Extended Unemployment Benefits Make Sense

This libertarian's case for extended unemployment insurance.

Of all the quixotic policy battles over the last few years, my nominee for the most misguided is the one over extended unemployment benefits. Although I opposed Obamacare, and wasn't overly enthusiastic about the stimulus (particularly its composition), the Democrats had the right of it on extended unemployment benefits: In the depths of the Great Recession, extending benefits was good policy. Our only mistake was not making them longer, and more generous.

Many of my readers will say that I am crazy, and probably announce that I am once again being expelled from the libertarian movement. Hear me out. Extended unemployment benefits may not be the best possible policy in the best possible world. Yet for the dilemma that the U.S. faced in 2009, they were the best policy we had.

You're asking, what about moral hazard? If you give people money to be unemployed, won't people stay unemployed?

In normal times, yes, there is a moral hazard problem with unemployment benefits. Economists have been finding a link between the duration of benefits and the duration of unemployment for a long time. I don't think that this finding has been overturned: In general, if you subsidize something, you get more of it. In this specific recession, however, the big labor market problem wasn't getting people to take jobs; it was finding jobs for them to take.

Observe this graph from the Bureau of Labor Statistics. It comes from their JOLTS (Job Opening and Labor Turnover Survey) data. What it shows is an unprecedented spike in the ratio of job-seekers to job openings:


As you can see, the ratio has recovered from its peak -- but it has recovered to the high point of the last recession, when yours truly was having a hell of a time finding a job, despite many, many resumes sent and networking lunches attended.

Moreover, it's not even clear that this is a real recovery; as Peter Orszag recently pointed out in our pages, while job openings have improved dramatically, hiring has not.


Somewhat surprisingly, the problem is not that people are being fired/ laid off at an unusual rate -- separations actually fell. That's because everyone's clinging to their job like a deranged limpet:


So even though layoffs have actually been pretty steady, except for a small spike in early 2009, we have a high unemployment rate, because labor market churn has basically stopped. Since no one's quitting their job, there's nowhere for the displaced workers to go. The current labor market is like a series of musical chairs: For one person to find a job, someone else has to lose out.

So what's the deal with all those job listings? I find the explanations offered at Outside the Beltway plausible, notably that companies are either putting up pseudolistings to justify an H1B visa, or they're "fishing," as Dave Schuler calls it -- willing to hire the perfect candidate on the cheap, but otherwise prepared to limp along with their current workforce. Either way, it's cruel to the folks who are still desperately looking.

But at least they're still looking, which is why I support extended unemployment benefits for the recent emergency. Some people will definitely take the benefits while pretending to look for work. But as long as they're pretending to look for work, some of them will find it. The alternatives are things like disability insurance, which seems to have absorbed a lot of unskilled workers over the last decade or so. And to paraphrase a smart economist of my acquaintance, "unemployment benefits, even extended ones, do eventually end. Disability almost never does." Pushing people off of unemployment benefits -- and on to SSDI -- is bad policy.

Now, maybe you think that SSDI shouldn't exist, or should only be given to people who have a fatal disease or a missing limb. The fact remains that right now, it does exist, and almost anyone who is persistent about applying seems to be able to get onto the system eventually, after which they get back benefits to cover the cost of their appeals. And once you get onto the system, it's very dangerous to exit -- earning more than a token amount of money risks not only your guaranteed disability payment, but also your Medicaid and other in-kind benefits. (Hopefully, that incentive declines somewhat once Obamacare kicks in.) So virtually no one ever leaves the system once they get on.

Given that reality, unemployment benefits were a cheap way to keep people attached to the labor force, however marginally, rather than the welfare system. While they may have nominally raised the unemployment rate, there's a very decent chance that they actually lowered the number of people who detached themselves from the labor force.

It's not that the moral hazard argument is wrong. It's just that over the last five years, we've been in extraordinary circumstances. During normal times, or even a normal recession, lengthy unemployment benefits were a bad idea. But in an emergency, when the ratio of job openings to seekers suddenly spiked to record levels, it made sense to try to keep people looking for work, rather than worrying that they might collect an extra few months' worth of checks they didn't deserve.

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

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