On the face of it there should be nothing contentious about the international labor standards movement. It is meant to be a global effort to raise the working conditions and living standards of workers, primarily in developing countries. What is curious is that the biggest opposition to labor standards has come from its alleged beneficiaries—to wit, Third World workers, unions, and governments. The fear in the South is that once such a global monitoring scheme is brought into existence, it will get diverted into an instrument of protection for the North. In the name of international labor standards, arbitrary and inflexible trade sanctions will be imposed on Third World countries. This fear gets heightened if the labor standards are imposed through the World Trade Organization (WTO), via a "social clause" provision, which would allow the WTO to use trade sanctions against any nation that violates minimal labor standards.¹ The other concern stems from the adjective "international," which suggests a uniform global standard for all nations.

Fung, O'Rourke, and Sabel have come up with an ingenious suggestion for international labor standards that gets around some of these criticisms. They christen this contribution "Ratcheting Labor Standards" (RLS). They try to bring in flexibility by keeping labor standards away from formal global organizations. Instead, they propose a system of collecting and publicizing information about the labor practices of firms, and encouraging consumers, journalists, and other ordinary citizens to use social sanctions, such as product boycott and public criticism, against firms that violate minimal standards. In addition, they recommend that the minimal standards that monitors seek and publicize be different across different nations, depending on their levels of economic development.

I believe that Fung, O'Rourke, and Sabel air the right concerns and steer the debate in the right direction. Nevertheless, I think that, as a practical proposal, RLS is flawed. It will not attain the objectives that the authors (rightly) uphold.

One novel and attractive aspect of their scheme, though this is left latent in their statement, is that they place the burden of responsibility on the firm that violates labor standards, rather than the country where the labor standard is violated. We know from economic analysis in other areas that responsibility, no matter where it is initially placed, can be partly deflected. But holding the firm responsible will have the advantage of the firm not being able to freely play one poor nation against another and thus drive down standards.

The main weakness of RLS is rooted in its relying so heavily on social sanctions and citizen action. Such a policy has a nice, progressive ring to it, and a certain kind of flexibility—but its shortcomings outweigh these advantages. First, this will handicap small producers. Consider the soccer ball industry in Pakistan. Soccer balls used to be stitched by thousands of small producers, working out of their homes, which doubled as residence and factory. Even if such a production unit did not use child labor, there would be no way for it to "prove" this to outside monitors. A large producer, on the other hand, can easily centralize production in a big factory, stop children from entering the premises, and prove to outside monitors that production is free of child labor. Indeed, Reebok has done just that in Pakistan. Hence, a by-product of this scheme is that it creates a competitive advantage for large firms by making it virtually impossible for small producers to get certification.

Second, once information on firms is collected and publicized and citizens are encouraged to take action, there is no way of ensuring that this system will achieve the right kind of flexibility. In several domains of our civic life, social sanctions play a useful role. But social sanctions have also been the basis of witch-hunts and the persecution of harmless behavior that happens to deviate from the mainstream. Once consumers in a rich country are given the moral responsibility to enforce standards and they are told that in Ethiopia workers are paid ninety cents for a day's work, it is easy for the consumers to believe that this is not a living wage and begin a boycott of Ethiopian goods, unmindful of the fact that such a boycott could cause unemployment and drive the incomes of many workers down to zero. As Joan Robinson, the eminent British radical economist, once noted, in some situations what is worse for a worker than being exploited is not being exploited.

Also, in a world of informal control, corporations will soon be competing by deliberately using adverse publicity against firms that sell cheap.

It is not surprising that, while all societies rely on informal mass action for curbing a variety of undesirable actions, there are also domains where we prefer to use more centralized methods for addressing issues of justice, for instance, through the courts. I would argue that labor standards do not belong to the first category—they must not be enforced through informal mass action.

Does this mean that we must take institutional (this includes "governmental") action for raising labor standards? Before answering this, note first that not doing so does not mean that labor standards will necessarily be abysmal. As productivity and wages rise, workers can demand higher standards and have many of these demands met, not through government action but by virtue of the standard forces of the market—namely, the implicit threat that the worker will not otherwise accept the job. Very few people, however, argue that all labor standards should be left to what workers can achieve through market mechanisms. Most countries have laws against workplace sexual harassment, for example; they do not leave it to workers to ensure that they are not harassed by threatening to quit if they are. Likewise, there is reason for public action in certain domains of labor standards in developing countries. The question is: How can this be ensured without hurting the very workers that this is meant to help (by causing unemployment or by impoverishing further some already-poor nation)?

In the debate on international standards, so much attention has been directed at the alleged (and largely unsubstantiated²) conflict of interest between First and Third World workers that the tension among workers in different nations of the Third World is often overlooked. In this age of mobile global capital, it is easy for corporations to move their capital from one nation to another. Hence, each developing country hesitates to take action to raise its labor standards for fear of driving capital away to another developing country. Hence, there is need for collective action on labor standards among the Third World nations.³ If we take this seriously, then we need to allow Third World nations to develop their own agenda—a consensus from the tropics—of what constitutes minimal labor standards.

The trouble with the WTO is that it is viewed by most developing nations, not entirely without reason, as controlled by rich nations. Even though it runs on the principle of one-country-one-vote, the "green room" where the agenda is set is, in practice, controlled by industrialized nations. There is a great need to encourage the reorganization of international organizations so that they represent the interests of all nations democratically, and to provide a forum where poor countries can develop their own agenda for what constitutes labor standards.

Action for international labor standards is desirable, and such action must be carried out through global institutions, not informal mass action. But if the institutions do not have democratic representation, the process can work to the detriment of the developing nations. Hence, while we strive toward this goal of global action, we should also be prepared for the possibility that, given the current structure of global governance, we may for the time being prefer to resist globalization in this one area and leave labor-standards policy and intervention to individual nations. It is sobering to recall that in the United States the attempt to bring all states under a common labor code was on the agenda for decades (actively from 1906) before it could finally be implemented in the form of the Fair Labor Standards Act in 1938. •

Kaushik Basu is professor of economics and Carl Marks Professor of International Studies at Cornell University. His most recent book is Prelude to Political Economy.

Read all eight responses to "Realizing Labor Standards," by Archon Fung, Dara O'Rourke, and Charles Sabel.

² As I argue in "International Labor Standards and Child Labor," Challenge 42 (September/October 1999): 80-92.