Payrolls rose 223,000 last month, beating expectations of 190,000, and the unemployment rate ticked down to 3.8 percent, its lowest level since April 2000, and before that, a level much more commonly seen in the 1960s. (At 3.75 percent, the jobless rate just missed falling two-tenths).
[Before the release, President Trump tweeted that he was looking forward to the jobs numbers. Since certain top officials, including the president, see the report on Thursday night, his tweet telegraphed the positive report, a highly unusual occurrence.]
The unemployment rate for African-Americans fell to 5.9 percent, an historical low point by a wide margin. Typically, the black unemployment rate is twice the white rate. But persistently tight labor markets are especially helpful for minority workers, as they make it more costly for employers to discriminate. In May, the black/white ratio was 1.7, still too high, but lower than average, underscoring the relative gains to less-advantaged workers.
Given the noisiness of these monthly data, our patented jobs smoother looks at average monthly employment gains over 3, 6, and 12-month intervals. As shown below, the trend in payroll growth is running at around 180K-200K per month, a solid trend that, if it persists, is strong enough to continue pushing down the unemployment rate.
Wage growth picked up slightly, up 2.7 percent overall and 2.8 percent for middle-wage workers. This too is a positive sign, as the tight labor market pushes up wage growth. The figures show yearly wage gains for all private sector workers and for the 82 percent that are blue-collar production workers and non-managers in services. The smooth trend in the first figure shows little by way of recent acceleration. Hourly wages were up 2.7 percent last month, a bit faster than the latest reading on consumer inflation of 2.4 percent.
The other figure, however, for middle-wage workers, shows a bit of a trend increase, as wage growth has accelerated in recent months and was 2.8 percent in May. This is once again consistent with the tight labor market disproportionately helping the least advantaged.
If it sticks, this “trend is our friend,” as is the solid payroll jobs’ trend. But is there anything out there that could whack it? The Fed could raise interest rates too quickly, but, barring a sharp acceleration in prices, which I judge to be unlikely, I believe they will be careful not to make this mistake. Trump’s trade war could, and probably will, escalate. That’s slightly worrisome, but remember, relative to other countries, the US is somewhat insulated to trade shocks as our imports as a share of GDP are only 15 percent, compared to at least twice that in Europe.
The biggest constraint to the jobs trend is labor supply. If the supply of available workers dries up, that will definitely constrain both job and overall economic growth. However, I’ve argued that this constraint may be less binding than many economists believe to be the case (yes, the May labor force barely budged, but these monthly numbers are especially noisy).
Employment rates of prime-age workers (25-54) were flat last month, but they’ve been climbing and have recovered 4.4 out of 5.5 percentage points, or 80%, of their losses since the recession. Historically, this indicator has flattened before recessions, but, May’s result aside, it has been growing lately for both genders, suggesting more room to run. We also know that there is considerable geographical variation in labor market tightness, so while some cities may be close to tapped out, supply-wise, other places are clearly not. At least thus far, these dynamics, combined with low productivity growth and weak worker bargaining power, have constrained wage and price growth.
I recently pointed out the prime-age employment rate is a better predictor of recent wage growth (nominal, i.e., before inflation) than the unemployment rate. The figure below (which does not include this month’s data) shows the results of a simple statistical model that predicts the annual wage growth of non-supervisory workers (the one that grew 2.8 percent over the past year). I run the model through 2014 and then predict wage growth based on a slack variable and lagged wage growth.
What it shows is that variables that are more inclusive of slack do a better job of predicting wage growth. The unemployment rate says wages should be growing about 3.5 percent right now. The more slack-inclusive underemployment rate (U6) is a little more pessimistic/realistic but the men’s prime-age employment rate, which shows the most slack, does the best.
There are many caveats to this simple exercise–the differences are all within a margin of error and a more complete model would include the slow productivity growth that is putting downward pressure on wage growth. But it does provide some useful information. The notion that labor supply is fully tapped in the U.S. is not well supported by these monthly jobs reports. First, the persistently strong monthly payroll numbers are inconsistent with seriously binding supply constraints. Second, the employment rate for prime-aged workers doesn’t appear to have topped out. Third, while some price and wage pressures are building, these capacity indicators are not flashing red by a long shot.
Thus, especially from the Fed’s perspective, the assumption that there’s still room to run–that labor supply is not clearly exhausted–is the right one to make. The gains to African-Americans must be preserved and built upon. Same with that tick up in wage growth for mid-wage workers. Remember, in an economy with little union power, tremendous finance power, and thereby, far too much inequality, the best friend working people have is a persistently tight labor market.