Rare Bad Krugman Column

So I like most Krugman columns, but I think today's is a bit of a stretch. Or rather, its conclusion is probably right, but his evidence doesn't really get you there. Krugman basically argues that a lot of the flaws in the Bush administration follow naturally from conservatism, and are not deviations from it. I think that's true, but Krugman has a bad habit of citing not-so-persuasive evidence for his argument. So for instance, he quotes Irving Kristol exhibiting disdain for fiscal responsibility, and later he cites an article in the American Prospect arguing that privatization had gone too far under Reagan and Bush.

I use the word "cite" because I think Krugman is making the same mistake I often make when I'm starting to research an assignment at work. I'll find a case that says exactly what I want to argue, and I'll be elated. What happens next, though, is crucial. I check to see which court issued the opinion. I check to see if it's been overruled ("shepardize" it). I check to see if other opinions undermine its usefulness.

Ultimately, I think this is where Krugman goes wrong in columns like this one. He doesn't give any indication that he hasn't cherry-picked a few examples that happen to support his case. It's not that his examples are useless. It's that they're at best suggestive - it's like finding a bankruptcy case from the northern district of Illinois and trying to use it in a case in the southern district of New York. Sure, the judge is going to give it consideration, but it's not binding. Likewise, maybe Irving Kristol was a maverick or idiosyncratic about deficits. Maybe the American Prospect will write anything critical of Republican administrations.

Maybe not, of course. These data points are worth something, but it's hard to say how much. You can't really assess the column without making judgments about how valid and representative his examples are, and so it ends up being the kind of thing that doesn't really change minds.