I Can't Stop Looking at These Terrifying Long-Term Unemployment Charts

Unemployment looks normal for everyone except those out of work for six months or longer. If we don't act soon, the long-term unemployed will become unemployable.


There's a new cliff in town, and it's much scarier than the fiscal cliff. It doesn't have anything to do with expiring tax cuts or sequesters. It has to do with people who have been out of work for six months or longer. It's the worst cliff of them all: the Unemployment Cliff.

Our unemployment crisis is also an unemployment enigma. When jobs openings go up, unemployment should go down. This relationship is captured by the Beveridge Curve, seen below. The diagonal red line says that when there are more vacant job openings, the unemployment rate should be lower. But as you can see in the bottom right hand corner, something strange (and very bad!) is happening. More job openings haven't produced more jobs. That suggests a mismatch between jobs and skills ... the dreaded "structural unemployment."


Look again. This might be the most important chart you'll see. If unemployment really is structural, there's not much more policymakers can do to bring it down. If it's not, policymakers should be tearing their hair out to put people back to work. So, is it? No. A pioneering paper out of the Boston Fed pretty definitively shows that we have a long-term unemployment problem, not a structural unemployment problem. 

There's always a story when it comes to structural unemployment, and it's almost always a story about old workers needing new skills for our brave, new economic world. The Boston Fed paper, by Rand Ghayad, a Ph.D. candidate in economics at Northeastern and Visting Fellow at the Federal Reserve Bank of Boston, and William Dickens, a professor of economics at Northeastern and visiting scholar at the Federal Reserve Bank of Boston, looks at the Beveridge curves for different ages, industries and education levels to figure out exactly who is getting left behind nowadays. The answer is ... everybody. The Beveridge curves for young and old, blue-collar and white-collar, and high school and college graduates all look alike -- there's the same upward tick in all of them. There's a word for this, and that word is flabbergasting. As Ghayad and Dickens point out, the last time we had a structural unemployment problem was during the deindustrialization of the 1970s and 1980s, when Beveridge curves for blue-collar workers, and only blue-collar workers, moved up. Did we all wake up in 2008 and suddenly lose our skills?

Not exactly. Ghayad and Dickens broke down Beveridge curves along one more axis -- length of unemployment. Here's what it looks like for people who have been out of work for less than six months. This is what normal looks like.


This chart is worth approximately 20 words. People out of work for less than six months haven't had a harder time finding work than they usually do. But the Beveridge curve has shifted up for all workers, so that implies all of the shift must have come from people out of work for six months or more. The chart below shows us that this is indeed the unhappy case. Unemployment is a cliff that's hard to climb out of after six months.


It's hard to imagine a big skills or incentives gap between people unemployed for five months and people unemployed for six months. But it's not hard to imagine companies treating their resumes differently. Overrun HR departments might just toss the resumes of applicants who have been out of work for six months or more, because they assume there must be something wrong with people who have been out of work that long. Sadly, this isn't a hypothetical. Scott Pelley reported on firms that won't consider the long-term unemployed -- or the unemployed, period -- for 60 Minutes earlier this year. It's depressingly legal to discriminate against the unemployed, and a depressing number of companies do just that.

Circles don't get more vicious than this. The people who need work the most can't even get an interview, let alone a job. It's a cycle that could end with the long-term unemployed becoming unemployable. It's what economists call hysteresis, the idea being that a slump, left untreated, can make us permanently poorer by reducing our future ability to do and make things. You should be scared anytime you see the words "permanently" and "poorer" together in a sentence -- especially if you're a policymaker. We need more stimulus, and we need it now. That means the Fed needs to figure out its thresholds for forward guidance and Congress needs to not only undo the fiscal cliff, but also, please, give us some more infrastructure spending. Heck, Larry Summers and Brad DeLong think fiscal stimulus might even pay for itself with interest rates so low by preventing hysteresis from happening.

We can do better, if we want to. As Paul Krugman points out, people told themselves structural unemployment was to blame during the Great Depression too, only to discover that all the people who supposedly didn't have the right skills suddenly did once the military buildup started. Funny how adequate demand works. The best thing we can do for long-term growth is to forget the long-term and get the long-term unemployed back to work now.

In the long run, we can't afford to worry about the long run.