September 1, 2017

August 2017 Employment Situation Summary (090217): Unemployment Rate Rises to 4.4 Percent; 156K Payroll Jobs Added, Prior Months Revised Down by 41K; Wage Growth Flat, Average Hours Drop

Filed under: Economy,Taxes & Government — Tom @ 7:26 am

Predictions:

Didn’t get to the BLS tables soon enough to bet benchmarks for the raw data on jobs added, so we’ll have to deal with that after the report’s release.

Last-minute thoughts: Besides the overriding importance of the raw numbers, watch for how much the labor force grows or contracts, hourly wage growth, weekly earnings growth, and full-time employment growth or contraction. Also recall that BLS, according to Mark Zandi at Moody’s and other data watchers, has been notorious for issuing an initially small job-growth number followed by significant upward revisions in the next two months and during the annual comprehensive revision.

The report will be here at 8:30.

HERE IT IS (permanent link with tables): Well, it’s in line with diminished expectations discussed on Wednesday —

Total nonfarm payroll employment increased by 156,000 in August, and the unemployment
rate was little changed at 4.4 percent, the U.S. Bureau of Labor Statistics reported
today. Job gains occurred in manufacturing, construction, professional and technical
services, health care, and mining.

Household Survey Data

In August, the unemployment rate, at 4.4 percent, and the number of unemployed persons, at 7.1 million, were little changed. After declining earlier in the year, the unemployment rate has been either 4.3 or 4.4 percent since April.
Among the major worker groups, the unemployment rates for adult men (4.1 percent), adult women (4.0 percent), teenagers (13.6 percent), Whites (3.9 percent), Blacks (7.7 percent), Asians (4.0 percent), and Hispanics (5.2 percent) showed little or no change in August.

The number of long-term unemployed (those jobless for 27 weeks or more) was essentially unchanged in August at 1.7 million and accounted for 24.7 percent of the unemployed.

The labor force participation rate, at 62.9 percent, was unchanged in August and has shown little movement on net over the past year. The employment-population ratio, at 60.1 percent, was little changed over the month and thus far this year.

… Establishment Survey Data

Total nonfarm payroll employment increased by 156,000 in August. Job gains occurred in manufacturing, construction, professional and technical services, health care, and mining. Employment growth has averaged 176,000 per month thus far this year, about in line with the average monthly gain of 187,000 in 2016.

Manufacturing employment rose by 36,000 in August. Job gains occurred in motor vehicles and parts (+14,000), fabricated metal products (+5,000), and computer and electronic products (+4,000). Manufacturing has added 155,000 jobs since a recent employment low in November 2016.

In August, construction employment rose by 28,000, after showing little change over the prior 5 months. Employment among residential specialty trade contractors edged up by 12,000 over the month.

Employment in professional and technical services continued to trend up in August (+22,000) and has grown by 262,000 over the last 12 months. In August, job gains
occurred in computer systems design and related services (+8,000).

Health care employment continued on an upward trend over the month (+20,000) and has risen by 328,000 over the year. Employment in hospitals edged up over the month
(+6,000).

Mining continued to add jobs in August (+7,000), with all of the growth in support activities for mining. Since a recent low in October 2016, employment in mining has risen by 62,000, or 10 percent.

Employment in food services and drinking places changed little in August (+9,000),
following an increase of 53,000 in July. Over the year, the industry has added 283,000 jobs.

Employment in other major industries, including wholesale trade, retail trade, transportation and warehousing, information, financial activities, and government, showed little change over the month.

The average workweek for all employees on private nonfarm payrolls declined by 0.1 hour to 34.4 hours in August. In manufacturing, the workweek declined by 0.2 hour to 40.7 hours, while overtime was unchanged at 3.3 hours. The average workweek for
production and nonsupervisory employees on private nonfarm payrolls was 33.7 hours
for the fifth consecutive month.

In August, average hourly earnings for all employees on private nonfarm payrolls rose by 3 cents to $26.39, after rising by 9 cents in July. Over the past 12 months, average hourly earnings have increased by 65 cents, or 2.5 percent. In August, average hourly earnings of private-sector production and nonsupervisory employees increased by 4 cents to $22.12.

The change in total nonfarm payroll employment for June was revised down from +231,000 to +210,000, and the change for July was revised down from +209,000 to +189,000. With these revisions, employment gains in June and July combined were 41,000 less than previously reported. (Monthly revisions result from additional reports received from businesses and government agencies since the last published estimates and from the recalculation of seasonal factors.) After revisions, job gains have averaged 185,000 per month over the past 3 months.

