Unemployed but not Counted
The official unemployment rate of 7.9% is about the same as it was in January 2009, and a broader measure of joblessness that also includes recent labor force drop-outs has only risen from 8.8% to 9.3%. These figures give the appearance that the number of jobless workers today is about the same as it was 46 months ago. This misconception is largely the result of the government’s official definitions of unemployment and labor underutilization. Specifically, the Bureau of Labor Statistics (BLS) can and should adapt its measures of underutilized (marginally attached or discouraged) workers to reflect the current economic reality.
To understand the problem, first consider how the BLS defines unemployment and underutilization. The BLS official definition of unemployment requires jobless workers to be currently looking for work to be counted as unemployed. This means that once jobless workers give up their job search, they are not considered unemployed or counted as labor force participants. In its measures of labor underutilization, the BLS only attempts to count labor force drop-outs who stopped looking for work within the past year. The problem is that because of the severity of the economic downturn and the relative weakness of the recovery, many jobless workers are not considered unemployed or underutilized because they gave up looking for work more than one year ago. Although many frustrated jobless workers gave up their job search and dropped out of the labor force prior to October 2011, none are included in official labor underutilization statistics for October 2012.
The BLS reported 2.4 million people were “marginally attached to the labor force” in October because they weren’t currently looking for a job but: (1) wanted a job and were available to work and (2) either worked, or looked for work, within the past twelve months. A subset of these workers who dropped out of the labor force because “they believe there are no jobs available or there are none for which they would qualify” are counted as “discouraged workers.” Adding the 2.4 million marginally attached workers to the 12.3 million officially unemployed workers means there were 14.7 million jobless workers in October who were either looking for work or recently gave up their job search. Using this broader measure the “U-5 unemployment rate” was 9.3% last month.
But even this broad measure of labor underutilization undercounts jobless workers. The number of people “marginally attached to the labor force” has fluctuated narrowly between 2.0 and 2.8 million people since January 2009. There has been no significant trend in these data because jobless workers stop being counted as “marginally attached” if they don’t resume their job search within twelve months. On the other hand, the labor force participation rate has fallen by 2.3 percentage points in the last five years, the equivalent of 5.6 million adults leaving the labor force. Part of this decline is due to the aging of the workforce. The majority of the decline is due to the weakness of the economy, however. Even among men age 30 to 54 (and too young to retire) the participation rate has fallen by 2.0 percentage points in the past five years as jobless men stopped looking for work.
I estimate that there are over four million fewer labor force participants than what would have occurred if age-adjusted participation rates maintained their pre-recession trend. In this recovery, the official BLS count of “marginally attached” workers underestimates, by 40%, the number of people who left the labor force because they stopped looking for work. Although BLS figures suggest that marginally attached workers are a minority of the 5.6 million adults who left the labor force, a more plausible estimate is that 72% of these non-participants stopped searching for work in the past few years.
If official underutilization measures included jobless workers who gave up searching for work within the past 36 months the labor force underutilization rates reported by the BLS would be higher by about 0.9%. For example, the underutilization measure that includes unemployed and marginally attached workers would have been 10.2% (rather than 9.3%) last month.
The BLS could change its methodology and improve its measures of labor underutilization. They have made other adjustments because of the severity of the most recent economic downturn. Prior to January 2011 unemployment durations were measured up to two years in length but are now reported up to five years in length. A jobless worker who reports that she is still looking for work will be counted as unemployed regardless of when she last worked. In contrast, an individual who gave up looking for work more than one year ago is merely counted as “out of the labor force” and treated no differently than retirees or full-time students even if she worked more recently than many unemployed workers.
Between one and two million jobless workers who gave up their job search after twelve months of frustration are no longer counted in official figures. The BLS should use a more accurate measure of labor underutilization that reflects the economic reality that many jobless workers stopped looking for work more than one year ago but will quickly return to the labor force when the economy improves.
Stephen Bronars, a Senior Economist at Welch Consulting ia an Adjunct Professor of Economics at Georgetown. He has a Ph.D. in Economics from the University of Chicago,