circumstances" in the case of child labor. Further, countries may differ in the extent to which labor unions
and collective bargaining are guaranteed as an absolute right. To illustrate further, Aggarwal (1995, pp. 4-5) has proposed that a distinction be drawn between standards related to labor processes and standards related to labor outcomes. This distinction would apply some definition of what constitutes a "minimum" standard to the determination of basic worker rights in terms of labor processes. Presumably, the point of taking labor processes, rather than outcomes, into account is to make allowance for differences and changes over time in the level of Economic Development
and related factors. What remains ambiguous, however, as Aggarwal acknowledges, is the difficulty of deciding whether the identification and guarantee of labor processes lead to the improvement of labor outcomes. The foregoing discussion is by no means intended to deny the desirability of improving working conditions through higher labor standards. The issue, rather, is how this can best be accomplished. Before discussing existing institutions and mechanisms for monitoring and enforcement of labor standards, it may be useful first to consider the central theoretical issues and empirical evidence involved in analyzing their effects. III. Economic Effects of Labor Standards: Theoretical Considerations Two main issues are to be considered in what follows: (1) the diversity of labor standards and the case for free trade and (2) the effects of standards and the international harmonization of standards on the
4 economic welfare of individual nations. Also to be considered are whether labor standards constitute private or public goods and issues of political economy associated with enforcing standards. Diversity of Standards and the Case for Free Trade As noted in the preceding discussion, labor standards may vary across nations depending on their level of development, per capita incomes, and a host of political, social, and cultural conditions and institutions. The issue is whether such diversity of standards alters the case for free trade. Srinivasan (1995, 1996) has investigated this issue theoretically. He concluded that the diversity of labor standards between nations will reflect differences in factor endowments and levels of income, and that such diversity is consistent with the case for free trade. He noted further that if minimum international labor standards are to be attained, it will be necessary to have arrangements for international income transfers and domestic tax/subsidies. This will be the case as well when consumers in countries with high standards have a moral preference to raise standards in their trading-partner countries with lower standards. Further, if there are market failures that prevent the attainment of minimum labor standards, income transfers and domestic tax/subsidies will be required to achieve optimal conditions for resource allocation and consumer welfare.3 Finally, the use of trade intervention could hinder the attainment of higher labor standards, and it may accordingly be in the collective interests of countries to cooperate in setting labor standards. The implications of Srinivasan's conclusions will be examined below in considering the different options for dealing with international differences in labor standards. International Harmonization of Labor Standards Brown, Deardorff, and Stern (1996) have analyzed the effects of labor standards on economic welfare and do not concern themselves directly with issues of the diversity of standards and the case for free trade. They concluded that economic welfare is best served when countries act on their own to correct their domestic (labor) market failures. But, since these market failures will likely differ between countries, there 3 Srinivasan points out that the case for promoting labor unions and collective bargaining, which is considered to be a core labor standard, is by no means obvious in many developing countries, especially where unions are concentrated in the organized manufacturing and public sectors rather than in agriculture where a relatively large proportion of the
5 is no obvious case on welfare grounds for pursuing universal standards and the international harmonization of standards that this may imply. This conclusion is consistent with that of Srinivasan, namely that diversity of working conditions between nations is the norm and is by no means in itself "unfair" so long as the extant labor standards are consistent with efficient resource use.4 Further, despite the good intentions of government, it may turn out that the imposition of labor standards will not correct a market failure if the preferences of workers differ with respect to what they consider to be acceptable levels of, say, health and safety conditions in the workplace. Brown, Deardorff, and Stern (BDS) further assessed arguments for having standards imposed on low-income countries. They noted that low-income countries might conceivably benefit in case a government is unable for domestic political reasons to enact legislation on its own, although this presumes that the policy in question will indeed correct a market failure. It is also possible that requiring the guarantee of such standards as the right of workers to organize may serve to reinforce development of democratic institutions. Finally, they asked if there is any justification for high-income countries to take countervailing actions against the ostensibly unfair labor standards of their trading partners. They answer in the negative so long as resources are being employed efficiently. If, nonetheless, a high-income country imposes a tariff or quota on labor-intensive imports from a low-income country, this will obviously be harmful to the economic interests of workers in the low-income country. In general then, BDS concluded that the case for international harmonization of labor standards appears rather weak, and it is quite possible that harmonization could have unintended adverse consequences for the very people who are in the greatest need for assistance. It is difficult therefore to generate much theoretical support for pursuit of core labor standards that would have universal application. population may be employed. 4 An exception arises here in cases of slave labor and what may be considered to be egregious treatment of child labor.
