Tuesday, May 8, 2018

Free-Trade Economists Begin To See the Light on Trade and Globalization



The great economist John Maynard Keynes once observed that people who think themselves practical and realistic are often slaves of “some defunct economist.” One long-dead economist who is poised to become very defunct is old David Ricardo. His free trade views have dominated thinking on international trade for decades. But the tide is beginning to turn among economists, as some highly respected academic economists move away from their doctrinaire position that all free trade is always good everywhere and begin to acknowledge that free trade and large trade deficits have inflicted substantial harm to the US economy. 


Between 1945 and the 1970s, free trade (i.e. lower tariffs and growing world trade) contributed to rising US prosperity. From the mid-1970s on, growing trade has also meant the growing loss of good-paying jobs to competitor nations. Around the year 2000, the process accelerated. Globalization entered a new phase, which Harvard economist Dani Rodrik has labeled “hyperglobalization.” Three significant effects launched hyperglobalization: in 1993 the US signed onto NAFTA, in 1999 the European Union launched the euro, and in 2001 China joined the WTO. The NAFTA agreement effectively erased the border between the affluent US and low-wage Mexico. Creation of the euro tied together 19 economies with a single currency, effectively forcing 18 other economies to compete with Germany without the tool of an independent exchange rate. The admission of China to the WTO gave that nation a guarantee of long-term low tariffs that opened the US to Chinese imports and opened China to foreign investment by multinationals eager to produce in China. 

The tsunami of imports into the US decimated the US manufacturing base. Between December 1999 and December 2007, we lost 34% of our manufacturing labor force.  This was actually worse than the Great Depression. According to a Congressional Research Service study[1], the US lost 31% of its manufacturing workforce between 1929 and 1933. Of course, there are large differences between the 1930s and today. The Depression was more broadly-based than today’s industrial decline. On the other hand, in 1937, four years after the Depression’s nadir, manufacturing employment was back up to the 1929 level of 10.7 million. It’s now eight years since the 2010 nadir and there is no sign we will get back to the 17 million level of 2000 anytime soon. 

To those in affluent parts of the US, the economy seems better than ever. Real GDP is up, and unemployment is down. We have been in an economic recovery for nine years. Yet wage growth has stagnated for years, especially for the lower two thirds of the working population. Figure 1 shows that the real (inflation-adjusted) weekly income of the average American production and nonsupervisory worker peaked in 1972. Today, at $756 a week, it is 10% below its 1972 peak. A 10% decline in 46 years! That’s unprecedented in modern American history. 

 
Figure 1: Real Weekly Earnings of US Production/Nonsupervisory Workers in 2018 dollars. Source: BLS via Advisor Perspectives.

A large part of the reason for declining real wages for production workers is straightforward competition from low-wage countries, led by China and Mexico. Other nations in Asia, Latin America, and eastern Europe also play roles in the downward pressure. The wage pressure has been exacerbated by a full suite of predatory practices pursued by many nations to favor their export and manufacturing industries. Those practices include: Chinese government support and subsidies worth billions of dollars to industries the Chinese government aims to grow into world leaders; currency manipulation to favor exports and restrain imports by numerous nations; and policies to favor saving and restrain consumption by China, Germany and other nations, which have the effect of supporting growth in their economy at the expense of importing nations like the US. 

The effect of hyperglobalization in the US is to drive down the incomes of millions of low and middle-income workers working in sectors that compete internationally, drive millions of employees out of tradable sector jobs and into service jobs, some of which pay well but the majority of which offer worse pay and benefits than jobs in tradable sectors, and finally drive millions of people out of the labor force altogether. Another important effect is to increase income inequality in the US. A 2016 report from the bipartisan Congressional Research Service[2]used Census Bureau data to show how income inequality has increased in the US in the last 40 years (see Figure 2). According to the authors: “Between the mid-1970s and 2000, high-income households experienced rapid real income growth relative to middle- and low-income households, but incomes grew on average for all quintiles. Between 2000 and 2015—a period that includes two economic recessions—average incomes fell in the bottom three quintiles of the distribution, and the previous rapid growth in mean incomes enjoyed at the top of the distribution stalled.”
 
Figure 2.  US real household income by quintile 1967-2015. Data shows large shift in income to top quintile over the period, minimal progress since 1967 for bottom three quintiles, and a slowdown for all five quintiles after 2000. Source: Congressional Research Service.

