The Macroeconomic Role of Unemployment Compensation

The standard motivation for unemployment compensation is consumption smoothing, and most papers in the literature have analysed trade-offs involving consumption smoothing and moral hazard. In a recent paper, CEP associate Eran Yashiv and co-authors Tomer Blumkin and Yossi Hadar investigate a macroeconomic motivation for unemployment insurance. The authors show how unemployment insurance can increase output by enhancing the assignment of workers to jobs in the face of firm productivity heterogeneity and skill-biased technological change.

Research on the macroeconomic effects of unemployment compensation is rare, although CEP member and Nobel Prize winner Chris Pissarides has worked in this sparse area himself; see Pissarides (1998) and Mortenson & Pissarides (1998). However, the import of this paper comes also from its analysis of macroeconomic effects when relating them to the productivity distribution across firms, two previously unrelated areas.

This paper uses a model where firms have heterogeneous productivity and workers are ex ante identical. There are more firms than workers. The benefit of well designed unemployment insurance in this model comes from its effect on which firms choose to enter the market. Well designed unemployment insurance can encourage the lowest productivity firms to leave the market, which leaves matches to occur between workers and better firms.

This happens through equilibrium effects on both worker and firm behaviour. One very important effect is that having unemployment insurance fall with duration of unemployment breaks Diamond's (1971) result that wage offers collapse to a singleton. With declining unemployment insurance workers are ex post different, depending on how long they have been unemployed, creating an equilibrium where many wage offers can coexist. This equilibrium features high productivity firms posting high wages and low productivity firms posting low wages. Therefore if a worker receives an offer from a low productivity firm early on in his unemployment (when his benefits are still high) he may reject it to wait for a better offer from a high productivity firm.

Thus high productivity firms always have their offers accepted in equilibrium, while low productivity firms face 'vacancy risk' as they will have their offers rejected by anyone who hasn't been unemployed for very long. This cost borne by low productivity firms induces those with the lowest productivity to leave the market. This improves output in equilibrium by improving the average quality of matches, but comes at the cost of increased short-term unemployment from workers waiting for better offers. Optimally designed policy takes both of these effects into account.

The authors show that key to designing optimal policy are the features of the productivity distribution of firms. In particular, the variance of productivity determines the extent to which heterogeneity is important, and the skewness determines the relative density of firms with low productivity. This then affects how many firms will be discouraged from entering. Overall, a key lesson of their paper is that even in the absence of moral hazard there is a key role for a declining profile of unemployment compensation. The declining time profile induces worker sorting, enhancing worker matching and increasing output and efficiency.

To read more about Eran Yashiv's work on The Macroeconomic Role of Unemployment Compensation see

• "The Macroeconomic Role of Unemployment Compensation" [Full document in Adobe PDF] (Tomer Blumkin, Yossi Hadar and Eran Yashiv), CEP Discussion Paper 0909, February 2009