Payrolls rose 136,000 last month and the unemployment rate dipped to 3.5 percent, its lowest rate since the late 1960s. Though the payroll number missed analysts’ expectations (~145,000), the more reliable 3-month average came in at a healthy 157,000, strong enough to put downward pressure on unemployment (the prior two months of payroll data were revised up by 45,000 jobs).

Our monthly smoother takes 3, 6, and 12-month averages of monthly job gains to help pull out the underlying trend out of the noisier monthly data. Over the past 6 months, payroll gains have average 154,000, a deceleration from the 12-month number (179K), but such a pattern is expected in an economy closing in on—though not yet at—full employment.

Wage growth for private-sector workers was up 2.9 percent over the past year, a slightly slower rate than in previous months. The wage pace was stronger for middle-wage workers (production, non-supervisors) at 3.5 percent, but in both cases, as the figures reveal (note especially the 6-month moving averages), the trend in wage growth is not accelerating, even given the low unemployment rate. For the “all” group (first figure), there’s even some evidence of decelerating wages, a possibility that is now on my watch list. I return to these important observations below.

State and local government hiring was important in September, adding 24,000 jobs. Though analysts expected hiring for the decennial Census to be a factor in these data, that was not the case, as the BLS reported such hiring only accounted for 1,000 jobs last month. The factory sector is clearly stressed, with manufacturing losing 2,000 jobs in September. The GM strike is certainly in the mix here, but thus far this year, the factory sector has added an average of fewer than 5,000 jobs per month, compared to 22,000/month last year. That’s much more trade-war than strike.

As noted, the Household survey showed greater signs of job-market strength last month. Along with unemployment at a 50 year low, the underemployment rate (the “u6” rate, which includes part-timers who want full-time work) fell to 6.9 percent, close to its all-time low of 6.8 in October of 2000 (this series only starts in 1994). The closely watched employment rate (“epop,” for employment-to-population ratio) for prime-age workers ticked up one-tenth for both men and women. Women’s prime-age epop–74 percent last month–has handily surpassed its 2007 peak, while men’s–86.4 percent–is still below their 2007 peak of 88 percent.

However, as the next figure shows, since the 1970s, men’s epop’s have moved like a ratchet–highly cyclical, but never quite regaining prior peaks. One conclusion is that men (and women) respond to employment opportunities but, at least for the men, they’ve been losing more in the downturns than they’ve gained in the expansions. My analysis suggests that if the cycle persists, prime-age epop’s will regain their prior peak, pushing back on the long-term ratchet.

Consider the following:

–Wage growth is not speeding up and probably decelerating;

–The pace of job gains has attenuated but remains solid, even this late in the expansion;

–Labor supply continues to grow, as per the epop discussion above.

–Price growth shows little pressure, even at historically low unemployment.

Put these facts together and one, strong conclusion is that even in year 11 of this long expansion, the U.S. labor market is not yet at full employment. Thus, the Federal Reserve has little cause to tap the growth brakes and good reasons to try to keep the recovery going, which in the current context means pushing back on pressures from the trade war, slowing global growth, and political chaos.

[Huge hat-tip to Katie Windham for stepping up and helping with the above!]