The US income distribution can also be analyzed on a regional basis. An insightful 2016 study by the Pew Research Center looked at Census data on household income for America’s 229 largest metropolitan areas. Pew divided each metro’s households into three categories: upper-income, middle-income, and lower-income. They found that a decline in the share of middle-income households was widespread across the majority of metropolitan areas, along with stagnation or real declines in income: 

“Among American adults overall…the share living in middle-income households fell from 55% in 2000 to 51% in 2014…The decline was pervasive, with median incomes falling in 190 of 229 metropolitan areas examined…The decline of the middle class is a reflection of rising income inequality in the US.”[3]

Surprisingly, some of the 190 metros with declining median income included cities with reputations for regeneration through attracting new industries. For example, Raleigh, North Carolina has successfully built a sizable high-tech software industry. Yet in Raleigh, the middle income group fell from 55% to 50% of the metro population.  The cities where median income rose tended to be located on both coasts. A separate data point showing the wide regional breadth of dissatisfaction with the state of the economy was the geographic split in the 2016 presidential election results: although winning the popular vote, Hillary Clinton only won 487 counties, as compared to Trump’s 2,626 county wins. 

In summary, in the last 45 years the US has suffered growing income inequality and for the last 18 years income stagnation on top of continued growing inequality. Much of this was the product of imports eliminating good-paying manufacturing jobs. Nevertheless, the majority of economists have continued to defend free trade. The nation’s foremost trade economist, former Princeton professor, Nobel Laureate and New York Times columnist, wrote[4]in 1997: “The idea of comparative advantage—with its implication that trade between two nations normally raises the real incomes of both—is, like evolution via natural selection, a concept that seems simple and compelling to those who understand it.” This idea continues to be taught as absolute truth in virtually every undergraduate economics course in every college in America. And, on May 3rd, more than 1,100 economists signed a letter[5]to President Trump and Congress claiming that “fundamental economic principles” showed that Trump’s tariffs and his focus on the trade deficit would hurt the nation’s economy. “Economists urge you not to repeat that mistake,”they said. 

Cracks in the Consensus
However, below the radar, a growing pile of academic studies are providing evidence that trade and in particular imports from China and Mexico are the primary cause of the deteriorating outcomes in employment, wages, and other key economic variables in the 21stcentury, including even the rise in early deaths of midde-aged Americans. Many economists have contributed to this literature but none more so than MIT professor David Autor.  In a series of studies over the last five years, Autor and his co-authors have used statistical techniques to show the close correlation between imports from China and economic and social costs inflicted on the affected communities. 

These studies reached a tipping point within the economics profession with a 2016 paper entitled The China Shock: Learning from Labor-Market Adjustment to Large Changes in Trade[6], written by Autor, David Dorn of the University of Zurich, and Gordon Hanson of the University of California at San Diego. China Shockused statistical analysis of Census data on 722 US “commuting zones” (CZs), with measures of the impact of Chinese imports on each CZ to estimate the influence of those imports on unemployment and other key outcomes. Autor, Dorn, and Hanson (ADH) found that Chinese imports were associated with higher levels of unemployment, longer periods of unemployment, greater job churn, and lower lifetime earnings for affected individuals. They also found that China import penetration was significantly correlated with greater use of government social welfare programs including medical care, disability payments, and income support. ADH estimated Chinese imports were directly responsible for 2.4 million job losses, and the total figure could likely be higher. The job losses, income reductions, and other negative outcomes were all strongly concentrated among relatively lower-wage employees. From the study:

“In response to a given trade shock, a lower-wage employee experiences larger proportionate reductions in annual and lifetime earnings, a diminished ability to exit a job before an adverse shock hits, and a greater likelihood of exiting the labor market.”

