The Lingering Unemployment Problem

Aside from setting monetary policies, the Federal Reserve, through its regional banks, also provides some very interesting research.  In some cases, this research actually shows how the Fed's policies have been relatively ineffective at curing the problems that have appeared in the American economy since the beginning of the Great Recession.  This article by two researchers at the Federal Reserve Bank of St. Louis explains quite clearly how the Fed has failed America's long-term unemployed.

Here is a graph showing both the unemployment rate (in orange) and the median number of weeks in unemployment (in black) going back to 1970:

While the headline unemployment rate has fallen and even though we are more than 6 years into the post-Great Recession recovery, the median duration of unemployment for American workers is still at the highest level since the early 1980s.  The median duration rose to as much as 25 weeks in 2010, more than twice the highest level over the last 45 years and has since fallen to 12 weeks.

As we all know, by definition, workers are considered to be long-term unemployed when they are unemployed for 27 weeks or longer.  As shown on this graph, the number of long-term unemployed workers is still higher than it was prior to the Great Recession:

Let's look at a breakdown of these long-term unemployed workers and how the situation has changed for various age cohorts.  From the research by Alexander Monge-Naranjo and Faisal Sohail, it appears that long-term unemployment (LTU) has impacted Americans far differently  when age is taken into consideration.  The authors found that LTU is concentrated in two main groups:

1.) those who are about 50 years of age

2.) those who are about 23 years of age

They note that the average of a worker in LTU rose from 38.3 years in 2006 - 2007 to 40.1 years 2012 - 2013.

The authors divided the long-term unemployed onto one of four age-based cohorts; ages 16 to 24, 25 to 44, 45 to 54 and 55 plus.  Here is a table showing how the likelihood of entering long-term unemployed over time varied for each age cohort:

As you can see, the likelihood of entering LTU prior to the Great Recession increased with age from 16 percent for the youngest workers to 27 and 25 percent for the oldest workers.  By January 2010, the likelihood of entering LTU rose to between 28 percent for the youngest workers to a high of 53 percent for workers aged 55 and older.  By January 2015, the likelihood of being long-term unemployed had dropped to between 20 percent for the youngest workers to a high of 41 percent for workers 55 and older.  While that seems like a reasonable recovery, in fact, it isn't that great since none of the age cohorts have reverted to their January 2005 levels.  As you can see from the last column on the table, there is a very significant "recovery gap" between the share of long-term unemployed for each age cohort from prior to the Great Recession.  For older workers, aged 55 and older, the share of long-term unemployed is 60 percent higher than before the Great Recession and for prime-age workers between the ages of 25 and 44, the share of long-term unemployed workers is 73 percent higher than before the Great Recession.

This graph shows how the age distribution of long-term unemployed workers has shifted from 2006 - 2007 to 2012 - 2013:

As the research shows, long-term unemployed workers are getting older.

This research on the changing pattern of long-term unemployment leads us to two possible conclusions based on the higher likelihood of being long-term unemployed for certain age groups:

1.) Prime-age workers who were driven into long-term unemployment during the Great Recession are likely to suffer permanent negative impacts on the remainder of their working careers.

2.) Older workers who were driven into long-term unemployment during the Great Recession will find it far more difficult to regain employment in their chosen profession because they may not have the skills currently in demand by employers.  This will lead to postponed retirements.

The Fed's own research is showing that their "heroic" monetary policies since 2008 have still left millions of long-term unemployed Americans far worse off than they were prior to the Great Recession.  The impact of long-term unemployment has affected both older and prime-age workers between the ages of 25 and 44 more than their younger counterparts, a feature of today's job market that will be felt by these workers for years to come.

The Impact of Joblessness on America's Children

Updated December 2014

While we always think of the impact of unemployment on the working age population of the United States, the impact reaches further into the family unit.  Statistics gathered by the Economic Policy Institute look at how difficult long-term unemployment has been for American families.

Here is a graph showing the number of American children with an unemployed parent between 1994 and the end of 2013:

Of course, the population of the United States has grown substantially since 1994 as has the number of children; between 1994 and 2013, the number of American children grew from 65.96 million to 69.93 million, an increase of 3.97 million or 6 percent.  What we would not expect is the sharp increase in the number of children with an unemployed parent that took place during the Great Recession when the number more than doubled from around 3.5 million in late 2007 to over 7.5 million in 2010.  Since that time, even though the number of children has dropped by just under 900,000, the number of children with an unemployed parent has only dropped to 5.42 million, still elevated above its pre-Great Recession level. 