6 Labor Standards as Private/Public Goods As already indicated, there may be a moral basis motivating the pursuit of higher labor standards.5 Thus, in his analysis noted above, Srinivasan made allowance for moral considerations so that consumers could express their concern by a willingness to pay relatively higher prices for goods and services that reflected higher labor standards. In this connection, there is an issue of whether labor standards are to be considered as public or private goods. As long as the same standards are shared generally for all consumers, the standards are public goods. But individual consumers may have a sense of virtuousness and derive pleasure from believing that the good they are consuming embodies some acceptably high level of labor standards. If individual consumers care only about their own satisfaction and not about others, labor standards can be treated as private goods. This view of higher standards as private goods has been stressed by Freeman (1994), who argues that a market solution based on labeling may be an especially effective way to raise labor standards internationally. He makes the point that labeling has the advantage that consumers pay more for what they consider morally acceptable, and at the same time foreign suppliers are compensated for their increased costs. Labeling may also undercut protectionist influences.6 It is not altogether clear, however, that labor standards should be considered to be private goods that lend themselves to a market-based treatment dependent on supplying all relevant information to consumers. If instead, labor standards take the form of public goods, Freeman (p. 30) acknowledges that some type of government intervention may be called for.7 5 Granting this, it is nevertheless important to stress that concern about labor standards ought to be motivated by improving the welfare of workers in the developing countries, and not the workers with whom they compete in the advanced industrialized countries. It is this latter view that motivates many of the advocates of labor standards. What these advocates may not realize or acknowledge is that taking actions against alleged violators of labor standards will normally make the "exploited" workers worse off, not better off. That will be true whether the sanctions are applied by government policy
or by individual consumers responding to labeling. Therefore, if we wish to make workers and their families better off, we must find a way to raise their incomes, not take their incomes away. 6 But, as just noted, labeling does not in itself raise the incomes of foreign workers and their families. 7 Freeman's argument for consumer labeling may therefore be limited insofar as it rests on treating labor standards as private goods. He does not make clear, moreover, what role, if any, the government should play in providing information to consumers and in facilitating labeling and preventing private labeling arrangements from being co-opted by producing interests.
7 Political Economy Aspects of International Labor Standards In discussing the sources of support for governmental action on labor standards, it is important to identify the constituent interest groups involved. Thus, in the United States for example, it would appear that organized labor, import-competing firms, and human rights and public-interest groups are the main proponents of stricter labor standards applied to developing countries. These interest groups may often recommend policies, including sanctions and import restrictions, which are presumably designed to change the behavior of trading-partner governments. By the same token, interest groups are influential in many developing countries, especially among unionized workers in manufacturing sectors, employees of state enterprises, and owners/managers of import-competing firms. These groups may seek to protect and enhance their own ends and to resist foreign intrusion in setting standards. Krueger (1997, p. 283) characterizes the protectionist motivation as the "prevailing political economy view of international labor standards."8 The issue then is how governments choose to respond to the various interest groups. This will be addressed in Section V below. IV. Economic Effects of Labor Standards: Empirical Evidence In this section, we consider the empirical evidence on the effects of labor standards on wages and trade, foreign direct investment, and the role of interest groups. Labor Standards, Wages, and Trade In the earlier discussion, "core" and "other" labor standards were distinguished. The question then is the extent to which international differences in the various standards affect wages and trade performance. Brown (2000) has surveyed the relevant empirical evidence on wages and trade in the industrialized countries. First, with respect to the recent experience of the trade of the industrialized countries with the developing countries, she stated (p. 16) that: "...the impact on wages and unemployment in industrialized countries is difficult to gauge." Nonetheless, she noted that there is a preponderance of evidence 8 While Krueger's characterization may apply to unions and import-competing firms, it may not apply to the activities of human-rights and public-interest groups that are not motivated by protectionist considerations.
8 suggesting, for the United States at least, that trade was not the major reason for the observed widening of the skilled/unskilled wage differential. This suggests that biased technical change rather than trade may have increased the demand for skilled workers, thus widening the U.S. wage gap. This further suggests that, since imports from developing countries account for a relatively small proportion of total industrialized country imports, trade may more generally have a limited impact on the wages of unskilled workers in the industrialized countries. Brown, Deardorff, and Stern (2002) have reviewed the anecdotal and econometric evidence on the effects of multilateral production on wages in developing countries. They concluded (p. 51) that: ...there is virtually no careful and systematic evidence demonstrating that, as a generality, multinational firms [do not] provide incentives to worsen working conditions, pay lower wages than in alternative employment, or repress worker rights. In fact, there is a large body of empirical evidence indicating that the opposite is the case. Brown also reviewed the empirical studies of the effects of variations in labor standards across countries in determining the volume of trade, competitiveness, and comparative advantage. She emphasized the importance of taking into account the empirical determinants of trade performance, not just labor standards, citing Rodrik (1996) as an exemplar. What is noteworthy is that Rodrik found in his study that only the factor endowment (comparative advantage) variables were statistically significant and that none of the labor standard indicators were statistically significant. Another study of interest is Aggarwal (1995), who investigated in detail the relationships of labor standards and the pattern of U.S. imports from ten major developing countries that accounted for 26.5 percent of U.S. imports in 1994. Her major findings (p. 7), as follows, further dissociate low labor standards with improved trade competitiveness: Sectors typically identified as having egregious labor conditions do not occupy the only or even the primary share of these countries' exports. Comparisons across more export-oriented and less export-oriented sectors indicate that core labor standards are often lower in less export-oriented or non-traded sectors such as agriculture and services. Similarly, within an export-oriented sector, labor conditions in firms more involved in exporting are either similar to or better than those in firms that are less involved in exporting.
