International trade and labor standards: A proposal for linkage

Cornell International Law Journal

Volumn 39, Issue 3, 2006, Pages 545-639

  Barry, Christian ^a   Reddy, Sanjay G ^b  

^a AUSTRALIAN NATIONAL UNIVERSITY   (Australia)

^b Och Spine at New York Presbyterian Hospitals   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 34047155526     PISSN: 00108812     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Conference Paper

References (262)

1. Other such proposals have been made, all of which share some of the features of the proposal sketched below. E.g., Daniel S. Ehrenberg, From Intention to Action: An ILO-GATT/WTO Enforcement Regime for International Labor Rights, in HUMAN RIGHTS, LABOR RIGHTS, AND INTERNATIONAL TRADE 163, 168 (Lance A. Compa & Stephen F. Diamond eds., 1996);
2. Int'l Confederation of Trade Unions, Building Workers' Human Rights into the Global Trading System 66 (1999), http://www.icftu.org/www/english/ els/escl99BWRGTS.pdf [hereinafter ICFTU (1999)];
3. Pharis J. Harvey et al., Developing Effective Mechanisms for Implementing Labor Rights in the Global Economy, Workers in the Global Economy § II(A), www.laborrights.org/projects/globalecon/ilrf/index.html (last visited Dec. 13, 2006).
4. It is possible that other valued objectives, such as promoting environmental quality or respect for human rights norms more broadly may also be promoted through linkage. We take no stand on these issues here.
5. We are assuming that global institutional reforms that promote better working conditions and living standards for less advantaged persons in the world without placing significant burdens on more advantaged persons advance the ends of justice. Widely varying conceptions of justice would affirm this view.
6. A number of preferential trading agreements, e.g. NAFTA, CAFTA, United States-Cambodia trade agreement, United States-Jordan trade agreement, U.S.-Chile trade agreement, contain provisions regarding labor standards. See also Tobias Buck, EU to Offer Rewards to 'Good' Poor Countries, Financial Times, Oct. 20, 2004.
7. Specifically, we do not assume that free trade is always the policy that maximizes the gains from trade.
8. We deliberately formulate this proposition abstractly to accommodate the broad range of views that are present in the literature.
9. Although the way in which individual advantage is conceived will undoubtedly influence the specific policies and institutional arrangements that are deemed desirable, the arguments for and against linkage discussed in this essay are largely independent of the choice of a particular conception, within reasonable bounds of variation. See, e.g., UTILITARIANISM AND BEYOND (Amartya Sen & Bernard Williams eds., 1982);
10. QUALITY OF LIFE (Martha C. Nussbaum & Amartya Sen eds., 1993);
11. JOHN RAWLS, A THEORY OF JUSTICE (1971) [hereinafter RAWLS (1971)], for a discussion of different conceptions of individual advantage.
12. INTERNATIONAL LABOR ORGANIZATION, DECLARATION ON FUNDAMENTAL PRINCIPLES AND RIGHTS AT WORK § 2(a)-(d) (June 1998), available at http://www.ilo.org/public/english/standards/relm/ilc/ilc86/com-dtxt.htm (last visited Dec. 13, 2006).
13. It may be helpful to contrast proposals to further the achievement of basic labor standards with proposals to further the achievement of labor standards as such. Whereas the former are concerned with minimally adequate labor standards (however defined), the latter may seek the attainment of still higher labor standards regardless of the levels already attained. Proposals of the latter kind are not our focus here.
14. Many recent critics of linkage have characterized the idea of linkage much more narrowly than we have. Arvind Panagariya, for example, has claimed that "[t]he trade-labor link effectively requires countries to raise standards to the level desired by importing countries or face trade sanctions by the latter. It is argued that a country that adheres to higher labor standards within its national boundaries has the moral right to suspend trade with another country that does not adhere to equally high labor standards." ARVIND PANAGARIYA, LABOR STANDARDS AND TRADE SANCTIONS: RIGHT END WRONG MEANS 5 (Jan. 15-16, 2001), available at http://www.columbia. edu/-ap2231/Policy%20Papers/Hawaii3-AP.pdf (last visited Dec. 13, 2006) [hereinafter Panagariya (2001)]. Clearly it is possible to oppose linkage as Panagariya has characterized it without opposing linkage as we understand it. Indeed, few (if any) linkage proponents would endorse linkage as Panagariya understands it, and in what follows we express full agreement with the criticisms that have been advanced by Panagariya and others of this type of linkage. Similarly, Srinivasan frames disagreement about linkage in terms of differing views regarding whether diversity in labor standards among nations is legitimate.
15. T.N. SRINIVASAN, DEVELOPING COUNTRIES AND THE MULTILATERAL TRADING SYSTEM: FROM GATT TO THE URUGUAY ROUND AND THE FUTURE (2d ed. 2000). This is misleading, since many linkage proposals (such as the one sketched below) not only allow that diversity of labor standards amongst nations is legitimate, but insist that attempting to "harmonize" them under present conditions would be illegitimate. See ICFTU (1999), supra note 1, at 31;
16. Pharis J. Harvey et al., Developing Effective Mechanisms for Implementing Labor Rights in the Global Economy, WORKERS IN THE GLOBAL ECONOMY § II(D).
17. The distinction between sanctioning and offering additional opportunities depends on having specified a baseline. Such a baseline can be identified on the basis of various (empirical and normative) criteria. A common misunderstanding is that it can only be defined on the basis of empirical considerations, such as whether a measure restricts or expands the opportunities possessed ex ante by the parties.
18. A system that provides a country with additional permissions (e.g., to impose tariffs on foreign products) if it adequately promotes labor standards is also a form of linkage as defined by Proposition L since the rights to trade that it accords to other countries are made conditional on the promotion of labor standards. We do not focus on linkage of this kind below but note that proposition L accommodates it.
19. n). It does not necessarily follow that this objective function is additively separable. As a result, it is often impossible to evaluate the marginal contribution of each attainment to the master-goal, and thus of the optimal combination of attainments to be pursued, without determining the extent of other attainments.
20. This agent could be individual or "collective," i.e., a group or an organization.
21. Rights-linkage as we understand it is linkage between the possession of rights and the undertaking of certain conduct (the exercise of their rights in a specific way) and is distinct from another type of linkage that might also deserve the term rights-linkage: that in which an agent's possession of one right is made conditional on his or her possession of some other right.
22. Possibly more than one: agency-linkage requires that at least some agents be charged with promoting distinct ends, but it does not require that each end be promoted by only one agent.
23. We are referring here to what is needed to demonstrate Proposition L to those who accept Proposition O. Different arguments may be necessary to persuade those who reject Proposition O to accept Proposition L.
24. See, e.g., CONSUMER UNITY & TRUST SOC'Y, THIRD WORLD INTELLECTUALS AND NGOs' STATEMENT AGAINST LINKAGE ¶¶ 1, 3 (Sept. 6, 1999), http://cuts-international.org/twin-sal.htm [hereinafter TWIN-SAL].
25. Some individuals have been hostile to such conditionality in both the domestic and international contexts, e.g., libertarians such as Robert Nozick, and consequentialists who have favored the "unfettered" free market such as Milton Friedman, but this is distinctly the view of a minority. Dani Rodrik notes a useful point that is related to, but distinct from, that which we raise above: trade may be viewed (in the abstract) as a technology that allows inputs to be transformed into outputs. Dani Rodrik, Labor Standards in International Trade, in EMERGING AGENDA FOR GLOBAL TRADE 35, 41(Robert Z. Lawrence et al. eds., 1996) [hereinafter Rodrik (1996)]. It is widely agreed that production technologies may not be employed in the domestic economy without regard to whether or not they violate minimal labor standards. Those who advocate some restrictions on the production "technologies" that may be used in the domestic economy but reject such restrictions in the global economy must, Rodrik argues, make clear why these two spheres should not be treated in a like manner. Id. at 42. It may, however, be argued that this comparison obscures the potentially morally relevant distinction between production at home and production abroad. Some moral conceptions may disvalue violations of minimal labor standards that affect workers within a country to a greater extent than they disvalue such violations abroad.
26. Although it is certainly true, as Richard Freeman has argued, that the debate about trade and labor standards is often simply " 'one of a set of running battles between diose who believe the unfettered market can do no wrong and those who believe govern-mental regulations can make things better,'" Rodrik (1996), supra note 19, at 37 (quoting Richard Freeman, A Hard-Headed Look at Labor Standards, in INTERNATIONAL LABOR STANDARDS AND ECONOMIC INDEPENDENCE 80 (Werner Sengenberger & Duncan Campbell eds., 1994)), it nevertheless does seem that many linkage critics heartily affirm the role of various domestic regulations of labor and product markets.
