OK, back on track. Anyway, let's just start out by getting right to the heart of Godwin's law. Germany elected Hitler because its economy was in tatters, its international standing was weakened, and so forth. It wasn't precisely an expression of intrinsic evil at the level of the country itself, so much as a demonstration of what happens when a country is economically and geopolitically downtrodden. People turn to authoritarians, not because they want to give up freedom and have authoritarian leaders do horrifically evil things, but because it can seem, in some dire circumstances, like a response to those dire circumstances.
This is sort of a question about human nature. How far do you have to push people before they turn to evil? From Hobbes versus Rousseau to Milgram, put people in the right circumstances with the right pressure and they do bad things, even if, yes, Milgram messed with his data, but let's not start on that fraud, Zimbardo, shall we?
This all leads to the appeal of the ubiquitous "a crappy economy caused Hitler" explanation, and extrapolations to the generalized theory of the rise of authoritarianism.
Trump is not Hitler. Not even close. He is, however, a racist authoritarian, and I'm not going to bother arguing the point here. However, I do want to pick up on a point I made in a recent post. Consider the actual circumstances of 2016. Donald Trump made some bizarre speeches about how horrible everything supposedly was in 2016, and, um... no. With respect to crime waves, and such, his speeches were little different from what Democrats say today, as long as you substitute the phrase, "gun violence," for the word, "crime," and none of it bears any resemblance to reality. Really, people, check the FBI Uniform Crime Statistics and CDC mortality rates. Minor changes in terminology don't change actual facts.
Economically, though, what was happening in 2016? In each quarter of 2016, here was the GDP growth rate: Q1: 1.5%, Q2: 2.3%, Q3: 1.9% : Q4: 1.8%. I'm just pulling these numbers from Federal Reserve Economic Data. This was after growing since Q3 of 2009. Unemployment had fallen to 4.7% by November of 2016, which wasn't quite where it is now, but still quite low, especially given the worst of the "great recession." Inflation was negligible. The S&P had been growing solidly. Economic rhetoric was weird, but the economy itself? It was doing pretty damned well in 2016.
When Britain voted for Brexit, their economy was hardly in tatters. Go around the world, and while you can make a case for Bolsonaro, the recent rise of anti-democratic authoritarians and trend towards democratic backsliding is not the result of collapsing economies. It's just... not.
Of all the countries that should have gone full-fascist by this model, we would have expected Greece, and yes, there has been a rise in support for fascism in Greece, but Tsipras? I think he's a moron, and an embarrassment to game theory,* but he's no fascist.
In How Democracies Die, Levitsky & Ziblatt write about the political processes by which a party ideologically colludes with an anti-democratic demagogue, and democratic norms erode, and OK, I'm not being fair to Britain by calling the Brexit vote that, although it did demonstrate a rise in anti-intellectual isolationism, which is connected to the rise of authoritarian demagogues elsewhere.
And that's really my point. The standard story about how you get anti-democratic authoritarians is that the economy collapses and people want a strongman to fix things, but... empirically, we have been seeing countries turn to a bunch of phony strongman wannabes when there's not much to fix. In the US in 2016, the economy was strong and growing, and crime was low. The factors that were supposed to be there to explain the rise of the authoritarian strongman weren't there.
Why did Trump win? Well, as I have written many times before, at the level of the general election, there is a pretty clear pattern. DDRRDDRR... Alan Abramowitz's "Time for a Change" model shows that the incumbent party doesn't tend to win a third term. The Republicans did in 1988, but the economy was REALLY strong then. GDP was growing at 5.4% in Q2 of 1988, and Q2 of the election year is the quarter that Abramowitz uses in his model. In 2016, growth wasn't strong enough to overcome the two-term penalty, which led to the model actually favoring the Republicans. As for why Trump won the Republican nomination, remember how this blog began-- with a series called: "Trump to Political Science: Drop Dead." It was me calling bullshit on my colleagues, telling them to wake up and realize that Trump was winning the nomination because their nomination models just sucked. Why did Trump win?
Basically, white identity politics, and this has been borne out by plenty of analysis. Support for Trump in 2016 was highly correlated with negative attitudes towards minorities. And this gets us at the core of... "the authoritarian personality." Or should I type... The Authoritarian Personality, by Theodor Adorno et al. There is variation at the individual level with respect to traits like anti-intellectualism, belligerence, racism/misogyny and a preference for order over liberty, all of which lead to a preference for a strongman-type leader. Sound familiar?
The model of the authoritarian personality has always been controversial, especially as originally conceived, but the central point I am making here is that there is variation at the individual level in preferences for a certain type of leader, independent of economic or political circumstances. Those with the individual characteristics, specified more clearly than in Adorno et al., who prefer authoritarians, will do so regardless of external economic or political circumstances because that's just what they want. It doesn't necessarily have a lot to do with the state of the economy.
In the wake of the 2016 election, one of the standard dodges that commentators offered for Trump supporters was that they were the downtrodden "working class" who had been somehow hurt by "globalization," and that Trump somehow "spoke to them" on the basis of "economic" issues, or some such nonsense. The problem is that the survey data never supported that story. Oh, sure, if you commit what we call the "ecological inference fallacy," you can convince yourself of some bullshit, but my job as a statistician is to bring the ruler down on your knuckles when you do that. Remember how the "rust belt" states and the regions with higher levels of opioid overdoses voted for Trump?
Does it follow that the individuals who had lost their jobs or gotten addicted/had family addicted voted for Trump? No. That's ecological inference. Ecological inference is the attempt to draw inference about individuals from group-level associations. It doesn't work. (At least, not without some special techniques, developed by Gary King, but even that is messy.) At the individual level, what do we know about who voted for Trump and who voted for Clinton?
Republicans voted for Trump, and Democrats voted for Clinton. Party ID. As it always has been, as it ever shall be. Once you press the people who first lie and claim to be "independents," you can explain, depending on the survey and the year, between 85% and 90% of all votes cast in any presidential election, with just party ID, and 2016 was no different. Within that, what's the pattern with respect to wealth? That's conditional on region. Andrew Gelman did some great work here in Red State, Blue State, Rich State, Poor State, and I wish I could come up with titles like that. Essentially, once you stop committing the ecological inference fallacy, if you live in a poor state, the richer you get, the more Republican you get. E.g., Mississippi. If you live in a rich state, like Connecticut, wealth is not closely connected to party ID or vote choice.
Trump wasn't appealing to the downtrodden. He was appealing to Republicans. That breaks down along the same lines it always has.
The story about Trump appealing to the downtrodden working class on the basis of economic issues? No. He appealed to Republicans. The basic party cleavages, at the level of individual voters, were not fundamentally different in 2016 than in past years. The only thing Trump did differently is turn up the volume on white identity politics, and appeal more directly to "the authoritarian personality."
And he won, as an overtly authoritarian candidate, when the economy was doing just fine, with low crime rates. This was not unique to the US.
The story that you need a collapsing economy to elect an authoritarian is a nice one, if you want to console yourself about how people are basically decent unless they are under some dire economic pressure, or something like that. It just doesn't hold up to empirical scrutiny.
Here's some bonus Richard Thompson. The original studio version is from his classic, Shoot Out The Lights, but I thought I'd go with a really cool live, acoustic version. Nothing is better than Richard Thompson solo and acoustic.
*Brief explanation: He let Greece's banks close during a financial crisis, paying all of the pain the country would have paid to leave the Euro, and then stayed on the Euro, caving completely to Germany's demands, and leaving the country in semi-permanent depression because they can't devalue their currency and have no negotiating leverage, having caved so completely. Tsipras is a moron.
