What's the best way to help the long-term unemployed? There's new concern about how difficult it is for the long-term unemployed to find jobs in light of an interesting study by Rand Ghayad, a visiting scholar at the Boston Fed and PhD candidate. Ghayad sent out resumes that were identical except for how long the candidate was unemployed, and the longer they were unemployed, the less likely it was they would get called back. Matthew O'Brien has a great writeup of the study here, and there's additional thoughts from Megan McArdle, Paul Krugman, Felix Salmon, and Matt Yglesias.
The impact of long-term unemployment on human lives is very real, and I think the government should be combating it using every tool it has. However, I want to push back on a few of the economic ideas that tend to hover in the background of these discussions; specifically, the idea that we should consider the long-term unemployed uniquely in trouble in this economy. Because, based on my interpretation of the evidence, the best approach to handling this problem is to aim for full employment.
It's well known that it is harder for those who have been out of the labor force the longest to find jobs. It would be weird if Ghayad hadn't found that result. There is a large debate in the literature about whether this is driven by employers or job candidates, and Ghayad provides a very useful study finding that employers are a key part here.
But let's look at the likelihood of finding a job in three different economic scenarios (2000, 2007, and 2012) by duration of unemployment:
But notice that when the economy is much stronger, as it was in 2000 when unemployment averaged 4 percent, the rate at which the long-term unemployed find jobs jumps up. Let's zoom in on the last category, the job-finding rate of those who have been searching for a job for 53 weeks or longer, and chart it back to 1995. (Since the data, provided by the BLS, is not seasonally adjusted, the number here is a 12-month rolling average.)
As you can see, it's much easier for the long-term unemployed to find jobs when there's a tight labor market, like there was in the late 1990s. This rate collapses in a recession, and with years of 7+ percent unemployment, it has stayed depressed.
A lot of people are drawing conclusions that something has broken in long-term unemployment based on a previous paper by Rand Ghayad, where he disaggregates the Beveridge Curve by unemployment duration. I've been critical of that paper. I think, strictly speaking, that the disaggregation just tells us that the long-term unemployed have become a larger percentage of the unemployed, which we knew. Meanwhile, the labor market is depressed for everyone, even short-term unemployed (also see SF Fed for more evidence of this). As the long-term unemployed are less likely to drop out of the labor market than in normal times right now, the dramatic increase in the long-term unemployed hasn't turned into a large drop in labor force participation like many worry about.
We should do things that are smart policies that target the long-term unemployed. Amy Taub of Demos has done convincing work on why ending credit checks as part of the job interview process would be a good idea. Extending unemployment insurance is also important. But the idea that we should change course away from boosting the general economy strikes me as a bad idea. The long-term unemployed experience the worst impact of a generally weak economy. But its that weak economy that is doing the damage. If unemployment was actually brought down, which we could do with more expansonary policy, then employers couldn't afford to be so choosy.