Already, there's the potential for an outcome other than excess capacity followed by a crash. A perfectly informed trader would know the minimum efficient scale of a pizzeria as well as the long-run supply elasticity of pizzerias, and that trader would be able to exploit the error others are making. A less than perfectly informed trader, or, for that matter, a loan officer at a bank, might remember enough about fitting u-shaped cost curves under the demand curve to look very carefully at business proposals for new pizzerias. The economics of exuberance is something a bit too challenging for introductory economics, although "Rational Expectations and the Dynamics of Hyperinflation" (International Economic Review 14, 2, June 1973: 328-350) will reward careful study. The pizza bubble is more likely to persist if people expect tomorrow's demand for pizza to exceed today's demand at the same price, ad infinitum. That's a rational expectations tulipmania, and there's a lot of very technical work on ruling out such behavior both in theory and in practice.
1. Start at equilibrium with zero economic profit. As pizza demand goes up due to, say, the opening of a new university in the area, price rises leading to profits in the short-run (this is the points labeled sr in the diagram).
2. It is widely reported on local financial pages that the industry is booming. In response to these reports, and buttressed by their own analysis, firms enter. In fact, and there's nothing in theory that says this won't happen, it's possible that too many firms enter. That is, the demand for pizza goes up and twenty firms enter, but there's only room for fourteen to survive in the long-run. Extra firms, i.e. firms that won't survive in the long-run can also enter if the increase in prices is higher than justified by the underlying economic conditions (i.e. there is a bubble in the technical sense due to over exuberance or other reasons) and false signals are delivered to the market.
In the simplest model of perfect information, excess capacity won't happen. In a relatively simple model with learning, excess capacity will be transitory.
3. Makers of products such as pizza ovens, pizza boxes, and pizza trucks are getting the word out to all who will listen. They're building a new university and it's now open! There's a boom in the pizza industry! Get in now while profits are high, before it's too late. Pizza's never been better! After all, the suppliers stand to profit from every firm that enters and they have an incentive to encourage as many firms as they can to join in the boom. While not actually telling falsehoods, at least in most cases, they make the opportunities in the industry sound as rosy as possible to all who will listen, offer enticing good credit terms to entering firms, and so on.If it sounds too good to be true, it is ...
The post goes on to note that subprime mortgages are subtly different.
My point is that the housing bubble is an enlarged, very exaggerated version of a process we see regularly when a market opportunities open up. In the case of housing markets, for example, financial innovation allowed a segment of the market to be served that had not been well served in the past, and it also allowed existing markets to expand creating highly profitable opportunities. The result was just what you would expect when there are profits to be made, a rush of resources into the industry. Real estate agents, loan companies, builders, etc. all entered, in fact too many entered and now we are seeing the clean out that is the equivalent of the failure of the six excess pizza firms (we shouldn't forget that this is the excess that is being cleaned out, at least in the example above, the industry itself has grown and now serves more people than before).The problem might be as simple as no loan officer stopping to ask how many firms could fit under the new demand curve. It strikes me as a bit of a stretch to suggest, as Marginal Utility's Ken Houghton does, that efficient markets are definitionally suboptimal. Not quite. The simplest model of markets assumes away, a priori, the kind of mistakes that Professor Thoma invokes to create his pizza bubble. Those mistakes can be corrected if, under costly information conditions, some traders are sufficiently alert to the arbitrage opportunities the mistakes make possible. That same alertness renders the cattle-cycle model a commenter invokes inoperative.
We teach the cob-web function in one lecture in intro-economics and then forget it. But the cob-web does a better job of explaining the system than the mainstream theory that academics teach.Not really. The cattle cycle might be a first approximation of price cycles under costly information. Then, however, traders are operating in a regime of randomly occurring surprises of unknown magnitude and duration. There are counters to changes of known magnitude or duration, summarized in the efficient market maxim, "if we know the price is going to go up tomorrow, it will go up today." In like manner, if you know that today's price is near the top end of the cattle cycle, you have incentives to exploit it today (or to go short for the future) in such a way as to damp, eliminate, or reverse the fluctuations. The underlying arguments haven't yet been distilled into a principles-friendly form.