On Up with Chris Hayes today, one of the topics we discussed was the extension of unemployment insurance benefits. I’d like to take this opportunity to revise and extend my remarks on when and whether we should worry that UI benefits are inducing unemployment. Particularly, though I favor a continuation of extended 99-week unemployment benefits—for now—we should be very careful about rhetoric along the lines that “we don’t need to worry about extended UI because unemployed people want to work.”
UI benefits are payments that people get only if they are not working, and therefore they are a disincentive to find work. Of course, they are not an absolute disincentive. People on UI often have good reasons to look for new jobs—they would make more working than they can draw in UI benefits; they avoid damaging career gaps; they get fulfillment out of work.
If you have something like five job seekers for every job listing, like we do today, you don’t need to worry about this very much. Let’s say that half of unemployed workers are eligible for UI, and half of those prefer to draw UI rather than return to work. You’re still left with several interested job seekers for every job listing, and so UI benefits should not have a big effect on unemployment rates.
The effect on unemployment is nonzero because workers are not interchangeable cogs. A modestly tighter (but still loose) labor supply means that it will take employers somewhat longer to find suitable candidates, and the job market in some regions and professions is tight even while the overall market is loose. A 2010 study from the San Francisco Federal Reserve compared cohorts of workers that are eligible and ineligible for UI benefits, and suggested that extended UI benefits might be raising the unemployment rate by 0.4 percentage points.
The considerations would look very different in a strong economy. The negative incentives from UI benefits would be a lot more important, because businesses would want to hire more employees than are interested in working—you would see a substantial amount of induced unemployment. And the positive effects would be much less—there would be no need for economic stimulus, and the humanitarian benefits of extended UI would be reduced.
So, that’s why I said the concerns about incentive effects of UI benefits should not be dismissed. This is a policy area that requires cost-benefit analysis, and at some point, as unemployment falls, that analysis will shift in favor of shortening the UI window from 99 weeks. Just not today.