Globalization, Job Loss, and Stagnant Wages: The Evidence Is Changing

For decades, economists resisted the conclusion that trade – for all of its many benefits — has also played a significant role in job loss and the stagnation of middle-class incomes in the United States. As recently as 2008, for instance, Robert Lawrence of Harvard, one of the country’s most respected trade experts, concluded that trade explained only a small share of growing income inequality and labor market displacement in the United States.

Rather than focusing on trade, economists argued that other factors – especially “skill-biased technical change,” technological innovation that puts an added premium on skilled workers – played the biggest role in holding down middle-class wages. But now economists are beginning to change their minds. Responding to The New York Times’ recent survey about the causes of income stagnation, many top economists have cited globalization as a leading cause.

While the evidence is still not conclusive, it is pretty strong. Trade’s effect on jobs and income, which was probably modest through the 1990’s, now seems to be growing much larger. Among the recent studies:

• In “The Evolving Structure of the American Economy and the Employment Challenge,” my CFR colleague Michael Spence looked at job growth from 1990 to 2008 in sectors of the United States economy. He found almost no net job growth in sectors, like manufacturing, in which global trade played a large role. Nearly all of the net gains occurred in sectors in which trade plays a minor role. Government and health care, in which trade plays almost no role, accounted for more than 40 percent of all new jobs.

• David Autor, David Dorn, and Gordon Hanson looked at regions in the United States where companies are competing most directly with China. From 1990 to 2007, they found that regions that faced growing exposure to Chinese competition had higher unemployment, lower labor-force participation, and lower wages than might otherwise be expected. And the effects grew over that period. In 1991, just 2.9 percent of United States manufacturing imports came from low-wage countries; by 2007, that had risen to nearly 12 percent, mostly from China.

• In the Council on Foreign Relations Task Force on U.S. Trade and Investment Policy, my colleague Matthew Slaughter looked at employment at multinational companies with headquarters in the United States, companies that account for roughly 60 percent of American exports and imports. From 1989 to 1999, those companies created 4.4 million jobs in the United States and 2.7 million jobs at their foreign affiliates overseas. From 1999 to 2009, however, those same companies eliminated a net of nearly 3 million jobs in the United States while adding another 2.4 million jobs abroad.

The usual rebuttal to these findings is to argue that they stem mostly from manufacturing. And manufacturing, the argument goes, is facing a long-run, secular decline in employment that is largely technology-driven, not unlike the story of agriculture in the 20th century. The job losses in manufacturing may seem as if they have been caused by trade, according to this view, but they have actually been caused by technological change.

Through the 1990s, that story was largely plausible. But over the last decade it is not. Manufacturing output in the United States is no longer growing as rapidly as it once was (and as you would expect if technology had simply been replacing workers in factories). Robert Atkinson and colleagues have shown convincingly that the loss of more than five million jobs in manufacturing in a decade was not primarily a technology and productivity story. Real manufacturing output grew just 15 percent in the 2000s, compared with more than 35 percent in each of the 1970s and 1980s and more than 50 percent in the 1990s. And one sector where the statistics significantly overstate output — computers and electronics – accounts for almost all of the recent gains, even though the U.S. trade deficit has actually grown sharply in this sector over the past decade. In thirteen of nineteen manufacturing sectors, real output declined over the last decade, in some industries quite sharply. There is no question that in recent years United States manufacturing has declined, taking away jobs and driving down wages for those who are still employed.

The real-world evidence makes it surprising that it has taken economists so long to catch on. The recent strike in Joliet, Illinois, at Caterpillar – a true global company — ended with union workers being forced to accept an agreement that includes a six-year wage freeze, even as the company is earning record profits. Elsewhere, two-tier agreements, in which new hires earn wages and benefits roughly half as large as those in the old union contracts, have become standard in many of the manufacturing industries that remain in the United States.

One reason that economists may be uncomfortable talking about trade’s impact on jobs and wages may be concern that it could set off protectionist responses. And expanded trade has certainly been good for the United States. It has brought us better and cheaper consumer goods, opened new export markets, lifted up many poor countries and strengthened American alliances around the world. But I think the fear of protectionism is overblown. One unexpected feature of the Great Recession was how little protectionism it led to, especially in the advanced economies. The lesson of the Great Depression – that protectionism is counterproductive – seems to have been learned.

Instead, the evidence should produce some soul-searching about the causes of this country’s declining competitiveness. The list is discouragingly long: crumbling infrastructure, inadequate educational performance, stifling regulation, and a cumbersome tax system. But it might not take that much to tip the scales in favor of the United States. The Boston Consulting Group, which has looked at the slight uptick in the nation’s manufacturing employment over the last two years, argues that rising wages in China, high transportation costs and falling United States energy costs should bring more manufacturing back home.

With the rapid growth of middle classes abroad, trade should be an opportunity for the United States to sell into growing markets, increasing opportunities and wages for many Americans here at home. But over the last decade, that has not been the story.

A version of this post appeared in The New York Times Economix blog on August 29, 2012.