Payrolls rose 160,000 last month, less than the 200,000 we’ve come to expect and the smallest monthly gain since last September. Revisions to the prior two months data shaved 19,000 off of their previously reported gains.
However, though this slower pace could represent a downshift in the rate of job creation, it is far too soon to jump to that conclusion. These monthly numbers are jumpy and require averaging a few months’ gains to get at the underlying trend. In fact, the monthly trend over the past three months is precisely 200,000, as shown in my monthly smoother below. So, even while one can point to other slowing indicators, especially the 0.5% GDP growth in the first quarter of the year, do not assume the job market is softening.
The rest of the report provides indicators that bounce both ways. On the soft side, the slipping of the labor force participation rate was a real disappointment and a reversal of a recent upward trend in this closely watched metric of movements in and out of the labor force (see figure). After rising from a low of 62.4% last September to 63% in April, the LFPR ticked back down in May to 62.8%. That’s still significantly off its lows, and again, the monthly numbers are jumpy, but this was the number I liked least in today’s report, especially since the same 0.2 percentage point decline was seen among prime-age workers, meaning the drop can’t be pinned on aging retirees.
On the other hand, both average hourly wages and weekly earnings continue to beat (very low) inflation (weekly hours ticked up slightly last month), with both earnings measures up 2.5% over the past year, while inflation’s running around 1%. Importantly, from the Fed’s perspective, even while the job market is tightening, the extent of wage acceleration remains mild. Essentially, average wages from the establishment survey have climbed from around 2% growth in the first half of last year to around 2.5% this year. That’s what we’d expect in an improving job market and a pace that remains below the 3-3.5% Fed chair Yellen has suggested is “equilibrium” wage growth.
The smoother shows 3-, 6-, and 12-month averages of monthly gains. As noted, the recent trend is 200K, a slight deceleration from the longer term trend, but still a solid number that will move the job market towards full employment if it persists.
Underemployment, which includes about six million part-time workers who’d rather work full-time (and are thus under-employed), ticked down slightly but remains elevated at 9.7%. I’ve argued that an underemployment rate about a percentage point lower than this is consistent with full-employment, meaning there’s still slack left to be squeezed out of the job market.
The goods-producing side of the job market was notably weak last month. Even as oil prices have come up a bit, extraction industries continue to shed jobs, and manufacturing remains weak. Factories added 4,000 jobs last month, an improvement over losses of 45,000 jobs over the prior two months. But the strong dollar continues to weigh on the international competitiveness of the sector.
As noted, GDP rose only 0.5% and productivity fell 1% in the first quarter of this year, obviously weak indicators. Now we can add a jobs report that’s off its recent pace. Is the U.S. economy, just about to hit year seven of an expansion that begin in mid-2009, heading toward recession?
That’s unknowable, but I would strongly avoid reading too much into these indicators. On a year-over-year basis, GDP is up about 2%, which is about its trend in the recovery. Similarly, I’d definitely discount the negative handle on the Q1 number—these data are noisy and I don’t think we’re forgetting how to make stuff. Yr/yr, productivity is up 0.6%, which is too slow but also right at its recent trend. And, as I’ve shown, underlying job growth at 200K is also at trend.
In other words, filtering out some statistical noise, we’re growing at trend. Productivity is too low, but the job market is tightening and wage growth is getting a bit of a boost. We certainly want to take note of the weaknesses in the “high-frequency” data, but a month or a quarter does not a new trend make.