9 Changes in technology and the structure of international trade are leading developing countries to compete in a race upward in terms of product quality rather than a race downward with respect to price. ...Wages and working conditions in developing countries have been exhibiting positive trends. In general, these have been in line with productivity changes. Finally, some of the main conclusions from the OECD study of Trade, Employment and Labour Standards (1996, pp. 12-13) are worth citing:9 ...empirical research suggests that there is no correlation at the aggregate level between real-wage growth and the degree of observance of freedom-of-association rights; ...there is no evidence that low-standards' countries enjoy a better global export performance than high-standards' countries; ...a detailed analysis of US imports of textile products (for which competition from lowstandards countries is thought to be most intense) suggests that imports from highstandards' countries account for a large share of the US market. Moreover, on average, the price of US imports of textile products does not appear to be associated with the degree of enforcement of child labour standards in exporting countries.... While the studies cited above may not constitute the final word, the conclusion seems inescapable that there is little compelling empirical evidence suggesting that low labor standards have a significant impact on wages and trade. Labor Standards and Foreign Direct Investment (FDI) It is often alleged that multinational enterprises may be attracted to locate in countries with lower labor standards to take advantage of lower costs. The available empirical evidence actually indicates the opposite to be the case. This is borne out, for example, in Rodrik (1996, p. 22), who concluded that low labor standards may be a hindrance, rather than an attraction, for foreign investors. Aggarwal (1995, p. 7) reached a similar conclusion, as did the OECD (1996, p. 13). Along these same lines, Brown, Deardorff, and Stern (2002, p. 51) concluded in their literature review
that "...there is no solid evidence that countries with poorly protected labor rights attract FDI." Thus, the empirical evidence suggests rather convincingly that there is no basis for claiming that allegedly low labor standards in developing countries have a significantly adverse effect on wages in 9 See also OECD (2000, esp. pp. 31-42) for further discussion and references.
10 industrialized countries, provide developing countries with an unfair advantage in their export trade, or distort the geographic distribution of FDI between low and high standards countries. Labor Standards and the Role of Interest Groups As mentioned above, there is a common view that support for international labor standards reflects protectionist interests in the United States and other industrialized countries. In an effort to test this proposition empirically, Krueger (1997) analyzed the determinants of sponsorship in the U.S. House of Representatives
for the Child Labor Deterrence Act of 1995. If approved, this legislation would have prohibited imports of goods produced abroad by child labor under specified circumstances, including by children under 15 years old and subject to a review of child labor practices by the U.S. Secretary of Labor.10 Krueger (p. 289) found that "...Congressmen from districts with a high concentration of high school dropouts are less likely to cosponsor the Child Labor Deterrence Act. ...This finding is contrary to what I would expect from a simple political economy model...." But Krueger also found that higher rates of unionization were associated with support for the Act as were representatives who were Democrats and also had voted against NAFTA and the Uruguay Round negotiations. In interpreting his results, Krueger (p. 293) argued that unionized workers who tend to be more highly skilled and thus may not benefit directly from a ban on imported goods made with child labor may in this case be acting as a matter of principle to pursue policies that strengthen worker rights more generally rather than pursuing their own narrow self interest. He went on more broadly to state (pp. 29394): "Indeed, in many instances I am surprised that the AFL-CIO used its limited political capital
to press for international labor standards that are of little benefit to its members, when instead it could pursue policies that are of much greater direct benefit to its membership." While Krueger's results are suggestive, they are by no means definitive and have been questioned in particular by Srinivasan (1996, 1998) on the grounds that Krueger mismeasured the legislative sponsorship and the demographic characteristics of the labor constituencies of the U.S. Congressmen covered in his study. 10 While this legislation never came to a vote, the U.S. 105th Congress subsequently enacted the Bonded Child Labor
11 The foregoing review of labor standards and trade strongly suggests, in this writer's judgment, that the weight of the theoretical analysis and empirical evidence argues against taking an activist position to mandate and enforce international labor standards on developing countries.11 Nonetheless, issues of labor standards continue to have a high profile in the current trade-policy environment. It is essential accordingly to consider the alternative arrangements that exist for the monitoring and enforcement of labor standards.. V. Monitoring and Enforcement of Labor Standards Labor standards are presently dealt with in a variety of settings: global; regional; national/ unilateral; and other, including private, arrangements. I shall discuss each of these in turn.12 Global Arrangements The main international organization that is concerned with labor standards is the ILO, which was established as part of the Treaty of Versailles of 1919 following the end of World War I. The methods and principles set out in the ILO constitution deal with all conceivable aspects of labor standards. As stated in ILO (1988, p. 4), ILO action designed to promote and safeguard worker rights takes three main forms: (1) definition of rights, especially through adoption of ILO Conventions and Recommendations; (2) measures to secure the realization of rights, especially by means of international monitoring and supervision but not by imposition of trade sanctions; and (3) assistance in implementing measures, particularly through technical cooperation and advisory services. Elimination Act, and it was signed by President Clinton on October 10, 1997. 11 There is evidently a marked difference in worldview between most advocates of international labor standards and trade and most other, particularly trade, economists. Labor standards advocates seem to see the world in terms of a struggle between capital and labor for the rewards from production, without much regard to the size of the output that they will have to share. They see the outcome as depending on power, not on economics. Trade economists see the world in terms of how resources are allocated to production with a view to maximizing the total output. They see the distribution of that output between capital and labor as depending on scarcity and productivity, not on power. Therefore labor standards advocates favor the use of intervention to tilt the balance of power in favor of labor, believing then that labor will get a larger share of a fixed pie. Trade economists see those same policies as shrinking the pie while altering the slices not by changing power but by changing the markets within which scarcity determines the rewards to capital and labor. 12 Elliot and Freeman (2003) provide a comprehensive analysis of the issues that are discussed in what follows. See also CUTS (2003) for a useful survey of the issues from the developing countries' perspective.
