Author:Shin, Hee-Young
  1. Introduction

    The standard labor force statistics has been widely used as an important indicator of the overall health of an economy, and both economists and economic policy makers in the U.S. have conventionally relied on a narrowly defined single official unemployment rate concept whenever they propose and make an important policy decision. It is well known, however, that the official labor force statistics (i.e., the very definition of employment and unemployment) that most governments in advanced capitalist economies have used is based upon an outdated theoretical framework and is of no use in analyzing a certain quality aspect of employment.

    The goal of this paper is to revisit some of the limitations associated with the conventional labor force statistical framework that has been uncritically used by the US Bureau of Labor Statistics (BLS) and to propose a series of advanced and superior measures for analyzing certain quality aspects of employment into the government's formal labor force statistics. For this purpose, this paper introduces two alternative frameworks to overcome the limitations in the existing standard labor force statistics. The first one is the US BLS's multiple underemployment rate series and the other is the International Labor Office (ILO)'s labor (under-)utilization framework. The labor utilization framework adopted in the ILO's international conferences emphasizes the importance of capturing quality aspects of labor utilization and introduces the concept of income-, time-, and skill-related labor underutilization.

    The paper argues that the ILO's labor utilization framework is broader in scope and much more consistent in analyzing the labor market performance. The framework also provides better insights into analyzing the labor market dynamics than any other formal labor force statistical frameworks, including the US BLS's standard labor force statistics. Even though the ILO's labor utilization framework is in no way a complete solution to the task of analyzing the capitalist labor market dynamics and of constantly improving the government's labor force statistics, it is far more superior to the conventional unemployment rate and any other narrowly defined labor force statistics. As a conclusion, the paper urges the US BLS to seriously consider incorporating labor utilization framework into its formal labor force statistics.

  2. The Official Definitions in the BLS Labor Force Statistics and Their Limitations

    The governments in most advanced capitalist economies have provided a series of statistical information that captures a certain quantity aspect of labor markets. This statistical indicator includes the number of working age population (economically active population), civilian labor force, the number of employed and the number of unemployed, etc.

    In the US, the BLS has provided a variety of statistical indicators of the labor market in order to show how well the labor market fares with the overall macroeconomic activities. In order to do so, the BLS economists and statisticians have relied on the current population survey (CPS) and classified the economically active population and civilian labor force according to their own operational concepts (BLS, 2010; BLS and Census Bureau, 2002).

    According to the BLS, the employed is a person who did any work for pay or profit during the week before the survey is conducted by the Census Bureau or who worked 15 hours or more in a family-owned business with/or without receiving pay. Unemployed person, in turn, is the one who is not currently employed, but who is actively seeking a job, and who is immediately available for work once the job opportunity is given.

    These operational definitions that the BLS relies on have its particular historical origin dating back to 1954. Many labor economists and statisticians at that time agreed upon the need of compiling labor force data in a coherent manner, and they proposed these operational definitions of economically active population, civilian labor force, employment and unemployment that we introduced just before. (See Figure 1)

    It has been widely recognized, however, that the formal definition of employment and unemployment is too narrow and in some cases extremely arbitrary. For example, there are no particular working hour criteria in the definition of the employment. Thus, even if someone worked for a second, (s)he is still classified as the employed person as long as (s)he earned profit or wage income during the survey week. The "15 hours or more" work criteria used in defining employment in the family-owned enterprise is also an arbitrary working hour criteria, since there is no particular reason to choose this "15 hours or more" working hour duration when the BLS classifies someone as the employed.

    The BLS's official definition of unemployment rate, which is the percentage ratio of the number of unemployed people relative to the total civilian labor force, is not an adequate measure to capture dynamic aspects of labor markets, either. In this narrow definition, those who were once actively looking for a job but gave up seeking work--maybe because they believed that the current economic situations do not allow them to obtain a job opportunity --are no longer classified as the unemployed but as not-in-the labor force. Given potentially large numbers of "discouraged" and "marginally attached" workers are forced to give up job searching activities under particular economic circumstances, the official unemployment rate significantly underestimates the actual underemployment of labor or involuntary unemployment situation.

    Furthermore, those who are working part-time--not because they prefer to work on a part-time basis, but because there is no full-time job opportunity for them--are simply classified as the employed. If we account for the marginally attached workers and involuntary part-time workers in the official unemployment statistics, the actual size of workers who are experiencing serious underemployment of labor situation would be significantly higher than the single official unemployment rate suggests.

    The most serious problem, however, is in the fact that the BLS's official employment statistics do not convey any useful information at all about the quality aspect of job and employment. Even if workers are employed (whether part-time or full-time), if their earned labor income (wages and salaries) is not sufficient to support their material well-being, employees will not be able to maintain a reasonably good quality of life. In addition, if the job does not provide any meaningful chance of utilizing their vocational skills or expertise, or if the job requirement enforces the workers to work over an extended hour, their employment situation may not be considered meaningful, even if workers are classified as the employed.

  3. The US BLS Economists' Alternative Series of Underemployment Rates

    For these reasons, many prominent economists and labor statisticians have proposed a series of alternative labor market indicators, and historically there have been two different initiatives in this regard. One is the BLS economists' alternate and multiple series of underemployment rates, and the other is the ILO's labor (under-)utilization framework.

    First, a group of labor economist working at the BLS has developed alternate series of underemployment rates in order to examine and compare what they called "real" labor market situations among advanced industrialized countries. Initially, the BLS researchers began developing "international comparisons of labor markets" projects in 1962. Like those of the ILO's and the OECD's comparable measurements of unemployment rates, the BLS gathered various labor statistics from selected countries (Australia, Canada, Japan, and six European countries) and generated internationally comparable labor force statistics series for these countries1 (Meyers and Chandler, 1962; Sorrentino, 1983; McMahon, 1986; Moy, 1988; Sorrentino, 1993; Fullerton, 1999; and Sorrentino, 2000). While doing so, they developed a series of alternate measures of underemployment rates, ranging from U-1 to U-7. The reason for developing these multiple series of unemployment rates was that no single unemployment rate measurement served various "analytical and ideological purposes" (Shiskin, 1976; Sorrentino, 1993).

    According to this series, the U-1 rate measures the number of persons who are unemployed 13 weeks or longer as a percentage of the civilian labor force. Initially, the BLS economists used "15 weeks or longer" as the main criterion for this U-1 rate. However, they later shortened the duration of being unemployed to "13 weeks or longer" in order to reflect rapidly changing labor market situations. Regardless of this change, the...

To continue reading