The Goldilocks Approach to Unemployment

Were extended unemployment benefits a big mistake?

Still searching.

Photographer: Joe Raedle/Getty Images

A new working paper from the National Bureau of Economic Research argues that extended unemployment benefits probably raised the unemployment rate by about 50 basis points. This is of particular interest to me for two reasons:

  1. Like all Americans, I am passionately interested in having a strong labor market that employs as many people as possible.
  2. In my book, I argued that extended unemployment benefits were a good idea during the depths of the Great Recession. If you look at the data on separations and job turnover, it seems obvious that what was driving unemployment higher was the fact that there were way more people who wanted jobs than there were jobs available.

So was I wrong? Were extended unemployment benefits a big mistake?

I've thought about this pretty carefully. I've also read other papers that tackled the same question. And though I am open to revising my beliefs, so far I stand by them: Extending unemployment benefits was good policy in 2008. But let me propose an addendum: Ending the extension was good policy in 2013. And it maybe should have been done even earlier.

Before I proceed, however, the standard caveats: This is only one paper. Its results could be wrong, not through any fault of the authors, but because the world is complicated, and it's hard to tease true economic effects out of noisy data. However, the methodology seems sound, and I have no reason to think that it is wrong; I'm just applying my standard caution against relying too heavily on the results of a single paper, even a good one.

Having gotten that out of the way, assume that it is correct. Why doesn't that change my mind about the merits of extending unemployment benefits in 2008 and 2009?

Because the labor market was very different in 2008 and 2009. Back then, the ratio of job seekers to job openings had more than tripled; at its peak, there were nearly seven people searching for every one available job. The dominant factor in unemployment was clearly not the availability of benefits; it was the lack of employers who had places to employ new workers. Even a diligent job seeker was in for a long search, and extended unemployment benefits did a lot of good things: injected some cash into a shattered economy; stabilized family finances so that more people did not join (and make worse) the cascade of defaults, bankruptcies and foreclosures; protected people from the paralyzing despair that can actually make it harder to look for a job. At that time, I think extensions were the right policy.

By December 2013, when Congress decided to end the extension program, the job market looked very different. The ratio of job seekers to job openings had fallen much closer to its historical average. At that point, extending unemployment benefits could make a plausibly large difference in the unemployment rate.

As I've written before, exiting unemployment unsurprisingly seems to be closely connected to how much time you spend looking. Unfortunately, looking for a job is miserable, and it makes unemployed people anxious -- and anxious people may put off looking, because the constant rejection, the feeling of doing your best and not getting anywhere, ranges from unpleasant to terrifying.

This is a huge mistake. The longer you're off the market, the harder it is to find a job. It's an entirely understandable mistake, one I myself made during my two-year attempt to find a permanent job after grad school. But a policy that helps people make that mistake is a bad policy. And while a 0.5 percent increase in the unemployment rate may not sound like that, we're talking about hundreds of thousands of people who we seem to have helped to commit this grave error.

That's why I think that America generally has the right idea about unemployment insurance: We provide benefits, but they're sharply time-limited. It's a cushion to tide people over, not a means of helping them avoid the psychological pain of a job search (or, yes, in some cases, to put off looking for work because you'd rather not work). Countries that provide lengthier and more generous support tend to have higher unemployment rates -- though to be sure, they also tend to have other labor market policies, such as rigid job protections, that also depress employment.

But as sound as our policy is, we need a safety valve for extraordinary circumstances like the Great Recession, when even the most dedicated job seekers had slim chances of landing another slot right away. And again, there, I think we hit on the right policy: Extend benefits in emergencies, then terminate those extensions when the labor market inches back to normal.

So I don't think I was wrong to support extending unemployment benefits during the worst of the recession. But I also don't think that Republicans were wrong to want to curtail those extensions five years later. Both were good policy, and I hope they'll be a model for any future crises we face.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

    To contact the author on this story:
    Megan McArdle at

    To contact the editor on this story:
    Brooke Sample at

    Before it's here, it's on the Bloomberg Terminal.