The broken window fallacy as it relates to Japan

[bassrique]

For those not familiar with the broken window fallacy, here's an explanation.

In a nutshell, the idea is that the destruction of a store's window may help the window maker, but the shop keeper in the end has replaced a window he already had and did not better himself by using that money elsewhere.

In this month's New Yorker, a financial writer makes the case that there are times when broken windows can be good for an economy.

In a study of eighty-nine countries, the economists Mark Skidmore and Hideki Toya, after controlling for every variable they could think of, found that countries that suffered more climatic disasters actually grew faster and were more productive. This seems bizarre: it’s close to the broken-windows fallacy identified by the nineteenth-century economist Frédéric Bastiat—the idea that breaking windows is economically useful, because it makes work for glaziers. But Skidmore and Toya argue that disaster-stricken economies don’t simply replace broken windows, as it were; they upgrade infrastructure and technology, and shift investment away from older, less productive industries. (After the Kobe quake, the city’s plastic-shoe factories never returned.) In Horwich’s somewhat ruthless phrase, disasters can function as a form of “accelerated depreciation.” Something similar often happens on the level of the individual consumer: homeowners rebuilding after a disaster take the opportunity to upgrade, a phenomenon known as “the Jacuzzi effect.” In ordinary times, inertia keeps old technologies in place; it may be easier to make dramatic changes when you have to start from scratch.

Still, the impact of any given disaster depends on a variety of factors. Skidmore and Toya have found that geological disasters don’t seem to have the same effects on growth rates as climatic disasters. And growth rates seem to be resilient only for relatively wealthy, well-run countries, which can raise money easily and administer reconstruction funds efficiently.

The whole article.