Here is a graph showing the percentage of American children with an unemployed parent between 1994 and the end of 2013:

When we remove the impact of a change in the number of children, we see how dire the situation is for American children.  In the years before the Great Recession, the percentage of children with one unemployed parent ranged between 4.5 and 7 percent.  During the Great Recession, the rate shot up to just over 10.7 percent in mid-2010 and had dropped to 7.8 percent at the end of 2013, still well above the 14 year average between 1994 and 2007.

We all know that long-term unemployment which is defined as being jobless for more than twenty-seven weeks is proving to be a stubborn issue as shown on this graph:

Here is a graph showing the number of American children with a long-term unemployed parent between 1994 and the end of 2013:

As we noted on the first graph, the number of children with a long-term unemployed parent grew rapidly during the Great Recession, reaching a peak level of 3.32 million at the end of 2010.  By the end of 2013, the number had dropped to 2.06 million, still twice the maximum level seen during the period from 1994 to 2009.

Here is a graph showing the percentage of American children with a long-term unemployed parent between 1994 and the end of 2013:

During the period between 1994 and 2009, the percentage of children with a long-term unemployed parent hit a high of just over 1.5 percent in early 1994 and again in early 2004 and was frequently less than 1 percent over most of the 16 year period.  During early 2011, the percentage of children with a long-term unemployed parent hit a peak of 4.68 percent, over four times the rate experienced just prior to the Great Recession.  As the economy has recovered since the end of the latest Recession, the percentage of children with a long-term unemployed parent has dropped, however, the 2013 year-end rate of 2.94 percent is still nearly three times the background rate during the period between 1994 and the Great Recession.

As we can see from this data, millions of American children are suffering this extended post-Great Recession period of joblessness.  The prolonged joblessness of parents will have a substantial impact on the future of America's children; without jobs, parents cannot save for post-secondary school education let alone pay their monthly bills.  In this economy where many families require two incomes to survive, the issue of millions of family units with unemployed parents will have a long-term impact on America's future success.

The Long-Term Unemployment Trap

As has become apparent, the recovery in the American job market since the end of the Great Recession has been much different than in the past.  One of the key differences this time has been the elevated numbers of Americans that have been unemployed for the long-term (over six months) as shown on this graph:

The economy is nearly five years out of the last recession and the number of long-term unemployed workers is still nearly twice the level seen after the end of the 1991 and 2001 recessions.  The continuing level of high long-term unemployment is puzzling, particularly given that, as shown here, the number of job openings has increased substantially and is nearly at the same level that it was at before the glutinous brown matter hit the rotating cooling device:

Obviously, the elevated level of long-term unemployment has long outlived its original causes.

A paper, "The JoblessTrap" by Rand Ghayad examines what it is like to be one of America's long-term unemployed undertaking a job search.  The author uses a resume audit study in which he explores the extent to which employers forgive long periods of unemployment even when the applicant's merits appear to be strong because they've worked in the same type of firm as the prospective employer according to their resume.   The resumes were constructed to represent young applicants with six years of work experience after college.  By manipulating the length of time spent out-of-work and the relevancy of prior industry experience, the author is able to test the response of potential employers.  To complete his research, the author sent out 3360 fictitious resumes to 600 job ads in the financial activities, wholesale and retail trade, professional and business services and education and health services industries. 

Here are the findings of his study.  The number of interview requests per month drops as the months of unemployment pass, particularly sharply after six months of non-employment for both matching and non-relevant experience as shown on this graph:

Here is a graph showing the interview request rate for workers with matching and non-matching experience over time:

Note that the non-employed applicants with relevant industry experience have a substantially higher interview request rate for the first six months of unemployment.  You'll also notice that the candidates with matching experience see a drop off in the rate of interview requests after 6 months of unemployment and that by 7 months of unemployment (i.e. as the unemployed worker is officially long-term unemployed), the interview rate for both qualified and less qualified types of candidates are virtually identical, suggesting that the experience premium has completely dissipated.

In the case where the applicant has matching experience and is employed, on average, 14.66 percent of employed applicants received an interview request.  By comparison, on average, 10.09 percent of non employed applicants received an interview request.  In the case where there was no relevant experience, an average of only 5.83 percent of employed workers received an interview request compared to an average of only 4.35 percent of unemployed workers.  Short-term unemployed workers with same industry experience had a 16.11 percent interview request rate compared to 8.33 percent for those applicants with different industry experience.  The situation for long-term unemployed workers was far different; long-term unemployed workers with same industry experience had only a 1.94 percent interview request rate compared to a measly 0.56 precent rate for those applicants with different industry experience.

