NEAR-RATIONALITY VS. SIMPLE CONFUSION I just finished Peddling Prosperity by Paul Krugman and I want to say again that it is fabulous book. Go out and read it. In Krugman's tale of the resurgence of Keynesianism, he presents the "near-rationality" theories of George Akerlof. Why, asks Krugman, do housing markets, like Boston's in the early nineties, sometimes malfunction by failing to lower prices fast enough? In Boston it took several years of persistently high prices and unsold houses for sellers to finally lower their prices enough for the market to clear. Why the big delay? Why didn't prices adjust immediately? The traditional conservative explanation for a housing recession (or any recession) is that sellers are confused about the state of the market, and so thing their' prices are appropriate. The price correction comes as soon as the confusion abates. But a delay of several years is too long -- everyone knew a recession was happening, but still prices did not adjust. Enter George Akerlof. Akerlof made the point that in many cases, acting with what economic model would consider perfect rationality (which would involve marshalling all available information about the state of the world before making a decision) is actually not rational at all for real people. Real people have jobs, and deadlines, and other things to do, so it is probably best for them to act with "near-rationality," in other words, so gather some information and make a best guess. Akerlof believes, and Krugman agrees, that a market full of approximators, cognizant of the recession but not responding to all cues, might have kept prices irrationally high.

But at least to my eyes, there is nothing in Krugman's story of Akerlof's work to explain why the thousands of approximations always err in the same direction, the too-expensive direction. If everyone is just winging it, shouldn't about as many land below the market equilibrium as above it? Shouldn't near-rationality lead simply to a greater spread of prices in a non-homogenous market, not a recession?

There are probably a number of ways to counter that argument. For instance, it may on the whole be safer to remain in one's current house longer than one wants than to be out on the street with less money than one should have gotten -- it may be safer for families to err on the side of high prices. Also, there may be an effect whereby families are biased against re-analysis. If the cost of doing a new analysis is high, it may be best to stick with the old, which in this case would mean to ask a higher price. Still, as soon as one talks about the choice not to do an analysis, the story begins to sound like the conservatives' tale of persistent confusion. But there is one crucial difference: here the confusion is intentional. Therefore, intentional analysis and action, as by a government, can still help.

This whole discussion brings up an idea I've been tossing around for a while. The idea is that perhaps you don't need a population confused or near-rational people at all to get a recession. Perhaps all you need is a few. A few people behaving irrationally might change the market enough that rational people would be forced to change their behavior too. For instance, suppose a house that once sold for $200,000 now ought to sell for $150,000.If everyone was rational and everyone knew everyone else to be rational, the price would drop and the house would be sold. But suppose that a small segment of the population is irrational, and despite lower wages and the like, would still buy the house at $200,000. Suddenly a rational seller's calculus chances because of the chance of encountering such an irrational person. It may now be rational to lower the price only to $175,000. Futhermore, if every rational seller makes this decision, and every like house is priced at $175,000, it may appear as if the market has equilibrated, reinforcing the behavior. Now all rational buyers who would have paid up to $180,000 for the house (and gotten a deal at $150,000) can now be sold to at $175,000, so rational sellers has little reason to drop much below the price everyone else is asking.

The above example begs the question, do we need the irrational people at all? Perhaps it is enough to have the perception among otherwise rational people that a few irrational people exist. The simple expectation that there are people who will buy at the wrong price may be enough to start the cycle. Furthermore, do we even need that expectation? Why not just the expectation that some have the expectation or, throwing off the layers of infinite regress, the simple expectation that the housing market will be slow to adjust. Since it is always slow to adjust, this seems a rational expectation, despite the fact that the expectation itself creates the sluggishness. Think of it as two equilibria: an unstable one where prices adjust quickly, and a stable one where prices adjust slowly.