Jagdish Bhagwati is a humble man. He will tell you so himself. Describing the effect of his book In Defense of Globalization during a speech at the John Hopkins School of Advanced International Studies (SAIS) last fall, the Columbia economist politely refused credit for single-handedly dampening growing concerns about the fallout from free trade. Fears of trade are "low-key," he said. "I won't say it's because of my book. I have colleagues who would say that. ... People believe I have a large claim to fame, so I don't have to do it myself."

But, for all his bluster, Bhagwati was in something of a defensive crouch. His talk was titled "Free Trade Policy Today: Why Is the United States in Retreat?" and, although he was tempted to pin the blame on idiots in the press--he called New York Times writer Louis Uchitelle a protectionist who "never saw a tariff [he] didn't like"--a wayward economics reporter or two could hardly explain the shift in public opinion. Although Washington's trade deal with Peru--largely modeled on the North American Free Trade Agreement (NAFTA)--managed to pass Congress in December 2007 on a bipartisan vote, many politicians have become increasingly vocal in their opposition to future deals. One need only look at the dustup that preceded the Ohio primary, in which senators Hillary Clinton and Barack Obama competed to out-anti-NAFTA each other, despite the fact that it was Clinton's husband who pushed NAFTA through a resistant Congress 15 years ago. Nor are concerns about the effects of trade agreements limited to Democrats campaigning in the hard-hit manufacturing belt. Poll after poll shows that a majority of Americans, skeptical of their benefits, oppose NAFTA-style deals. A Wall Street Journal/ NBC survey last year found that about 60 percent of Republicans believed foreign trade has been bad for the United States, up from 31 percent nine years ago.

The public's increasing wariness of trade agreements is easy enough to integrate into the narrative of eternal besiegement that free trade's advocates tend to construct. (Adam Smith's The Wealth of Nations was born, more or less, as an argument against the prevailing protectionist consensus of the day.) But what truly irritated Bhagwati was the perception that economists were rethinking the fundamentals of their pro-trade arguments. Two high-profile "defections" in particular merited Bhagwati's ire. In 2004, the father of modern trade theory, Nobel laureate Paul Samuelson, published a paper which argued that, as China developed economically--from manufacturing children's toys to, say, programming software--the net effect on the U.S. economy could be negative. Then, last year, Princeton economist (and former Federal Reserve governor and adviser to President Clinton) Alan Blinder circulated a working paper that calculated the number of U.S. jobs that could be shipped overseas. The number was headline-grabbing: 40 million.

Despite Bhagwati's assurances that these were minor blips misinterpreted by a man-bites-dog-hungry press, nearly all of the dozen or so economists I've spoken to have said that the academic conversation about trade has moved significantly. Fifteen years after NAFTA and ten years after the protests against the World Trade Organization in Seattle, economists can now look at the actual results from a host of multilateral free-trade agreements. The results do little to overthrow the basic theoretical argument about comparative advantage--economies do best when they specialize in producing the goods and services they can make most efficiently and trade for those goods outside their specialization--but they have led many economists to be far more skeptical of the actual "free trade" policies that have emerged from Washington over the last several decades. The evidence has forced academics to focus on the distribution of trade's negative effects, the role trade agreements play in rising inequality, and the failure of trade agreements to deliver the bounty of jobs their advocates predicted. The result is that, while they may still deride protectionism, laissez-faire economists who once sought to keep government from meddling in the market have begun to embrace an unlikely new partner: the welfare state.

***

Given how heated debates about trade have become on the political circuit, you might imagine that the academic landscape is similarly contested. But there's actually a surprising degree of consensus across ideological and methodological divides in the academy that the predictions of trade theory have been more or less borne out in the developed world: Trade agreements like NAFTA have produced concentrated pain (job losses in low-skilled industries) and diffuse gain (cheaper consumer goods), with the gains outweighing the losses. "I am actually a textbook purist on this, and I'm a lefty," says Josh Bivens, an economist at the liberal Economic Policy Institute. "The idea that trade raises total real income isn't in dispute," Princeton economist and New York Times columnist Paul Krugman wrote me in an e-mail. "But it definitely feeds inequality in rich countries, although the magnitude is in dispute."

It's this distributional component--the difference between the "winners" and the "losers" in a trade deal--that has begun to command far more attention in the United States. "It's not just workers in the importing sector that suffer the wage cut" when forced to compete with foreign workers, says Bivens. "It's everyone that looks like them. Landscapers don't get replaced by imports, but their wages are depressed by having to compete with laid-off apparel workers. ... Those without a college degree are going to take a hit; those with college degrees are going to get gains."

Just how much the losers have lost is a matter of debate, but most economists agree that the wealth gained from free trade has been redistributed upward, toward the skilled, and that low-skilled workers have suffered the most. They also agree that, as a portion of the total U.S. economy, the overall net benefit of NAFTA and other free-trade deals is too small to find with even the most powerful econometric microscope. What you're left with is a small gain in the nation's net income and a strong, lasting depression of wages that hits exactly the kinds of unskilled workers who had already been falling further and further behind.

