By Lars-Alexander Kuehn, Nicolas Petrosky-Nadeau and Lu Zhang
http://d.repec.org/n?u=RePEc:nbr:nberwo:17742&r=dge
Search frictions in the labor market help explain the equity premium in the financial market. We embed the Diamond-Mortensen-Pissarides search framework into a dynamic stochastic general equilibrium model with recursive preferences. The model produces a sizeable equity premium of 4.54% per annum with a low interest rate volatility of 1.34%. The equity premium is strongly countercyclical, and forecastable with labor market tightness, a pattern we confirm in the data. Intriguingly, search frictions, combined with a small labor surplus and large job destruction flows, give rise endogenously to rare disaster risks à la Rietz (1988) and Barro (2006).
What else can the labor search model explain? Here it solves the equity premium puzzle and rare disaster risk. These are exciting results. But skimming through the paper (I do not want to delay mailing the NEP-DGE report further) I could not find good intuition for this except for the following: with the search frictions, wages are less volatile than the marginal productivitry of labor, because of the outside option of unemployment in the Nash bargaining of the wage. That makes profits more volatile than output, and you have an equity premium. But it seems to me that it is the Nash bargaining that drives the result, not the search mechanism. Have a different wage determination, and the result may not hold. Correct me if I am wrong.
Thanks for selecting our paper to be on this blog. Our latest draft dated January 2012 available at http://fisher.osu.edu/~zhang_1868/SearchAP_2012Jan.pdf
contains (hopefully!) clearer exposition of the intuition. See the last two paragraphs on page 2.
See, in particular, the last paragraph on page 2 on the role of search frictions (the cost of vacancy posting) in driving our key results.
And yes, Nash bargaining is important too. We use it to break the tight link between marginal product of labor and wages.
Note that nowhere did we claim that the model “solves” the equity premium puzzle. But we’re excited in that the quantitative results seem to go to the right direction. And the results are more positive than those in prior studies that rely on capital accumulation only.
Forgot to mention in my previous post about the wage determination. Yes, it’s key to break the marginal product of labor = wage relation in a frictionless labor market. The reason is that procyclical wages dampen the procyclicality of profits. It’s highly likely that Gertler and Trigari 2009 style sticky wages will do the job too. But we stick with the period-by-period Nash bargaining because it’s much more tractable…
With all the papers that have claimed to have solved part or all of the equity premium puzzle, the real question now is why the equity premium is so low.
We’re still climbing out of a recession. So, the equity premium is unlikely to be low. Also, expected returns (such as the equity premium) and average realized returns in the recent past are two different things. The fact that average realized returns are low, indeed even negative, in the recent past often means that going forward the risk premiums are high. I.e., investors on average require a higher rate of return to hold the riskier stocks…