Friday, January 7, 2011

Unemployment Statistics - What don't they tell us?

We all hear the monthly unemployment statistics but what goes into calculating them, who is included in them and, most importantly, who is statistically excluded from the workforce by the Bureau of Labor Statistics?

First, let’s look at four types of unemployment so near and dear to the hearts of your friendly neighbourhood economist:

1.) Structural unemployment occurs when there is a long-term imbalance between the supply of labourers and the demand for those labourers; it can be economy-wide or specific to certain sectors of the economy (i.e. there can be a permanent oversupply of auto workers when auto manufacturing plants shutter).  Structural unemployment can also occur when the skills of the unemployed workers do not match the demands of the market, for example, as companies increased their use of technology, those workers who were not trained in those specific fields found themselves out of work.  Again, the use of robots to build automobiles resulted in greater efficiency and the need for fewer workers.  Structural unemployment can also occur when the demand for workers is in a geographically different part of the country than where the workers live.

2.) Cyclical or Keynesian unemployment is the most familiar type of unemployment.  This type of unemployment occurs during certain parts of the business cycle when there is simply not enough demand in the economy to supply jobs for all people who wish to work.   Simply put, the number of unemployed workers exceeds the number of job vacancies.  Economists fall into two camps when it comes to cyclical unemployment.  Some classical economists do not believe that cyclical unemployment exists because when demand for labour falls and the supply of labour remains steady, wages will decline and the lower wages will result in additional hiring resulting in a reduction in unemployment (the good old supply-demand curve in action).  Keynesian economists believe that the lack of jobs can be resolved by government intervention including the use of deficit spending to stimulate the economy (this sounds familiar, doesn't it?)

3.) Classical unemployment occurs when the wages for certain jobs are set above the market level which causes the number of those seeking that particular job to exceed the number of jobs available.  Some economists argue that this is one result of minimum wage laws where the minimum wage is set above market level making it less likely that employers will hire.

4.) Frictional unemployment (who thought of this name?) is a type of voluntary unemployment that arises because of the time needed to match those seeking jobs with the jobs that are available.  This can occur when employees quit their current job with the deliberate intention of looking for a better one.

Okay, enough of the textbook stuff. 

The United States Bureau of Labor Statistics (BLS) is responsible for measuring and releasing labor data.  On a monthly basis, they interview a random sample of roughly 60,000 households for the Current Population Survey (CPS) which is used to calculate the official unemployment rate as defined by the International Labour Organization (ILO).  Only household members that are over 15 years of age qualify.  Here is the ILO's official definition of unemployment:

"The "unemployed" comprise all persons above a specified age who during the reference period were:

"without work", i.e. were not in paid employment or self-employment, as defined in paragraph 9;

"currently available for work", i.e. were available for paid employment or self-employment during the reference period; and

"seeking work", i.e. had taken specific steps in a specified reference period to seek paid employment or self-employment. The specific steps may include registration at a public or private employment exchange; application to employers; checking at worksites, farms, factory gates, market or other assembly places; placing or answering newspaper advertisements; seeking assistance of friends or relatives; looking for land, building, machinery or equipment to establish own enterprise; arranging for financial resources; applying for permits and licences, etc.”

The BLS classifies unemployed persons as those who ' not have a job, have actively looked for work in the prior 4 weeks and are currently available for work."  Employed persons are those who did any work for pay or profit during the survey reference week, those who did at least 15 hours of unpaid work in a family-owned enterprise operated by someone in their household and those who were temporarily absent from their regular jobs".

It may surprise some readers to know that BLS has 6 different measures of unemployment as follows:

1.)  U-1  persons unemployed 15 weeks or longer, as a percent of the civilian labor force.
2.)  U-2  job losers and persons who completed temporary jobs, as a percent of the civilian labor force.
3.)  U-3  total unemployed, as a percent of the civilian labor force (the official unemployment rate).
4.)  U-4  total unemployed plus discouraged workers, as a percent of the civilian labor force plus discouraged workers.
5.)  U-5  total unemployed, plus discouraged workers, plus all other marginally attached workers, as a percent of the civilian labor force plus all marginally attached workers.
6.) U-6  total unemployed, plus all marginally attached workers, plus total employed part time for economic reasons, as a percent of the civilian labor force plus all marginally attached workers.

