The Bureau of Labor Statistics (BLS) announced on Friday, October 3, 2014 that the economy added 248,000 non-farm jobs and that the unemployment rate had fallen to 5.9%.  Professional analysts and the media however disagreed on the information content of these announcements, finding good and bad news by parsing the results. One negative observation cited was the decline in the participation rates of the civilian labor force to 63.2%, the lowest since 1978. A positive observation cited was the decline in the number of discouraged workers, defined as individuals who have given up looking for jobs, from 852,000 to 698,000.

Interpreting survey results can be tricky especially when the results are mixed and inconsistent. For the above two citations, the interpretation can change by the simple process of extending the timeline. For example, if we expand the results of the labor force participation rates all the way to 1955, as shown below, a different picture emerges. Labor participation rates show a gradual growth from the 1950s to the late 1990s (peaking in 1998) followed by a slow descent to 2014.

Civilian labor force participation rate (in percent). Source: Bureau of Labor Statistics.September Data.























Viewed from this perspective, the recent decline does not appear as alarming.

Similarly, if we examine the number of discouraged workers from 1994 (earliest data available from the BLS), the decline to 698,000 last month should not be encouraging because the average numbers prior to the financial crisis of 2008 was approximately 300,000, as shown below.

Discouraged workers (in 000s). Source: Bureau of Labor Statistics. September Data.
























Interpreting the monthly announcements is also compounded by the trust factor; are these results reliable? Fortunately, the survey methodology employed by the BLS is rigorous and consistent, a fact not widely reported in the financial press. Below is a brief outline of the methodology employed by the BLS.

BLS Survey Methodology

The BLS selects about 60,000 households or 110,000 individuals across the country from 2,025 geographic units covering a variety of rural, urban, industrial and residential areas. Each month, twenty five percent of the households are dropped. Each household is interviewed for four months, then left alone for eight months, and interviewed again for eight months before being dropped forever. This ensures that 75% of households remain in the sample month to month and 50% year to year. In contrast, many of the polls conducted by the media, including CNN, Fox, ABC, NBC, and the Wall Street Journal survey between 500 and 2000 individuals.

In addition, approximately 2,200 specially trained BLS workers interview the households, initially in person and then by telephone. The important thing to note is that the respondent is not giving the opportunity to determine his or her status of employment. Instead, the responses to a specially designed set of questions determines their status; even the interviewers are not directly involved in the determination of the status. The results are categorized into three status:

  1. Employed
  2. Unemployed
  3. Not in the Labor force

A few examples of how a person’s status is determined are listed below, taken from the BLS site:

  1. Mrs. Jenkins tells the interviewer that her daughter, Katherine Marie, was thinking about looking for work in the prior 4 weeks but knows of no specific efforts she has made. Katherine Marie does not meet the activity test for employment and will be classified as not in the labor force.
  2. John Stetson has been checking for openings at a local superstore for each of the past 3 weeks, but his wife reported that last week he had the flu and was unable to seek work. John will be classified as unemployed because he took steps to look for work and would have been available for work during the survey reference week, except for his temporary illness.
  3. Joan Howard told the interviewer that she has filed applications with three companies for summer jobs. However, it is
    only April and she doesn’t wish to start work until at least June 15, because she is attending school. Although she has taken specific steps to find a job, Joan is classified as not in the labor force because she is not currently available for work. Students are treated the same as other persons; that is, they are classified as employed or unemployed if             they meet the criteria, whether they are in school on a full- or part-time basis.

Among those “not in the labor force,” the following additional questions determine whether the worker is “marginally attached to the labor force.”

  1. Do you currently want a job, either full or part time?
  2. What is the main reason you were not looking for work during the last four weeks?
  3. Did you look for work at any time during the last 12 months?
  4. Last week, could you have started a job if one had been offered?

Finally, these responses determine whether an individual is a “discouraged worker.”

  1. They believe no job is available to them in their line of work or area.
  2. They had previously been unable to find work.
  3. They lack the necessary schooling, training, skills, or experience.
  4. Employers think they are too young or too old, or they face some other type of discrimination.

Several features of the methodology employed by the BLS provide confidence on its accuracy and reliability.

  1. The sample size is large and fully representative of the population.  With 60,000 households in the sample, even if whole households were dishonest in their replies (and assuming they all lied in one direction), the overall numbers are unlikely to impact the final numbers.
  2. The methodology is rigorous and consistent, perfected by the BLS over several decades, and instead of asking direct questions, the responses to a range of questions determine the status of employment.
  3. The methodology makes it very difficult for any individual interviewer from biasing the results.

Given the rigorous methodology, economists can safely use the monthly announcements as a factor in evaluating the current performance of the economy and to make forecasts.

The gradual drop in the unemployment rate over the last 24 months provides reliable evidence of households migrating from the unemployed to the employed. Although the civilian labor participation rate continues to fall, it has yet to reach its historical lows. The rates can be expected to fall further as baby boomers continue to retire.  However, it suggests that there is still time to initiate long-term policies to reverse the trend. Suggestions have included providing incentives for late retirements and reducing the need for two-wage earners to afford large families.

As for discouraged workers, a recent St. Louis Fed study concluded that discouraged workers rarely stayed in that category for long.  Approximately, 90% of the discouraged workers are new entrants and approximately 50% of them move within a month to the status of employed or seeking employment.  Since the BLS methodology ensures 25% are newcomers to the pool, the results suggest that a large number of discouraged workers reported last month are already employed or seeking employment.

An additional measure that could be of use is a broader estimate of the underemployment rate. Currently, the BLS produces a 4-quarter moving average measure termed involuntary part-time employment, defined as those working less than 35 hours but are willing to work full-time. A broader measure of underemployment would also include workers willing to work in higher skill jobs. The International Labor Organization calls it skill-related inadequate employment. An exaggerated example would be a scientist with a Ph.D. working in as a busboy in a restaurant. Currently, there is less consensus on whether the underemployment is due to economic slack or lack of good paying jobs.

Although designing questions to capture skill-related underemployment will be challenging, as benchmarks for different levels of skill sets have to pre-determined, a reliable underemployment measure is necessary for good policy decisions and should be developed by the BLS.