12 What might be considered to be ILO core labor standards was noted in Section II above. It is interesting that formal ratification of the ILO Conventions on core labor standards differs considerably among ILO members, apparently because particular Conventions may be at variance with national laws and institutional practices. Thus, for example, as noted in OECD (2000, pp. 22-23), the United States has ratified only two of the eight core labor standards Conventions, whereas several other industrialized and developing countries have ratified a significantly larger number. Ratification of ILO Conventions may therefore not be an accurate indicator of existing national regulations governing labor standards, and there are many cases in which ratified Conventions are in fact not enforced.13 It is interesting in this connection, as Charnovitz (1986, pp. 566-67) has noted, that issues of alleged unfair competition involving labor standards were addressed in Article 7 of Chapter II of the 1948 (stillborn) (Havana) Charter of the International Trade Organization (ITO). Since the GATT was conceived with a more narrow mandate as compared to the ITO, it did not address labor standards, except in Article XX(e) that provides for prohibition of goods made with prison labor. Charnovitz (p. 574) notes further that as early as 1953 the United States proposed (unsuccessfully) adding a labor standards article to the GATT. This would have empowered GATT members to take measures against other countries under the provisions of GATT Article XXIII (Nullification and Impairment). The United States continued, again unsuccessfully, to push for negotiation of a GATT article on labor standards in both the Tokyo and Uruguay Rounds of Multilateral trade negotiations
in the 1970s and 1980s. But the international community
was put on notice in April 1994 at the Marrakesh signing of the Uruguay Round accords that the United States intended to pursue issues of labor standards in future multilateral negotiations. Subsequently, there were efforts at drafting a so-called social clause dealing with core labor standards and including trade sanctions for noncompliance that might eventually be incorporated into the WTO. As noted in Aggarwal (1995, p. 38), in June 1994, the ILO began a research program dealing with the integration of social welfare and trade policy. A central objective was to develop a stronger 13 A detailed discussion of the observance of core labor standards in 75 selected countries is provided in OECD (1996, pp. 39-70) and OECD (2000, pp. 21-30).
13 enforcement mechanism. The ILO Secretariat proposed that the ILO and WTO work jointly on the oversight of international core labor standards, with the ILO concentrating on international monitoring and the WTO responsible for enforcement by means of trade-related sanctions. But because of disagreements among the country representatives of the ILO Working Party on the Social Dimensions of the Liberalization of International Trade, it was decided in early 1995 to suspend further discussion of the use of trade sanctions for alleged noncompliance with core labor standards. Instead, as noted in OECD (1996, p. 7), the ILO initiated a program of research on the effects of trade liberalization on core standards and a review of ILO means of action for the promotion of standards. Nonetheless, the United States, with some support from European Union
members, Canada, and Japan, continued to pursue the issue of trade and labor standards in the context of the WTO Singapore Ministerial Meeting in December 1996. The Declaration issued following the Singapore Meeting acknowledged the importance of international labor standards. But according to Williams (1996, p. 4), there was an apparent understanding that labor standards would remain the responsibility of the ILO. However, this understanding was not accepted by the United States, which interpreted the Singapore Declaration as leaving the issue open. The United States continued thereafter to press for inclusion of labor standards on the agenda of the Geneva 1998 WTO Ministerial Conference and was again denied. The United States raised the issue of labor standards still again in December 1999 at the Seattle WTO Ministerial Conference. At that time, President Clinton gave a speech and press interview in which he called for WTO trade sanctions in cases of violation of core labor standards. The U.S. position provoked concerted opposition by the developing countries and together with such other divisive issues as agricultural protection and subsidies and antidumping policies, the Seattle Ministerial Conference was terminated without agreement on a negotiating agenda. At the next WTO Ministerial Conference that was convened in Doha, Qatar, in November 2001, it was agreed to launch a new multilateral trade round in January 2002 that was to be concluded by the end of 2004. Issues of labor standards were again acknowledged in the Doha Declaration, but the developing countries maintained their opposition to including labor standards as part of the negotiating agenda. Thus,
14 labor standards are unlikely to be addressed multilaterally in the current Doha Round. But, as we now discuss, labor standards are and will continue to be included in the regional and bilateral trade agreements negotiated by the United States especially and some other industrialized countries and in such other contexts as corporate codes of conduct and monitoring and labeling arrangements. Regional Arrangements European Union (EU) Issues of worker rights have been a focus of attention in the EU because of concerns over low-wage competition from some EU member countries, persistent unemployment, and wage stagnation. Sapir (1995b) noted that the first efforts to address the harmonization of social policies in Europe can be traced back to early stages of European integration prior to 1958. According to De Boer and Winham (1993, p. 17), the issue of a Community-wide Social Charter was first broached in 1972. Subsequently, with the issuance in 1985 of the white paper signaling the intention to remove remaining barriers to trade and creation of a Single Market, a Community Charter of Fundamental Social Rights for Workers was drafted in 1988. This Charter, which is quite comprehensive and encompasses the "core" and "other" labor standards noted in our earlier discussion,14 has been adopted by all EU members. In his evaluation of the EU Social Charter, Sapir (1995a, pp. 742-743; 1996) concluded that harmonization of social policies was not a pre-condition of successful European trade liberalization and integration. He noted further that: ...differences in labour standards between member states remain substantial and `social harmonisation' remains a distant reality. ...whatever harmonisation has been achieved in Europe, it could not have occurred without redistributive mechanisms between countries. In the absence of such mechanisms, the harmonisation of social policies cannot be contemplated internationally. Granted the foregoing, a perhaps more fundamental reason why EU social harmonization is not an appropriate model for the world economy is that the EU aspires ultimately to some form of political union. 14 The highlights of the Charter of Fundamental Social Rights are summarized in De Boer and Winham (1993, pp. 3637), and the full text is to be found in Commission of the European Communities (1990).
