It’s not a puzzle if American workers oppose trade agreements

Yesterday, NPR’s Chris Arnold wrote the latest in what has become a very long line of “explainer” pieces about economic globalization and the presidential campaigns. Nearly all of these pieces seek to resolve an alleged puzzle: nearly all reputable economists argue that policy efforts to boost trade are good for the U.S. economy, yet many (if not most) American workers strongly oppose trade agreements signed in recent decades.

Arnold puts forward a pretty common solution to this alleged puzzle: “trade’s benefits are diffuse, but the pain is concentrated.” In this view, the only losers from trade are those workers directly displaced by imports. Every other consumer in the economy benefits from lower prices. But because the losers are small and concentrated, they can organize to oppose trade agreements. And while the winners are numerous and widespread, the benefits (e.g., slightly more affordable clothing and DVD players) are hard to notice, so no one organizes to support these agreements.

This is a common way of describing the effects of using policy to expand trade, but on the economics, it is certainly not correct. In textbook trade models, using policy levers (lower tariffs, for example) to boost trade with poorer countries will indeed cause total national income in the United States to rise. But these same textbooks also predict that the resulting expansion of trade will redistribute far more income than it creates. And the direction of this redistribution is upward. So it is perfectly possible to have policy efforts to expand trade lead to higher national income yet leave the majority of workers worse off. Importantly, the losers in these textbook models are not just workers directly displaced by imports—they’re all the workers in the entire economy who resemble the trade-displaced in terms of education and credentials.

To put it simply, landscapers might not lose their jobs to rising imports, but their wages are hurt if they have to compete for jobs with workers who were laid-off from apparel factories as imports replaced domestic clothing. In the end, for many trade flows (particularly, say, trade flows between the United States and countries like Mexico and China) the common formulation is exactly backwards. The losers are the majority of American workers who still lack a 4-year college degree, while the winners are the minority of workers who do have these degrees.

In the NPR story, Arnold and MIT’s David Autor provide some intuition behind the effects of trade. Their example has two countries (France and the United States) and two goods (cars and cheese). What it’s missing is two kinds of workers—those with and those without substantial human capital (think a 4-year college degree). After all, if there is only one kind of worker, then worrying about redistribution makes no sense. With two kinds of workers, the intuition gets clearer.

Take another example, one that swaps out cars and cheese for two goods that will sharpen the intuition a bit. In this example, the United States produces and exports more human capital-intensive goods (say airplanes) and imports more labor-intensive goods (say apparel). By focusing on what we’re relatively better at producing (capital-intensive airplanes) and trading this extra output for what our trading partners are relatively better at producing (labor-intensive apparel), we can see total national incomes rise in both countries. This specialization in the United States requires shifting resources (i.e., workers and capital) out of apparel production and into airplane production. But each $1 in apparel production lost requires more labor and less capital than the $1 in airplane production gained—causing an excess supply of labor and an excess demand for capital. Capital’s return rises while labor’s wage falls.

What’s particularly striking in so many of these trade-explainer articles are the claims that wage pressures stemming from policy efforts to expand trade are some huge surprise. They shouldn’t be—again, textbook economics says clearly that the likely outcome of efforts to expand trade with lower-wage nations should be a redistribution of income to highly credentialed workers from the rest. And this prediction has been borne out.

Besides nodding to the wage effects of policy efforts to expand trade, most of Arnold’s piece is about job loss in manufacturing. He notes that the 5.4 million jobs lost in U.S. manufacturing between 1998 and 2015 have not been entirely driven by trade with China, which is largely true. He points the finger at other suspects like technology, but the real job-killer in manufacturing over that time has been the 2001 and 2008-09 recessions. (A side note on NPR’s graphic on this—it shows too smooth a decline between 1998 and 2015. It’s clearly not based on real data, just a straight line interpolation. This obscures the fact that the decline is recession driven, which may be why less time is spent on that interpretation.) The enormous rise in the U.S.-China trade deficit absolutely contributed to manufacturing employment’s failure to sharply rebound after these recessions. This trade deficit in turn absolutely has intentional policy roots: China decided to pursue a policy of buying U.S. assets at enormous scale, which boosted the value of the dollar and made U.S. exports less competitive in global markets. We in turn made the policy decision to accommodate this.

In the end, there’s no real puzzle about American workers’ attitudes towards our trade policy in recent decades. Genuine free trade with poorer nations would have been rough on the majority of American workers’ wages. So even if the trade agreements we signed in recent decades had been good-faith exercises in simple liberalization, it would not have been a shock that lots of workers opposed them. This is particularly true given that we have done nothing in the way of complementary policies to re-redistribute the income these workers lost due to expanded trade.

And it is even a lot worse than that. Trade agreements in recent decades have not been simple good-faith exercises in trade liberalization. Instead, they have exposed some American workers to fierce international competition while locking in rules that expanded protections for others (patent and copyright enforcement for industries like pharmaceuticals, software, and entertainment). Predictably, the beefed-up protections were offered to the already affluent, while the greater competition was foisted on those already struggling. Further, there’s no reason why expanded trade has to mean hugely unbalanced trade, with jobs displaced by imports not being matched by jobs created through exports. Given all of these policy failures, as well as the underlying economics of trade, it’s no surprise at all that many American workers see our recent decades’ trade policy as nothing but a regressive insult to their livelihoods.