This analysis shows that employers strongly discriminate against long-term unemployed workers with related industry experience far more than their apparently less qualified peers as time passes.  Once six months of unemployment passes and workers officially become long-term unemployed, the chances of getting an interview is essentially the same for candidates with both relevant experience and those with no relevant experience.   It appears that firms adjust their screening thresholds to avoid the chances of hiring what they believe to be a low productivity worker, even though that worker has relevant experience and is less likely to use company resources for training.

America's long-term unemployed are caught in a trap.  The longer that they are unemployed, the more likely it is that they will remain unemployed for even longer since the odds that they will get an interview request decline to near zero as the months pass even if they have relevant experience that might benefit a potential employer.  Even as the economy creates a growing number of job openings, America's long-term unemployed are unlikely to benefit. 

Who Are America's Long-Term Unemployed?

As has been repeatedly pointed out, the employment picture in this recovery is different than other recoveries and, as shown on this graph, is largely due to the elevated number of Americans that are unemployed for long-periods of time:

A paper "Are the Long-Term Unemployed on the Margins of the Labor Market" by Alan B. Kruger, Judd Cramer and David Cho looks at what has happened to America's long-term unemployed since the so-called "end" of the Great Recession and who they are.

America's long-term unemployed are a persistent problem for the economy as you can readily see in the first graph from FRED, and are pushing up the overall unemployment rate.  As you can see on this graph, a very significant 37 percent of America's total unemployed workers have been unemployed for 27 weeks or more:

While this is down from its post-Great Recession peak of 45.3 percent in April 2010, it is still three times the average level of long-term unemployed seen between 1948 and the beginning of 2008. 

In contrast, as shown on this graph, the number of civilians unemployed for 5 to 14 weeks (the short-term unemployed) is very close to normal levels looking back to the mid-1970s:

Here is a graph showing the percentage of total unemployed Americans that have been unemployed for 5 to 14 weeks:

It's interesting to see that the percentage of workers who have been out of work for 5 to 14 weeks is nearly the smallest proportion of the total unemployed looking all the way back to 1948.

Certain parts of the economy have a higher proportion of long-term unemployed than others; 36 percent were previously employed in sales and service and 28 percent were employed in blue collar jobs.  When these long-term unemployed do return to work, they tend to return to similar occupations that they held prior to being laid off.

Among workers who had been unemployed for 27 weeks or longer in a given month in the period between 2008 and 2012, fifteen months later:

- 30 percent were unemployed and looking for work

- 34 percent weren't working and weren't looking for work

- 36 percent were employed

Drilling down further into the data, of the 36 percent that were employed:

- 11 percent had been employed full-time for at least 4 months

- 14 percent had been employed for at least 4 months but at least 1 month was part-time

- 11 percent had been employed for some but not all of the previous 3 months

Long-term unemployed workers have problems finding jobs in all states, even those with low unemployment rates; among the 14 states with the lowest unemployment rate (average of 4.4 percent), long-term unemployment grew to 4.5 times its historical average.  In general, the authors found the following about the long-term unemployed:

1.) They tend to be younger with 40 percent of the long-term unemployed being between the ages of 16 to 34, 29 percent being between the ages of 35 and 49 and 31 percent being older than 50 years of age.

2.) They tend to be less educated with 18 percent having no high school, 35 percent having a high school diploma, 20 percent having some college and 18 percent having a Bachelor's Degree or higher.

3.) They tend to be unmarried with 44 percent having never been married, 37 percent being married and 19 percent being widowed, divorced or separated.

4.) Racially, there is a strong connection with long-term unemployment as shown here:

African-Americans make up 22 percent of the long-term unemployed compared to just under 13 percent of the total population and Hispanics make up 19 percent of the long-term unemployed compared to 16 percent of the total population.

In general, the longer that a worker is unemployed, the less time they spend looking for a job as the frustration over not having employment builds, compounding the problem.  Because of this, they become increasingly marginalized, particularly since the data shows that, in a given month, only 11 percent had achieved full-time employment for a continuous four month period since they returned to employment.  Sadly, many of these discouraged workers simply choose to leave the workforce simply because there are not sufficient job openings to accommodate them.