Wednesday, October 20, 2010

The Problem With "Potential GDP"

A key concept in Keynesian theory is that of "potential GDP", as opposed to actual GDP. That is supposed to be GDP if there is full employment and capacity utilization.

There are several problems with this. First of all, surely there are more things which could be organized in a more rational way than just employing more workers or using unused existing capacity. For example there could be higher growth in capacity through higher investments, improved skills of workers, improved use of existing skills and better technology. To only include some possible ways to boost output and not others is an arbitrary distinction.

A second problem is that potential GDP is in fact unknowable. The first reason for this is that we don't know just what rate of unemployment and capacity utilization constitutes full employment and full capacity utilization. In practice 0% unemployment and 100% capacity utilization is never reached, so that's not the answer. But what is the answer then? No one knows.

Furthermore, it is not certain what the marginal product of more workers or higher utilization really is. It will probably not increase output proportionately.

Some Keynesians like Paul Krugman choose to simply extrapolate previous growth. But there is no way of knowing that "potential GDP" really continues to increase at the same rate. And in fact there are good reasons to believe that potential output is lowered.

First of all, a weak labor market will limit immigration and make many workers lose contact with the labor market, reducing the size of the labor force. And lower investments during the recession will reduce industrial capacity. Official U.S. industrial capacity was for example 1.1% lower in October 2010 compared to Ovtober 2008, while during the previous boom industrial capacity rose 1-2% per year. And the real level of capacity that will ever be used again has probably fallen more.

Contrary to what Krugman claims, there are thus good reasons to believe that this unusually long and deep slump will result in a permanent reduction in output. It is however not possible to know just how big this reduction will be. For this reason, we should be very skeptical toward Keynesian estimates of "potential GDP".