Paul Krugman wrote a blog post this morning, Varieties of Error. In it, he highlights two kinds of predictive failures: those that indicate the fundamental thinking (or “model”) is wrong and those that just show that predictions are predictions. Think of football, for example. The Texans is a much better team than the Chiefs. In general, the Texans will beat the Chiefs. But if the Texans commits a number of turnovers, the Chiefs will win. The prediction was wrong, but the basic idea is correct: the Texans is a better team than the Chiefs. On the other hand, if your model was that the Chiefs is better than the Texans, there is something wrong with the model. What’s more there is something wrong with you if, after 10 match-ups between the Texans and the Chiefs in which the Texans won 9 times, you still claim that the Chiefs is the superior team.
We see this a lot in politics and economics. Conservatives have been screaming for years that hyper-inflation is just around the corner. Every time someone writes a column about budget hysteria, there are always loads of comments reading, “But Greece!” We have explained for years why Greece is not the same as the United States. (We have our own currency!) I think that the United States could continue to not become Greece for another 100 years and this would still be a common conservative retort.
What I find fascinating is the divide between the certainty of liberals versus conservatives. This brings to mind Nate Silver. Even when his model was predicting a 90% likelihood of Obama winning, he was still cautious. He always pointed out that his model was basically just a poll aggregator and that the polls might be wrong. Contrast this to the conservative loons who were just certain that they were right.
For years, Krugman has been saying that we would not see inflation so long as the economy was depressed. But during that time, he has always added caveats. In particular, he’s noted that there may have been some kind of economic mechanism that he’s missing. Contrast this with, say, Niall Ferguson. For years, he would brook no doubt: inflation was on its way. (Eventually, he had to admit that he was wrong. Since then, he’s just become a general purpose conservative hack.)
This divide is not due to the humble characters of people like Nate Silver and Paul Krugman. As far as I can tell, both these men are as (rightly) filled with themselves as anyone. I think the divide comes from the acceptance of facts and the rejection of authority. The modern Republican Party is an authoritarian group. The truth is whatever their leaders say it is. Democrats still expect proof. And this gets to the heart of what it is to be a member of either group. Liberals could not live with themselves if they didn’t think that what they believed was in some fundamental sense true. Conservatives depend upon group identification. This is why Fox News exists: people want to be told that belonging to their group is right and true. Their position in the hierarchy is what matters.
As a result of this, we should not be surprised that conservatives continue to hold long discredited ideas. As long as the party accepts it, all is good. On the other hand, if the party suddenly believes something different, the members will change on a dime. This will eventually happen to the Republican Party on the issue of global warming. Of course, by then, it will be too late. In fact, it already is.
I’m generalizing here a bit. !rtists and scientists are liberal. Those who aren’t tend to be libertarians, rather than the more vile forms of conservatism. It is hard to be a creative thinker when you feel obliged to groupthink.