Paul Krugman is an economist at Princeton and a columnist for the New York Times.

Kevin O’Rourke follows up on some of what I’ve been writing, and argues that doing good macroeconomics depends crucially on knowing a fair bit about economic history. Indeed....

I’d also say that there’s history and then there’s history. Knowing the time-series properties of US quarterly data since 1947 isn’t what I mean. In macro, in particular, you need to know about drastic events. I don’t have it in front of me, but in his book on the German hyperinflation Frank Graham said roughly this: “Disorder is, for the social sciences, the sole substitute for the controlled experiments of the natural sciences.” That means knowing about prewar experience; it also means knowing about international experience — for there have been numerous crises even since World War II, just not in the United States....