27. In particular, it is argued that influential interests (such as labor unions and employers in some industries) in developed countries would benefit from reduced competition from countries with low-cost labor and therefore press for such opportunistic misuse. See, e.g., GOTE HANSSON, SOCIAL CLAUSES AND INTERNATIONAL TRADE 34-38 (1983);
28. Jagdish Bhagwati, Policy Perspectives and Future Directions, in INTERNATIONAL LABOR STANDARDS AND GLOBAL ECONOMIC INTEGRATION 57, 60 (Gregory K. Schoepfle & Kenneth A. Swinnerton eds., 1994);
29. JAGDISH BHAGWATI, IN DEFENSE OF GLOBALIZATION 122-34 (2004) [hereinafter Bhagwati (2004)];
30. JAGDISH BHAGWATI, THE WIND OF THE HUNDRED DAYS 274 (2000) [hereinafter Bhagwati (2000)];
31. Drussila K. Brown, Labor Standards: Where do They Belong on the International Trade Agenda?, 15 J. ECON. PERSPECTIVES 89, 102-03 (2001) [hereinafter Brown (2001)];
32. T.N. Srinivasan, International Trade and Labor Standards from an Economic Perspective, CHALLENGES TO THE NEW WORLD TRADE ORGANIZATION 219, 239 (1996); Panagariya (2001), supra note 10 at 9;
33. Kaushik Basu, Child Labor, 37 J. ECON. LITERATURE 1083, 1092 (1999) [hereinafter Basu (1999)];
34. GEORGE TSOGAS, LABOR REGULATION IN A GLOBAL ECONOMY 27-28 (2001);
35. Jose M. Salazar-Xirinachs, The Trade-Labor Nexus, 3 J. INT'L ECON. L. 377, 380-81 (2000);
36. Martin Khor, Director of Third World Network, Rethinking Liberalisation and Reforming the WTO § 5 (Jan. 28, 2000), available at http://www.twnside.org.sg/title/davos2-cn.htm (last visited Dec. 13, 2006);
37. MUCHKUND DUBEY, SOCIAL CLAUSE: THE MOTIVE BEHIND THE METHOD, available at http://www.aidc.org.za/?q=book/view/71&PHPSESSID= 4be77ce78ffc2de752f57c783c9c1145 (last visited Dec. 13, 2006);
38. Gregory Shaffer, WTO Blue-Green Blues, 24 FORDHAM INT'L L.J. 608, 621 n.44 (2000).
39. See, e.g., TWIN-SAL, supra note 18, at 5; Bhagwati (2000), supra note 21, at 320; Jagdish Bhagwati, Free Trade and Labor (2001), http://www.columbia.edu/-jb38/ft_lab.pdf [hereinafter Bhagwati (2001)].
40. See Nicholas D. Kristof & Sheryl WuDunn, Two Cheers for Sweatshops, N.Y. TIMES MAG., Sept. 24, 2000, at 70;
41. Vivek H. Dehejia & Yiagadessen Samy, Trade and Labour Standards - Theory, New Empirical Evidence, and Policy Implications 20-25 (CESifo, Working Paper No. 830, 2002).
42. See, e.g., CARNEGIE ENDOWMENT FOR INTERNATIONAL PEACE, BREAKING THE LABOR-TRADE DEADLOCK 3 (Carnegie Endowment, Working Paper No. 17, 2001).
43. See, e.g., KEITH E. MASKUS, SHOULD CORE LABOR STANDARDS BE IMPOSED THROUGH INTERNATIONAL TRADE POLICY 33-35 (World Bank, Working Paper No. 1817, 1997), available at http://ssrn.com/abstract=44605 (last visited Dec. 13, 2006); Brown (2001), supra note 21, at 100.
44. See LYN SQUIRE & SETHAPUT SUTHIWART-NARUEPUT, THE IMPACT OF LABOUR MARKET REGULATIONS 7-11 (World Bank Pol'y Dept., Working Paper No. 1418, 1995); Panagariya (2001), supra note 10, § 2.5; Carnegie Endowment, supra note 24, at 3;
45. Pranab Bardhan, Some Up, Some Down, 26 BOSTON REVIEW (Feb./Mar. 2001);
46. PRANAB BARDHAN, SOCIAL JUSTICE IN THE GLOBAL ECONOMY 13 (2000).
47. See, e.g., CARNEGIE ENDOWMENT, supra note 24, at 3; Bhagwati (2000), supra note 21, at 143-44.
48. In particular, it will be undesirable according to those views which hold that the level of advantage is directly or indirectly influenced by the extent of inequalities (for example, in working conditions or command over resources).
49. T.N. Srinivasan, International Labor Standards Once Again!, in INTERNATIONAL LABOR STANDARDS AND GLOBAL ECONOMIC INTEGRATION 34, 37 (1994) [hereinafter Srinivasan (1994)]; see also Basu (1999), supra note 21, at 1093; MASKUS, supra note 25, at 22.
50. We deal in this section with the outcomes that are expected to arise as a result of alternative institutional arrangements. We are concerned here with "comparative static" comparisons of distinct outcomes. Comparisons of the cost of transition from the status quo to distinct institutional arrangements are dealt with below, under the heading of feasibility.
51. Note that the set of "superior means" (SM) arguments and the set of "inconsequential/self-defeating" (I/S) arguments overlap, and it is not necessarily the case that one is contained in the other. Any argument that recognizes linkage as being at least partially effective in attaining its goals, though less effective than the alternatives, is an SM argument and not an I/S argument. Any argument that views linkage as unable to achieve its aims, but also views all other feasible policy interventions as being unable to achieve those aims, is an I/S argument and not an SM argument. The "tragic sense of life" underlying this perspective is one that does not appear to be prevalent in the debate on linkage. Most I/S arguments are also SM arguments.
52. E.g. Jagdish Bhagwati, Trade Liberalisation and "Fair Trade" Demands, 18 WORLD ECONOMY 745, 757 (1995); Bhagwati (2000), supra note 24, at 160-62; Bhagwati (2001), supra note 22, at 5 ("[A] good tongue-lashing . . . can unleash shame.");
53. Jagdish Bhagwati, Free Trade: Why AFL-CIO, the Sierra Club and Congressman Gephard Should Like It, 43 AMERICAN ECONOMIST 3, 11 (1999) [hereinafter Bhagwati (1999)]; see also MASKUS, supra note 25, at 67;
54. Ajit Singh & Ann Zammit, Labor Standards and the "Race to the Bottom", 20 OXFORD R. OF ECON. POL'Y 85, 102 (2004).
55. Archon Fung et al., Realizing Labor Standards, in CAN WE PUT AN END TO SWEATSHOPS? 3, 4-5 (Archon Fung et al. eds., 2001).
56. See, e.g., Rodrik (1996), supra note 19, at 60; KIMBERLY ANN ELLIOT & RICHARD B. FREEMAN, CAN LABOR STANDARDS IMPROVE UNDER GLOBALIZATION? 27-48 (2003); Panagariya (2001), supra note 10, § 3.2; MASKUS, supra note 25, at 21.
57. Srinivasan (1994), supra note 29, at 35.
58. Bhagwati (1999), supra note 32, at 10; Bhagwati (2000), supra note 24, at 277-78.
59. Professor Jagdish Bhagwati writes, "[Linkage] wind[s] up harming both trade liberalization (which is the true objective of the WTO) and advancement of the social and moral agendas. . . . The underlying reason for such an unsatisfactory outcome is that you are trying to kill two birds with one stone. Generally, you cannot. . . . [By] trying to implement two objectives, the freeing of trade and the advancing of social and moral agendas, through one policy instrument such as WTO, you will undermine both. You will miss both birds." Bhagwati (2000), supra note 24, at 277-78. See also T.N. SRINIVASAN, DEVELOPING COUNTRIES AND THE MULTILATERAL TRADING SYSTEM (2d ed. 2000) [hereinafter Srinivasan (2000)].
60. Bhagwati (2001), supra note 22, at 4; Bhagwati (1999), supra note 32, at 10-11; Bhagwati (2000), supra note 24, at 278. See also Srinivasan (2000), supra note 37, at 72-73.
61. See Kyle Bagwell & Robert W. Staiger, The Simple Economics of Labor Standards and the GATT, in SOCIAL DIMENSIONS OF U.S. TRADE POLICY 195, 197 (Alan V. Deardorff & Robert M. Stern eds., 2000). See generally Srinivasan (2000).
62. See Bagwell & Staiger, supra note 39; see also Kyle Bagwell & Robert W. Staiger, The WTO as a Mechanism for Securing Market Access Property Rights, 15 J. ECON. PERSPECTIVES 69 (2001).
63. In principle, there is a corresponding rule involving export items and export subsidies, which the authors do not discuss.
64. See, e.g., Kristof & WuDunn, supra note 23. Kristof and WuDunn's article is cited at lengdi approvingly in Bhagwati (2004), supra note 21, at 175. See also Basu (1999), supra note 21, at 1114.
65. For examples of critics who have made that argument, see Maskus, supra note 25, at 49; Brown (2001), supra note 21, at 105-06; Arvind Panagariya, Trade-Labour Link: A Post-Seattle Analysis, in GLOBALIZATION UNDER THREAT 101, 110 (Zdenek Drabek ed., 2001) [hereinafter Panagariya (2001b)].
66. See, e.g., Singh & Zammit, supra note 32, at 95-96.
67. Kristof & WuDunn, supra note 23; Bhagwati (2004), supra note 21, at 175; see also Carnegie Endowment, supra note 24, at 3.
68. See Bhagwati (2004), supra note 21, at 172-73 (citing the hypothesis that U.S.-run factories provide higher wages because they have higher productivity).
