IT HAS become commonplace for politicians of both political parties to trot out rhetoric about how we need free-but-fair trade. Expanding markets through trade liberalization, it is urged, is a win-win situation. How is it, then, that in spite of assertions that everyone benefits from trade, there is so much opposition, in both developed and developing countries? Is it that populists have so misled ordinary citizens that, though they are really better-off, they have come to believe they are doing worse?

Or is it because trade liberalization has, in fact, made many people worse off, in developed and developing countries alike? Not only can low-skilled American workers lose their jobs or be paid less, those in developing countries suffer, too. They end up having to take the short end of the stick time and time again in trade agreements because they have little leverage over the big boys. And the links between trade liberalization and growth are far weaker than liberalization advocates claim.

A closer look at both data and standard economic theory provides further insight into the strength of the opposition to trade liberalization. In most countries around the world, there is growing inequality. In the United States, not only is there a steady uptick in poverty, but median household income has been falling for at least eight years. There are many factors contributing to these changes: technology, weakening of social mores, labor unions and, lest we forget, trade liberalization. More than sixty years ago, prominent economists Paul Samuelson and Wolfgang Stolper explained that trade liberalization in high-income countries would lower wages of unskilled workers. The economists showed that even a movement toward free trade brought wages of unskilled workers around the world closer together, meaning, for example, that America's unskilled workers' pay would fall toward that of India and China. Although their model stems from the mid-twentieth century, some of its assumptions hold even more true today. In particular, globalization has greatly reduced disparities in knowledge and technology between the developed and developing world. Lower-paid workers in the developing world now often have the tools, and increasingly, even the education, to perform the same tasks as their counterparts in developed countries. American workers simply get paid more to do the same task. Quite obviously, this can hurt even the higher-paid skilled American worker.

More generally, standard economic theory does not say that everyone will be better-off as a result of trade liberalization, only that the winners could compensate the losers. They could take a portion of their gains, give it to the losers and everyone could be better-off. But, of course, the winners, which in much of America are the very well-off, haven't compensated the losers; indeed, some have been arguing that to compete in the new world of globalization requires cutbacks in government spending, including programs for the poor. The losers then lose doubly.

These results of traditional economic theory are based on assumptions like perfect information, perfect-risk markets, perfect competition and no innovation. But, of course, we do not live in such a perfect world. Modern economic theory has shown that in the imperfect world in which we live, trade liberalization can actually make everyone worse off. For instance, trade liberalization may expose individuals and firms to more risk. In the absence of adequate insurance markets, firms respond by shifting production away from high-return risky activities to safer, but lower-return areas, thereby lowering national income.

Careful studies have found, at best, weak links between trade liberalization and growth. Many studies do show that countries that have increased their levels of trade-China is a good example-have grown faster. But these countries did not liberalize in their earlier stages of development. They promoted exports and restricted imports. And this export promotion worked.

A standard argument for reducing tariffs is that it allows resources, especially labor, to move from lower-productivity sectors into higher-productivity ones. But all too often, it results in moving workers from low-productivity employment into zero-productivity unemployment. For example, workers in Jamaica's dairy industry cannot compete with America's highly subsidized milk exports, so when Jamaica liberalized, opening up its markets to these subsidized imports, its dairies were put out of business. But the dairy workers didn't automatically get reemployed elsewhere. Rather, they simply added to the already-high unemployment rolls. In many countries, where there is high unemployment, there is no need to "release" resources to expand exports. There are a variety of impediments to expanding exports-including internal barriers to trade (such as the absence of infrastructure, which highlights the need for aid-for-trade) and, on an even-more-basic level, the absence of capital. Ironically, under today's rules, trade liberalization may again make matters worse. That is because countries are being forced to open up their markets to foreign banks, which are more interested in lending to multinationals and national monopolies than to local small- and medium-sized businesses, the sources of job creation.

Waxing Politic

THE CASE for trade liberalization is far weaker than most economists will admit. Those who are more honest fall back on political arguments: it is not that trade liberalization is such a good thing; it is that protectionism is such a bad thing. Inevitably, it is argued, special interests prevail. But in fact, most successful economies have evolved with at least some protection of new industries at critical stages of their development. In recent work, my colleague from Columbia University, Bruce Greenwald, and I have built on that idea by developing an "infant-economy argument" that looks at how using protection as countries grow can encourage the industrial sector-the sector most amenable to learning and technological progress. The benefits of that support then diffuse throughout the economy. Such policies do not require governments to "pick winners," to identify which particular industries are well suited to the country. These policies are based on a recognition that markets do not always work well, particularly when there are externalities, where actions in one part of the economy affect another. That there are huge spillovers from successful innovation is incontrovertible.

Politicians, of course, are not constrained by economics and economic logic. Even if we see in our model that safeguarding nascent sectors is the best way to support economic growth, trade advocates claim, for instance, that trade creates jobs. But exports create jobs; imports destroy them. If one justified trade liberalization on the basis of job creation, one would have to support export expansion but simultaneously advocate import restrictions-these days, typically through nontariff barriers called dumping and countervailing duties. This is the curious position taken by many politicians who say they favor free trade.  George W. Bush, for instance, while bandying about terms such as free trade and free markets, imposed steel tariffs at a prohibitive level even against desperately poor and tiny Moldova. This in spite of the fact that Moldova was struggling to make the transition from communism to a market economy. American steel producers could not compete and demanded these kinds of tariffs-they couldn't compete, not because of unfair competition from abroad, but rather because of failed management at home. In this case, eventually the World Trade Organization (WTO) ruled against the United States, and this time, the United States complied.

The important point missed by these politicians-and the economists who serve them ill by using such arguments-is that trade is not about job creation. Maintaining the economy at full employment is the responsibility of monetary and fiscal policies. When they fail-as they have now done once again-unemployment increases, whatever the trade regime. In reality, trade is about standards of living. And that raises an important question: whose standards of living, exactly?

Double Standards

IN DEVELOPING countries, there is another set of arguments against the kind of trade liberalization we have today. The so-called free-trade agreements being pushed by the Bush administration are, of course, not free-trade agreements at all. If they were, they would be a few pages long-with each party agreeing to eliminate its tariffs, nontariff barriers and subsidies. In fact, they go on for hundreds of pages. They are managed-trade agreements-typically managed for the special interests in the advanced industrial countries (especially those that make large campaign contributions, like the drug industry). The United States keeps its agricultural subsidies, and developing countries are not allowed to impose countervailing duties. And the agreements typically go well beyond trade, including investment agreements and intellectual-property provisions.

These investment agreements do far more than just protect against expropriation. In a perfect show of how all of this is supporting the developed countries while hurting the developing, they may even give American firms operating overseas protections that American firms operating domestically do not have-such as against loss of profits from new regulations. They represent a step backward in creating a rule of law: disputes are adjudicated in processes that fall far short of the standards that we expect of others, let alone of ourselves. Even worse, the ambiguous provisions can put countries in crisis in an impossible bind. They have given rise to large lawsuits, forcing developing countries to pay out hundreds of millions of dollars. In a particularly egregious example, Indonesia was forced to pay compensation for profits lost when it abrogated an almost-surely corrupt contract that then-President Suharto signed. Even though the abrogation of the agreement took place when Indonesia was falling into crisis and receiving support from the International Monetary Fund, the country was still held responsible for repayment of anticipated profits, which were unconscionably large because of the very corruption that many believe contributed to the country's problems in the first place.