June 1, 2018

May 2018 Employment Situation Summary (060118): 223K SA Payroll Jobs Added, Raw Results 75K Stronger; Unemployment Rate Drops to 3.8 Percent; Black Unemployment Plunges Below 6 Pct. For First Time in History; Asian Unemployment Hits Record Low 2.1 Pct.; Hourly Pay Up a Decent 8 Cents

Filed under: Economy,Taxes & Government — Tom @ 8:22 am


  • Associated Press — 190,000 payroll jobs added, unemployment rate remaining at 3.9 percent.
  • Bloomberg — Same (scroll down to after the article at link).
  • Zero Hedge says “It’s finally time for a solid beat.” We’ll see. Its posted expectation is 188,000 payroll job additions and a 3.9 percent unemployment rate.

Not seasonally adjusted figures: I’ll have to evaluate them after the report is released. May is a big month for actual job adds, so big raw numbers are needed.

What to hope for:

  • People coming in off the sidelines, which might even push the unemployment rate up a bit.
  • A long-overdue pickup in average hours worked (full-time jobs have accounted for more than 100 Household Survey job additions since the end of 2016 (that’s because part-time jobs have contracted slightly, so the average work week should be increasing, and hasn’t done so meaningfully).
  • An hourly wage pickup of 8 cents or more.

The report will be here at 8:30 (permanent link with all tables will be here).

HERE IT IS: A pretty decent beat on jobs, with pretty small upward prior-month revisions, and the unemployment rate dropped —

Total nonfarm payroll employment increased by 223,000 in May, and the unemployment rate edged down to 3.8 percent, the U.S. Bureau of Labor Statistics reported today. Employment continued to trend up in several industries, including retail trade, health care, and construction.

Household Survey Data

The unemployment rate edged down to 3.8 percent in May, and the number of unemployed persons declined to 6.1 million. Over the year, the unemployment rate was down by 0.5 percentage point, and the number of unemployed persons declined by 772,000.

Among the major worker groups, the unemployment rates for adult men (3.5 percent), Blacks (5.9 percent), and Asians (2.1 percent) decreased in May. The jobless rates for adult women (3.3 percent), teenagers (12.8 percent), Whites (3.5 percent), and Hispanics (4.9 percent) changed little over the month.

The number of long-term unemployed (those jobless for 27 weeks or more) was little changed at 1.2 million in May and accounted for 19.4 percent of the unemployed. Over the past 12 months, the number of long-term unemployed has declined by 476,000.

Both the labor force participation rate, at 62.7 percent, and the employment-population ratio, at 60.4 percent, changed little in May.

Establishment Survey Data

Total nonfarm payroll employment increased by 223,000 in May, compared with an average monthly gain of 191,000 over the prior 12 months. Over the month, employment continued to trend up in several industries, including retail trade, health care, and construction.

In May, retail trade added 31,000 jobs, with gains occurring in general merchandise stores (+13,000) and in building material and garden supply stores (+6,000). Over the year, retail trade has added 125,000 jobs.

Employment in health care rose by 29,000 in May, about in line with the average monthly gain over the prior 12 months. Ambulatory health care services added 18,000 jobs over the month, and employment in hospitals continued to trend up (+6,000).

Employment in construction continued on an upward trend in May (+25,000) and has risen by 286,000 over the past 12 months. Within the industry, nonresidential specialty trade contractors added 15,000 jobs over the month.

Employment in professional and technical services continued to trend up in May (+23,000) and has risen by 206,000 over the year.

Transportation and warehousing added 19,000 jobs over the month and 156,000 over the year. In May, job gains occurred in warehousing and storage (+7,000) and in couriers and messengers (+5,000).

Manufacturing employment continued to expand over the month (+18,000). Durable goods accounted for most of the change, including an increase of 6,000 jobs in machinery. Manufacturing employment has risen by 259,000 over the year, with about three-fourths of the growth in durable goods industries.

Mining added 6,000 jobs in May. Since a recent low point in October 2016, employment in mining has grown by 91,000, with support activities for mining accounting for nearly all of the increase.

In May, employment changed little in other major industries, including wholesale trade, information, financial activities, leisure and hospitality, and government.

The average workweek for all employees on private nonfarm payrolls was unchanged at 34.5 hours in May. In manufacturing, the workweek decreased by 0.2 hour to 40.8 hours, and overtime edged down by 0.2 hour to 3.5 hours. The average workweek for production and nonsupervisory employees on private nonfarm payrolls remained at 33.8 hours.

In May, average hourly earnings for all employees on private nonfarm payrolls rose by 8 cents to $26.92. Over the year, average hourly earnings have increased by 71 cents, or 2.7 percent. Average hourly earnings of private-sector production and nonsupervisory employees increased by 7 cents to $22.59 in May.

The change in total nonfarm payroll employment for March was revised up from +135,000 to +155,000, and the change for April was revised down from +164,000 to +159,000. With these revisions, employment gains in March and April combined were 15,000 more than previously reported. … After revisions, job gains have averaged 179,000 over the last 3 months.

Not Seasonally Adjusted Analysis:

The seasonally adjusted results understate the relative strength of today’s result compared to the seasonal adjustments of the past four years by about 75,000:


If I’d have had the chance ahead of time, I would have benchmarked 975,000 total nonfarm jobs and 1.025 million private-sector jobs. The figures seen above fell short, but only by 30,000 or so.

Nonetheless, today’s beat is bigger than the seasonally adjusted numbers would indicate. A result genuinely reflecting the past four years’ seasonal conversions would have been at or close to 300,000.