15 NAFTA At the time that the NAFTA was being negotiated, some observers urged that it should include a Social Charter for North America as a possible means of protecting the interests of workers.15 Instead of including a Social Charter, however, and since the NAFTA had already been signed by the member countries in the summer of 1992, the newly elected Clinton Administration opted to pursue a separate side agreement covering labor and Environmental Issues
. At the time, the negotiation of these side agreements may have been helpful in obtaining Congressional approval for the NAFTA, especially in the U.S. House of Representatives
. Aggarwal (1995, p. 34) has summarized the main features of the labor side agreement as follows: First, the NAFTA supplemental agreement contains a more comprehensive list of labor standards than the five typically present in U.S. trade programs [which are noted below]. The agreement commits each party to the promotion of eleven broad labor conditions ranging from freedom of association to migration policies. Second, the agreement does not attempt to apply U.S. standards or...common uniform criteria in its evaluation of labor conditions in other countries. Instead, the agreement contains different enforcement mechanisms for different standards. The complaint process consists of three stages--filing a petition with the domestic National Administrative Office (NAO), Ministerial consultations, and lastly consultation with the Evaluation Committee of Experts (ECE). Complaints pertaining to freedom of association, the right of collective bargaining, and/or the right to strike can only be taken to the second stage of the complaint process...[and] sanctions cannot be utilized to encourage enforcement of laws pertaining to these rights. Of the eleven labor principles, only the implementation of those pertaining to child labor, minimum employment standards, and occupational health and safety can be supported by sanctions.16 According to the U.S. Department of Labor (2003b), as of April 13, 2003, 25 submissions have been filed under the NAFTA labor side agreement. Sixteen were filed with the U.S. NAO, of which 14 involved allegations against Mexico and 2 against Canada. Thirteen of the 16 U.S. NAO submissions have involved issues of freedom of association. Three submissions were withdrawn before hearings were held or the review process completed, four were declined for review, and hearings were held and public reports released on nine submissions. Seven of the U.S. submissions have gone to ministerial level consultations. 15 A useful reference is Lemco and Robson (1993). 16 More complete details on the North American Agreement on Labor Cooperation are provided in U.S. Department of
16 The Mexican NAO filed six submissions against the United States. Five submissions resulted in ministerial consultations and one submission is pending for consultations. The Canadian NAO filed one submission against Mexico for which consultations are pending and two against the United States that have been declined. The NAFTA side agreement on labor has thus involved a number of submissions especially by the United States regarding violations of labor standards aimed principally at Mexico. As mentioned, the objective of the submissions has been to address the labor issues through a consultative process, and it is noteworthy that sanctions have not been used to enforce any of the labor rights at issue. The side-agreement process might thus be viewed as a convenient political expedient to placate U.S. labor activists while at the same time avoiding confrontation with NAFTA partners over the use of trade sanctions. It is interesting to note that this defusing of labor opposition has been carried over to the inclusion of the NAFTA labor provisions in the bilateral free trade agreements (FTAs) that the United States has negotiated with Jordan (2001), Chile (2003), and Singapore (2003).17 There is a presumption that future bilateral and regional FTAs that the United States is currently negotiating or planning will contain these same labor provisions. Thus, while the United States has been frustrated in including labor standards in the WTO multilateral negotiating agenda, it has actively pursued the inclusion of labor standards in its bilateral and regional trade agreements. It is an interesting question accordingly whether at some future time this may open the door to including similar provisions in the WTO multilateral rules and procedures that could be coupled with trade sanctions. Labor (2003a). 17 The texts of the provisions of the U.S. FTAs with Chile and Singapore are available on the USTR website [www.ustr.gov]. As in the NAFTA side agreement, the bilateral FTAs contain a provision for monetary penalties in the event that a party fails to conform to a ruling on a dispute settlement panel on a labor issue. It may be noted that Canada had labor provisions and a procedure for permitting fines in the NAFTA side agreement and its FTA with Chile. But in the recent Canadian FTA with Costa Rica, the administrative/institutional provisions are less strict, there is greater provision for technical cooperation, and no fines. Canada is currently negotiating a FTA with four Central American countries, and the labor provisions will contain an even greater emphasis on cooperative programs. The EU association agreements, as for example between the EU and Morocco (2000), contain language affirming the democratic principles
and fundamental human rights established by the Universal Declaration of Human Rights, but do not provide for sanctions. Japan's FTA with Singapore and other FTAs in process do not contain explicit labor provisions.
