The Washington PostDemocracy Dies in Darkness

Full employment: a potent antidote to racial gaps in jobs and wages

Source: Valerie Wilson, “The impact of full employment on African-American employment and wages.”

While last week’s jobs report came in well below expectations, the underlying trend in labor market improvement remains pretty solid, and one bad month does not a new trend make. The overall job market is still on a tightening path, but we’re not yet at full employment, and last Friday’s underwhelming numbers serve as an important reminder that the path remains bumpy.

A key word in this analysis is “overall.” There is great and persistent variation in job market indicators, particularly by race. While March’s headline unemployment rate clocked in at 5.5 percent, the unemployment rate for white workers was 4.7 percent while the rate for black workers was more than twice the white rate, at 10.1 percent.

Moreover, as the figure below shows, that difference has persisted for as long as we’ve been tracking the data. The figure plots black and white unemployment along with their ratio, given on the right axis. During this 43-year period, the black rate has averaged 2.2 times the white rate, with a standard deviation of 0.2, meaning relatively little variation around that mean.

What explains this persistent difference?

Most people’s first guess is education. Since more highly educated workers have lower unemployment rates, the black-white unemployment rate gap is often assumed to be a function of differences in educational attainment that favor whites.

But, in fact, educational attainment, while obviously important in terms of jobs and earnings, actually explains very little of the gap. First of all, as the next figure shows, we see that same gap in black-to-white unemployment rates even within education groups. The figure, for those 25-years-old and over (that’s how the BLS provides these data), shows that even while education and unemployment are negatively correlated for each group, even college-educated blacks have a jobless rate that’s almost twice that of whites (4.1 percent versus 2.2 percent).

Another way to make this point is to construct a simple thought experiment in which we set educational attainment levels for black workers equal to those of whites and recalculate the black unemployment rate (again, for those 25 and up). If blacks had white education levels, their overall unemployment rate would be 8.1 percent instead of 8.6 percent. Thus, by this admittedly simple exercise, higher educational attainment for whites explains only 10 percent of the difference between racial jobless rates.

Unquestionably, facilitating college access and completion for black students is a worthwhile endeavor, but it may not have nearly enough of an impact on the racial unemployment gap. There’s lots of evidence that labor market discrimination is alive and well. Also, relative to whites, blacks have less mobility out of poverty, less wealth, and fewer networks.

There is, however, one important labor market dynamic that disproportionately helps African-Americans and other minorities: full employment.

As economist Valerie Wilson shows in an important new paper, a change in the aggregate unemployment rate of 1 percentage point between 1979 and 2014 was associated with a change in the black unemployment rate of 1.7 percentage points, compared to a change in the White unemployment rate of 0.9 percentage points. That finding is consistent with the change over the past year: while the white rate has fallen 1 percentage point since March of 2014, the black rate, which stood at 12.2 percent a year ago, has declined by 2.1 percentage points.

In what I found to be her most important finding — one I hope policy makers pay a lot of attention to — Wilson finds that these unemployment differences play out in real wages as well. She shows that for a given increase in the unemployment rate, real median wages for black workers fall more than twice as much as those of white workers.

The figure above this post, from her paper, indicates that a doubling of the unemployment rate would be expected to cut real hourly wages by 8 percent for black workers but only by 3 percent for white workers. Excluding the recent recession and subsequent recovery from the data suggests an even larger impact on black median wages of 10 percent. Changes in aggregate unemployment have disproportionate impacts on both the quantity of jobs available to Black workers and to the quality of wages provided in those jobs.

Last month’s jobs report showed that if our destination is, as it must be, truly full employment, we’re not there yet. As Wilson stressed in a recent talk on her new paper, getting “there” takes longer for minorities than for whites. A brief look at the benefits of doing so, especially regarding wages, underscores just how important this trip really is.