Not seasonally adjusted results:

  • Total nonfarm payrolls actually added were 213K. I would have benchmarked at +400K if I had the chance to do so before the report, so this is disappointing.
  • Total nonfarm payrolls actually added were 51K. I would have benchmarked at +200K if I had the chance to do so before the report, so this is also disappointing.
  • The seasonal conversions to +156K overall and 165K for the private sector look reasonable in the context of the past two years, but a bit generous (by maybe 40K) in the context of the three preceding years. But more recent years are supposed to carry more weight in seasonal adjustments.

More in a bit.

UPDATE (references are to seasonally adjusted numbers unless otherwise indicated) —

  • Labor force growth and participation indicators were disappointing. Only 71K new workers entered the workforce. Household Survey employment fell by 74K. Not in labor force jumped 128K to 94.785 million. The number of officially unemployed increased by 151K. These are the kinds of numbers which, if repeated for a couple more months, would lead you to believe that the economy has peaked.
  • The black unemployment rate has increased from 7.1 percent to 7.7 percent in the past two months. All of the bad news is among 20-and-over men (from 6.3 percent to 7.8 percent), while the rate for 20-and-over women has dropped from 6.8 percent to 6.4 percent.
  • Full-time employment dropped by a disappointing 166K, while part-timers increased by 34K. Self-employment increased by 142K, while multiple jobholders fell by 243K. If that all seems like a jumbled mess, it’s because it’s a jumbled mess. But it does seem to indicate overall softening.
  • Because of the declining work week, weekly earnings dropped by $1.60 to $907.82. The hourly wage increase of 3 cents was obviously unimpressive.
  • The jobs news in the goods-producing sectors was quite good: Mining +6K, Construction +28K and manufacturing +36K. The guess here is that we haven’t seen the goods-producing sectors add 70K jobs in a single month in quite a long time, and certainly not as consistently as they’ve been added during calendar 2017.
  • The rest of the economy had a pretty weak month. Especially worth watching in the next couple of months are retail and leisure/hospitality (particularly food service establishments). The theory here is that minimum wage laws in many parts of the country are starting to slow the economy down.

Overall, this was a mediocre month which came pretty close, in my view, to waving a couple of serious red flags.

That said, there is still an unexplained difference of 377K in private sector jobs added per ADP during the first 8 months of this year (1.783 million) vs. BLS (1.406 million). That’s a roughly two-month difference between the two measurements. Something’s got to give. I really don’t think it’s ADP that’s wrong. Even though the person who runs that survey is a partisan Democrat who if anything would be inclined to fudge the numbers lower, I don’t think he’s doing anything but reporting what ADP is really finding (he’s got to justify getting paid by clients). Meanwhile, in my view, BLS is still infested with Obama-partisan holdovers who have every incentive to understate the numbers until they get forced into reality by the comprehensive revision taking place early next year (to be preliminarily announced, if past form holds, in October or November).

Share

6 Comments

  1. We’ve had decent/good job reports under Trump and I’ve seen no trends about it on Twitter, we get one poor report today and all of a sudden #JobsReport is trending along with a helpful description of why it is a poor report. Coincidence?

    I fully expect today’s negative report to get more MSM coverage than all of the positive reports under Trump combined.

    Comment by zf — September 1, 2017 @ 11:12 am

  2. Initial reactions were surprisingly positive and forgiving, but I’m certainly watching for what you’re concerned about. Zandi at ADP pointed out Wednesday that BLS usually initially screws up August (not those words, of course, but that’s what he meant) and that he expected a private sector reading 50K to 100K below ADP’s 237K. We got 72K (private sector was +165K).

    Point is that anyone trying to claim today’s report stinks (as opposed to disappointing, which I still think it is) is going to do so ignoring the six or so years of history Zandi identified, and if they do, the pouncing will begin.

    Comment by Tom — September 1, 2017 @ 3:01 pm

  3. #2, Figures. They were forgiving and positive just to make my prediction wrong. :)

    Comment by zf — September 1, 2017 @ 6:26 pm

  4. I think Zandi sort of maneuvered them into having to concede that position. AP’s coverage echoes it: “One reason why few analysts expressed concern about last month’s slower job gain is that monthly employment reports can be volatile – especially figures for August.”

    Comment by Tom — September 1, 2017 @ 7:29 pm

  5. While it is true that the Not in Labor Force # is up, however when taken in context using the graph, this is not something to be concerned about yet. The trend is essentially flat which is significant compared to the Obama and Bush years:

    https://data.bls.gov/timeseries/LNS15000000

    Comment by dscott — September 3, 2017 @ 3:29 pm

  6. We also need to appreciate the U-6 dropping to 8.6%:

    https://unemploymentdata.com/current-u6-unemployment-rate/

    Again, the graph put the data in context shows a good trend.

    Comment by dscott — September 3, 2017 @ 3:36 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.