69. Panagariya (2001), supra note 10, §2.1.
70. A recent Cato Institute publication puts this point rather gushingly as follows: The threat of using trade restrictions to advance human rights is fraught with danger. Free trade is itself a human right and rests on an individual's rights to life, liberty, and property - rights the U.S. Founding Fathers regarded as inalienable and self-evident. When the federal government closes U.S. markets to countries with governments that deny their citizens certain civil liberties, it robs those citizens of one more freedom and undermines the market dynamic that in the end is the best instrument for creating wealth and preserving freedom. . . . Free Trade Is a Human Right: The proper function of government is to cultivate a framework for freedom by protecting liberty and property, including freedom of contract (which includes free international trade) - not to use the power of government to undermine one freedom in an attempt to secure others. The right to trade is an inherent part of our property rights and a civil right that should be protected as a fundamental human right. The supposed dichotomy between the right to trade and human rights is a false one. Market exchange rests on private property, which is a natural right. As moral agents, individuals necessarily claim the right to liberty and property in order to live fully and to pursue their interests in a responsible manner. The freedom to act without interference, provided one respects the equal rights of others, is the core principle of a market economy and the essence of human rights. Without private property and freedom of contract, other rights - such as free speech and religious freedom - would have little meaning, because individuals would be at the mercy of the state. The human-rights fabric is not made stronger by unraveling economic liberties in the hope of enhancing other liberties. Protectionism violates human rights. It is an act of plunder that deprives individuals of their autonomy - an autonomy that precedes any government and is the primary function of just governments to protect. . . . The danger of buying into the argument that restricting trade with China will increase human rights is that such an argument diminishes the significance of the moral case for free trade, politicizes economic life, and weakens the market-liberal vision - a vision that needs to be strengthened in order to protect civil society and human liberty. James A. Dorn, Trade and Human Rights: The Case of China, 16 CATO J. (1996), available at http://www.cato.org/pubs/journal/cj16n1-5.html (last visited Dec. 13, 2006).
71. ROBERT NOZICK, ANARCHY, STATE, AND UTOPIA 163 (1974).
72. Personal conversation with Kamal Malhotra.
73. The narrower objective of promoting basic labor standards of those who are employed will in this case allegedly conflict with the broader objective of improving the level of advantage of less advantaged persons more generally. The implicit premise of the argument is that the gain in attaining the latter objective justifies the loss in attaining the former.
74. Srinivasan (2000), supra note 37.
75. This assumption is common in the literature. See, e.g., Basu (1999), supra note 21, at 1100, 1103-04.
76. See, e.g., Bhagwati (2000), supra note 24, at 163; Paul Krugman, One in the Eye with an American Pie, BANGKOK POST, Feb. 17, 2000; Shaffer, supra note 21, at 624-25; Srinivasan (2000), supra note 37.
77. Srinivasan (2000), supra note 37, at 73-77. The literature on "Asian values" is replete with such claims.
78. See, e.g., Brown (2001), supra note 21, at 91.
79. See, e.g., Basu (1999), supra note 21, at 1089.
80. See, e.g., Carnegie Endowment, supra note 24, at 4; Jagdish Bhagwati, The Question of Linkage, 96 AM. J. INT'L L. 126, 128, 131 (2002).
81. See, e.g., Theodore H. Moran, Trade Agreements and Labor Standards 6 (Brookings Institution Policy Brief No. 133, 2004), available at http://www.southern.org/content/sc/sc.asp?edition_date=062204.txt (last visited Dec. 13, 2006).
82. Bhagwati has often cited a Human Rights Watch report on the United States' violation of the right to organize in connection with this claim. See, e.g., Bhagwati (2004), supra note 21, at 177, 192, 247, 251.
83. See, e.g., Mark Levinson, Wishful Thinking, in ARCHON FUNG ET AL., CAN WE PUT AN END TO SWEATSHOPS? 54, 54 (2001).
84. We are not committing ourselves to the view that no imposed institution could under any conditions be legitimate. However, it is our judgment that under present conditions it is highly unlikely that an imposed system of linkage would be legitimate.
85. This idea is associated with the work of Hans Kelsen. Cf. RAWLS (1971), supra note 7, at 567.
86. See Philip Alston, Labor Rights Provisions in U.S. Trade Law, in HUMAN RIGHTS, LABOR RIGHTS, AND INTERNATIONAL TRADE 71, 73-83 (Lance A. Compa & Stephen F. Diamond eds., 1996) for an interesting discussion of the reasons to reject the system of linkage imposed by the United States through its General System of Preferences (GSP).
87. See id. ("[T]he United States is . . . imposing its own, conveniendy flexible and even elastic, standards upon other states."); see also Terry Collingsworth, International Worker Rights Enforcement, in HUMAN RIGHTS, LABOR RIGHTS, AND INTERNATIONAL TRADE 227, 229-33 (Lance A. Compa & Stephen F. Diamond eds., 1996) for a discussion of some of the abuses by the United States of the GSP.
88. The baseline against which these costs can be measured can, of course, be specified in different ways, such as the status quo ex ante or an appropriate counterfactual, for example, what would have occurred pursuant to the previous rules, or some other morally appropriate benchmark.
89. The Kyoto Protocol has a burden-sharing component (Article 11), as did the framework convention on climate change agreed to at the Rio conference in 1992. As Srinivasan describes this later convention, "[It] set explicit goals under which several rich nations agree to emission-level-reduction targets (i.e., to return, more or less, to 1990 levels), whereas the commitments of the poor countries were contingent on the rich nations' footing the bill." Srinivasan (2000), supra note 37. Technical cooperation and financial aid are provided to countries so that they can comply with WTO rules. Bilateral trade agreements, such as the US and Cambodia textile and apparel agreement (Article 10 (E)) also offer examples of this kind. The burden sharing that took place with respect to reforming the labor practices of the Bangladeshi garment sector is discussed in Elliot & Freeman (2003), supra note 34, at 113. The linkage proposals developed by the ICFTU (1999) and the ILRF (2001) both demand burden sharing.
90. Unlike Srinivasan, who seems to view international burden sharing through income transfers and linkage as alternative and mutually exclusive means of promoting labor standards, we view such transfers as essential to any plausible linkage proposal. Srinivasan (2000), supra note 37, at 74. We argue below that there are reasons why a scheme that employs income transfers alone as a means of promoting labor standards will likely be inferior to one that combines these with trade incentives, and that a plausible linkage scheme will combine trade and non-trade incentives. For a discussion of the kinds of complementary policies that may be necessary to combat objectionable forms of child labor, see Drusilla K. Brown et al, Child Labor: Theory, Evidence, and Policy, in INTERATIONAL LABOR STANDARDS 195, 225-37 (Kaushik Basu et al. eds., 2003) [hereinafter Brown (2003)].
91. Cf. ROBERT E. GOODIN, PROTECTING THE VULNERABLE 186 (1985);
92. Peter Singer, Famine, Affluence, and Morality, WORLD HUNGER AND MORALITY 26, 26-27 (William Aiken & Hugh LaFollette eds., 2d ed. 1996);
93. PETER UNGER, LIVING HIGH AND LETTING DIE 62-72 (1996). Even those who reject this claim may have reason to be sensitive to the distributional consequences of proposed institutional reforms if they believe that the present unequal distribution of advantages has emerged from a historical process in which rights that they wish to see respected have been violated.
94. See, e.g., Hillel Steiner, An Essay on Rights 266 ("Redress transfers are redistributions which, very broadly, undo the unjust redistributions imposed by encroachments on rights: they restore just distributions."); THOMAS W. POGGE, WORLD POVERTY AND HUMAN RIGHTS 14 (2002).
95. See, e.g., Christian Barry, Applying the Contribution Principle, 36 METAPHILOSOPHY 210, 211-13 (2005). For examples of other principles, see Goodin, supra note 69,
96. and David Miller, Distributing Responsibilities, 9 J. POL. PHIL. 453 (2001).
97. See, e.g., Barry, supra note 70, at 213-14.
98. This concern is not unique to a system for promoting labor standards, as mentioned earlier. Similar issues arise in other areas of international cooperation such as the promotion of environmental standards, as is ably discussed in Raghbendra Jha & John Whalley, Migration and Pollution (2003), available at http://rspas.anu.edu.au/economics/publish/papers/ wp2003/wp-econ-2003-07.pdf. In discussing the conditions under which India might be willing to agree to a system of linkage, Rob Jenkins emphasizes the importance of developed countries sharing substantially in the burdens of linkage (as well as making greater progress in fulfilling their commitments under the Uruguay Round Agreements).
99. Rob Jenkins, India and the Trade-and-Labour-Standards Controversy, 4-7 (2001), available at http://www.gapresearch.org/governance/ The%20Politics%20of%20Trade%20in%20India.pdf.
100. A similar point is made in reference to other developing countries by Kevin Kolben in The New Politics of Linkage: India's Opposition to the Workers' Rights Clause, 13 IND. J. GLOBAL LEGAL STUD. 225-259 (2006).
101. The level and nature of burden sharing required to make a linkage scheme feasible may be different from that required to make it morally legitimate. The level of burden sharing that is adequate will have to be determined in light of both considerations.
