Daron Acemoglu responds to Dean Baker

Politics is indeed a major factor in the plight of American workers. But the challenges they face are more multifaceted and in some ways different than the ones Dean Baker describes.

Baker focuses on one economic driver behind the transformation of the U.S. labor market: globalization. But the more potent source of change is technology. Labor economists have documented extensively the impact of automation on wages and jobs. If our goal is to isolate causes, then we need to account for technological change. And if we want to do more than that—if we want to enact new policy—such accounting is doubly important. If we leave out technology, we will not be able to pursue our best options for creating equitable growth, and we will not handle well the winds of change buffeting the economy.

New technologies and globalization—the surge in trade and offshoring—share two key characteristics: both have generated considerable GDP growth by reducing the costs of certain production tasks and both have created losers along with the winners. Costs go down when the labor of American workers is replaced either by machines or by that of foreign workers. Consequently, consumers can buy more goods or save more money. But some American workers, particularly those with manufacturing jobs, get laid off or lose wages. Thus globalization and technological change together contribute to increasing income inequality in the United States and helped cause the plight of American manufacturing workers.

In emphasizing the role of automation, I do not mean to absolve the political process. Baker deserves credit for orienting our thinking toward the political underpinnings of economic distribution in general. Perhaps the fact that they have not been on the agenda was also a political choice. Thus too little attention has been paid to the role of deregulation and the gradual erosion of the real value of the minimum wage in fostering inequality. At the same time, the oversized political influence of Wall Street drives huge growth in financial-industry profits, executive pay, and traders’ earnings. And Baker correctly argues that it was a political choice to stand aside and let the losers suffer while only some workers and firms benefited from globalization. But one should add technological change to this critique: policy also shapes the distributional consequences of technology.

There are several ways a society can modify the distributional effects of new technologies and globalization. The most obvious is to block them, to protect potential losers. Though nobody is currently calling for blocking the development and adoption of new technologies, rolling back globalization appears palatable to many, and trade barriers and the repudiation of America’s trade deals are a critical campaign promise of Donald Trump. But blocking trade in new technologies forgoes the considerable gains these opportunities would bring. Throughout history, blocking new technologies has been a surefire way of condemning nations to poverty. When Russian tsars prevented railway construction in the first half of the nineteenth century, or when fourteenth-century Ming emperors banned not just overseas trade but also shipbuilding, they set back economic development in their countries by several generations. The situation is no different for the United States and other Western countries today. To his credit, Baker does not advocate this response. Nevertheless, it is important to articulate alternative, effective solutions. Otherwise, emphasis on the adverse distributional implications of trade, and perhaps soon those of new technologies as well, will inevitably strengthen those wishing to block them.

Unfortunately, Baker’s proposals likely won’t be very effective. He singles out policies aimed at reducing the extent of the U.S. trade deficit or changing its composition. I believe Baker is too optimistic about the ease with which the trade deficit could be reduced. A basic identity in economics links the current-account deficit (which is slightly broader than the trade deficit because it includes factor payments) to a surplus in the financial or capital account. Put simply, a country can only import more goods than it sells abroad if foreigners are willing to lend it enough money (e.g., by buying or investing in its domestic assets) to generate a sufficient surplus in its financial relations with the rest of the world. Similarly, if foreigners are pouring money into a country’s domestic assets, the same identity applies and makes a current-account deficit inevitable. Therefore, policy can reduce the current-account (and trade) deficit only if it simultaneously reduces the financial-account surplus. The remarkable fact of the last two decades is that foreigners have been willing to invest in and hold U.S. assets, including U.S. government bonds, despite virtually no real returns on many of these assets. Given voracious demand for U.S. assets, changes in trade policy probably couldn’t have significantly affected the current-account deficit.

Perhaps judicious policy could have changed the composition of the deficit even if it did not eliminate it. Baker’s proposal for greater openness to trade in services, such as health care, is definitely a step in that direction. Yet it is not clear how effective it would be. Many services, such as those provided by physicians, cannot be imported. Advances in technology might change this in the years to come, but at the moment, the only way to import medical services is via immigration. I agree with Baker—and this must be emphasized in these times of strident anti-immigrant rhetoric—that increased immigration of health, education, and engineering professionals is not just desirable but in fact an imperative for the U.S. economy to maintain its edge as the world’s technology leader. Yet even a considerable increase in immigration of skilled professionals would not have greatly altered the composition of the current-account deficit. The United States would still have spent recent decades importing far more manufactured goods than it exported.

Rather than target the current-account deficit directly, I suggest three other responses. In combination, they can do much to increase U.S. benefits from the productivity gains of technology and trade while redressing some, if not all, of their adverse distributional consequences.

The first proposal is the most straightforward: a social safety net and redistribution via the fiscal system can ensure a more equitable distribution of gains from automation, offshoring, and trade. So far, American politicians have rejected this approach, despite the fact that the United States has one of the weakest social safety nets in the Western world. A modern social safety net would not only help workers recoup lost wages but also help the next generation of Americans from poorer backgrounds acquire the higher-quality education and job skills they will need to compete in the labor market of the future.

In this sense, a social safety net is related to the second type of response, which focuses on investing in workforce skills and capabilities so that people can take advantage of new opportunities and shield themselves from adverse consequences of change. This may be the area in which the United States has failed most miserably. To put it bluntly, the United States prepares its workers for the technology and jobs of the twenty-first century using an education system that is for the most part a remnant of the mid-twentieth century. What is sorely needed is not just more resources but also a total revamp of the organizational structure and curriculum of middle and high schools. Investment in new educational directions should be complemented with on-the-job and vocational training programs, both for workers seeking new skills related to their long-term occupations and for those looking to switch occupations and industries. Training, however, is another area in which U.S. investments have been sadly inadequate. Politics on both sides of the aisle have conspired against change. Conservatives are unwilling to invest tax revenues for public services and deal with the plight of less-fortunate Americans; liberals have been unwilling to confront teachers unions whose stubbornness leaves students less prepared to rise to the challenge and opportunity of twenty-first century jobs.

The third type of response starts by recognizing that technology is not a force entirely out of our control; it doesn’t act upon a passive society. It is shaped by institutions, policies, and the choices of government officials, scientists, entrepreneurs, and businesspeople. The United States has poured considerable resources into technologies that automate the jobs of its middling workers and raise the productivity of its most highly paid. The distributional effects of technology were not preordained. It was and remains possible to support new technologies that generate employment opportunities and increase the productivity of workers, rather than just replacing them.

The path to economic growth in Germany, and to some degree Scandinavia, illustrates an alternative to the current direction in the United States. German manufacturing has thrived while U.S. manufacturing has folded. This was achieved by deploying all three of the responses I suggest. Germany has a more protective social safety net and a more flexible education system strongly supplemented by worker training. And German firms have invested in increased manufacturing productivity without eliminating too many low- and medium-skill jobs. (It should be noted, though, that German manufacturing has benefited from the relatively low value of the euro compared to German productivity, boosting the country’s exports.) These choices have not prevented inequality, which has risen sharply in Germany since the early 2000s. But they have helped to spare manufacturing workers and low-paid service employees large-scale job losses and earnings collapse. Though it is not possible or desirable for the United States to emulate Germany or Scandinavia wholesale, there is still much we can learn from their experiences.