17 National/Unilateral Arrangements As noted in Brown, Deardorff, and Stern (1996, p. 229), since the 1980s it has become increasingly common to include international labor standards criteria in U.S. foreign economic legislation.18 The Trade Promotion Authority Act of 2002 contains labor clauses that establish U.S. negotiating objectives and criteria. The 2002 Act includes renewal of the: 1971 Generalized System of Preferences (GSP); the 1984 Caribbean Basin Economic Recovery Act; the 2000 Caribbean Basin Trade and Partnership Act; and extension of the 1991 Andean Trade Preference Act. There is also the 2000 African Growth and Opportunity Act.19 Under U.S. legislation, countries can be removed from beneficiary status on submissions to the Office of the U.S. Trade Representative (USTR) by labor or non-governmental organizations or by the USTR itself. GSP eligibility has in fact been revoked at times for a number of developing countries until they have shown evidence that the offending actions have been or are in the process of being eliminated.20 It is noteworthy that the European Union (EU) also has a GSP program,21 which includes special incentive arrangements for additional tariff preferences for countries that comply with the eight ILO core labor standards noted in Section II above.22 18 The standards include: (1) the right of association; (2) the right to organize and bargain collectively; (3) freedom from forced labor; (4) a minimum age for employment of children; and (5) acceptable conditions of work, including a minimum wage, limitations on hours of work, and occupational safety and health rights in the workplace. 19 The United States has a Memorandum of Understanding (MOU) with Cambodia that links special quotas for most Cambodian textile exports with improving worker rights in textile factories. As noted in USTR (2002), this MOU was extended for an additional three years at the end of 2001 and runs through 2004. The quota for 2002 was fifteen percent higher than in 2001 and can be increased further provided that Cambodia demonstrates "substantial compliance" with internationally recognized core labor standards. It may be noted that when Vietnam was granted "normal trade relations" in the July 2000 U.S.-Vietnam Bilateral Trade Agreement, there were no provisions for labor rights. It may be noted, further, that there are provisions for worker rights in the authorization for the U.S. Overseas private investment
Corporation (OPIC) and in the bilateral investment treaties that the United States has signed with 34 countries. As reported in U.S. Department of the Treasury (2003), the U.S. Treasury is directed to evaluate the recognition of workers rights and to promote such rights by borrowing member countries of the International financial institutions
. . 20 The Worker Rights Case History under the U.S. GSP as of April 11, 2003 is documented and is available from the U.S. Department of Labor, Bureau of International Labor Affairs. Elliot and Freeman (2003, esp. Ch. 4) provide an assessment of the effectiveness of the U.S. GSP investigations and suspensions. 21 See European Union (2003) for a user's guide to the EU GSP. 22 According to the World Trade Organization, WT/DS246/4, on December 6, 2002, India requested establishment of a dispute settlement panel to investigate whether the EU tariff preferences accorded under the special incentive arrangements for the protection of labor rights and the environment may be inconsistent with Article 1.1 of the GATT. Article 1.1 requires that extension of any advantage or privilege for imports provided to any individual WTO member country be extended to all WTO member countries. India also claims that the EU special preferences violate the GATT "ena-
18 While there may be instances in which countries have improved their labor standards to maintain their U.S. GSP eligibility or qualify for EU special preferences, the issue is that the extension or removal of GSP eligibility is essentially decided unilaterally by the United States and the EU, both of which are obviously very powerful entities in the global trading system. It might nonetheless be argued that the experiences with quid-pro-quo actions under the GSP programs can possibly provide some useful insights into the design and implementation of policies and procedures governing trade-linked labor standards in other contexts. But the question remains that the GSP programs may offer only limited benefits to the developing countries involved because of the restricted product coverage and the administrative rules that are applied in determining rules of origin.23 . Also, in the future the value of GSP will be eroded by the tariff reductions previously negotiated in the Uruguay Round and those to be negotiated in the ongoing Doha Development Agenda
round. Developing countries might therefore be advised to be wary of entering into preferential arrangements that may not be in their national interests.24 Other Arrangements There are a number of other arrangements that deserve mention in addition to those already discussed above. For example, as noted in OECD (1996, pp. 161-69), the OECD, ILO, UNICEF, and other UN agencies have been active in promoting cooperative programs of economic development in which practical measures backed up often by multilateral and bilateral financial assistance can be devised to deal with some of the underlying causes of poverty in poor countries
that may be reflected in the employment of children and the absence or relatively weak enforcement of core labor standards. The OECD and ILO have also developed international codes of conduct applicable to multinational enterprises (MNEs) that may assist in bling clause" that calls for establishing nonreciprocal and nondiscriminatory preferences under GSP schemes and that preferences should foster the trade and other needs of developing countries. See Howse (2003) for an analysis of India's challenge in the WTO of the EU special preferences. 23 See Brenton and Manchin (2003) for an assessment of the EU GSP. 24 Srinivasan (1997) characterizes the GSP as "`crumbs from the rich man's table' which the developing countries should do well without."
19 improving labor standards and working conditions in MNE affiliates in host developing countries. Individ- ual firms can attempt to develop codes of conduct on their own, as Elliot and Freeman (2003) have noted and has been done by several U.S. MNEs and those of other countries. These cooperative efforts and codes of conduct are essentially voluntary in nature, and, of course, there is no guarantee that they will be effective in all circumstances in developing countries. Nonetheless, they serve an important role insofar as they help to focus attention on the importance of the root causes of underdevelopment and the types of business practices that may help developing countries to raise per capita incomes and improve conditions of work.25 Finally worth mentioning is the use of consumer labeling in providing a market-based method for helping to improve labor standards when these standards can be treated as private goods. The advantage of labeling is that it provides information about production process
es being used and allows consumers in making their consumption choices to reflect the satisfaction that they derive from the presumed realization of higher labor standards internationally.26 When labor standards are considered to be public goods, there will be a need for governmental policies. What is important is that these various private and public actions can be carried out without the coercion that may be involved when efforts are made internationally to influence governments to change their domestic labor-market policies.27 25 See Brown, Deardorff, and Stern (2002) for an elaboration of the case to be made for voluntary codes of conduct. 26 Aggarwal (1995, pp. 39-40) cites the example of the Child Labor Coalition, which was formed in 1989 by several religious, human rights, and union groups for the purpose of informing consumers in high-income countries about child labor conditions used in producing goods such as rugs in South Asia. The Coalition has sponsored the so-called Rugmark campaign that provides producers with a certifying label that they can attach to their exports indicating that they do not employ child labor. It has been suggested that the Rugmark labeling system could be extended to clothing and other products. In this connection, see U.S. Department of Labor (1996) for a report on codes of conduct for the U.S. apparel industry. Voluntary codes of conduct in the apparel industry have become increasingly common since the early 1990s. A recent example is the U.S. Presidential task force agreement to "end" apparel sweatshops worldwide and give a seal of approval to companies that comply with the code of conduct. The Fair Labor Association (FLA) was created to implement this code of conduct. As reported in the June 10, 2003 Financial Times by Maitland (p. 8), the FLA audited nearly 50 MNE factories globally in 2001-02 and reported a number of abuses that have been or are being corrected. For further analysis of the design and objectives of the FLA and the Worker Rights Consortium (WRC), both of which have been actively monitoring labor rights in developing countries on behalf of a large number of U.S. universities and colleges that sell MNE produced clothing bearing their academic logos, see Brown, Deardorff, and Stern (2002). 27 It should be noted, however, that in a "quiet maneuver" led by Representative Bernard Sanders, (Ind-VT) and Senator Tom Harkin (D-IA), the 105th U.S. Congress enacted the Bonded Child Labor Elimination Act, and it was signed by President Clinton on October 10, 1997. The Act is designed to prohibit imports of goods made by indentured child laborers, that is, children who are sold into bondage by their parents and who must work for an extended period of time to gain their freedom. But it is not clear how this Act is being enforced, and it is unlikely in any case to effect significant improvement in the conditions of poverty that are characteristic of the families involved.