102. This is borne out by studies of attitudes toward linkage among trade unions in the South, which show that they possess much greater receptivity to linkage proposals than is widely believed. See the results of the remarkable survey of developing country trade unions reported in Gerard Griffin et al., Trade Unions and the Social Clause: A North South Union Divide? 8-11 (Nat'l Key Ctr. in Indus. Relations, Working Paper No. 81, 2002) (finding that 95% of union members in the global South favored international trade agreements that protect core labor standards), and the sources cited therein, reporting similar conclusions.
103. See Jenkins, supra note 72, at 3; see also Srinivasan supra note 10, at 70.
104. See supra note 63 and accompanying text.
105. For our argument to succeed, it is sufficient to identify one class of proposals for linkage that satisfies the standard objections. It is therefore no embarrassment to fail to identify all the classes of proposals that satisfy the standard objections.
106. Different models can both be observed in the world and in the proposals that have been advanced about how best to promote labor standards. See, e.g., Ehrenberg, supra note 1; G.B. Nath, Linking International Labour Standards with Trade: Implications for India, 41 INDIAN J. OF LABOUR ECON. 1005, 1011 (1998) (contrasting the structural weakness of the ILO with the WTO's capability for enforcement);
107. Rohini Hensman, World Trade and Workers' Rights, 33 ANTIPODE 427, 442-46 (2001); ICFTU (1999), supra note 1, at 53; Harvey et al., supra note 1, §III; ELLIOT & FREEMAN, supra note 34, at 90-92;
108. Kevin Kolben, Trade, Monitoring, and the ILO: Working to Improve Conditions in Cambodia's Garment Factories, 7 YALE HUM. RTS. & DEV. L.J. 79 (2004).
109. We draw here on the terminology developed in AMARTYA SEN, THE STANDARD OF LIVING (1987).
110. It does not follow from this rejection, of course, that we are indifferent to the losses of those in the focal group who do worse under the new system. We favor measures that minimize these losses.
111. It should be pointed out that Staiger himself resists the description of his argument as an argument for linkage by attempting to distinguish between the economic rationale and the moral or political rationale for maintaining a floor for labor standards. Robert W. Staiger, A Role for the WTO, in INTERNATIONAL LABOR STANDARDS 273, 277 (Basu et al. eds., 2003). However, it is clear that the argument Staiger provides is an argument for linkage as we define it (in Proposition L);
112. Josh Ederington, Trade and Domestic Policy Linkage in International Agreements, 43 INT'L ECON. REV. 1347, 1361 (2002) (using game theory to show that, at least under certain circumstances, linkage can increase the ability to enforce the domestic provisions of an international agreement, in this case higher labor standards).
113. Bagwell and Staiger have presented an alternative way of avoiding such problems, in which countries are required to abide by specific rules when adjusting their labor standards but are not required to promote them to any specific extent. Bagwell & Staiger, supra note 39, at 225-26. We discuss their proposed possible solution in detail below, presenting reasons that why some of its elements may be beneficially incorporated into a linkage system.
114. "Race to the bottom" is a widespread but unfortunate name for the more general concern that competitive pressures will undermine efforts to secure basic labor standards. It is unfortunate because it suggests that, absent evidence of deterioration of labor standards over time, competitive pressures that undermine efforts to raise labor standards are not present. This conclusion would be false, because even if labor standards were everywhere improving, it would not follow that the threat of being undercut by others with lower labor standards was not exerting downward pressure on labor standards. Indeed, it is entirely consistent with the fact that much more rapid improvements in labor standards might be obtainable were stronger incentives provided to countries to do so. In such a case, there would be (to use Staiger's phrase) a "regulatory chill" but not an observable race to the bottom. The former concept depends on a counterfactual comparison, while the latter depends on a purely empirical one.
115. Such incentives are already part of the European Union's Generalized System of Preferences. See Buck, surpa note 4. It is interesting to note that the linkage proposal of the International Confederation of Free Trade Unions calls for "urgent removal of tariffs and import quotas for least developed countries respecting core labour standards." ICFTU (1999), supra note 1, at 23.
116. See also Jenkins, supra note 72, at 3, 5-7 (predicting that India, which has opposed linkage, might accept it if wealthy countries abide by Uruguay Round agreements that require them to open their markets in certain cases). The Indian Government Commission on Labour Standards and International Trade issued a report expressing an open-minded position on linkage, especially if appropriate supports were to be offered by developed countries. The chair of the commission, Subramaniam Swamy, argued in a subsequent book that India could be a net gainer should linkage be implemented, in part because of the competitive advantages it would gain relative to countries which would be unlikely to adequately promote labor standards. See generally Kolben, supra note 72.
117. It is far from clear that the motivation behind linkage is in fact protectionism. Krueger attempted to examine this assumption by identifying the constituencies whose representatives supported the Harkin Bill in the U.S. Congress, which proposed the imposition of specific trade sanctions on countries exporting goods produced with child labor. He concluded that self-interested material motives were not discernible from the empirical profile of these constituencies. Alan Krueger, Observations on International Labor Standards and Trade 13-23 (Nat'l Bureau of Econ. Research, Working Paper No. 362, 1996) [hereinafter Krueger (1996)]. Krueger's argument is cited in Basu (1999), supra note 21, at 1092, and criticized in Bhagwati (2004), supra note 21, at 244-45. Sandra Polaski argues that where trade treaties have contained a labor clause, it has generally not been misused for protectionist ends.
118. SANDRA POLASKI, TRADE AND LABOR STANDARDS 14 (2003), available at http://www.carnegieendowment. org/pdf/files/Polaski_Trade_English.pdf (last visited Dec. 13, 2006) [hereinafter POLASKI (2003)]. A similar finding is presented by Elliott and Freeman, who discuss in detail the existing evidence on such motivations. ELLIOTT & FREEMAN, supra note 34, at 84. Critics of linkage often argue as if the mere fact that support for linkage may be due in part to protectionist concerns rules out the possibility that it is desirable. This is false, since it is possible that agents may do the right thing for the wrong reasons or from questionable motivations, just as they may do the wrong thing for the right reasons and from pure motivations.
119. As Alan Krueger, The Political Economy of Child Labor, in INTERNATIONAL LABOR STANDARDS: HISTORY, THEORIES AND POLICY 251 (Kaushik Basu, et al. eds., 2003) rightly points out, "[e]ven if international labor standards were motivated by self-interest, they nonetheless may raise welfare in less developed nations. . . . And the converse is also true: even if international labor standards were motivated by humanitarian concerns they may hurt those they are intended to help in developing countries. Of course, the motivations of agents are certainly relevant in forming predictions about how they will act. It may be quite reasonably feared that if protectionist motives drive at least some of those who support linkage, any system of linkage that emerges will likely be used for protectionist purposes. However, even agents with largely self-serving motivations can be expected to comply with a system of rules if it provides them with the right incentives to do so. The WTO is built on the premise that situating trade negotiations within a transparent and negotiated system of rules with a binding and impartial dispute settlement mechanism can promote a fairer world trading system, notwithstanding the often self-seeking motivations of the states who participate in the system. Indeed, it seems implausible that any system of international trading rules can be created that could effectively guard entirely against such opportunistic misuse.
120. Examples include the National Labor Relations Board in the United States and comparable bodies in other countries, the WTO's Dispute Settlement Body (DSB), existing free trade agreements with labor provisions such as the U.S.-Jordan Free Trade Agreement, the North American Agreement on Labor Cooperation, and the Canada-Chile Free Trade Agreement. See POLASKI (2003), supra note 86, at 13-14; Sandra Polaski, Cambodia Blazes a New Path to Economic Growth and Job Creation 14 (Carnegie Endowment for Int'l Peace Paper No. 51, 2004) [hereinafter Polaski (2004)];
121. Sandra Polaski, Protecting Labor Rights Through Trade Agreements: An Analytical Guide, 10 J. INT'L L. & POL'Y 13, 17-20 (2004) [hereinafter Polaski (2004a)].
122. See generally Kolben, supra note 78. Ehrenberg, supra note 1, at 168 proposes that an "Admissibility Committee" comprised of nine members appointed jointly by the ILO and the GATT/WTO and charged with determining whether complaints with respect to the observance of labor standards (specified in terms of eight publicly stated criteria) are admissible for consideration by a linkage enforcement scheme. See also ICFTU (1999), supra note 1, at 52-53 (describing how the ILO could report to the WTO on violations of core labor standards); Harvey et al., supra note 1, §III (proposing that the ILO interact with independent monitors of labor violations).
123. Cf. Hensman, supra note 78, at 433. Indeed, many prominent critics of linkage defend the WTO on similar grounds.
124. It may also be feared that the standard of proof required for establishing that labor standards violations have taken place may be set so low as to make it easy for rich countries to establish claims that labor standards have been violated and difficult for poor countries to deny such claims.
125. This is consistent with the widely discussed concept of the progressive realization of human rights. Existing international treaties do often make allowances for the level of development of countries. Examples include the Kyoto protocol and the TRIPS agreement.
126. Indeed, the Ehrenberg, ICFTU, and Harvey, Collingsworth, and Athreya proposals cited above all insist on such a requirement.
