An Analysis Of Government Statistics Part 2: The Unemployment Rate

by: Ben Kramer-Miller

This is the second in a series of articles in which I analyze key pieces of economic data produced by the Bureau of Labor Statistics (BLS) and the Bureau of Economic Analysis (BEA), which are both part of the U.S. government. I am specifically interested in their methodologies in determining these statistics. There is no single way of calculating figures such as the unemployment rate or the size of the economy, yet they are presented to us in the media as official and incontestable.

In Part 1 I argued that government statistics are prone to bias for two reasons. The first is that the government isn't driven by a profit motive in doing its research, and so there is no "fear-of-failure" insurance that comes with a private for profit company. If a private company comes up with erroneous data it will lose customers and go out of business, and it thereby has incentive to be as accurate as possible. The government has no such incentive.

The second is that the government has an interest in maintaining order so that it can retain its power. If people believe that the economy is generally "good" then they will continue to re-elect the politicians in power, and they will be less likely to revolt. Thus the government has incentive to make the numbers appear to be better than they really are. This bias doesn't necessarily manifest itself in outright lying, but it can lead to methodological choices that favor particular outcomes.

As investors we should have an idea as to what these numbers represent and go on from there to decide if they are useful in making decisions with our money.

In this article I focus on the unemployment rate. The unemployment rate, probably more so than any other piece of government data, has a meaningful impact on mass psychology. Markets react sharply when this data is released on the first Friday of each month at 8:30 Eastern Time. The president often makes a statement in response to the number later in the day. Furthermore, the results usually make the front page of the business sections of major newspapers the following morning.

The unemployment rate is supposed to tell us how "the man on the street" is doing economically. If more people are employed then "Main Street" is doing well, and we are supposedly more productive as a nation.

Ironically a lot of the people who are employed would rather be unemployed -- they work because they need the money, and if they didn't need the money they wouldn't work, they would work less, or they would do something else. Were we to imagine a utopian economy in which everybody was rich and we were serviced by self-maintaining robots, the unemployment rate, as calculated by the BLS, would probably be much higher than it currently is.

Thus, as an economic indicator, the unemployment rate should ideally calculate what percentage of people doesn't have work that needs it to survive. But even with this criterion there are a lot of grey zones. First, this lack of work may not mean no work -- a family of six living in Manhattan may have two employed parents, but if they make minimum wage then from the standpoint of meeting their income needs they might as well be "unemployed." Meanwhile a single individual who lives in a rural area might be able to survive quite comfortably on the same salary. Furthermore the 16 year old child of a wealthy family might enter the work-force with a minimum wage job in order to earn himself some extra spending money, whereas he may be working alongside someone his age -- say a child from the aforementioned six member family -- who needs the money to help his family survive.

The point is that employment-status might be an "either/or" categorization from a legal standpoint, but it certainly is not from an economic standpoint.

U3 Vs. U6

The BLS does something to the effect of making this distinction insofar as it calculates the unemployment rate in many different ways. The number that we hear in the news and that is cited by economists and politicians in the mainstream media is the U3 number, defined as follows:

U-3 Total unemployed, as a percent of the civilian labor force (official unemployment rate)

The U3 figure makes no distinction between the various scenarios listed above, and so long as everybody is working at least 15 hours per week they are all employed. Furthermore, the fact that all of these people are employed is supposed to signify strength of the U. S. economy when in fact the employment status of the two parents of six that are making minimum wage have jobs that perhaps indicate the contrary. This point becomes even more acute if we assume that one or both of them has an education beyond high school and the qualifications to earn more money by doing something more productive than flipping burgers at a fast food restaurant.

The U6 figure does a better job in discriminating between the different ways that people can be employed:

U-6 Total unemployed, plus all persons marginally attached to the labor force, plus total employed part time for economic reasons, as a percent of the civilian labor force plus all persons marginally attached to the labor force

If any of the government's figures should be used by those wishing to draw a conclusion about the state of the economy it should be the U6 figure, although even this has issues as I explain below. The U6 figure will make the following distinction. Suppose that one of the parents in the aforementioned six member family was employed in the banking industry before the 2008 financial crisis, and he now works at McDonald's (NYSE:MCD) making minimum wage. U6 will count this person as unemployed whereas U3 will not.

