Broken Windows
August 29, 2011
Bastiat’s “Parable of the Broken Window,” shouldn’t be controversial at all, but it is. Your more libertarian-type economists say that it’s a fallacy committed by Keynesians all the time, as they attempt “stimulus” while ignoring the costs. On the other hand, you have Matt Yglesias and others countering that the parable only applies when we have full employment, not vast quantities of idle resources, as we have today. As is so often the case, the problem is that everything got all screwed up when we tried to do the analysis in terms of money, as opposed to the perfectly comprehensible but non-existent barter economy. Here’s what Bastiat writes:
The glazier comes, performs his task, receives his six francs, rubs his hands, and, in his heart, blesses the careless child. All this is that which is seen. […] It is not seen that as our shopkeeper has spent six francs upon one thing, he cannot spend them upon another. It is not seen that if he had not had a window to replace, he would, perhaps, have replaced his old shoes, or added another book to his library. In short, he would have employed his six francs in some way, which this accident has prevented.
The way Bastiat writes it, it would be easy to interpret this passage as saying that the wasted resource is the six francs that get used up in repairing the window.* But you don’t have to think about it very long to realize that that can’t be true. The six francs are as usable as ever, it’s just that now they are in the hands of the glazier instead of the shopkeeper.
Clearly, the real loss here is the glazier’s time and effort, which could have been applied to some other valuable activity had the window not been broken. Among the more important ideas in economics is that when resources have alternative incompatible uses, there is always a cost to employing them in some particular use: namely, the next-best alternative that you must give up.
Those who claim the Broken Windows Fallacy is not really a fallacy during recessions are abandoning too much of classical economics. The economic problem remains the same: we are trying to find and communicate information about the ways in which to employ our resources. Their main beef with the Classicals is that they assume that this information is always transmitted quickly and accurately; critics argue that that assumption is way off during recessions.
Still, that doesn’t mean that you should employ “idle” resources in just any old way you please. When you engage in Keynesian pyramid-building you can’t argue that it was a wise policy because “Look, now we have all these great pyramids that we built out of resources that were formerly unemployed.” No, you say, “It’s true, we built all these crappy pyramids, but actually it was worth it because building these pyramids was the only way to get that Neoclassical information-transmission machine up and running again.”
The Broken Windows Fallacy, which is always a fallacy, is simply to ignore opportunity cost. If, for the sake of argument, World War II was what cured the Great Depression, it wasn’t because we built a bunch of tanks and ships and blew a bunch of stuff up. Perhaps doing all that made it possible for the Depression to end, but we have to acknowledge that it would have been even better if we could somehow end the Depression and also use the resources that went into building that war stuff in ways that people valued.
*Matt Yglesias tries to rescue this strange interpretation by claiming, strangely, that in the 1850s silver pieces really were a scarce resource.
William's Continued Adventures 
August 29, 2011 at 1:14 pm
I interpreted Matt Yglesias differently, to be saying that even though Bastiat is right and breaking windows reduces over all wealth (or, in the case of stimulus spending, future, possible wealth), the tradeoff is that it increases the growth rate, and gets more people working again. In effect, it becomes a way of performing wealth redistribution while also trying to lessen the problems of a large, aging, idle work force.
And the fact that, in this case, people aren’t breaking widows (reducing present wealth) but are allocating unused resources in a way that might be less than optimal (so possibly giving up future wealth) makes things more palatable.
But the main issue he’s getting is at, and why I’m kind of convinced by his point, is that there doesn’t seem to be a super-compelling neo-classical explanation for why so many resources are going unused right now, so the strictly neo-classical policy recommendations kind of loose their persuasive policy.
But, of course, maybe I’m way out of my league here 🙂
August 29, 2011 at 2:05 pm
Matt’s posts on this subject are very unclear, the result of him writing way too many posts and thus not devoting sufficient time to any of them.
In “The Anti-Keynesian Two-Step“, he does say that employing glaziers increases “growth,” but he doesn’t actually explain why. If the government is investing in durable goods and equipment which will deliver benefits for years to come, that’s great, but it’s completely separate from the point that the “Broken Window Fallacy” is meant to address. I think he just wrote the word “growth” in haste, without thinking about whether he really wanted to say “growth” or “GDP.”
Okay, but here’s the point: Yglesias thinks that the problem during a recession is that people don’t have enough money. So, logically, he proposes that the Fed give them more money. But then, in both posts, he adds on this condition, for no obvious reason, that the government should at the same time be hiring unemployed people to do “useful work.” Before we were just correcting a lack of money, but now we’re using up real resources, and we’re getting into Broken Window Fallacy territory.
Again, what is supposed to happen is that prices give people the information they need to know what to do with themselves. If there’s not enough money, this information doesn’t get transmitted. But once monetary policy has dealt with this problem, there is no need for fiscal policy. We don’t need to tell people to “renovate schools or repair bridges” (as Matt says) unless we somehow know that this is the most valuable use for their time and effort.
Fiscal policy as a recession cure is based on the idea that when people and resources are unemployed, we need to tell them what to do.* But that doesn’t make sense unless we think that the government has some idea of what people should be doing. And if it already had that idea, it should have been doing it anyway, recession or no recession.
*This isn’t entirely true. Liquidity trap people say fiscal policy is necessary because the Fed is powerless to raise nominal GDP. Matt Yglesias is not a Liquidity trap person, and explicitly says that this claim is absurd.
August 29, 2011 at 2:17 pm
I understand the point that prices point people to efficiently allocate resources, but I think Matt is trying to address a couple of different points at the same time through the single policy recommendation (or, at least, this is what I take as the most persuasive version of his argument)
– There are huge amounts of unused resources in the economy now.
– The government could put some of these resources to use by, in effect, borrowing against future income at the better price the govt can get, and paying it to people who are not working now.
– While its possible to try and limit unemployment by writing checks and hoping that the money employs the currently unemployed, there are a number of reasons to expect that this policy won’t happen
a. if the money is transferred directly to the unemployed, there is the political issue of people not supporting the govt writing checks to the poor
b. there are reasons to think money given straight to sectors of the economy won’t increase employment among the currently unemployed (ex, transfers to banks haven’t increased lending on the low end [i think…], or the class gap between the unemployed and the employed)
– Since its not likely that the govt will be able to get support behind a “send more checks to the unemployed” program (see how difficult it was to extend unemployment benefits), the next best case is for the government to write those same checks, contingent on the unemployed doing something useful (cleaning a park, etc), even if that activity isn’t the most efficient possible use of their labor. Since, while not ideal, cleaning a park is more productive than sitting at home unemployed (not to mention to social consequences of being unemployed)
I’m nervous about the policy, but I think that there is more to it than Bastiat