127. Even the WTO DSB, which has not arisen from a truly transparent and participatory process, often makes decisions that are not in the interests of member countries, including the most powerful. See, e.g., PANEL REPORT, UNITED STATES - TRANSITIONAL SAFEGUARD MEASURE ON COMBED COTTON YARN FROM PAKISTAN, WT/DS192/R (May 31, 2001).
128. For a description of the recent ruling on cotton, which Brazil won against the United States, see U.S. Loses Cotton Fight with Brazil, BBC NEWS, Mar. 3, 2005, available at news.bbc.co.uk/2/hi/business/ 4316671.stm (last visited Dec. 13, 2006).
129. Cf. Gene M. Grossman & Alan B. Krueger, Environmental Impacts of a North American Free Trade Agreement, THE MEXICO-U.S. FREE TRADE AGREEMENT 13, 48 (1993). Alan Krueger emphasizes this point, arguing that for this reason it would be difficult for developed countries to exploit labor standards to achieve protectionist ends. Krueger (1996), supra note. 86, at 12. See also POLASKI (2003), supra note 86, at 13 (stating that currently competition for manufacturing occurs only among developing countries, since textile, apparel, footwear, electronics, etc. are produced almost exclusively in low-wage countries). Basu (1999), supra note 21, at 1114 acknowledges the desirability of collective coordination of standards. Robert Pollin et al. find that increasing the cost of labor, by increasing wages or raising labor standards, does not consistently lead to job losses.
130. Robert Pollin et al., Global Apparel Production and Sweatshop Labour: Can Raising Retail Prices Finance Living Wages?, 28 CAMBRIDGE J. ECON. 153, 156-60 (2004).
131. See Krueger (1996), supra note 86, at 12; cf. Grossman & Krueger, supra note 93, at 48.
132. Jagdish N. Bhagwati & V.K. Ramaswamy, Domestic Distortions, Tariffs and the Theory of Optimum Subsidy, 71 J. POL. ECON. 44, 44-50 (1963).
133. See Sanjay G. Reddy, Pareto-Improving International Labor Standards Agreements: A Simple Model (Working Paper, 2006), available at http://ssrn.com/abstract=930113.
134. Note that free trade with the wage subsidy and enhanced labor standards and free trade without the wage subsidy or enhanced labor standards are each Pareto superior to autarky (in principle) but that the two free trade alternatives may not be Pareto comparable, because some may do strictly better under the former scheme, and others may do strictly better under the latter scheme, depending on the nature and extent of the ex-post taxes and transfers that are implemented. The Pareto ranking of free trade and autarky depend on the (1) existence or absence of a domestic distortion; and (2) correction of a domestic distortion at the source or failure to correct the domestic distortion at the source. The ranking is as follows (assuming the existence of efficient tax and transfer instruments and specializing, for simplicity, to the case of production distortions): (1) Without a domestic distortion: free trade is Pareto superior to autarky, because there are gains from trade which can be redistributed in a lump sum fashion. This is because each unit produced for export satisfies the condition that the world price exceeds the true domestic cost of production; (2) With a domestic distortion; (2.1) If the domestic distortion is corrected at the source then free trade is Pareto superior to autarky, because there are gains from trade which can be redistributed in a lump sum fashion. Again, this is because each unit produced for export satisfies the condition that the world price exceeds the true domestic cost of production. [The correction of the domestic distortion at source ensures that producers' perceived costs are equal to the true domestic costs of production]; (2.2) If the domestic distortion is not corrected at the source, then it is ambiguous whether or not free trade is Pareto superior to autarky, since it is ambiguous whether there are gains from trade. Whether or not there are gains from trade will depend on the nature and extent of the distortion. Consider the following illustrative example. A country produces an exportable commodity (say oil) with a great deal of attendant pollution per unit produced. This externality is not internalized. [If it were, say through an appropriate Pigouvian tax (case 2.1), then the marginal unit produced would represent its true domestic (social) cost of production. In that case there would be gains from trade, since oil would be produced for export if and only if the world price were greater than or equal to the true domestic cost of production. Since the externality is not internalized (case 2.2), oil will be exported as long as the world offer price is greater than or equal to the domestic producer cost. However, since the marginal domestic producer cost is lower than the true marginal domestic cost, there will be a social cost created by these units produced for export. At the margin, the revenue garnered by the sale of these units on the world market will exceed the true domestic cost of their production. Whether there are gains from trade will depend on whether, on average, for the additional units sold due to trade opening the average true domestic cost of their production is higher or lower than the world price. In principle, whether this is so is ambiguous and depends on the level of the world price and the nature and extent of the domestic distortion.
135. Rohini Hensman reports that while "[o]n the whole, the proposal for a workers' rights clause in WTO agreements too has been greeted in a positive spirit by informal sector activists. . . . Once again, this does not mean that all aspects of the proposal are accepted without criticism - for example, the suggestion that it will apply only to export production is seen as a defect - but, rather, that these activists are open to the possibility of using international pressure to secure rights for workers who have little hope of getting them through purely domestic action." Rohini Hensman, The Impact of Globalisation on Employment in India and Responses from the Formal and Informal Sectors 21 (CLARA, Working Paper No. 15, 2001).
136. Systems of linkage are often, but need not be, justified on the grounds that failure to respect basic labor standards gives an "unfair advantage" to some countries that must be corrected. Instead, linkage can be justified on the ground that the trading system provides an effective means of altering the incentives faced by countries, creating an environment that better enables and urges them to promote basic labor standards. By justifying linkage in terms of "unfair trade," and referring to countries that do not respect basic standards as "free riders," IFCTU (1999), supra note 1, at 43, proponents of linkage fail to place enough emphasis on a very important class of persons who are harmed by these failures, namely the workers in countries that fail to promote basic labor standards. For this reason, our proposal for linkage (sketched below) differs from others, such as Ehrenberg's, whose would allow penalties only against countries whose exports are produced in a way that disrespects basic standards (and only against such exported goods). See, e.g., Ehrenberg, supra note 1, at 172-73.
137. This must be true if the labor supply curve is upward sloping or entirely inelastic, as typically assumed.
138. See, e.g., Richard B. Freeman, Spurts in Union Growth: Defining Moments and Social Processes 10 (Nat'l Bureau of Econ. Research, Working Paper No. 6012, 1997), available at http://papers.nber.org/papers/w6012. v5.pdf (last visited Dec. 13, 2006);
139. Karl Moene & Michael Wallerstein, Social Democracy as a Development Strategy, GLOBALIZATION AND EGALITARIAN REDISTRIBUTION (P. Bardhan et al. eds., 2005), available at http://discuss.santafe.edu/files/globalization/SFI_dvifigs.pdf (last visited Dec. 13, 2006). Note also that the opposite conclusion could result, for example if labor unions represent a "labor aristocracy" that successfully demands that public resources be put to purposes other than those that benefit the most disadvantaged.
140. See generally BHAGWATI & RAMASWAMY, supra note 95. For discussion, see Arvind Panagariya, Bhagwati and Ramaswami: Why it is a Classic 13 (Nov. 8, 2000), available at http://econwpa.wustl.edu/eps/it/papers/0308/0308004.pdf (last visited Dec. 13, 2006).
141. See Appendix.
142. Implicit in this proposition is the application of an appropriate social welfare (or "aggregation") function. An example of an aggregation function to which it clearly applies is that defined by the total wage bill. See, e.g., Martin Rama, The Consequences of Doubling the Minimum Wage: The Case of Indonesia (World Bank, Working Paper No. 1643, 1996), available at http://ssrn.com/abstract=604935 (showing that in Indonesia, a ten percent increase in average wages, induced by an increase in the minimum wage, was associated with a two percent decrease in wage employment, implying a significant increase in the overall wage bill as a result of the minimum wage increase). The seminal detailed empirical research presented by Harrison and Scorse concerning the impact of "anti-sweatshop" activism on labor market outcomes in Indonesia comes to the conclusion that such activism has had ambiguous results, causing decreases in employment and wages in some sectors and increases in employment and wages elsewhere (including some of the firms targeted by activists).
143. See Ann Harrison & Jason Scorse, Moving Up or Moving Out? Anti-Sweatshop Activists and Labor Market Outcomes 32-35 (Nat'l Bureau of Econ. Research, Working Paper No. 10492, 2004), available at http://www.nber.org/papers/w10492.pdf.
144. See, e.g., DEREK PARFIT, EQUALITY OR PRIORITY? (1995).
145. See, e.g., Kaushik Basu & Pham Hoang Van, The Economics of Child Labor, 88 AM. ECON. REV. 412, 413 (1998) [hereinafter Basu & Van (1998)]; Basu (1999), supra note 21, at 1115 ("[A] large scale withdrawal of child labor can cause adult wages to rise so much that the working class household is better off."). Similarly, under certain conditions, eliminating the right of workers to enter into bonded labor contracts may benefit such workers, since the availability of such contracts may prevent more beneficial kinds of credit contracts from emerging.
146. See Garance Genicot, Bonded Labor and Serfdom: A Paradox of Voluntary Choice, 67 J. DEV. ECON. 101, 119-22 (2002).
147. This is true on a subjective preference-based conception of welfare as employed in Basu & Van (1998), supra note 106, at 36-37. However, the assumption that household welfare can be conceived of without further disaggregation ought to be questioned.
