More and more voices point at the irrationality of markets, institutions, and traders. It is claimed that markets are dominated by bubbles, fads and frenzies and that institutions and investors take risks that they do not understand. The recent book by George Soros is a good example of the point of view which goes as far as hinting that established financial theory is obsolete. How can the past Internet bubble be explained otherwise? The current crisis, the argument goes, is the final proof that markets do not aggregate information efficiently. We are closer to Keynes’ casino than to the view of the market as a marvel of Hayek. Market operators miscalculate, are overconfident on their information and overreact to news. However, as argued in a recent book of mine, there is another explanation which is based on rational calculation and information processing by institutions and traders. When information in the market is dispersed enough there will be different views of the evolution of stock prices. This is particularly the case for stocks related to new technologies like the Internet, as was the case, for example, with stocks related to railways more than a century ago. And, by the way, sophisticated loan packages are a new and complex financial product over which information dispersion is bound to be wide. In those circumstances, booms in volume traded and in prices do happen. Prices may get far from the fundamentals as assessed by a hypothetical collective wisdom that would pool all the information in the market. Indeed, trading on the momentum may be self-fulfilling and rationally calculating traders may well decide to “ride the bubble” while it lasts. The boom is fostered furthermore by the asymmetry between those that bet that the market will go up and buy, and those that bet that it will go down, and stay out since to sell short is costly. The same goes for crashes with the aggravating factor that many traders have to sell to pay off debt. This means that important and relatively persistent departures of prices from fundamentals values are possible, and even likely, but a correction that aligns them with reality will follow. Over the long haul stock prices do reflect the fundamentals of the economy.
Let us turn to institutions. How can one explain the overexposure of many institutions to subprime risk and the collapse of the interbank market? Were irrational the banks that chose to securitize subprime loans instead of keeping them on balance? Keeping the loans in the book a bank had to monitor its performance at a cost and incur a capital requirement. Securitizing it avoids such costs and can place the package advantageously -with the complicity of the rating agency that has a stake in the business- and profits from the inexperience, read lack of information, of investors in this type of products. Furthermore, when things turn ugly, equity holders are protected by limited liability and the executives keep a generous bonus. The collapse of the interbank market is due to informational failure. This is a well-studied phenomenon for which George Akerlof got the Nobel Prize. When someone tries to sell you a second hand car you realize that he knows much more about the car than you and he may be trying to sell you a lemon. Banks still do not trust each other since each wonders how many corpses the others keep in the closet. Again, there is no irrationality here, it is just that each individual institution does not have enough information. This market failure is what explains the calls for transparency in the disclosure of losses.
The debate over the role of irrationality in the present crisis, as well as in other crisis, is not only academic. Indeed, if economic actors behave in an irrational way policy measures have to be paternalistic and very broad since there are as many types of irrationality as people in the world. Taxes on financial transactions or outright prohibitions to invest in certain products for some people and institutions may be a way to put a brake to the perceived excesses. However, if we believe that economic actors respond to the incentives they face and process information the best they can, then we may attack the failures in the present regulatory frame with well-targeted measures. Restrictions on off-balance sheet vehicles, so that capital requirements can not be circumvented, tougher disclosure requirements, and control of the conflicts of interest in rating agencies, will change the incentives of the actors in the right direction. On the other hand, keeping the liquidity pump working may destroy the incentives to be prudent and lay the foundations for the next bubble. The belief in the irrationality of market participants may be self-defeating when it comes to policy measures.
Xavier Vives is Professor of Economics and Finance at IESE Business School and author of Information and Learning in Markets (Princeton University Press, 2008)