The Left and the Right disagree on just about everything, but if there’s one thing that people on both sides of the political spectrum believe, it’s that trade with China has hollowed out North America’s manufacturing sector and devastated once-thriving communities in the industrial heartland.
Indeed, after China joined the World Trade Organization in 2001, U.S. manufacturing employment dropped sharply while imports from China surged. However, this empirical relationship mostly depicts correlation, not causation, and the conventional narrative about the so-called “China shock” is more fiction than fact.
Graphic credit: Janice Nelson.
Economists David Autor, David Dorn, and Gordon Hanson coined the term “China shock” to describe how China’s WTO accession disrupted U.S. labour markets. Their research shows that regions in the United States that were relatively more exposed to competition from Chinese imports experienced job losses, lower wages, and other adverse labour-market outcomes following China’s accession to the WTO relative to less-exposed regions. Andrew Sharpe and Alexander Murray, economists at the Centre for Living Standards, documented similar findings for Canada.
The widespread belief that the China shock is responsible for putting millions of Americans out of work is due in no small part to the research of Autor and his coauthors. This is exemplified by a recent op-ed in the Wall Street Journal by U.S. Treasury Secretary Scott Bessent, who cites their work as evidence that “3.7 million Americans lost their jobs” as a result of trade with China, accounting for “59.3 percent of manufacturing job losses.”
However, this isn’t what Autor and his coauthors’ findings say.
Their research is designed to answer a very specific question: how did regions that were more exposed to Chinese import competition fare relative to less-exposed regions? The word “relative” is crucial; their findings don’t say anything at all about the total number of jobs lost, only the differences in job losses across regions. In fact, their findings are consistent with a narrative in which trade with China actually increased overall employment, destroying jobs in import-competing regions while creating jobs elsewhere. It’s even possible that trade increased manufacturing employment everywhere, just less so in import-competing regions.
The evidence indicates that while industries that compete with Chinese imports may have been hurt, other industries, especially those that use inputs that China produces, have indeed gained. Nicholas Bloom and coauthors found that while trade with China did cause job losses in manufacturing, it also caused job gains in the service sector. Another study by Zhi Wang and colleagues found that these service-sector gains boosted overall employment, even in regions that were highly exposed to industries that experienced losses due to import competition.
Determining just how much of the decline in manufacturing employment during the 2000s was caused by trade with China is challenging because there are many other confounding factors at play. Among the most important are the steady long-term decline in manufacturing employment that began decades before China joined the WTO in 2001, and the recession the U.S. economy experienced that very same year. My own research, which tries to separate the effects of these various forces, indicates that trade was responsible for only 15 percent of the decline in manufacturing employment during this period, a quarter of the figure cited in Bessent’s op-ed. Another study by Caliendo, Dvorkin, and Parro reports similar results.
Regardless of how many manufacturing jobs the China shock destroyed, the United States remains a manufacturing powerhouse. The U.S. manufacturing sector’s real output grew rapidly from 2000-2007 and is currently at its all-time high. This is due largely to the fact that productivity in this sector has grown more quickly than in the rest of the economy; fewer workers are needed each year to produce the same quantity of goods. My research indicates that this trend, rather than trade, is responsible for the bulk of the decline in manufacturing employment.
Graphic credit: Janice Nelson.
Empirically, China’s WTO accession looks more like the culmination of a long period of gradual integration with U.S. markets than a sudden shock to the world trade order. U.S. tariffs on China didn’t change at all—Chinese exporters had gained access to the most-favoured-nation (MFN) tariff rates enjoyed by WTO members 20 years earlier—and the share of U.S. imports coming from China grew much faster during the 1980s and 1990s than in the 2000s.
Some economists have argued that joining the WTO reduced uncertainty about trade policy by making it more difficult for the U.S. to revoke China’s MFN status, but my research indicates that the probability of this actually happening did not materially change.
Graphic credit: Janice Nelson.
Regardless of how much truth there is to the China-shock narrative, focusing on the labour-market effects misses the big picture entirely. For the vast majority of households, trade with China has led to lower consumer prices, not dislocation from employment. A recent study from the Federal Reserve Board of Governors finds that “trade with China increased U.S. consumer surplus by about $400,000 per displaced job.” Moreover, low-income households have reaped the greatest benefits: “product categories catering to low-income consumers experienced larger price declines.”
Perhaps it’s no surprise, then, that despite its populist pretext, most Americans oppose President Trump’s tariff policy.
