Keynesians, Earthquakes, and Broken Windows

Keynesians like Paul Krugman, believing that government spending is needed in a recession to stimulate the economy, have been arguing that the previous round(s) of stimulus weren’t big enough. There has been some interesting discussion across the economic blogosphere in recent months, but in the past few days some of the discussions have been escalating almost to the point of parody.

Some of this recent debate has to do with the “broken window” idea considered by Bastiat a long time ago. The Keynesian argument (ok, Keynes came after Bastiat, but it’s modern Keynesians that are bringing up this broken window) is that if a shopkeeper’s window is broken, he has to spend money to replace it, which increases revenue for the window-fixer, who then has money to spend on something else, and it basically “trickles down” from there. Voila, stimulated economy! The Bastiat response is that this does not consider what is “unseen” – namely, what more efficient resources the shopkeeper might have spent his money on if he hadn’t had to fix his window. The Keynesians respond that while the shopkeeper may be losing short-term wealth he is shifting future consumption to present consumption which stimulates the economy now, and so on and so forth. At this point we start to get into complicated economic models where both sides rely on abstract theory and cherry-pickings from historical examples to prove their points.

But the Keynesians have been fervently arguing for the broken window lately. Paul Krugman actually suggested that preparations for a fake alien invasion would stimulate the economy.  Matt Yglesias just said that somehow you can increase growth by breaking windows but nobody ever said you could increase wealth that way. But here’s what tops them all: When the East Coast was hit by its largest earthquake in more than century yesterday, jokes went around Twitter that Paul Krugman said it wasn’t “big enough.” But then Krugman’s Google+ account actually said, “People on twitter might be joking, but in all seriousness, we would see a bigger boost in spending and hence economic growth if the earthquake had done more damage.”

Woah. That statement ignited the conservative interwebz! Was Krugman so attached to his Keynesian theories that he wished for greater natural disasters to stimulate the economy more?

Then we all found out that Krugman doesn’t have a Google+ account and that the whole thing was a hoax. But the lie in the hoax was that it made Krugman sound like a mean guy who wanted big disasters to stimulate the economy – not that Krugman doesn’t think big disasters really could stimulate the economy. The real Krugman’s response was that things like war aren’t “desirable,” but the idea that they do stimulate is still “correct.” Krugman doesn’t want a major earthquake to happen, but it sounds like he really does think it would stimulate the economy through all the money that would have to be spent to rebuild everything. (Krugman’s Keynesian fans apparently thought so, too, because before the hoax was revealed and the post was deleted, they were engaging in fervent comment debate about how the broken window fallacy is not really a fallacy.)

Now maybe I’m just a dumb armchair post-libertarian who doesn’t understand the nuances of the complicated economy (I hope I’m not adding to the misrepresentations of Krugman here), but if great natural disasters stimulate the economy, then why did interest rates immediately plummet after the giant March earthquake in Japan? (I remember this because we were house-hunting and we locked in a lower interest rate for a mortgage a couple of days later.) All the world’s investors didn’t jump out of stocks and into bonds because they thought growth was about to be stimulated.

But, hey, maybe the Keynesian technocrats are smarter than the world’s investors (irrational market theories and all that). Or maybe the broken window fallacy really isn’t a fallacy under certain conditions. But I’m not convinced that a 9.0 magnitude earthquake is one of those conditions.

Let’s start with the basic condition of a shopkeeper’s broken window. Here’s why I don’t believe this leads to long-term stimulus of the economy. Consider the time just before the window was broken. The shopkeeper was thinking about things to spend his money on, and what if upgrading a to new window was not at the top of his priority list as a use for his limited resources? Or, in the unlikely event that upgrading to a new window was previously at the top of his priority list, consider if he had just replaced the window and that replacing the window again was no longer at the top of his list. Either of these scenarios seem far more likely than a scenario where the shopkeeper’s top priority is to replace the window but he hasn’t yet done it when the window is broken. Thus, when the shopkeeper replaces the broken window, he is most likely using his resources in what he believes to be a less efficient manner than if the window was never broken. Since economic progress comes from all the individuals seeking the most efficient uses of their limited resources, this leads to less economic progress. Destruction causes a loss. End of story. Go Bastiat!

But the Keynesian simply disagrees. The shopkeeper may have been saving the money for some future use. (Krugman believes we’re in a “liquidity trap” where, as I understand it, there is too much global saving going on.) The broken window forces the shopkeeper to give the money to the window-fixer. It will continue to be used in the economy and it otherwise would not have for some time, and that extra use leads to more economic progress.

In this basic condition, this is a simple philosophical disagreement about the economic impact of replacing a broken window. It’s a complicated matter to discuss what would have happened to the money or what will happen to it now. I have a bias causes me to side with the idea that inefficient spending does not lead to greater growth, and Krugman and other Keynesians have a bias that causes them to side with the idea that spending now, though possibly “inefficient,” leads to greater growth than spending later (please correct me if I’m wrong?).

Now this simple philosophical disagreement becomes a big philosophical disagreement very quickly, because Krugman believes that if one broken window is a tiny stimulus, then two broken windows is a bigger stimulus, and ten thousand broken windows is an enormous one! And Keynesians have a giant historical broken window as evidence: “World War II was expansionary because it led to a large increase in public spending,” Krugman says. This is a reference to the economic argument that America thrived after World War II because giant government spending ended the Depression. (Hayek fans argue that we rationed food during the government spending of World War II, and that it was only when government spending shrunk afterwards that we thrived. But the war at least looks like a powerful piece of evidence.)

Yet I think the greater the destruction, the harder it is to make a case that the stimulus will be greater. If a shopkeeper has to replace one broken window, his short-term profits go down a little but he still stays in business, and he gives business to the window-fixer. If the Keynesian is correct, the shopkeeper has sacrificed a little profit to increase the growth of the overall economy. But what if he has to replace all of his broken windows? Or what if his shop burns down completely and he has to rebuild it? Isn’t there some point at which decreases in profit become enormous losses that lead to bankruptcy? Isn’t there some point at which the opportunity cost of the less efficient use of resources becomes so great that nobody can even afford to replace the broken windows? What kind of economic growth comes from that?

Yglesias says, “the conservative accuses Keynesians of wanting to break windows or believing that window-breaking increases wealth. But nobody ever said that! The point is that we have very good reasons to think smashing windows would be a bad idea.” It is important to make the distinction that Keynesians don’t want to break windows. But Yglesias also says, in the very same post, “the Keynesian accurately points out that you could, in fact, increase growth by breaking windows.” I’m just not convinced that such a statement is really accurate.