The distinction is made by introducing a category: marginal attachment to the labor force to a broader set of categories. The first is the civilian labor force, which includes those people who have a job or who are actively looking for a job (i.e. they are unemployed). U3 takes all of the people who are looking for a job and divides it by the number of people in the civilian labor force. Our educated burger flipper would be counted in the civilian labor force and as an employed person. But as we have established this person's employment status is different than it was when he worked for a bank. So the BLS calculates U6, and it says he is marginally attached to the labor force, meaning that instead of just being counted in the denominator of the equation "unemployed persons/civilian labor force," he is now also counted in the numerator.

This is where the U6 figure differs from U3. As it stands as of the end of November the U3 rate was 6.6%, whereas the U6 rate was 12.7%, or nearly twice the official number. One thing to note that is not so apparent given the media's emphasis on the U3 figure is that the difference between U3 and U6 has been steadily growing. In 2003 the economic downturn caused the U3 figure to rise to the current level, but the U6 figure was only about 10%. This points to economic weakness that those who just follow the U3 figure would overlook.

Disgruntled Workers and the Shrinking Labor Force

The BLS unemployment data fails to account for what are often referred to as "disgruntled workers," or those workers who have given up looking for work. They drop out of both the numerator and denominator of the unemployment ration which brings it down. This makes the number look better when the economic outcome is clearly negative.

Let's look at an example.

Suppose there is a family with two working professionals as parents who work while raising some kids. Their salaries enable them to live a particular lifestyle. One day the father is laid off due to outsourcing or so that the company can hire somebody younger and pay him less. The family can still survive on just the mother's salary so the father begins looking for work, but he feels it is not worth it to take a job at Wal-Mart (NYSE:WMT) or in fast food, and so he is counted as unemployed by both the U3 and U6 measurements. But after two years his unemployment benefits run out. He has still not found work being in his 50s and so he simply gives up to retire early. Suddenly he is dropped from the data: he is not unemployed, nor is he a part of the labor force. His economic situation, along with his family's clearly worsened, and the American economy is less productive without his contribution, and yet the way the BLS calculates the unemployment rate the data looks better!

This is an ongoing phenomenon that has been exacerbated by the fact that many Baby Boomers have been forced into early retirement because they have lost their well-paying jobs and have simply given up trying to find another one. Some have enough savings and perhaps a small pension on which to live modestly but their lifestyles have taken a hit.

Both the U3 and U6 figures fail to account for this. What we can do, however, is attempt to correct for the decline in the labor participation rate, which measures the percentage of people who comprise the labor force relative to the general population. A chart of this data is highly revealing.

(Click to enlarge)

Let us analyze the implications of this. As the labor participation rate falls so does the denominator in the unemployment rate ratio. Thus in order to get a more accurate account of the unemployment rate in the U.S we might want to change our general equation to the following: the number of unemployed persons divided by the number of people in the labor force divided by the labor participation rate. As the labor participation rate falls the "true" unemployment rate rises. It is for this reason that analysts who account for the decline in the labor force have actually reported an increase in the unemployment rate over the past few years. Private, for profit economist John Williams of Shadow Government Statistics has done just this. He provides a very useful chart with U3 data, U6 data, and U6 data that is corrected for the decline in the labor force.

(Click to enlarge)

The rapid decline in the size of the labor force is clearly illustrated in the growing disparity between the U6 figure and Williams' alternative.


In calculating the unemployment rate the BLS is forced to make a lot of choices in determining what constitutes employment. We can turn a blind eye to the decision that the U3 number is "official" -- writing it off as mere propagandizing; and we can even laud the BLS for admitting that using one measure of the unemployment rate we get a number -- 12.7% -- that is rather alarming.

However the decision to omit "disgruntled workers," which is actually quite recent (1994) should raise a red flag as it appears to be little more than an attempt to lower the unemployment number and to make the economy look better. There is no apparent reason in my mind not to include them, and there certainly is reason to include them as the above example demonstrates.

Whether my (and Williams') concern that so-called disgruntled workers are omitted from BLS data is valid is a debatable point, but only from a purely academic standpoint. From a practical standpoint we can clearly see how this data point makes the economy look as though it improved when in fact something happened in the economy to make it less productive.

I believe that at this point I am justified in stating that we are now beginning to see a pattern. As we saw with the CPI in Part 1, it is evident that government choices are leading to more optimistic results that make the economy look better. This pattern, at least in my mind, casts a shadow of a doubt regarding the academic merit of BLS statistics and the motivation behind several of this organizations methodological decisions.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

About this article:

Author payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500. Become a contributor »
Problem with this article? Please tell us. Disagree with this article? .