Now we get to the heart of the matter.  When we see the unemployment rate reported in the mainstream media (i.e. 9.8 percent), that statistic represents what BLS terms "U-3".  If you look at the other "U" measures though, you'll see that they are much more inclusive and are probably far more representative of the real employment/unemployment world to most Americans.  For instance, if you were to use the U-6 measure to define what most of us think of as "unemployment", it would be a great deal higher, in fact, the U-3 rate in January 2011 was 9.0 percent but the U-6 rate was a whopping 16.1 percent.  In fact, the U-6 measure has ranged from a low of 15.4 percent in January 2009 to a high of 17.2 percent in November 2009 and 16.7 percent.  Note that the U-6 rate is pretty consistently 7 percentage points above the officially quoted unemployment statistic.  I guess the government would rather report the U-3 statistic because the U-6 rate is getting uncomfortably close to "Great Depression" territory and, despite the so-called economic recovery, it has still not dropped either!  Could you imagine the uproar if newspaper headlines read "Unemployment drops to 17 percent?".

Here's a table from BLS comparing the 6 unemployment rates now and for the past few months:

A wonderful website called Shadow Government Statistics publishes great charts showing the U-3 and U-6 unemployment rates and their own SGS Alternate rate which roughly approximates the old pre-1994 BLS method of calculating long-term discouraged workers who are no longer counted in BLS statistics.  Here is their chart showing the three unemployment rates:

As noted above, long-term discouraged workers are not included in the official unemployment rate.  These workers are not in the labour force but want and are available for a job and have looked for a job in the past 12 months but who are not looking for a job currently because they perceive that they have no prospect for employment.  This is most closely reflected in the U-4 number and, as shown in the chart above, is 0.8 percentage points higher than the official U-3 unemployment rate and, according to BLS statistics, totalled 1.3 million workers in the latest period.  If that worker has not looked for a job for more than 12 months, they no longer exist in the BLS statistics.  That is, in large part, why the SGS Alternate unemployment rate hovers between 22 and 23 percent.

BLS also does not measure underemployment even though, for example, there are many workers who are over-qualified for their low-wage jobs but are unable to get a job for which they are qualified (i.e. the engineer with a PhD driving a taxi).  The Bureau's rationale is quoted here:

"Because of the difficulty of developing an objective set of criteria which could be readily used in a monthly household survey, no official government statistics are available on the total number of persons who might be viewed as underemployed. Even if many or most could be identified, it would still be difficult to quantify the loss to the economy of such underemployment." 

Yes, it would be difficult to statistically quantify underemployment but it would certainly be interesting to see just how many workers out there are way over-qualified for the job they are currently performing simply because the economy isn't allowing them to get their job of choice. 

From this little primer on unemployment, you can readily see that employment statistics are just like all other government economic statistics, they can make them say pretty much whatever they want them to say and they can pretty much hide what they think we don't need to know.


  1. Listening to the newsradio this morning, I was struck by the announcer's confusion over the fact that jobs created were lower than expected, but the unemployment rate unexpectedly went down. Now, maybe because I am a faithful follower of this blog, but I knew right away what had happened. And it infuriated me that the person delivering the news has no clue unemployment has so many categories.

  2. ...and that's why I write this stuff. Thanks Jenn. An informed voter is a better voter.

  3. This blog is another example of non-print news sources providing more information (news) than traditional news outlets.

    On a separate subject, I am encouraged the BLS interviews households instead of businesses in determining if there is an unemployment problem. imagine what the statistics would look like if instead they interviewing business to find out if they had enough qualified applicants to meet their needs. . .

  4. Wouldn't that be frighteningly inaccurate! Thanks Steve.

  5. To be honest, what bugs me most about how the U-# is read isn't in how the Government handles it, but how those who just find out a bit about the situation handle it.

    You deal with it well to basically note that the numbers are VERY easy to manipulate at will. Others, though, will do the very thing you suggest, such as:

    "OMG! The 'REAL' Unemployment is 23%! The government doesn't want you to know that we're close to Depression era unemployment!"

    Aka, they use the U-6 number of today and compare it to the U-3 number of the depression. It gets tiring to reply to such posts saying that, under the same 'REAL' Unemployment statistic, the Depression era sat at around 50%.

    With just a few exceptions, the U-#s pretty much go in lock step to each other. Good U-3 numbers are around 3% while good U-6 is around 7%. Honestly, you could use any of the numbers to judge the economy the same way so long as you STICK to the same statistic. Since the Government always sticks to U-3, they work out much better than folks who try to scare you into saying "we're facing depression era unemployment!" by switching stats as they see fit.

    To put it another way, if my clock is 40 minutes ahead, I can STILL use it to get to work on long as it STAYS exactly 40 minutes ahead every day.