Just how detrimental is international trade to American jobs? Spoiler alert: not as much as some may think.
Protectionists often warn that engaging in trade with countries that offer lower wages will erode employment opportunities for American workers. In a colorful metaphor reminiscent of Ross Perot, they predict a “giant sucking sound” as jobs vanish from the U.S. to these low-wage nations.
Theoretically, however, there’s scant evidence to support this notion. Wages are primarily influenced by the marginal productivity of workers. Since workers in different countries are not interchangeable, firms will not automatically chase after cheaper labor. The reality is that workers in other countries earn less than their American counterparts not merely because they accept a lower standard of living, but because their marginal productivity is lower. To illustrate, it’s akin to the New England Patriots passing over my non-existent football skills in favor of a fresh-faced 23-year-old quarterback from UNC—why pay for someone who brings little to the field?
According to conventional trade theory, significant job offshoring leading to wage declines only occurs under specific conditions: when productivity and technology levels are identical, and one country focuses on labor-intensive goods while the other specializes in capital-intensive products. (In such scenarios, we might hear that “giant sucking sound” more likely coming from Canada instead of low-wage countries like Mexico or China.) Even if trade does impact jobs, the expectation is that overall employment levels remain stable as workers shift between sectors, with some jobs disappearing while others emerge.
Standard economic theory provides compelling reasons to question the claims of protectionists. Yet they present an alternative narrative. To determine which theory holds more water, we can examine empirical data related to job security.
One effective measure of job security, developed by economist Steven J. Davis from the University of Chicago, is the ratio of initial jobless claims to the total employed population. Jobless claims represent instances where individuals separate from their jobs involuntarily, indicating a lack of security—unlike those who leave voluntarily. A higher percentage denotes a less secure job environment, while a lower percentage suggests greater stability. The following graph illustrates this trend from 1967 to 2019 (I opted to stop the graph before the pandemic, as the data became erratic during that period, but stabilized once lockdowns ended).
Two key observations arise:
First, since the 1980s (excluding recessions), there has been a consistent decline in the average percentage of jobless claims, indicating that American workers are generally experiencing increased job security. Even during the recession of 2008, the percentage of jobless claims was lower than during previous recessions in the 1970s and 80s.
Second, significant trade agreements like NAFTA (1994) and China’s accession to the WTO (2000) did not alter overall job security levels. The downward trend since the 1980s remains intact, disrupted only by recessions. If these trade deals wreaked havoc on American employment as protectionists suggest, we would expect to see some evidence of this in the data. Instead, the fluctuations we do observe correlate with economic downturns rather than trade policies.
“Hold on!” I can almost hear some protectionists exclaim. “Of course, it wouldn’t show in the aggregate! People lose manufacturing jobs but find themselves more secure in lower-wage positions. That’s the real concern!” Fair point, but the same reassuring trend is evident within the manufacturing sector as well.
The next graph provides insights into American manufacturing jobs, showcasing layoffs and discharges as a percentage of total manufacturing employment. The timeline is shorter because the Bureau of Labor Statistics only began collecting this data in 2000.
Once again, aside from recessions, it’s clear that job security for manufacturing workers has improved over the nearly two decades represented. In fact, the data works against the protectionists’ narrative; job security in manufacturing has actually strengthened since the so-called China Shock, with little change following NAFTA.
Protectionists may counter by arguing, “You’re considering too broad a perspective. It’s the regional impact that’s significant. Just look at the devastation of the Rust Belt! That’s all the evidence you need regarding the perils of international trade!”
Regional effects are a valid concern. In his insightful 2017 book, Clashing Over Commerce, Douglas Irwin highlights that U.S. trade policy has historically been influenced by regional dynamics. However, even the Rust Belt doesn’t fully support the protectionist argument. A recent study published in the Journal of Political Economy by Simeon Alder, David Lagakos, and Lee Ohanian found that “labor conflict accounts for half of the decline in the region’s share of manufacturing employment. Foreign competition plays a smaller role, and its effects are concentrated after most of the region’s decline had already occurred.” Essentially, they concluded that labor market factors, rather than international trade, are the primary drivers of the Rust Belt’s challenges.
To address every argument against international trade would require a tome. Adam Smith’s Wealth of Nations serves as a solid foundation; he effectively debunked protectionist claims of his time, and remarkably, those arguments have remained unchanged since the 1700s. Similarly, Douglas Irwin’s Free Trade Under Fire provides further insight. The evidence suggests that international trade isn’t the “gutting” force many protectionists allege. Instead, macroeconomic factors primarily dictate the number of jobs within a country, with international trade playing a secondary role. This isn’t to say that trade has no impact; some jobs have indeed vanished while others have emerged, but the overall net job effect has hovered close to zero.
Readers may notice I’ve sidestepped the protectionist claim that trade has “deindustrialized” the United States. This omission is intentional. There are two interpretations of “deindustrialization”: absolute and relative. I term these absolute deindustrialization and relative deindustrialization.
Absolute deindustrialization occurs when manufacturing output declines in absolute terms. This is typically the interpretation used by politicians and activists opposing free trade. By this definition, trade has not deindustrialized the U.S. According to the Federal Reserve’s monthly report on Industrial Production, U.S. manufacturing output is currently below its 2007 high but is still substantially higher than prior to China joining the WTO or the implementation of NAFTA. Manufacturing may have plateaued, but it is far from declining.
Relative deindustrialization, on the other hand, refers to a decrease in manufacturing output relative to other nations. Trade economist Richard Baldwin employs this definition (see, for instance, this blog post). In this sense, the U.S. is indeed experiencing relative deindustrialization due to trade, primarily because the economy is transitioning toward a service-oriented model. Although we continue to produce significant quantities of manufactured goods (including high-value items—U.S. manufacturing achieved a seasonally-adjusted annual rate of $7.5 trillion nominal dollars in early 2026), we are producing a relatively larger share of services in technology, healthcare, finance, education, and similar sectors.
America’s competitive edge lies in knowledge-based goods rather than traditional manufacturing. Our economy effectively capitalizes on our relatively high skill and expertise levels. Even our manufacturing output focuses on products that demand high skill levels to produce. This brings us to a final consideration: reshoring manufacturing could be a costly endeavor, as a skills mismatch exists.
