Employers added only 138,000 jobs last month, well below expectations for 175,000. Revisions to payrolls for the prior two months reduced employment gains by 66,000. The unemployment rate fell to 4.3 percent, its lowest level since 2001, but for the wrong reason: labor force participation fell by two-tenths of a percent. In other words, this is a considerably weaker-than-expected jobs report.
Given the noise in the monthly data, the question is: does this report signal a real downshift in job growth or is it a blip? Also, if we’re really at full employment, we should expect some slowing in payroll gains as employers bump up against supply constraints. And what does this all mean for the Federal Reserve when they meet in a few weeks to consider another rate hike that is firmly priced into the markets?
A good place to start is by smoothing out the monthly noise with the official JB smoother, which takes monthly averages over the past 3, 6, and 12 months. It shows a marked deceleration in job growth, from about 190,000 over the 12-month period to 121,000 over the past three months. While this is suggestive of a softening of the job market, it is also consistent with supply constraints.
The labor force participation rate is another important place to look in this regard, but it is a) a very noisy monthly indicator, and b) the overall rate is down in part due to retirement of aging boomers. A better indicator is thus the prime-age (25-54 year-old) employment rate. After falling 5.5 percentage points in the last recession, this rate has slowly been climbing back–score one for the tightening narrative. However, it ticked down a bit in May and, more importantly (this indicator is also noisy), is still about 2 percentage points below its pre-recession peak.
The underemployment rate, which includes 5.2 million involuntary part-time workers who’d rather be full-timers, fell to a cyclical low of 8.4 percent, though this too reflects May’s labor force exits. On the other hand, involuntary part-time work is down a solid 1.2 million over the past year.
Industry employment patterns reveal job losses in retail trade, down about 80,000 over the past four months, as brick and mortar stores lose demand to internet sales. Manufacturing remains soft, and state/local government shed 17,000 jobs in May. Health care continues to deliver, but gains in the sector have averaged 22,000 per month so far this year, compared to 32,000 per month last year.
So which is it: tightening or softening? I’ve got one more indicator to bring to bear before I render my judgement on this Talmudic question: core inflation.
Prior to the release of the jobs report, the futures market was assigning a 93.5 percent probability to another quarter-point rate hike by the Federal Reserve at their mid-June meeting. However, as the figure below shows, while unemployment is clearly below the Fed’s full-employment-unemployment rate of 4.7 percent, core inflation has been going the “wrong” way, i.e., slowing, not speeding up (see its down-tick at the end of the figure).
So, putting it all together, I think there’s enough evidence that the Fed’s tightening campaign has slowed the job market down for them to pause in their June meeting. I recognize that this will shock markets, but the Fed’s client is not the stock market. It’s the macroeconomy, and the dual mandate: full employment at stable prices. Given at least some evidence of softening in the job market in tandem with slower core price growth, a data-driven Fed should pause and take stock of where we are.
The members of the central bank apparently think the recent slowdown in price growth is transitory, and that at some point, price pressures will reflect the tightening of the job market. That’s certainly plausible, but the last figure shows that they’ve been missing their 2 percent inflation target for years now. More likely, the historical correlation between inflation and unemployment is much diminished, such that the Fed cannot reliably estimate the lowest unemployment rate consistent with stable, 2 percent inflation.
From this perspective, the Fed is being less data driven than they like to claim, and is in more of an ad hoc mode. They’re raising because the governors broadly believe that in year eight of an economic expansion closing in on full employment, rates should be at a more normal level. By doing so slowly, they can monitor the impact of their campaign, and I suspect that, as far as they can see, it looks to be going well–they’re “normalizing” the rate, yet, May’s results aside, not much dampening the pace of output or job growth.
There are two problems, however, with this approach. One, since they’re not data driven, at least as far as the inflation data are concerned, it is hard for observers to know where they’re going next. Ad hockery is tougher on expectations than following the data. Two, and this is a more serious concern, for the least advantaged workers to get ahead, the job market has to run very hot, and even small taps on the brakes push the wrong way in that regard.
Given some evidence–uncertain and shaky to be sure–that the job market has cooled a bit while inflation is non-threatening, from what I see from here, the Fed should revert to data-driven mode and punt on a June hike.