The employment picture in May was surprisingly weak, with employment outside the agriculture sector growing only 138,000. April employment growth was also revised lower, adding to the general malaise in the report.
The deceleration in hiring was broad-based, as employment in the goods, services and government sector all slowed. All major sectors of the economy showed slower employment growth, except for construction and temporary help services.
{mosads}So what’s going on, and should we be worried? On one hand, an increase in payroll employment of 138,000 is more than enough to keep overall labor market conditions improving and the unemployment rate moving lower. In other words, we’re still moving in the right direction, but the rate of improvement is just slower. That’s not all bad.
On the other hand, and this is the important part, history says slowing employment growth is often a sign for worry. Labor markets, in the eyes of economists like myself who follow the ebb and flow of the data quite closely, are “information rich”; they tell us a lot about the overall health of the economy in real time, and there is much more signal than noise in labor market data.
How good are labor markets as a signal about the health of the economy? Since 1960, a slowdown in employment growth has preceded every recession. Note, I didn’t say “coincided with”, but “preceded”.
One simple way to see this is to compare the current pace of employment growth against the average during any particular expansion. Across multiple business cycles, we find that employment growth gradually increases following any recession and then largely moves sideways for the duration of the expansion.
Yes, payroll growth can be volatile from month-to-month, but the change in employment tends to fluctuate around a fairly stable trend during expansions. This pattern is apparent during the long expansions of the 1960s, 1980s, and 1990s and the current recovery.
When employment growth starts to slow below the recovery-level average, watch out. Historically, employment growth drops below the average change in employment during the expansion about a year before the onset of the next recession. The average gain in employment has been about 185,000 per month since 2010. Viewed through this lens, the slowing to 138,000 in May could be legitimate cause for concern.
We think not. Instead, we think you should blame it on college kids. Ok, not really. But we believe there is sufficient evidence to suggest that the timing of the survey week meant that college-age workers, who return to the job market in large numbers beginning in May, were undercounted.
The survey week for the employment report is the week including the 12th of the month. Hence, the May employment report is really a picture of how mid-May hiring compared to mid-April. But the days of the month vary, and when the survey week for the May employment report falls in the second week of the month, like it did this year, reported employment growth tends to be suppressed since fewer numbers of college-age workers are captured.
However, when the survey week for the May employment report falls in the third week of the month, reported payroll growth tends to be stronger since the extra week means the survey has a greater likelihood of capturing their return.
We examined private employment growth in the month of May since 1980 and, excluding recessions, found that average growth in private payrolls was 110,000 when the survey week falls in the second week of the month (there were 15 such occasions). Of the 18 times the survey week fell in the third week of the month, private employment growth averaged 176,000, for an average difference of 66,000.
If calendar day effects and the timing of the survey week are a factor behind the weaker-than-expected May employment report, and we believe they are, then we are inclined to downplay any negative signal about heightened recessionary risks. Employment should rebound next month, and the economy should remain on its modest growth path.
We believe the Fed will see things the same way and look for it to raise the target range for the federal funds rate 25 basis points in June and proceed to balance sheet runoff in September.
Michael Gapen is the chief U.S. economist for Barclays.
The views expressed by contributors are their own and not the views of The Hill.