In the past, I have shown that most of the improvement in the unemployment rate since its peak of 10 percent in the fall of 2009—and all of the improvement in the unemployment rate over the last year—has not been for good reasons. It has been due to people either dropping out of, or not entering, the labor force due to weak job opportunities.

But what about other measures of labor market slack that the Bureau of Labor Statistics publishes? The most comprehensive direct measure of labor market slack published by the BLS is the U-6 measure of labor underutilization, the so-called “underemployment” rate. Like the unemployment rate, the underemployment rate has declined substantially in this recovery, from a peak of 17.1 percent in the fall of 2009 to its current rate of 13.6 percent. While still very elevated—the U-6 averaged 8.3 percent in 2007—that is significant improvement. Is it a sign of major healing?

Yet again, the answer is no. The U-6 includes three groups: the officially unemployed, people working part-time but who want and are available for full-time work, and the “marginally attached”—jobless workers who are available to work and have looked for work in the last year but have given up actively seeking work. By including the marginally attached, the U-6 measure captures some of the “missing workers”—workers who are neither employed nor seeking work but who would be in the labor market if it were healthy. However, the U-6 is capturing a smaller and smaller share of the total number of missing workers as the weak recovery drags on. One reason is that to be defined as marginally attached, a jobless worker must have looked for work in the last year. After more than 5 years of extremely elevated unemployment, there are more and more jobless workers who want to work but nevertheless haven’t actively searched for a job in over a year. Right now there are 2.3 million marginally attached workers, but I estimate that there are more than 5 million missing workers.

This figure shows the official U-6 over time, along with what the U-6 would be if instead of the including the marginally attached, it included the missing workers. Since mid-2010, the “Alt U-6” shows much less improvement than the official U-6. In other words, much of the improvement in the official U-6 in this recovery, and essentially all of the improvement in the official U-6 so far this year, is due to people dropping out of, or not entering, the labor force because job opportunities are so weak.