Tuesday, February 15, 2011

The natural rate of unemployment

Two economists at the Federal Reserve Bank of San Francisco (FRBSF) have written a paper in which they claim the natural rate of unemployment (i.e. the one consistent with price stability) is not around 5% anymore.  Now it is close to 6.7%.  It reminds me of a talk given by Solow at SCEPA in 2000, in which he said at the beginning  that the natural rate did not exist, and by the end he suggested it was 5.2%.

There are several ways of trying to measure the unobservable natural rate.  One is to use the inverse relation between vacancies and unemployment (the so-called Beveridge Curve).  The idea is that if there are more vacancies than regularly observed for a given level of unemployment, then the normal level of unemployment must be higher. Further, the idea is that job vacancies are completely driven by supply conditions (this is often omitted in explanations of the curve).

If that is the case, supply changes drive the vacancies up, and the level of unemployment consistent with price stability adjusts upwards.  So why did job vacancies went up?  For one, mainstream economists (the ones that missed the bubble and the recession) argue that it is a mismatch between skills required by new technologies and workers capabilities. Also, it is argued that the extension of unemployment benefits discourages unemployed workers to look for jobs.  The FRBSF economists argue that 50% of the increase in the natural rate is explained by unemployment insurance.  And, at least, they note that the increase in actual unemployment is explained by a cyclical fall in demand.

It is interesting to note that the natural rate changes with the actual rate of unemployment.  A period with high levels of unemployment on average leads to economists' claims that the natural rate is higher.  The Beveridge Curve is used to justify then the idea that it is the natural rate that drives the actual rate of unemployment.  Of course it could be a simple case of reverse causality.  Higher levels of unemployment on average imply that the normal (that somebody decided to call natural; Mr. Milton Friedman, by the way) would be higher.  In other words, what drives the natural rate is the actual rate.  Job vacancies can also go up as a result of higher levels of unemployment on average.  For example, it has been reported that many recruiters do not want to hire unemployed workers.  The workers may have the skills, but the perception is that if they are out of a job they must have done something wrong.

Logic suggests that it is unemployment that causes the increase in graduate student applications, but I would not be surprised if an economist claims it is the other way round. The point is simple. Perhaps one way of bringing down both the normal unemployment level, and the Beveridge Curve is to promote the creation of jobs now!


  1. Well, it's not clear that we cannot simply assert the NAIRU. While I normally like FRBSF reports, I assert NAIRU is 3.5. Get thee going, Fed.

  2. I knew graduate students were evil...


Godley versus Tobin on Monetary Matters by Marc Lavoie

  The 4th Godley-Tobin Lecture given by Marc Lavoie, a co-author of Wynne Godley, and one of the leading Post Keynesian authors.