Phantom jobs and job losses

Public Interest, Wntr, 2005 by Tim Kane

The problem with the critique offered by Krugman, the EPI, and others is that it is not supported empirically. The BLS itself has been collecting data on the number of discouraged workers since 1994, and even includes this group in calculating its alternative "underemployment rate." This statistic, known as "U-4," has followed the same trend as the traditional unemployment rate, suggesting that an increase in underemployment has not been fueling the decrease in full-blown unemployment. On the contrary, BLS data show that there were fewer discouraged workers as a percentage of the labor force in 2004 than in 1994. The labor force (the denominator in the unemployment rate) grew by 200,000 the same month that the EPI issued its press release, having grown by 1.2 million over the year leading up to EPI's announcement. Between the 2000 and 2004 elections the labor force grew by four million people. Given that the denominator was increasing at this pace, it is a simple mathematical fact that the only way the unemployment rate could have stayed as low as it did

through the recession and the aftermath of terrorism was if the number of people with jobs grew, too.

The second line of attack on the household survey claims that even if the absolute size of the labor force was increasing, it was not keeping pace with population growth. The labor force participation rate (the percentage of all working-age people who are defined as being in the labor force) declined slightly during the recession, from a peak of 67.2 percent in January 2001 to 66.8 percent in October 2001, and has hovered at 66 percent for most of the past three years.

But again, a closer look at the demographic breakdown of labor force participation statistics reveals a simple explanation: Fewer teenagers are working. The decline in total participation rates since 2001 is largely driven by the unprecedented dropoff in teenagers aged between 16 and 19, from 52 percent in 2000 to 44 percent in 2004. One way to put this in perspective is to look at the gap between the participation rate of all civilians versus the rate for teens. This gap averaged 10.8 percentage points between 1950 and 2000, before climbing to 20.7 points in the months after September 11, 2001.

The decline in teen participation is a surprising puzzle, given that conventional theory holds that the improvement in the overall unemployment rate should lead to a steeper jump in the number of teens willing to work. But, as troubling as it is for theorists, is the decline really bad news on its own? Clearly, the teenagers of 2004 are generally not heads of households, and they have certainly not lost jobs and become discouraged. Rather, these are young adults who chose not to enter the labor force in the first place, or were restricted from doing so by their parents.

Whether or not the dropoff is bad news for teens, it is bad news for proponents of the discouraged worker hypothesis who intend to create doubts about the integrity of the household survey. In the end, the household survey provides clarity on the labor force, clarity on discouraged workers, and clarity on demographic trends that are not available elsewhere.