Tuesday, July 19, 2005

Thoughts on Hayekian Economics

Just some random observations...I had wanted to do something more systematic, but I don't have enough to say yet.

One thing is particularly odd, since Hayek views economics as a "science." He doesn't believe that an empirical economics is possible:
The hope of becoming more "empirical" by becoming more macroeconomic is bound to be disappointed, because these statistical magnitudes - which are alone ascertainable by "measurement" - do not also make them significant as the cause of actions of individuals who do not know them. Economic phenomena are not mass phenomena of the kind to which statistical theory is applicable.
This stems from Hayek's view that important economic information cannot be known, save through the price system. Each individual has particular bits of information about scarcities and preferences. Markets take this information and operationalize it in the form of prices. Since prices "regulate" the economy, the only important economic information must be those bits that only individuals possess. And only the price system, not statistics, can effectively aggregate this information.

I vaguely agree with this line of reasoning...that is, there ARE bits of information that only individuals possess; the price system DOES effectively aggregate this information; and there is no other system that can do so. This is mostly because the price system makes the information relevant to the individual who possesses it. They are forced by incentives to reveal it...a survey can't do that.

But, I also think that Hayek's conclusion is quite wrong. Ironically, the best argument against Hayek is empirical: using statistics in economics has been quite successful in practice. No one is taking surveys of people to see how much shaving cream they would buy at different prices and forming a demand curve. But you can look at the price of shaving cream under different conditions, exogenous factors affecting the demand or supply of shaving cream, etc. You can look at the consumption/saving decisions of households across time and relate that to the interest rate. And so forth. This type of research uses the price system as data rather than using other data to try and reconstruct the price system as Hayek feared.

Hayek's assertion that economics cannot truly be empirical is important because he argues that
...[Keynes' theory of unemployment] is widely accepted only because the explanation earlier regarded as true, and which I still regard as true, cannot by its very nature be tested by statistics. [emphasis Hayek's]
In other words, if it makes intuitive sense, the data be damned!

In Hayek's "true, though untestable" theory of business cycles, unemployment is caused by temporary distortions of relative prices. Some industries hire too much labor and some hire too little. Unemployment arises temporarily as a result of relative prices returning to their true, unfettered equilibrium. The theory is a lot more complicated than this. Hayek analyzes the "stages of production" of consumption goods. When too much credit is extended to producers, the money available for production increases but the money available for consumption doesn't. This necessitates more stages of production and a lengthening of the supply chain. As soon as suppliers have retooled for this new structure, consumers demand that the original ratio of consumption to investment goods be reinstated and the suppliers must once again retool, causing unemployment as labor is shifted about.

Paul Krugman shrewdly asks why there is not also a recession when the supply chain is initially revamped. I refer you to Uncle Paul for a proper discussion of Hayek's theory. Personally, I don't find this style of analysis compelling or useful.

Now onto the fun part: Hayek's refutation of socialism!


Post a Comment

<< Home