148. An early example of this approach to child labor is presented in ARTHUR C. PIGOU, THE ECONOMICS OF WELFARE 751-53 (4th ed. 1960). An exemplary overview of existing approaches of this kind is presented in Brown (2003), supra note 68, at 225-37.
149. E.g., Mexico's Progressa and Brazil's Bolsa Escola programs.
150. This could perhaps be recognized by the agent herself under appropriate conditions. To take a rather tired example, although an alcoholic may reveal his preference for beer over beans, it is far from obvious that this preference reflects his best interests. A public policy that hinders the ability of the alcoholic to indulge his preference without restriction may be viewed as enhancing welfare. For discussion of the principles underlying judgments of this type, see, e.g., T. M. Scanlon, Preference and Urgency, 72 J. PHIL. 655, 658 (1975);
151. Amartya Sen, Positional Objectivity, 22 PHIL & PUB. AFF. 126, 134-36 (1993).
152. On the outcomes that can arise under imperfect information, see generally Joseph E. Stiglitz, The Contributions of the Economics of Information to Twentieth Century Economics, 115 Q. J. ECON. 1441 (2000).
153. The 1996 WTO ministerial meeting, for instance, declared that "The International Labor Organization (ILO) is the competent body to set and deal with these standards, and we affirm our support for its work in promoting them." World Trade Organization, Ministerial Declaration of 13 December 1996, WT/MIN(96)/DEC (1996).
154. There is only one case known to the authors of the ILO having applied penalties to a country due to violation of labor standards: Myanmar in 1996. See ELLIOTT & FREEMAN, supra note 34, at 95.
155. For example at the 1997 International Labor Conference. The ILO remains relatively "toothless" even after the "strengthening" of countries' reporting requirements. See, e.g., Hensman, supra note 78; ELLIOTT & FREEMAN, supra note 34, at 96-100; International Confederation of Free Trade Unions, Belarus Is Once Again Censured by the ILO (June 14, 2005), available at http://www.icftu.org/displaydocumentasp?Index=991221860&Language=en (describing the ILO's inability to punish Belarus's violations of trade union rights).
156. See generally Howard Chang, Carrots, Sticks, and International Externalities, 17 INT'L REV. L. & ECON. 309 (1997).
157. See, e.g., Giancarlo Spagnolo, Issue Linkage, and International Policy Cooperation, (FEEM, Working Paper No. 49.96, Mar. 1999), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=163173.
158. For an argument in this direction with respect to trade and environmental standards, see generally Howard Chang, An Economic Analysis of Trade Measures to Protect the Global Environment, 83 GEO. L.J. 2131 (1995).
159. For example, it is widely believed that such restrictions played a role in the end of the apartheid regimes in Southern Africa (i.e. South Africa, Zimbabwe/Rhodesia, and Namibia).
160. See, e.g., S. PRAKASH SETHI, SETTING GLOBAL STANDARDS: GUIDELINES FOR CREATING CODES OF CONDUCT IN MULTINATIONAL CORPORATIONS (2003); ELLIOTT & FREEMAN, supra note 34, at 27-48;
161. Fung et al., supra note 61, at 5-6; KARL SCHOENBERGER, LEVI'S CHILDREN: COMING TO TERMS WITH HUMAN RIGHTS IN THE GLOBAL MARKETPLACE (2000).
162. For evidence on how codes of conduct fail (often due to competitive pressures) to be fully incorporated into firms' buying practices, see, e.g., OXFAM INT'L, TRADING AWAY OUR RIGHTS: WOMEN WORKING IN GLOBAL SUPPLY CHAINS 38-39 (2004), available at http://www.oxfam.org.uk/what_we_do/issues/trade/downloads/trading_rights.pdf (last visited Dec. 13, 2006);
163. HONG KONG CHRISTIAN INDUSTRIAL COMMITTEE, HOW HASBRO, MCDONALD'S, MATTEL AND DISNEY MANUFACTURE THEIR TOYS 29-31 (2001), available at http://www.cic.org.hk/download/ CIC%20Toy%20Report%20Web%20eng.pdf (last visited Dec. 13, 2006);
164. KENNETH A. RODMAN, SANCTIONS BEYOND BORDERS: MULTINATIONAL CORPORATIONS AND U.S. ECONOMIC STATECRAFT (2001).
165. Rodrik (1996), supra note 19, at 61.
166. Id.
167. Tinbergen uses the term "general interest" and refers to a "collective ophelimity function" that represents this general interest "in whatever sense that may be taken" and which is "the object to be maximised." This entity is apparently "a function of a certain number of variables which we shall call the target variables," select numerical values of which are referred to as the targets. The targets are presumed to be chosen so as to maximize the ophelimity function. In contrast, instruments are "variables under the command of the government." JAN TINBERGEN, ON THE THEORY OF ECONOMIC POLICY 1, 7 (1966).
168. Utilitarians, for example, may conceive of this master goal in terms of world welfare and, moreover, specifically understand welfare in terms of subjective preference satisfaction or pleasure. In contrast, Rawlsians evaluate social institutions in terms of the level of social primary goods they engender for their least advantaged participants.
169. See, e.g., Bhagwati (2002), supra note 58.
170. See, e.g., Jagdish Bhagwati & V.K. Ramaswami, Domestic Distortions, Tariffs and the Theory of Optimum Subsidy, 71 J. POL. ECON. 44 (1963);
171. Jagdish Bhagwati et al., Domestic Distortions, Tariffs and the Theory of Optimum Subsidy: Some Further Results, 77 J. POL. ECON. 1005 (1969).
172. Another reason that it may be impossible to attain the level of multiple targets despite the existence of at least as many instruments as targets is the existence of possible causal interdependencies amongst the targets themselves (as have been, for instance widely believed to exist between inflation and unemployment). See, e.g., TINBERGEN (1966), supra note 122; J. TINBERGEN, ECONOMIC POLICY: PRINCIPLES AND DESIGN (1967).
173. Jan Tinbergen himself strongly supported an integrated international policy to deal jointly with employment and growth objectives. Tinbergen identifies six broad areas in which "tasks should be performed on a world basis, although some may also be subjected to cooperation on a regional basis, under supervision on a world level." J. TINBERGEN, INTERNATIONAL ECONOMIC INTEGRATION 145 (1954). It is evident that Tinbergen views the areas suitable to be addressed at a centralized (world) level to be determined wholly on empirical grounds. Foremost among these empirical grounds is whether the instruments in question have a "supporting" or a "conflicting" role in the sense that "the use of such an instrument by one country will support the policies of the other countries" or whether its use by one country "conflicts with the objectives of other countries' policies." See id. at 98-99. The framers of the post-war institutional scheme also considered such integration of objectives quite attractive. For example, the International Trade Organization ("ITO") was originally proposed by John Maynard Keynes at the Bretton Woods Conference to further the expansion of world trade as a means to the ends of development, adequate wages, labor standards, and full employment. The Havana Charter, which provided for the creation of the ITO, contained an explicit reference to "Fair Labour Standards" (in Chapter II, Article 7 of the Charter) providing for the need to "take fully into account the rights of workers," recognizing that because "unfair labour conditions, particularly in production for export, create difficulties in international trade, each Member shall take whatever action may be appropriate and feasible to eliminate such conditions within its territory" and requiring that the ITO "consult and co-operate with the International Labour Organisation" toward this end. However, the ITO did not come into being because of the failure of the U.S. Senate ultimately to ratify it.
174. For the history of the ITO, see generally Richard Toye, Developing Multilateralism: The Havana Charter and the Fight for the International Trade Organization, 1947-18, 25 THE INT'L HIST. REV. 253 (2003) ;
175. Daniel Drache, The Short but Significant Life of the International Trade Organization: Lessons for Our Time (Centre for the Study of Globalisation and Regionalisation, Working Paper No. 62/00, Nov. 2000); THOMAS W. ZEILER, FREE TRADE AND FREE WORLD: THE ADVENT OF GATT (1999);
176. Howard M. Wachtel, Labor's Stake in the WTO, THE AMERICAN PROSPECT, Mar. 1, 1998, at 34;
177. Mark Levinson, Global Is as Global Does?, THE NATION, Dec. 18, 1999, at 42;
178. WILLIAM DIEBOLD, JR., The End of the ITO, in ESSAYS IN INTERNATIONAL FINANCE (1951);
179. WILLIAM ADAMS BROWN, JR., THE UNITED STATES AND THE RESTORATION OF WORLD TRADE (1950).
180. On the view of the parties to the ITO that labor standards concerns must play a role in the organization, see generally Toye, supra note 126; Drache, supra note 126. For a discussion of the wage-based view of labor standards endorsed in the early stages of GATT negotiations, see generally Elissa Alben, GATT and the Fair Wage: A Historical Perspective on the Labor-Trade Link, 101 COLUM. L. REV. 1410 (2001).
181. See, e.g., Hensman, supra note 78; Collingsworth, supra note 65.
182. See generally Spagnolo, supra note 115; see also Nuno Limão, Trade Policy, Cross-border Externalities and Lobbies: Do Linked Agreements Enforce More Cooperative Outcomes?, 67 J. OF INT'L ECON. 175 (2005).
