The Measurement of Unemployment

The unemployment figures published by the Bureau of Labor Statistics are often under attack for understating real unemployment. Why? People know that the BLS uses Unemployment Insurance (UI) data to determine who is unemployed. Thus people who have filed for UI or are receiving unemployment insurance payments are the unemployed. But what about people who never file or whose insurance has run out? Well the answer is that the BLS does use UI data—but it is not the only source of its information. The current figure, putting the unemployment rate at 9.2 percent in June 2011 comes for the hard counts (the UI databases maintained in every state) and from the soft counts derived from the Current Population Survey (CPS).

The CPS is conducted monthly and is based on a sample of around 60,000 households. These represent, using BLS’ own numbers, about 110,000 individuals—which contrasts favorably with the usual opinion poll sample of around 2,000 individuals. Like poll samples the 60,000 households are also selected very carefully to be representative of the nation as a whole. Each month a quarter of the sample is changed to keep it varied and representative. Thus, we might say, the CPS is a “super poll,” much larger than the usual, and therefore, from a statistical viewpoint, a pretty good indicator. Surveying every household, monthly—as these are surveyed one every 10 years in the census years ending in zero—would cost too much.

Questionnaires are built carefully to determine if the individuals interviewed meet the technical definition of “unemployed”—but without ever asking the person if he or she is unemployed. This is done so that arbitrary definitions of unemployment, whether in the minds of the person questioned or the interviewer, distort results. People interviewed are then, later, classified by status using their answer to questions.

The technical definition of someone unemployed, put briefly, is a person who is jobless, is looking for work, and is available to work. The interviews are also structured in such a way that individuals who are jobless and available for work but have stopped looking for work can also be identified and counted. These are the shadow-unemployed, those who’ve given up looking. The numbers we sometimes see for this cohort also come from the CPS and also have an objective basis based on a structured sample.

The CPS data, being a sample, can, however, be extrapolated to the population as a whole—using the same methods pollsters use to tell us what we, Americans, thing based on talking to 2,000 people. The 9.2 percent number from the BLS, therefore is a pretty good number. It approximates the actual number of people who are looking for work and available. That number, added to those countably employed, also produces the estimate of the civilian labor force.

The CPS actually produces six different categories on unemployed persons, labeled U-1 through U-6. Here they are:

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

Data I found for May 2008 at the same place where these definitions appear (link) show the following unemployment rates:

This shows that the maximum unemployed rate may be nearly double the official rate. These data are regularly published but not publicized. The government is in conflict with itself. It wants to project optimism but does not wish to publicize bad news in too loud a voice. Soon I will show what these numbers look like over time.