20 VI. Conclusions and Implications for Policy The motivation for this paper has been to consider whether international labor standards should be incorporated into trade agreements. While an argument can be made for devising WTO rules and disciplines to improve core labor standards in developing countries, the feasibility of effective action is limited because of the diversity of labor standards in countries with differing national characteristics, policies, and institutions.28 Furthermore, the theoretical and empirical literature summarized above suggests no compelling rationale for the international enforcement and harmonization of labor standards. What then should be done on the global level? The responsibility for addressing issues of international labor standards has historically been assigned to the ILO. But the ILO has often been criticized because it lacks a mechanism for enforcement of discipline to raise labor standards and because at times it has espoused an interventionist social agenda. While these criticisms may be valid, they miss the point in my judgment. If one looks at the economic development of the United States, Western Europe, Japan and other advanced industrialized countries over the past century, it is evident that the real incomes of workers have increased dramatically and that the conditions of work have improved concomitantly. To achieve these improvements in labor standards has required an active role for government together with broad public support in individual nations. In recent decades, there have been similar improvements in a substantial number of developing countries, especially in East and Southeast Asia as well as in Latin America. What the historical record suggests therefore is that it is not through the external enforcement of labor standards that improvements have been realized, but through internal economic and social development. With sufficient encouragement and increased Financial support
, the ILO can provide a multilateral forum that would serve to strengthen its role and authority in pursuing improved labor standards internationally. While the United States and some other industrialized countries have tried to link trade and 28 See Brown, Deardorff, and Stern (2001) for an analysis of the pros and cons of linking trade and labor standards in the WTO. They argue against such linking because the sanctioning procedures involved in dealing with alleged violations of labor standards in developing countries may be dominated by protectionist interests in the United States and other developed countries.
21 labor standards in the WTO, their efforts have been unsuccessful. The challenge then is to reinforce the institutional role for which the ILO has been designed.29 If the responsibility for monitoring and helping developing countries to improve their labor standards is centered in the ILO, there is no obvious case to be made for the United States and other industrialized countries to incorporate labor standards issues into their national and regional trade policies and trade agreements. It has to be acknowledged nonetheless that adherence to certain specified labor standards has been made a condition in U.S. and EU preferential trade arrangements, including the GSP, the labor side agreement in the NAFTA and in bilateral U.S. FTAs, and in other U.S. legislation and programs. It has not been demonstrated, however, that these various arrangements have led to significant increases in the economic welfare of developing countries and improvement of their labor standards. Indeed, the interests of developing countries might be better served if the financial and technical assistance provided by the United States and other industrialized countries were more effectively targeted to reduce poverty and if the industrialized countries dismantled their import barriers to encourage greater exports by developing countries. In the final analysis, there is pervasive historical evidence suggesting that as developing countries achieve higher levels of per capita incomes and in the process institute pro-active social policies, conditions of work and workers' rights can be significantly improved. 29 In this connection, see Srinivasan (1995, 1997) and Brown (2001).
22 References Aggarwal, Mita. 1995. "International Trade, Labor Standards, and Labor Market Conditions: An Evaluation of the Linkages," USITC, Office of Economics Working Paper
No. 95-06-C (June). Brenton, Paul and Miriam Manchin. 2003. "Making EU Trade Agreements Work: The Role of Rules of Origin," The World Economy 26:755-69. Brown, Drusilla K. 2000. "International Trade and Core Labour Standards: A Survey of the Recent Literature," labour market
and Social Policy, Occasional Paper
s No. 43 (October), Paris: OECD. Brown, Drusilla K. 2001. "Labor Standards: Where Do They Belong on the International Trade Agenda?" Journal of Economic Perspectives 15:89-112. Brown, Drusilla K., Alan V. Deardorff, and Robert M. Stern. 1996. "International Labor Standards and Trade: A Theoretical Analysis," in Jagdish N. Bhagwati and Robert E. Hudec (eds.), Harmonization and Fair Trade: Prerequisites for Free Trade? Cambridge: MIT Press. Brown, Drusilla K., Alan V. Deardorff, and Robert M. Stern. 2001. "Pros and Cons of Linking Trade and Labor Standards," in Douglas Nelson (ed.), The Political Economy of Policy Reform, in process. Brown, Drusilla K., Alan V. Deardorff, and Robert M. Stern. 2002. "The Effects of Multinational Production on Wages and Working Conditions in Developing Countries," in Robert E. Baldwin and L. Alan Winters (eds.), Challenges to Globalization. Chicago: University of Chicago Press, forthcoming. Charnovitz, Steve. 1986. "The Influence of International Labor Standards on the World Trading Regime: A Historical Review," International Labour Review 126:565-84. Commission of the European Communities. 1990. The Community Charter of Fundamental Social Rights for Workers. European File 6/90. Brussels: Commission of the European Communities. Commission of the European Communities. 2000. "EURO-MED AGREEMENT establishing an association between the European Communities and their Member States, of the one part, and the Kingdom of Morocco, of the other part," Official Journal of the European Communities, L70/1 seq. (March 18). Consumer Unity and Trust Society (CUTS). 2003. "Directions on Trade and Labor Standards" [www.cuts.org]. De Boer, Elizabeth and Gilbert R. Winham. 1993. "Trade Negotiations, Social Charters, and the NAFTA," in Jonathan Lemco and William B.P. Robson (eds.), Ties Beyond Trade: Labor and Environmental Issues under the NAFTA. Canadian-American Committee: C.D. Howe Institute (Canada) and National Planning Association (U.S.A.). Elliot, Kimberley Ann and Richard B. Freeman. 2003. Can Labor Standards Improve Under Globalization? Washington, D.C.: Institute for International Economics. European Union. 2003. Generalized System of Preferences: User's Guide to the European Union's Scheme of Generalised Tariff Preferences (February) [http://europa.eu.int/comm./trade/gsp/index. en.htm].
