Trade and Wages

In a recent essay at American Compass, Michael Lind attempts to refute certain aspects of economists’ case for free trade.  Others have addressed the numerous empirical, factual, and theoretical issues with his essay.  I will focus on just one particular claim.  Lind writes:

For the sake of space, I will ignore his claim on left-dogma and focus on the claim of right-dogma: “deregulating trade and immigration to reduce wages for workers and thus reduce prices for consumers is the “efficient” and thus best policy, as long as the “winners” compensate the “losers”—preferably in the form of redistribution through the tax code.”  Lind provides no citations or links supporting his claim, so it is hard to tell who (or what), exactly, he is responding to here.  Open any trade textbook and you won’t find such dogmas he claims are there.  

Rather, I think he is referring to a potential outcome in international trade called the Factor Equalization Theorem, independently derived by Wolfgang Stolper & Paul Samuelson in 1941 and Abba Lerner in 1952.  Sparing you, dear reader, the technical details, I’ll note that the theorem states that, under certain conditions, when two countries trade the trading partner that is relatively labor abundant will see wages rise and returns to capital fall while the country that is relatively capital abundant will see returns to capital rise and wages fall.  According to this theorem, the factor prices (wages and returns to capital) will equalize across trading partners: wages will be equal between the two countries and returns to capital will be equal between the two countries.

Assuming I am correct that he is pulling off the Factor Equalization Theorem, Lind seems to be seizing on it, imparting to it certain claims no one actually holds, and claiming these as free-trade dogmas.  The problem is that the theorem hasn’t held up well to empirical scrutiny.  The assumptions in it are too strong.  In particular, the theorem requires that labor across the trading partners is virtually identical (the same with capital).  In reality, labor is not identical across many trading partners.  American workers are extraordinarily productive: we operate in a capital-fueled economy with powerful and stable institutions, high education, and generally high access to productivity-improving infrastructure.  Chinese workers, by contrast, do not: they are much less productive.  According to the World Bank, the average Chinese worker produces approximately $42,000 worth of goods and services a year.  Conversely, the average American worker produces approximately $150,000 worth of goods and services per year, making the average American worker 257% more productive than the average Chinese worker.  A Chinese worker is not a good substitute for an American worker.  We shouldn’t expect to see American wages fall toward Chinese wages with trade.  Indeed, we do not see American wages falling.

To make this point less abstract, consider the following: the New England Patriots, my hometown football team, is in desperate need of a good quarterback.  I, a 35-year old man, would love to play for the Patriots.  In fact, they could offer me the league minimum wage ($795k) and I would instantly quit my job and go play for the Patriots.  Millions of other New Englanders (and Americans, for that matter) would also take the Patriots up on that offer.  Why, then, did the Patriots offer 21-year old rookie Drake Maye from UNC almost $60 million ($36 million over 4 years plus $24 million signing bonus)?  The answer is obvious: He and I are not the same by a long shot.  No one would reasonably expect our wages to equalize (mine rise and his to fall).  The theorem doesn’t hold in this case.

Factor price equalization will more likely occur between very similar trading partners like the US and Canada: the technology is similar, there are lower costs to relocating, and workers are similar.  But factor price equalization will not occur as a matter of course from trade, as Lind seems to think.  In fact, just the opposite could just as easily occur.  Factor price equalization is a special case, not a general case, of trade.

P.S. In an article forthcoming in the Review of Economics and Statistics (ungated version here), Aaron Flaaten and Justin Piece of the Federal Reserve found that Trump’s tariffs resulted in “relative reduction in manufacturing employment.”  Depending on details, this reduction in demand could likely have led to a net relative reduction in wages too.  This would suggest that protectionism, not trade, is the bigger threat to American wages.