,
16 tweets,
7 min read
Read on Twitter
Time for charts!
For my earlier (chart-free) initial reaction, see this thread:
For my earlier (chart-free) initial reaction, see this thread:
The current pace of job growth (about 150k/month) isn't bad. It's enough to keep bringing down unemployment. But there no question growth is slowing. 
The slowdown is most acute in goods-producing sectors like manufacturing and energy, which are exposed to the trade war (and to slowing global growth more generally).
Meanwhile the service sector continues to look strong. Health care continues to add tens of thousands of jobs each month. Leisure has held on well.
The exception of course is retail, which continues to lose jobs month after month. Now down 60k jobs over the past year.
The glaring weakness in this report is wages. We all got excited last year when annual wage gains topped 3%. But the acceleration hasn't continued, and now the pace of growth is slowing (at least for headline -- slightly better news for nonsupervisory).
The combination of flat-to-slowing wage growth, slowing hiring and flat weekly hours means that total wage income (important for consumer spending) is growing more slowly. Though it's basically back to 2017 level -- not falling off a cliff.
Much better news in the household survey. The unemployment rate will get the headlines, but for my money the really exciting news is the resumed upward movement in the prime-age employment rate. Now at a cycle high.
Adjusting for the aging population, the employment rate is now clearly above its prerecession level, and labor force participation is finally back to where it was before the recession. (Both still well below 2000 level, though.)
Note that these more sophisticated demographic adjustments have been steadier than prime-age measures, which showed a troubling drop earlier this year.
This is the lowest the unemployment rate has been since December 1969. The broader "U-6" measure, which includes people who've stopped looking for work and those who are involuntarily part-time, tied its lowest level on record (only back to 1994).
Still, despite the low unemployment rate, the number of *long-term* unemployed (out of work more than 6 months) is still high compared to similar periods in previous eras.
As noted earlier, these numbers don't incorporate the preliminary revisions announced in August. Factoring those in, the slowdown in job growth started earlier and has been steadier than the headline numbers indicate. (Final revisions due early next year.)
nytimes.com/2019/08/21/bus…
nytimes.com/2019/08/21/bus…
A bit more on the manufacturing slowdown: The manufacturing diffusion index fell below 50, meaning more sub-sectors are cutting jobs than adding them. (This chart shows 6-month average, which is still just above 50.)
The breakdown is interesting. Biggest manufacturing job cuts are in trade-exposed sectors like machinery, autos, metals. Electronics still adding jobs, meanwhile.
More broadly: Goods-producing industries (esp. manufacturing & energy) surged early in Trump's term. Now, they're dragging down job growth.
Meanwhile, job growth in the (much larger) service sector has ebbed a bit but mostly held steady.
Meanwhile, job growth in the (much larger) service sector has ebbed a bit but mostly held steady.
