If our facts don't fit the theory, we need new theory. And why not? Ostensibly at least, we write these articles to understand the world we live in. Our theory serves a purpose if it helps us understand. If it doesn't fit the facts, how could it do that? If it doesn't fit the facts, we need new theory. In fact, of course, we don't. At least not necessarily-for facts are not always what they seem. As we all know, we rarely work with raw empirical phenomena. We use selected, sorted phenomena, and we cannot select and sort without theoretical priors. Sometimes, when the theory doesn't fit the facts, we need new facts. For years, in the academy we used "facts" from Japan to prove the culturally determined character of theory, and none more so than economic theory. Alas, we more often showed how tenaciously we could cling to even the most wildly implausible "facts." Where standard economic theory