The release of last week’s state-level unemployment data provides an inkling of why economic research can be both exhilarating and helpful to public policy.  As it turns out, this May the unemployment rate in Indiana fell beneath the threshold for the long-term extension of benefits.  That means that beginning next month, unemployed Hoosiers will be eligible for no more than 93 weeks of unemployment (down from 99 weeks).  This means that a large number of the states’ unemployed workers will lose their benefits in the coming weeks.  This begs the research question:  What will happen to the unemployed?

If history repeats itself, we will see an acceleration of new employment in Indiana in the coming months.  In the languid prose of one well known study of the issue, the ‘escape rate’ from unemployment tends to occur near the date of ‘benefit exhaustion.’  This finding is replicated across countries and across time by economists of all stripes.  Another finding is that for every five or six weeks of potential unemployment benefits available, the average length of unemployment grows by a week.  The economic research backing this finding is of high quality and performed by some of the most brilliant economists of the age.  In essence, the body of research tells us that longish periods of unemployment compensation tend to cause longish periods of unemployment.  This should be unsurprising; when you pay people not to work, the propensity to work declines.  If this trend holds—and it almost certainly will—we should see a big drop in long-term unemployment in Indiana in the coming months. 

The exciting part of the economic research is that we now have a country in which there is a wide range of variability in the length of available unemployment compensation.  This gives rise to a natural experiment that should inform our policy response in the next recession

The question also arises over the value of unemployment compensation to the economy.  While my anti-Keynesian friends will no doubt stir-fry over my comments, I for one think it a very valuable public policy. Certainly the role of layoffs over the past half century has altered the relevance of the program.  So, to survive it must change a bit. 

In past decades the most important role of unemployment compensation was as an automatic stabilizer of the economy during a downturn.  Today it is more important as a way of smoothing job transition for workers.  Because of this, we ought to vastly rework the system.  Instead of weekly payments that incentivize workers to take leisure, we should offer two types.  The first is a lump-sum payment to workers seeking new jobs.  This would tide folks over, but eliminate any incentive to pass on a job opening.  The second is a retraining or education grant that would promote retraining.  Neither of these are perfect fixes, but both are improvements on the current scheme of paying workers not to take available jobs just when we most need them to.