183. Specific subsets of possible strategies are considered in the formal analyses by Spagnolo and the other contributors to this emerging body of literature. See also Nancy Chau & Ravi Kanbur, The Race to the Bottom, from the Bottom, 73 ECONOMICA 193 (2006) (especially section IV) for a recent example of such a contribution which focuses specifically on labor and environmental standards, and which attempts to identify conditions under which international agreements on standards may be sustainable.
184. Spagnolo, supra note 115 focuses on the implications of the agents' valuations of different combinations of attainments whereas Limão, supra note 128 focuses on the implications of the causal interconnections between distinct issue areas. According to Spagnolo, there are two kinds of cases to consider. The first is that in which the outcomes are substitutes in the sense that increases in the level of achievement in one outcome dimension are valued less when the level of achievement in the other outcome dimension is higher. The second is that in which the outcomes are complements in the sense that increases in the level of achievement in one outcome dimension are valued more when the level of achievement in the other outcome dimension is higher. When the outcomes are substitutes, then the threat of withdrawal of future cooperation in one issue area alone may be relatively ineffective because cooperation in the other issue area may secure rest of the advantages that would have been achieved had there been cooperation in the two issue areas together. When the outcomes are complements, then the threat of withdrawal of cooperation in one issue area alone may be relatively effective, because when cooperation in the second issue area is taking place, then the threat of withdrawal of cooperation entails a significant loss of benefit. Paradoxically, the withdrawal of future cooperation in both issue areas together may not be as effective because the value attached to cooperation in each area diminishes when cooperation in the other issue area does not take place. From the above analysis, it follows that when the issue areas are substitutes, linking them will be advantageous with respect to both of the aspects of enforcement. When the issue areas are complements, whether linking them will be advantageous from the standpoint of enforcement depends on the empirical question of whether the gains from improved allocation of slack enforcement power are greater than the losses from lessened effectiveness of the threat of withdrawal of future cooperation. Therefore, there is no general reason to prefer disaggregated negotiations to linked negotiations from the standpoint of enforcement. For discussion of related issues, see generally P. Conconi and C. Perroni, Issue Linkage and Issue Tie-In in Multilateral Negotiations, 57 J. OF INT'L ECON. 423 (2002);
185. Josh Edderington, Trade and Domestic Policy Linkage in International Agreements, 43 INT'L ECON. R. 1347 (2002).
186. Arvind Panagariya, Trade-Labour Link: A Post-Seattle Analysis, in GLOBALISATION UNDER THREAT: THE STABILITY OF TRADE POLICY AND MULTILATERAL AGREEMENTS 101, 104 (Zdenek Drabek ed., 2001).
187. This narrowly welfarist normative framework is clearly insufficient for capturing the normative significance of labor standards. However, we cannot further address this issue in this section.
188. Other conditions are required, such as the existence of efficient tax and transfer instruments. In the absence of such instruments, there is no guarantee that redistribution of the gains from trade can produce a Pareto improvement.
189. See, e.g., T.N. Srinivasan, Comment, in SOCIAL DIMENSIONS OF U.S. TRADE POLICY 236 (Alan V. Deardorff & Robert M. Stern eds., 2000) .
190. See Jagdish N. Bhagwati, Lobbying and Welfare, 14 J. PUB. ECON. 355 (1980);
191. see also Jagdish N. Bhagwati, Directly Unproductive: Profit-Seeking (DUP) Activities, 90 J. POL. ECON. 988 (1982);
192. Anne O. Krueger, The Political Economy of the Rent-Seeking Society, 64 AM. ECON. REV. 291 (1974).
193. There are plausible exceptions to the idea that the conduct of an agent ought to be evaluated independently of others' conduct. For example, there is a long tradition of argument that has emphasized that the obligation of agents to refrain from armament or attack depends on whether they have assurance that other agents abide by corresponding obligations. But cases of this kind appear to involve special conditions, for example, that those whose well-being is put at risk by the agent's failure to meet the requirement themselves reciprocally put the agent at risk through their failures to abide by this same requirement.
194. Other proposals for linkage have failed to take adequate account of the responsibilities of countries that are the sites of registration, ownership or management of firms, directly or indirectly linked to violations of basic labor standards. A linkage system that fails to broaden its jurisdiction in this way inappropriately focuses on the punishment of developing countries by developed countries. Thus, it will lack legitimacy and effectiveness.
195. There is indeed a wide consensus that this is the case, as the widespread reference in the debate to ILO conventions and other international legal documents specifying international norms makes clear.
196. See, e.g., G.A. COHEN, HISTORY, LABOUR, AND FREEDOM 209-38 (1988); ALAN WERTHEIMER, EXPLOITATION 207-46 (1996).
197. It is common to all rights that they may be asserted without insisting on their absolute priority or unconditionality. This is also recognized in law. Famously, asserting that there is a right to free speech does not establish that people can everywhere and anywhere say what they want. See, e.g., Schenck v. United States, 249 U.S. 47 (1919) (holding that the most stringent protection of free speech rights will not protect a man who falsely shouts fire in a theater and causes a panic).
198. See Penny Abeywardena, Interview with Jagdish Bhagwati, RIGHTS NEWS, Fall 2004, at 2-3.
199. See, e.g., Amartya Sen, Weil-Being, Agency and Freedom: The Dewey Lectures 1984, 82 J. PHIL. 169 (1985) on consequentialist theories directly valuing rights fulfillment.
200. See, e.g., Partha Dasgupta & Debraj Ray, Inequality as a Determinant of Malnutrition and Unemployment: Policy, 97 ECON. J. 177 (1987);
201. Ross Levine & David Renelt, A Sensitivity Analysis of Cross-Country Growth Regressions, 82 AM. ECON. REV. 942 (1992);
202. HARVEY LEIBENSTEIN, ECONOMIC BACKWARDNESS AND ECONOMIC GROWTH (1957);
203. Michael J. Piore, International Labor Standards and Business Strategies, in U.S. DEPARTMENT OF LABOR, INTERNATIONAL LABOR STANDARDS AND GLOBAL INTEGRATION: PROCEEDINGS OF A SYMPOSIUM 21 (1994); Moene & Wallerstein, supra note 101.
204. See, e.g., DANI RODRIK, THE NEW GLOBAL ECONOMY AND DEVELOPING COUNTRIES: MAKING OPENNESS WORK (1999). It has been argued that governmental enforcement of labor standards has created incentives for technological and organizational innovation and thereby enhanced economic growth in Europe and the United States.
205. See generally Moene & Wallerstein, supra note 101 (studying European cases); Piore, supra note 143 (studying the nineteenth-century U.S. textile industry); Kenneth D. Boyer, Deregulation of the Trucking Sector: Specialization, Concentration, Entry, and Financial Distress, S. ECON J. 481 (1993).
206. We apply this concept broadly here, so as potentially to encompass those who may not hold formal citizenship rights, such as legal residents or long-term residents.
207. Rawls' concept of a "decent consultation hierarchy" is a concept of the latter kind. JOHN RAWLS, THE LAW OF PEOPLES: THE IDEA OF PUBLIC REASON REVISITED 71 (1999). Whether it is sufficient for regimes to be of this kind or whether regimes must have democratic characteristics in order to guarantee the (internal) moral legitimacy of the linkage scheme which they join is a question which we do not directly address here.
208. We leave open the question of whether these necessary conditions for legitimacy are also sufficient.
209. The tripartite model of decision-making in the ILO offers a suggestive instance of such a process, which has led to notable consensus of this kind.
210. See generally JOHN RAWLS, POLITICAL LIBERALISM (1993); JOHN GRAY, TWO FACES OF LIBERALISM (2000).
211. See, e.g., COHEN, supra note 139.
212. Not all legally binding contracts are morally binding. See David Singh Grewal Network Power and Globalization, 17 ETHICS & INT'L AFF. 89, 92-93; see also COHEN, supra note 139. We will not address these important concerns at much greater length here, as they seem not to be raised in present discussions of linkage. To the extent they apply to the linkage proposal we make here they would certainly seem also to apply to assessments of whether the entry of countries into the WTO and many other international treaty bodies is unduly coerced.
213. See JOHN GRAY, FALSE DAWN: THE DELUSIONS OF GLOBAL CAPITALISM 18 (1998).
214. Personal conversation with Kamal Malhotra.
215. For a discussion of the issues of legal interpretation involved, see, e.g., Kyle Bagwell, Petros C. Mavroidis, & Robert W. Staiger, It's a Question of Market Access, 96 AM. J. INT'L L. 56 (2002).
216. See generally Chang, supra note 116; Chang, supra note 114.
217. See generally THOMAS ROBERT MALTHUS, AN ESSAY ON THE PRINCIPLE OF POPULATION (1798) and DAVID RICARDO, ON THE PRINCIPLES OF POLITICAL ECONOMY AND TAXATION (1821) for a highly skeptical view of these prospects, based on the perspective that public supports for the poor would generate perverse effects (of sufficient magnitude to undermine the impact of the supports themselves) on population growth and work effort.
218. See, e.g., Amartya K. Sen, Rational Fools: A Critique of the Behavioral Foundations of Economic Theory, 6 PHIL. & PUB. AFF. 317 (1977);
219. AMARTYA K. SEN, ON ETHICS AND ECONOMICS (1988).
220. See, e.g., TWIN-SAL, supra note 18.
221. See, e.g., RAWLS (1971), supra note 7.
222. Indeed, it is the norm in international treaties to deem that they have come into force when there have been a sufficient number of signatories or ratifications.