23 Fields, Gary. 1995. Trade and Labour Standards: A Review of the Issues. Paris: OECD. Freeman, Richard B. 1994. "A Hard-Headed Look at Labor Standards," in U.S. Department of Labor, Bureau of International Labor Affairs, International Labor Standards and Global Economic Integration: Proceedings of a Symposium. Washington, D.C.: U.S. Department of Labor. Howse, Robert. 2003. "Back to Court after Shrimp/Turtle: India's Challenge to Labor and Environmental Linkages in the EC Generalized System of Preferences," in process. International Labour Organization (ILO). 1988. Human Rights: A Common Responsibility. International Labour Conference, 75th Session, 1988. Geneva: ILO. Krueger, Alan. 1997. "International Labor Standards and Trade," Annual World Bank Conference on Development Economics 1996. Washington, D.C.: The World Bank. Lemco, Jonathan and William B.P. Robson (eds.). 1993. Ties Beyond Trade: Labor and Environmental Issues under the NAFTA. Canadian-American Committee: C.D. Howe Institute (Canada) and National Planning Association (U.S.A.). Lyle, Faye. 1991. "Worker Rights in U.S. Policy," Foreign Labor Trends, 91-54, U.S. Department of Labor, Bureau of International Labor Affairs. Washington, D.C.: U.S. Government Printing Office. Maitland, Alison. 2003. "Big brands come clean on sweatshop labour," The Financial Times (June 10, p. 3). Office of the United States Trade Representative (USTR). 2002. "U.S.-Cambodian Textile Agreement Links Increasing Trade with Improving Workers' Rights," Press Release 02-03. Organization for Economic Cooperation and Development
(OECD). 1996. Trade, Employment and Labour Standards: A Study of Core Workers' Rights and International Trade. Paris: OECD. Organization for Economic Cooperation and Development (OECD). 2000. International Trade and Core Labour Standards. Paris: OECD. Rodrik, Dani. 1996. "Labor Standards in International Trade: Do They Matter and What Do We Do About Them?" In Robert Lawrence, Dani Rodrik, and John Whalley (eds.), Emerging Agenda for Global Trade: High Stakes for Developing Countries. Washington, D.C.: Overseas Development Council. Sapir, Andrй. 1995a. "Symposium: Does Free Trade Require Fair Trade and Harmonisation? Introduction," The World Economy 18:741-44. Sapir, Andrй. 1995b. "The Interaction Between Labour Standards and International Trade Policy," The World Economy 18:791-803. Sapir, Andrй. 1996. "Trade Liberalization and the Harmonization of Social Policies: Lessons from European Integration," in Jagdish N. Bhagwati and Robert E. Hudec (eds.), Fair Trade and Harmonization: Prerequisites for Free Trade? Cambridge: MIT Press.
24 Srinivasan, T. N. 1995. "International Trade and Labour Standards," in Pitou van Dijck and Gerrit Faber (eds.), Challenges to the New World Trade Organization. Amsterdam: Martinus Nijhoff/Kluwer. Srinivasan, T. N. 1996. "Post-Uruguay Round Issues for Asian Member Countries," in Asian Development Review 14:1-43. Srinivasan, T. N. 1998. "Trade and Human Rights," in Alan V. Deardorff and Robert M. Stern (eds.), Constituent Interests and U.S. Trade Policies. Ann Arbor: University of Michigan Press. U.S. Department of Labor, Bureau of International Labor Affairs. 1996. The Apparel Industry and Codes of Conduct: A Solution to the International Child Labor Problem. Washington, D.C.: U.S. Department of Labor. U.S. Department of Labor, Bureau of International Labor Affairs. 2003a. "North American Agreement on Labor Cooperation: A Guide," April 13 [www.dol.gov/ilab]. U.S. Department of Labor, Bureau of International Labor Affairs. 2003b. "Status of Submissions," April 13 [www.dol.gov/ilab]. U.S. Department of the Treasury, Office of the Undersecretary for International Affairs. 2003. 2002 Report to Congress on Labor Issues and the International Financial Institutions, Washington, D.C., March 31. Williams, Frances. 1996. "Singapore deals give a boost to WTO," The Financial Times (December 16, p. 8).