But even more important for the economics profession than the detailed evidence on the serious, persistent negative consequences of the China shock was ADH’s critique of standard economic theory: “If one had to project the impact of China’s momentous economic reform for the US labor market with nothing to go on other than a standard undergraduate international economics textbook, one would predict large movements of workers between US tradable industries, limited reallocation of jobs from tradables to nontradables, and no net impacts on US aggregate employment. The reality of adjustment to the China trade shock has been far different.”
Standard economic theory holds that if workers lose their jobs due to imports, they ought to be able to find other jobs in other industries and/or different regions of the country paying similar wages—and ultimately move to higher-paying industries as America exploits its alleged “comparative advantage,” as first laid down by David Ricardo 200 years ago. Autor, Dorn, and Hanson showed that this has not been true of America’s experience with Chinese imports. On the contrary, instead of moving to jobs in other industries nearby, ADH found that the decline in jobs in an import-affected region was actually greater than the decline in jobs in that industry. 
“Trade-induced manufacturing declines in CZs are not, over the course of a decade, largely offset by sectoral reallocation or labor mobility. Instead, overall CZ employment-to-population rates fall at least one-for-one with the decline in manufacturing employment, and generally by slightly more. These results run counter to a precept of general equilibrium trade theory that the local employment effect of sectoral demand shocks should be short-lived, as the forces of wage and price arbitrage and labor mobility dissipate these shocks nationally.”
The force of ADH’s evidence and the critique of economic theory is turning into a tipping point. In the 18 months since The China Shock was published, a number of leading economists have publicly admitted that they were wrong in dismissing imports as a problem for the US economy. For example, Brad DeLong of UC Berkeley, a former official in the Clinton administration, admitted at a public forum in April 2017: “I have had to substantially revise upward my own views about the negative U.S. domestic consequences of trade and the China shock that hit the U.S. starting the late 1990s as China’s industrialization shifted from something impressive for a very poor country to something that was of world historical consequence.”

Harvard economics professor Dani Rodrik published a book in 2017, Straight Talk on Trade[7], in which he blamed hyperglobalization for greater inequality and suggested that free trade agreements and World Trade Organization decisions were driven more by multinationals’ lobbying power and desire to maximize profit than by any rational economic concern for shared international prosperity. He also criticized the economic profession for ignoring real-world evidence: “Economists do not fully understand why expanded trade has interacted with the macroeconomy to produce the negative consequences for wages and employment that it has…we should not act as if our cherished standard model has not been severely tarnished by reality.”

But the most surprising admission of all came from Paul Krugman, the dean of American trade economists. In a little-noticed article[8]published in March 2018 on his academic website, he admitted he had been mistaken in the 1990s in dismissing the effects globalization could have on total employment and manufacturing employment. After noting that manufacturing employment “fell off a cliff” in the decade after 1997, he wrote that he now believes the trade deficit was a major factor in employment decline and in economic hardship in the local areas affected by imports. “The deficit surge reduced the share of manufacturing in GDP by around 1.5 percentage points, or more than 10 percent, which means that it explains more than half of the roughly 20 percent decline in manufacturing employment between 1997 and 2005…What did the 1990s consensus that the adverse effects of globalization were modest miss? A lot…This was, I now believe, a major mistake—one in which I shared.”

Krugman’s admission of the importance of the US trade deficit and the negative consequences of hyperglobalization put him in broad agreement with…President Trump and Senator Bernie Sanders on the trade issue.  Krugman may be reluctant to say that too publicly.

The admission by these influential academics of the importance of the trade deficit is just the beginning of growing recognition by economists that the traditional economics of David Ricardo’s Law of Comparative Advantage and the hands-off approach to free trade do not apply in today’s world. If employment does not adjust rapidly, as Ricardo’s theory says it should, and if deficits are large and persistent, a new approach to international trade is needed. That’s the subject of a future article.



[1]Linda Levine, Congressional Research Service, The Labor Market during the Great Depression and the Current Recession, 2009. Available here.
[2]Sarah Donovan, Marc Labonte, Joseph Dalaker, Congressional Research Service, The U.S. Income Distribution: Trends and Issues, December 2016. Available here
[3]Pew Research Center, America’s Shrinking Middle Class: A Close Look at Changes Within Metropolitan Areas, 2016. Available here.
[4]Krugman, Paul, Ricardo’s Difficult Idea, 1997. Available here.
[5]National Taxpayers Union, More than 1,100 Economists Join NTU to Oppose New Tariffs and Protectionism. Available here.
[6]Autor, Dorn, Hanson, The China Shock: Learning from Labor-Market Adjustment to Large Changes in Trade, August 2016, Available here.
[7]Available here.
[8]Krugman, Paul, Globalization: What Did We Miss?March 2018. Available here.

No comments:

Post a Comment