To Be Unemployed, or Not To Be Unemployed; That Is The Question

Speaking of unemployment (and we are all speaking of unemployment these days, are we not?), as you might suppose, it is not an easy thing to measure. Hence the resulting numbers must always be interpreted with a minute particle of sodium chloride. That, however, does not mean they should be ignored.

Every month the BLS, (for Bureau of Labor Statistics, not Bull…t, as one might suspect) polls 50,000 households and asks them a series of questions. Based on their answers, everyone in each of the households is placed into one of three categories: Employed, Unemployed, or Not in the Labor Force.

Into the first category (employed) goes everyone in the household who the respondent says has any form of paid employment. It does not matter if the job is only part-time, even an hour a week shoveling elephant droppings at the local zoo for minimum wage. Paid job, you’re in, no paid job, you’re out.

There are those who believe that virtually all part-time employees want to be full time employees, and hence should be counted as only a proportional fraction of a fully employed worker. That would, of course, raise the measured unemployment rate. The BLS does not make this adjustment. Many people prefer part-time employment, and we don’t really know what full time employment is anyhow. Is it an eight-hour day? If so we would have to count people who work more than that as more than one employed person. Gets too messy.

Into the second category (unemployed) goes everyone polled who is not employed but wants a job enough to have spent time recently in active job search; calling employers, putting in applications and such. The BLS 2 specifies a specific time frame for this, such as within the last ten days, two weeks, or whatever. No applications within that time frame and you go in the third category, even if you say you want a job.

The third category, Not In The Labor Force, is essential. It includes not just the so-called “discouraged workers” who flunk the job search question, but others who are not working simply because they are either unable or don’t want to work. Children and underage minors, cripples, stay at home moms, idle rich adults, and beach bums qualify. Some people not working should not be counted as unemployed.

Once the BLS has the numbers in each category for its household sample, it simply scales them up for the estimated total adult, civilian, non- institutional population (i.e. the adult population minus those in the military, or in prison, or in insane asylums, but with the insane criminals running the government itself somehow left in).

Then it computes the unemployment rate as follows:
where E = the estimated number of Americans employed,
U = the estimated number unemployed,
and F = E + U constitutes the Labor Force.

Then the percentage unemployment rate is W = U/F X 100 [or U/(E+U) X 100).

Note carefully that the measured unemployment rate is not the percent of the population unemployed. That is an elementary confusion. It is the estimated percent of the labor force that is unemployed.

Despite the seeming mathematical precision of this statistic, there are a lot of subjective elements in its construction. Some people argue that it persistently understates actual unemployment. They think part-time workers should be counted as partly unemployed, as discussed above. Plus, they argue that many actual long-term unemployed persons who become discouraged about finding a job, stop looking, and aren’t counted in the labor force, really should be included. Both of those adjustments would raise measured unemployment.

People on both the left and right use that argument for their own 3 purposes. When it was announced recently to the joy of Obama supporters that the unemployment rate had declined slightly, analysts on Fox News began a chorus that it had only declined because more workers had become discouraged and dropped out of the labor force. They might be right.

The problem with this popular argument, whomever uses it, is that it ignores factors working in the other direction. Consider: to be counted as unemployed a nonworking person only needs to say to the BLS pollster that he/she is looking for work. Many people have incentives to falsely report job search, or make insincere efforts (say to keep receiving unemployment compensation payments). That behavior artificially inflates the measured unemployment rate.

In addition, there is a significant amount of employment that is deliberately not reported. Some of it is criminal employment (“Yes, Mr. Pollster, I have a job; I import China White from the iron triangle and distribute it across three states. . .”). Some is simple off-the-books work, in which the employees and the employers agree on cash compensation that allows them to split the withholding and other taxes both would otherwise have to pay. The incentive to do this is rises with tax rates, and also during recessions.

Such unofficial employment and the output it generates is known as the underground economy. Though we cannot know its magnitude, some researchers estimate it at around ten percent of GDP.

The reason the BLS does not adjust for either set of factors, then, is simply that it does not know how they net out. Remember that the next time you hear a network talking head claim the unemployment rate statistic understates actual unemployment. Maybe he ought to be unemployed.