UPDATE (references are to seasonally adjusted data unless otherwise indicated):

  • The civilian labor force barely budge (up 12K), and the participation rate didn’t change meaningfully. Household Survey employment increased by 294K. Not in Labor Force is just shy of 96 million.
  • Both the black and Asian unemployment rates dropped by 0.7 points to 5.9 percent and 2.1 percent, respectively. Black unemployment never fell below 7.0 percent (a figure seen only once, in April 2000) until December 2017. Now it’s below 6.0 percent. It has fallen two percentage points in just 13 months.
  • I know these particular figures swing wildly from month to month, but it’s hard to ignore how full-time employment jumped by 904K, while part-time workers fell by 625K. Looking at a longer period, the economy has added 4.356 million full-time jobs in the past 17 months, while part-timers have fallen by 1.007 million. Yet the average work week has increased by only 0.1 hours during that same period, despite that significant change in the FT/PT mix. My back-of-napkin estimates indicate that this doesn’t compute.
  • The fully-loaded U-6 unemployment rate dropped to 7.6 percent. That’s the lowest figure since May 2001.
  • Goods-producing job additions have averaged 50K per month for the past 12 months, including 259K in manufacturing. Not bad for jobs that Donald Trump’s predecessor Barack Obama said were never coming back. (UPDATE: Rush Limbaugh made this very point on his Friday afternoon program.)
  • The 8 cents per hour wage pickup is where it should be, but that’s what needs to be sustained on average for 12 months to generate a full-year increase of 3.5 percent, which (assuming inflation stays relatively low) is the benchmark to shoot for.

Overall, this was a very strong report, and is really a strong beat considering the raw numbers. We need more of this.

Unfortunately — though it wasn’t a presumptively positive message — President Trump tweeted that he was “Looking forward to seeing the employment numbers at 8:30 this morning.” He should have stuck with protocol and remained silent.

So the press will likely try to make the news about him supposedly telegraphing a strong report. A failure to exercise proper restraint in these and other matters could burn him badly some day. His presidency is a high-wire act like no other.



  1. One more somewhat-expanded prediction via Yahoo Finance:

    - +190K jobs
    - 3.9% U-3
    - Wages +0.2% month-over-month, +2.6% year-over-year.

    Also, Trump appeared to violate protocol by insinuating good news on his Twitter feed earlier this morning.

    Comment by steveegg — June 1, 2018 @ 8:29 am

  2. Good grief, the guy won’t stop his high-wire act. Bet the press makes the story all about that.

    Comment by Tom — June 1, 2018 @ 8:45 am

  3. Or at least Sarah Huckabee Sanders’ comment that it was “appropriate” because “he didn’t put out the number.” That very nearly made me late with my Hot Air guest post.

    It wasn’t mentioned in the initial AP dispatch, which did a Jeckyl-and-Hyde approach that also included some good Q2 GDP news.

    Comment by steveegg — June 1, 2018 @ 10:05 am

  4. The tweet’s text was benign, but the fact that he tweeted about it was likely interpreted as a harbinger of good news which affected the markets, though I haven’t looked at how futures went and don’t have time to now.

    Comment by Tom — June 1, 2018 @ 10:15 am

  5. From what I heard, it was a marginal uptick in the futures.

    Comment by steveegg — June 1, 2018 @ 10:18 am

  6. Tom, the Troll Master in Chief is at it again, by tweeting about the stats before the “official” announcement, Trump gets to stir up the hornets nest and have the MSM continue to prove to the public that the MSM is negative about Trump. There is a method to his madness and it works for him. Remember, George W Bush was a man of class and graciousness but the MSM was less than kind to him. Obama was the golden child of liberalism and the MSM covered up (refused to cover) his obvious feckless incompetence.

    A note to all Never Trumpers, the MSM will never ever cut you an even break when it’s you against a Democrat; you suffer from battered wife syndrome. You are merely props, like McCain, for bad mouthing the GOP and Conservatism.

    Trump uses the obvious MSM distaste of him to his advantage. It’s like the movie, Roger Rabbit, all one needs to do is tap the tune to “A shave and a hair cut” and the MSM is set off reflexively to bad mouth Trump. Thus they confirm the stereotype of them hating Trump. Think about it, IF there is anything ever really bad reported about Trump, the public has already discounted the bad reporting as “yeah, boring, tell me something I haven’t already heard…” Hence his polling numbers continue to go up.

    Comment by dscott — June 2, 2018 @ 9:29 am

  7. The SNAP numbers haven’t been updated since February, maybe come June 4th or 5th we will have some updates:


    It shows we are down to less than 20 million adults on the rolls. Maybe this is why things aren’t computing?

    Also, I ask the question again on self employment, how is new business formation proceeding? Lot’s of new formation would clearly skew the math on hours worked.

    Comment by dscott — June 2, 2018 @ 9:39 am

  8. #6, Nothing wrong with trolling as long as you’re getting your job done too.

    But Trump’s pre-report tweet was not a judicious move, given that it could have moved the markets pre-announcement. What if someone in the WH knew what Trump was about to tweet and started trading in the markets on that knowledge? The tweet’s language was innocuous, but there should be NO communication EVER from an admin official, let alone the President, about an econ report before it comes out. Trump is doing many good things, but there are times when he has to hold his tongue and resist the urge to tweet.

    Comment by Tom — June 2, 2018 @ 9:55 am

  9. #7, If form holds, March will come out on Friday afternoon, June 8. The report is always two months behind.

    The failure of the food stamp rolls to come down significantly is infuriating. During relatively prosperous times they’re still more than double what they were during the early-2000s.

    Comment by Tom — June 2, 2018 @ 9:59 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.