223. We recognize that additional approaches may also exist.
224. This suggests that proponents of trade liberalization who are critical of its current pace should favor linkage. In the present political climate, it is far from obvious that adopting linkage will bring about a lesser degree of trade liberalization than would otherwise take place. See, e.g., Polaski (2004), supra note 87.
225. There is considerable evidence of such competitive pressures among developing countries, most recently as a result of the end of the Multi Fiber Arrangement. Unsurprisingly in this context, there is, contrary to popular impression, considerable evidence of support for linkage by developing country labor unions. See, e.g., Griffin et al., supra note 74. There is archival evidence that strong support for the labor standards provisions in the proposed post-war ITO was provided at the post-war Havana Conference by representatives of developing countries (in particular, Cuba and India). The authors were alerted to this evidence through conversations with Mark Levinson.
226. The literature on efficiency wages suggests that there may be productivity gains to be achieved as a result of higher wages, although whether this will result from general, as opposed to firm-specific or industry-specific, wage increases depends on the specific mechanism by which it is assumed that wages enhance productivity. Collective action problems among employers can lead to the failure to realize these gains in the absence of determined coordination (as emphasized, for instance, in the literature on the nutrition-productivity relationship).
227. The possible collective action problem among employers may otherwise prevent productivity-enhancing investments in the labor force.
228. Of course, on the other side, there is the risk to firms that linkage will reduce the ability of firms to threaten workers in developed countries with relocation as a means of gaining concessions. Such relocation threats may be a determinant of profits in industries in which rent-sharing takes place.
229. See, e.g., Elizabeth Becker, Low Cost and Sweatshop-Free, N.Y. TIMES, May 12, 2005, at C1.
230. It is hardly difficult to find flaws in these institutions.
231. See INTERNATIONAL LABOUR ORGANIZATION, supra note 8 (providing for periodic self-reporting by countries, reviews of those reports by experts, and the provision of technical assistance in response to needs identified in these reports).
232. The present Director General of the WTO, Pascal Lamy, has noted that there is at present an "imbalance of our international legal order" and has argued that it is therefore desirable to develop an understanding of WTO law as complementing and supporting other international legal orders which focus on non-trade concerns, as well as to strengthen these other legal orders. See Pascal Lamy, Director-General of the WTO, The Place and Role of the WTO (WTO Law) in the International Legal Order, Address before the European Society of International Law (May 19, 2006) available at http://www.wto.org/english/news_e/sppl_e/sppl26_e.htm. Our proposal appears not to be at odds with an emerging interpretation of WTO jurisprudence.
233. We use the phrase "labor intensive products" to denote goods that could be the subject of export-oriented production in labor abundant countries because their production in all countries involves the relatively intensive use of labor as compared to other factors of production. There are of course technical problems which are involved in presenting this definition, which following the custom we sidestep.
234. The question may quite reasonably be raised as to why a good is produced at all in developed countries if there are large cost disadvantages associated with production of the good in developed countries. One answer may be that the figures compared refer to variable costs. Developed country production of labor-intensive goods, which would otherwise be uneconomical, may occur due to the existence of prior investments in plant and fixed capital. Developed country production may also take place due to other advantages it may have, for instance proximity to markets (making it possible, for instance, to meet "just-in-time" production demands). See FREDERICK H. ABERNATHY ET AL., A STITCH IN TIME: LEAN RETAILING AND THE TRANSFORMATION OF MANUFACTURING 269 (1999). Finally, there may be less developed country production of the same goods as are produced in developing countries than is at first suggested by the generally available data, which covers highly aggregative categories of goods. Specific goods even within labor intensive production (especially those which require higher skills and specialized knowledge to produce) may still be most economical to produce in developed countries, whereas other goods may be wholly uneconomical to produce in developed countries at prevailing wage rates. Although the mass production of T-shirts is quite likely to take place in a developing country, the production of an expensive dress shirt in a small batch that follows the pattern of a fashion designer in a metropolitan capital is far less likely to be undertaken in a developing country.
235. SETHI, supra note 118, at 58.
236. Id.
237. Id.
238. Id. at 59.
239. See HONG KONG CHRISTIAN INDUSTRIAL COMMITTEE, supra note 119, table 2.
240. See Pollin et al., supra note 93, table 5.
241. See Ian Carson, The Tijuana Triangle: Mexico's Northern Border Is Modern Manufacturing on the Move, THE ECONOMIST, June 20, 1998, at S3.
242. U.N. INDUSTRIAL DEVELOPMENT ORGANIZATION, INDUSTRIAL STATISTICS DATABASE AT THE 3-DIGIT LEVEL OF ISIC (INDSTAT3 Rev. 2) (2004) [hereinafter UNIDO].
243. See Summary of Data on the Share of Direct Labor Costs in the Total Costs of Surveyed Manufacturing Enterprises (by industry and income level of country), available at http://www.alternatefutures.org.
244. Sources: UNIDO, supra note 180; see World Bank Country Classification, available at http://worldbank.org/data/countryclass.html.
245. An Excel spreadsheet containing the arithmetical model is available at http://www.alternatefutures.org.
246. Which assumptions are appropriate is far from obvious. There is some reason to believe that in developing countries, the stages of production antecedent to the final one may be more labor-intensive, but also that they may employ workers at lower wages. The net impact on the share of labor costs in total costs is ambiguous.
247. Of course, in practice the good may not be economical to produce in the North. In that case, this ratio is best interpreted as that which would prevail if the good were to be produced in the North (at the prevailing factor prices, using cost-minimizing technique).
248. This ratio is known for specific goods and countries. In 1997, the unit cost to retailers of a casual men's shirt produced in the U.S. was estimated $7.58 and the unit cost to retailers of a casual men's shirt produced in Mexico $4.45. Unit costs of producing clothing items in other developing countries have been deemed comparable to those in Mexico. See ABERNATHY ET AL., supra note 172, at 223-42; see also Pollin et al., supra note 93. In 2001, the unit cost to retailers ("unit price realization") of a men's shirt was $4.21 in Bangladesh and $4.02 in China. See Gopal Joshi, Overview of Competitiveness, Productivity, and Job Quality in South Asian Garment Industry, in GARMENT INDUSTRY IN SOUTH ASIA: RAGS OR RICHES? 8 (Gopal Joshi ed., 2002).
249. In 2003, the average export price for trousers, underwear, woven shirts and knit shirts was $1.84 in China, $7.63 in the U.S., and $4.42 in other exporting countries. See NAT'L COUNCIL OF TEXTILE ORGS., ANALYSIS SHOWS CHINESE APPAREL PRICES 76% BELOW U.S. PRICES AND 58% BELOW REST OF WORLD'S PRICES (Dec. 15, 2004), available at http://www.ncto.org/ newsroom/pr200414. asp. In December 2004, with the assistance of the UNITE union, the authors interviewed New York manufacturers of girls' specialty dresses engaged in global subcontracting of garment production. We were told that costs of production in Mexico are roughly one-third those in the U.S. and costs of production in China and Sri Lanka are roughly one-fifth of those in the U.S. The assumption of a ratio of unit costs of about two seems, in light of these reports, to be wholly reasonable.
250. See, e.g., George Wehrfritz & Alexandra Seno, Succeeding at Sewing, NEWSWEEK, Jan. 10, 2005, at 38 ("According to AT Kearney, labor for a shirt made in Bangladesh runs just $1.52, compared with $2.28 in China, but after factoring in materials and transportation, the total cost of the Chinese shirt is $11.15 - almost a dollar cheaper.").
251. Labor costs in the garment industry are lower in Bangladesh than in China but overall production costs are lower in China than in Bangladesh: Keith Bradsher, Bangladesh Survives to Export Again: Competition Means Learning to Offer More Than Just Low Wages, N.Y. TIMES, Dec. 14, 2004, at C1.
252. The relative unimportance of labor costs as compared to other considerations that play a role in the decision to source garments in one developing country rather than another is forcefully emphasized in a recent guide for garment industry buyers. DAVID BIRNBAUM, BIRNBAUM'S GLOBAL GUIDE TO WINNING THE GREAT GARMENT WAR (2000).
253. See Pollin et al., supra note 93, table 7.
254. Dehejia & Samy, supra note 23, at 32.
255. Jai S. Mah, Core Labour Standards and Export Performance in Developing Countries, 20 WORLD ECON. 773 (1997).
256. Dehejia & Samy, supra note 23, at 15; see Rodrik (1996), supra note 19, at 52-59.
257. Dehejia & Samy supra note 23, at 21.
258. Id. at 23.
259. Id. at 31 (emphasis added).
260. See generally Singh & Zammit, supra note 32.
261. See id. at 94.
262. "Beta" is the average share (for low-income countries) of direct labor costs in unit costs at the final stage of the production process. "Delta" is the increment by which the share of direct labor costs in unit costs is assumed sequentially to increase (i.e., in arithmetic progression) at each stage of production prior to the final stage. "Theta" is the share in unit costs of the total labor costs incurred directly and indirecdy over the entire production process. "N" is the number of stages in the production process.