Nobel Prize Winners: How We Can Reduce Unemployment

The need to find matches is present in all of economics. Buyers of goods must be matched with sellers, lenders must be matched with borrowers, workers matched with firms, and so forth. The work of Peter Diamond, Dale T. Mortensen, and Christopher A. Pissarides looks at markets with search frictions, i.e. markets where making matches is costly in terms of resources, search time, etc., and explains how markets work when the matching is not automatic.

The work is particularly relevant for labor markets. In particular, the work helps us to understand how it is that there can be workers willing to work, firms that are willing to hire, and yet still have an elevated unemployment rate. Just as importantly, the work helps us to design unemployment compensation and other programs that help to promote efficient matches between workers and firms. To get a flavor of this work, one result from Peter Diamond speaks to the benefits of unemployment insurance. This results states that labor markets can be made more efficient through unemployment insurance that allows some workers to pass up jobs. Without insurance, workers would be motivated to take any job that comes their way, even jobs they are overqualified for, and that leads to an inefficient match between the worker and firm and closes off the job to someone who is better qualified for it. Bu allowing workers the time to find the jobs they are suited for, they are much more productive and economic growth will be higher. Thus, the belief that higher unemployment compensation always leads to inefficient outcomes, a claim we've heard a lot in the recent debate over whether to extend unemployment insurance, is undermined by this work.

This work is also important for understanding the unemployment rate today, and the extent to which it can be considered structural due to matching problems, or attributable to deficient demand and hence cyclical. Structural unemployment is much harder to overcome with traditional policy. The solution to cyclical unemployment, increasing aggregate demand through monetary and fiscal policy, doesn't help much when the problem is structural. However, the work also implies that such distinctions may not be particularly useful, for example cyclical unemployment can turn into a structural problem if search persists for too long and skills deteriorate -- the deterioration in skills makes a match even less likely and hence creates a structural problem -- and these workers become part of the long-term or permanently unemployed.

While the current unemployment problem is mostly cyclical, that doesn't mean that there is nothing policymakers can do about the unemployment problem even if it is of the traditional structural variety. One key is to maintain workers' connection to the labor market by providing employment opportunities during the time when better opportunities -- those that provide a better match -- have not yet materialized. In the words of Christopher Pissarides:

Professor Pissarides said that he thought the work being honored had one lesson in particular for today's policymakers: "What we should really be doing is make sure the unemployed do not stay unemployed for too long, to try to give them direct work experience," so that they "don't lose their attachment to the labor force."

[For more on the work of Diamond, Mortensen, and Pissarides, see Marginal Revolution, Ed Glaeser, Paul Krugman, Real Time Economics, Rajiv Sethi, Econospeak, Cheap Talk, Edmund Andrews, NY Times, FT, and the WSJ.]

Update: Let me add this comment from Paul Krugman on the cyclical versus structural unemployment issue:

...With regard to current concerns, probably the most relevant paper is Blanchard and Diamond on the Beveridge Curve -- the relationship between job vacancies and unemployment.

What's the moral of that paper? It shows that structural unemployment is a real issue, and that the volume of structural unemployment shifts over time. It also shows, however, that short-term movements in unemployment are overwhelmingly the result of overall shocks to demand -- in effect, Keynesian business cycles.

And given the debate now underway about whether we're mainly facing a rise in cyclical or structural unemployment, it's definitely worth noting that they give us a simple way to make that distinction:

The economy, however, is subject to two types of shocks with quite different effects. Changes in the level of aggregate activity cause rates of job creation and job destruction to move in opposite directions, while changes in the intensity of the reallocation process cause them to move in parallel.

What do we see? ... Overwhelmingly, what we've seen is a simultaneous fall in vacancies and rise in unemployment, which tells us that this is an aggregate demand shock. ...

Deeply relevant work. And by the way, for those of us in the modeling business, Peter Diamond's work is breathtaking in its elegance -- nobody cuts through the complexities with such grace.

A happy day for economic theory.

Diamond's expertise in the labor market, which is of particular concern in the current economic environment, was noted when he was nominated for the Fed. However, his nomination was sent back to the White House. Several Republicans object to his candidacy on the grounds that he has limited macroeconomic policy experience. The White House renominated Diamond, but his nomination requires another vote of the Senate Banking Committee before it can go to the floor.

Senate Banking Committee Chairman Christopher Dodd (D., Conn.) has said he would hold a hearing on Mr. Diamond's nomination when the Senate reconvenes after the November mid-term elections.