A labor market tautology says that any change in labor usage can be decomposed into a movement along a marginal productivity schedule and a shift of the schedule. I calculate this decomposition for the recession of 2008, assuming an aggregate Cobb-Douglas marginal productivity schedule, and find that all of the decline in employment and hours since December 2007 is a movement along the schedule. This finding suggests that a reduction in labor supply and/or an increase in labor market distortions are major factors in the 2008 recession. The decline in aggregate consumption suggests that the reduction in labor supply (if any) is neither a wealth nor an intertemporal substitution effect. “Sticky real wages” or the emergence of significant work disincentives are possible explanations for these findings.
Critics have been quick to reject the idea that the rise in unemployment is largely due to labor supply shift (see here, here, and here). However, each of these criticisms assumes that the change in the labor supply is limited to voluntary choice (i.e. intertemporal substitution effects and/or wealth effects). Mulligan’s paper similarly refutes the idea that the shift in the labor supply can be attributed to these factors as they imply that consumption should be rising when, in fact, it is falling. He therefore argues that the shift in the labor supply can be explained by labor market distortions.
In contrast to the critics above, I actually find this argument somewhat compelling. For example, we are coming out of a period in which we experienced an unprecedented housing boom. Thus, from a macroeconomic perspective, it is likely that we are seeing an increase in the natural rate of unemployment as the economy goes through a significant restructuring. Whether or not this increase is substantial enough to explain the drastic rise in unemployment is questionable, but Mulligan’s paper seems provide some evidence for such a claim.