The question is, if this was more or less what economists expected, why doesn't it feel like what Americans were sold? Bill Clinton did not say that we needed to pass NAFTA so that every American could spend some $50 less per year on consumer goods. Rather, he said that NAFTA's passage would "consolidat[e] the victory of democracy and opportunity and freedom" and "recover the fortunes of the middle class." Richard Freeman, a labor economist at Harvard, says that he could have predicted the agreement's consequences, and that fellow economists who went along in selling NAFTA should have as well. "I don't have great feelings towards some of them. ... It was really oversold." Even strong supporters of NAFTA and increased liberalization agree. "Certainly, it was oversold," says Dartmouth economist and trade expert Doug Irwin about NAFTA. "Clinton spoke about all the job creation, etc. Academic economists tended not to be those cheerleaders."

***

Still, academic economists expected the gains to be more diffuse than they have been, and that reality has led them to reappraise the need for a helping hand from the government. "In the standard free-trade argument, it is assumed that positive impact will trickle down," says Ha Joon Chang of Cambridge University. "If the U.S. signs NAFTA with Mexico, some textile workers will lose their jobs, but, because it will sell more computers, that will create more jobs in the computer sector that some textile workers might get and more importantly expands the income of the national economy. The trouble is that, while this trickle-down does happen, it is not perfect. This is why many countries have policy mechanisms to deal with it. The most important is the welfare state." Bhagwati and Chang generally agree on little, but, on this point, Bhagwati is nearly on the same page. "The anxiety must be addressed," he said at SAIS. "We have to think of the system in a holistic way. ... Think of a trapeze artist: He's got to have a safety net."

Such rhetoric is a marked change from that of the ascendant neoliberal vanguard back in the 1990s. During that period, the most extreme advocates of free trade, free markets, and laissez-faire tended to wield globalization and international competition as the crowbar with which to dismantle the welfare state: Health care had to be scrapped as well as strong unions because developed countries were competing against low-wage workers around the globe. "Globalisation driven by technological and multilateral trends," one Italian economist observed, "weakens governments' power to enforce welfare state schemes."

The view expressed by Chang and others I talked to represents nothing less than a Copernican shift. Dani Rodrik, a professor of political economy at Harvard, says, "The conversation in the last ten years has really turned from saying, 'The effect of globalization on insecurity and inequality is rather minimal and therefore you need the least amount [of social safety net] that's politically required to get your trade agreements passed.' Now, it's: 'Globalization is probably contributing a lot more to these problems and therefore a healthy globalization requires a domestic agenda.'" Fred Bergsten, director of the Peterson Institute for International Economics, agrees, saying that the proper response to the anxieties about globalization "lies in changes in our domestic policy: universal portable health insurance, portable pensions, much better unemployment insurance. ... We just have to do a better job of dealing with the downsides, and the costs, and the losers."

Some form of this argument has long been in economists' arsenal when giving a measured defense of free trade: Yes, there are winners and losers, but government can first liberalize trade and then use the net benefits to compensate the losers. It's never quite worked out this way. The program created for this purpose, Trade Adjustment Assistance, is limited in scope and funding and not particularly effective, to say the least. In 2007, at the American Economics Association conference, economist Ramya M. Vijaya gave a paper which stated that a sample of out-of-work manufacturing laborers who enrolled in job retraining through the program ended up, on average, in positions with a lower wage than those who skipped the retraining.

And, if free-trade advocates have given lip service to the idea of strengthening the welfare state in order to offset uncertainty and inequality, they've spent little political capital doing so. Which is why it is interesting to see that Lawrence Summers, who served as President Clinton's treasury secretary during the headiest days of free-trade enthusiasm, is now having some very public second thoughts. Writing in the Financial Times, he noted that "[e]ven as globalisation increases inequality and insecurity, it is constantly and often legitimately invoked as an argument against the viability of progressive taxation, support for labour unions, strong regulation and substantial production of public goods that mitigate its adverse impacts." But Summers argued that such an attitude was a political non-starter, particularly as globalization "encourages the development of stateless elites whose allegiance is to global economic success and their own prosperity rather than the interests of the nation where they are headquartered." In a subsequent column, he concluded that the "domestic component of a strategy to promote healthy globalisation must rely on strengthening efforts to reduce inequality and insecurity. The international component must focus on the interests of working people in all countries, in addition to the current emphasis on the priorities of global corporations."

Summers's actual policy prescriptions--greater international cooperation in tax and regulatory policy--may have been weak brew, but it's important to note the arrow of casualty in the discussion here. It's not that academic ferment has changed the tenor of the political discussion over trade; rather, it's been the other way around. "I think, even on the right, you hear it more and more," says Mark Thoma, an economist at the University of Oregon who runs the popular blog Economist's View. "There's a growing perception that the political will to keep markets open or open them further depends on solving some of these distributional issues, health care, all of these things. I don't think there's complete buy-in on the welfare state, but what's new is the idea that opening trade further is going to require us to deal with the problem of winners and losers, rather than just acknowledge it. It won't just solve itself, and it won't happen quickly and easily."

Which is why, for all the economists' carping about anti-trade demagoguery, and the brain-deadness of trade discussions on the campaign trail, the popular and political opposition toward trade policy has had salutary intellectual effects. That is, academics may have scoffed at the supposed naïveté and short-sightedness of free trade's critics; now, they're admitting that they had a point.