Our Long-Term Unemployment Challenge (In Charts)

The reasons why employment has lagged in the recovery remains a central challenge for macro policymakers, influencing the fiscal debate as well as figuring prominently in Federal Reserve justification for its current unorthodox policies.  My colleague Dinah Walker points out a growing body of evidence that the problem of long-term unemployment is at the heart of the puzzle.

The Relationship Between Job Openings and Unemployment (The Beveridge Curve).

During most periods, there is a predictable relationship linking higher vacancies with lower unemployment.  Since 2009, that relationship has shifted up–i.e., a given level of vacancies is associated with higher unemployment, suggesting a less efficient labor market.

Its not unusual for there to be a shift of this sort at the start of the recovery, in part because firms may hold off on hiring until the recovery is more firmly entrenched.  But the shift this time has been larger and more persisitent than is usually the case.  Before 2009, a job openings rate (the number of openings divided by employment plus job openings) of 2.8% (February’s rate) would have been associated with an unemployment rate of about 5.5%, far below February’s unemployment rate of 7.7%.

Recent work by Rand Ghayad and Bill Dickens documents that, unlike past recessions, this time the breakdown in the relationship is concentrated in long term unemployment.

The authors did not find any evidence that the shift was concentrated in any particular sector of the economy, or across age or education.   One theory of long term unemployment is that there is a mismatch of skills between what the long-term unemployed have and what firms want–recent work by Federal Reserve economists finds some evidence to this effect in the U.S. and a number of European countries–but if that was prevalent we would expect to see variation across these categories.  Ghayad and Dickens go onto suggest though that a better explanation for the shift amongst those unemployed for longer than 27 weeks is that the long-term unemployed may be “searching less intensively–either because jobs are much harder to find or because of the availability of unprecedented amounts and durations of unemployment benefits”.  For any of these stories, a puzzle is why the effects don’t show up in shorter-term spells of unemployment as well.

Recently, Rand Ghayad ran a follow-up experiment, sending 4800 ficticious applications for 600 job openings. The applications differed by length of unemployment, how often they switched jobs, and experience.  What he found was long term unemployment dramatically lowers your chance of a callback.  In fact, long term unemployed with relevant experience (the red line) were less likely to get called back than those that did not have relevant experience (the blue line), but who had a shorter unemployment.

Where does this leave us?  Extended unemployment benefits appear to be contributing to the long-term unemployment challenge, and the rolling back of extended unemployment benefits should increase incentives to work. But this cannot be the entire story as there are good reasons to suspect that the loss of skills and connection to the job market provides a significant and lasting impediment to long-term unemployed reentering the workforce.  Those unemployed for long periods of time tend to lose skills, look less attractive, and have a harder time finding jobs than those who have been unemployed for shorter periods of time.  Expansionary monetary policy can boost overall employment and thus limit the rise in the pool of long-term unemployed, but has less ability to address those that have lost conenction to the workforce.  This puts the spotlight more on structural and fiscal policies, including measures aimed at helping those who are unemployed to find jobs and rebuild skills (earned income tax credit, job training).

This piece is cross-posted from Macro and Markets with permission.

6 Responses to "Our Long-Term Unemployment Challenge (In Charts)"

  1. Burk   April 18, 2013 at 4:15 pm

    Oh, Right.. blame the prejudice of employers on unemployment benefits. What a shameless piece.

    Anyhow, let me offer another model. Companies have personnel departments, with little to do during slack times. They also are always on the hunt for the perfect employee. So during these times of depression, employers are so spoiled that they advertise excessively, (compared to boom times), with insanely specific job requirements and little real need behind them. It is a bit of kabuki to keep the fishermen fishing, and the fish biting.

  2. SBG   April 19, 2013 at 9:43 am

    The noticeable piece missing is the effect of public policy on hiring. When the governemnet adds significant cost to an employee, employers will smartly opt to avoid hiring new hires to avoid the cost. The Obama administration has not only added significant cost per employee due to their social and regulatory policies, they have stated that they will continue on this trajectory. Combine that with your findings about the increase in unemployment benefits and you have a problem created purely by the Obama administration which won't change until that changes. Unfortunately so many of those that vote for him are the ones most affected.

  3. D. Madrone   April 19, 2013 at 10:49 am

    If you have been unemployed longer than six months, even with unemployment benefits, you have fallen down the poverty well. It takes money to present yourself everyday in clean, up to date clothes and shoes. After six months unemployed you may be without health care, including dental, and will certainly be desperate. And I remind you that even now there are about three times as many people looking for work as there are jobs available. This isn't about people sitting on their hands and living the good life on unemployment. It is about an economy that throws people away.

  4. benleet   April 19, 2013 at 11:07 am

    Corporate profits as a % of GDP are at a historical high. The employment to population ratio has retreated to the level of 1983, in 2000 it held at 64.7% and today it's at 58.5%. This 6.2% drop is overall almost a 10% reduction. And there has been a labor participation rate drop-off of about 5% from the 20 year (1988 – 2008) average, participation rate (from an average of over 66.6% to today's 63.3%). GDP is underperforming below potential at around 6% of GDP. The macro forecast would predict lower corporate profits in the near future. I think the EPI has it right, add a $200 billion per year stimulus until employment and purchasing demand rebounds, and private sector employment recovers. Private sector employment has added about 8% of expected jobs since 2000, that is, 92% of expected jobs did not materialize over a 13 year period, that's a backlog of nearly 15 million jobs. A net of 0 private sector jobs were created January 2000 to November 2011. My blog: http://benL8.blogspot.com

    • windriven   April 19, 2013 at 3:57 pm

      I'm right with you up to the point where you prescribe $200B in additional stimulus. The primary positive impact of stimulus to date has been to shore up Wall Street and restore bonuses there to prerecession levels. Theoretically I'm with you. But this administration is not competent to administer stimulus that stimulates the broad economy.

  5. Tom   April 20, 2013 at 9:46 pm

    You're absolutely write about long-term unemployed having a stigma with employers.

    But I don't see how the data as you describe it tells us that there isn't a mismatch of skills. It seems only to tell us that no particular group of long-term unemployed when broken down by field is more mismatched than any other. That could mean that all of them are equally mismatched in skills to what employers want. Given that the vast majority of long-term unemployed are not college educated, and most of those are former construction and manufacturing workers, that wouldn't be a surprising result. As for the relatively small numbers of college-educated long-term unemployed, there's all the more reason for their skills to be considered not only rusty but dated.