The proposal for "Ratcheting Labor Standards" (RLS) is a sophisticated elaboration of trends in corporate codes of conduct and private monitoring of workplaces that have developed over the past decade. It would certainly strengthen such monitoring, which has played a problematic but important role in the development of a global movement against sweatshops, and could improve some working conditions.

But ultimately the proposal steers that movement in the wrong direction. It builds on current conditions, which reflect the weakness in this era of globalization of the historic forces for improving workers' conditions—labor unions and state regulation. There is an important role for consumer pressure, pressure from nongovernmental organizations, and even independent monitoring, but it should be to strengthen unions and governmental regulation, not replace them with what ultimately would be a relatively weak, inadequate, and privatized regulatory system.

Fung, O'Rourke, and Sabel argue that a new system of regulation is needed because of problems with uniform standards (which they automatically but unjustifiably link to protectionism) and because the nature of production has changed, with decentralized global chains of production extending into the informal sector. But some uniform standards are legitimate, whatever the level of development. Certainly core International Labor Organization (ILO) standards of the right to organize and prohibition of child or forced labor or racial discrimination are relatively uniform and not protectionist—despite gripes from some politicians in developing countries. Does prohibition of exposure to toxic workplace chemicals have to await a higher level of economic development?

Wages should in some way reflect the development of a country, but a requirement that countries enforce reasonable minimum or living wage standards based on their development would not be protectionist. In any case, the companies in question are not struggling efforts run by entrepreneurs in poor countries hoping to break into a world market. These are operations run by major multinationals, like Nike, based in countries with a very high standard of development that are taking advantage of desperate workers in poor countries. If, as they claim, they are raising living standards in those countries, they can be held to higher standards than domestic industries in those countries. They can afford to pay for higher standards and so can their customers.

Who sets the standards? The authors suggest that firms "would be required to adopt a code of conduct"—but what code? And who would do the requiring? Individual governments could both require and set such standards—like adherence to core labor rights—as the basis for imports, but that would run up against World Trade Organization prohibitions on regulating production methods. The alternative is to make basic labor standards integral to international law and institutions governing trade and to strengthen institutions, such as the ILO. Even the Organization for Economic Cooperation Development (OECD) could develop mandatory codes of conduct for multinationals based in OECD countries.

Although production and ownership of factories has become decentralized, ultimate control over the global chain of production has become ever more centralized. Consumers and governments may not know where every part of every product they buy was made, but people who control corporations do—or they can know if they want to. Indeed, the key to restoring more traditional, effective regulation is to make those dominant corporations legally responsible for every infraction of labor rights at every stage of production, even if they are not the owner and direct employer in every workplace. They do have the power to determine conditions of their contractors and subcontractors. They should be required to make full disclosure of their production chain—that is, to be fully "transparent"—and they should be subject to meaningful penalties for all violations. Because of the internal management systems they already have, as the authors describe, they are capable of enforcing good standards. The authors' reference to the "hot cargo" provisions of American law and the leverage government gets through penalties on the ultimate controlling corporation illustrates how it is possible to regulate global chains of production.

Any good global regulatory system should emphasize both transparency and sanctions, as the authors propose. But there is reason to be skeptical about their mechanisms of competition and continuous improvement. Ultimately, they rely on "social and market pressures" as the power behind their regulatory system, not the authority of the state (and the quasi-state institutions on a global level) or the power of organized workers. They imagine that social pressures will lead monitors and firms to compete to ever-higher standards in order to appeal to ethical consumers. But history suggests that monitors—like today's financial auditors—will feel strong pressure to collaborate with businesses that employ them and that corporations will be motivated to do only what little they must in order to avoid bad publicity. (There are also many industries that are not very vulnerable to ethical consumer pressures.) If Fung, O'Rourke, and Sabel truly believed their model, they would not endorse any kind of governmental regulatory system—which is precisely what conservatives in this country want by dismantling occupational safety and health, environmental, and other enforcement, though not because they expect an ever-rising standard for corporate conduct.

Social and market pressures can be valuable, but there are limits to the power of the ethical consumer, even one who is fully informed by a new monitoring system. In any case, both workers in unions and ethical consumers would gain power by cooperating, not by posing as alternatives. Despite the authors insistence on listening to workers' views (who will be listening and deciding?), it is disconcerting that they think that unions, as an afterthought, might compete with nongovernmental monitors to see who does the better job. Shouldn't workers, as a matter of fundamental right, be actively supported in organizing themselves? If a monitor finds that the union isn't succeeding, workers can use the monitor's information to strengthen their hand in dealing with their employer.

There may be a role for nongovernmental organizations and independent monitors, but it is as a way of moving toward and then strengthening real protections of worker interests through governmental regulation and unions, not by creating a privatized, weak system that would take their place.