If we are going to play the correlation game, we can find dozens of correlates if we try hard enough: how about the higher incidence of opioid use in counties that Trump carried? (Photo: Notions Capital/Flickr/cc)
So much for the precise correlation between economic factors and political preference based on rational choosing to maximize one’s own material advantage.
Mike Lofgren / 07.13.2018
Ever since the 2016 election we all know that economic distress and anxiety out in the Great American Heartland caused white working people to vote for Donald Trump. How do we know that? The media have told us so repeatedly, from corporate NBC to the wonkish FiveThirtyEight.
There has been some pushback since then, but now comes a research paper in the Oxford Review of Economic Policy (the full paper here; a summary here). In it, three British economists claim that that the automation of jobs through robotics swung the 2016 U.S. presidential election in favor of Trump. They assert a very precise correlation, such as in their statement that a with 10 percent lower level of automation in the state, Hillary Clinton would have carried Michigan, and they further calculate the exact national electoral college count by which she would have won the election.
All very impressive, with lots of academic rigor and quantification.
But we should know that correlation is not causation, no matter how many times the rooster’s crowing precedes the sunrise. Economics is justly called the dismal science, but it is not even a science: pretended rigor and bogus quantification are the means by which we are meant to be convinced by an academic guild that is, at best, making hunches about complex human behavior based on selective evidence.
If we are going to play the correlation game, we can find dozens of correlates if we try hard enough: how about the higher incidence of opioid use in counties that Trump carried? It would certainly explain the irrationalism that runs rampant through our politics. Or, since the authors trace automation to the very beginnings of the industrial revolution and tie it to radicalism, aren’t we entitled to go back in history and pick another correlate? Suppose, in 1987, that the Fairness Doctrine had not been repealed, and the political propaganda of hatred and incitement did not swamp the American airwaves? By the standards of the Oxford Review, would a 10 percent lower exposure to Rush Limbaugh have changed the result in Michigan?
There are now contrary views to the economic distress thesis: a paper in Proceedings of the National Academy of Sciences argues that status anxiety, not economic anxiety, was the prime motivator for white working class Trump voters. (Evidently, “status anxiety” is approved academic jargon for resentment of other races.) The author’s conclusions may be more soundly based than inference from levels of automation or other such evidence: the voters told the interviewer what their motivation was.
That does not prove that status anxiety swung the election, but it provides a likely corrective to one of the great fallacies of modern economic theory. It also matches abundant evidence we have seen since the election.
There is a strong current in the academic study of political behavior asserting that, contrary to what you read from frustrated liberal pundits, nobody actually “votes against his interests.” This view of voting activity is based on public choice theory in economics, which “assumes that people are guided chiefly by their own self-interests and, more important, that the motivations of people in the political process are no different from those of people in the steak, housing, or car market.”
Given the shakiness of rational utility maximization even in purely economic domains – as the housing bubble clearly showed – the idea that it applies to selecting political candidates, a process fraught with emotionalism, propagandistic and misleading appeals, and tribalistic identifications, borders on the absurd. It is also a warning against letting economic dogma, with its conceit about rational actors, perfect equilibria, and free markets, infiltrate other disciplines.
How would the rational actor thesis explain why more than three-quarters of American farmers voted for a candidate who repeatedly told them he would start a trade war with the country that is their fastest growing market in virtually all US agricultural products, and already their largest market in several of them? Or a candidate who promised to mess with NAFTA, a trade agreement that, whatever its possible disadvantages for domestic manufacturers, has been a boon to farmers?
Now China has retaliated with tariffs on U.S. agriculture. Have farmers, already facing years of declining agricultural prices, increasing bankruptcies, and the highest suicide rate of any occupational group, abandoned Trump? At least so far, nope. While their trade associations may criticize his policies, the farmers themselves whistle past the graveyard with the pious wish that the president who deliberately created the mess will magically solve it.
If there was ever one constituency that was solidly pro-Trump, it was the Harley-Davidson crowd. Yet in the face of EU tariff retaliation, job losses, and Trump personally threatening the company, they retain their childlike faith in him. “. . . most [of the workers] happily volunteer that they voted for Mr. Trump and would do so again – tariffs or no tariffs.” Somehow, they rationalize, Trump is playing a 9-dimensional chess game they just can’t fathom, and it will all work out for the best.
So much for the precise correlation between economic factors and political preference based on rational choosing to maximize one’s own material advantage. This is not to say that economic conditions have no effect on political outcomes; there is certainly some connection. But the rise of European fascism in the 1930s cannot be explained solely, or maybe even mainly, by the Great Depression; why did Germany go fascist, and neighboring Switzerland, with a 70 percent German-speaking population, did not? And why did Sweden develop a comprehensive social democracy in the same period?
Perhaps James M. Buchanan, one of the founders of public choice theory, got it all wrong. Or maybe the presumed rationality of ordinary citizens in his theory was just window dressing for his real message to his true constituency: his patrons, the rich. They were the only group worth bothering about (as Buchanan’s own ideology made clear), and his message to them was to forget all the sob stuff about noblesse oblige or good corporate citizenship, and to ruthlessly press their advantage. In view of the contrasting behavior of the rich during the last several decades, and that of the mass of the American people, Buchanan was certainly onto something.