Jobs Report: June’s big number reveals welcome bounceback, but you’ve gotta smooth these noisy data.

July 8th, 2016 at 10:11 am

[Before I jump into analysis of the jobs report, I’m compelled to add a brief note about the violence and deaths in Dallas last night and the deaths of Alton Sterling and Philando Castile earlier this week. I will not try to hold forth on these tragedies other than to say that I’m both indelibly saddened and deeply enraged. That said, I’m sure my pain and anger are a fraction of that of many others directly affected by these murders. My heart goes out to all the victims.]

Payrolls posted a big 287,000 jump in June in stark contrast to revised gains of just 11,000 in May. Such monthly volatility provides an extremely clear example of why we should never over-interpret one month’s worth of jobs data. You have to “smooth,” or average out the gains over numerous months, as I do below. When you do so, you get a picture of a solid job market, adding somewhat fewer jobs than a year ago, but still making progress towards full employment.

The unemployment rate ticked up to 4.9% as more people entered the labor market, leading to a small but welcome uptick in the participation rate, up one-tenth to 62.7%. Year-over-year wage growth ticked up slightly as well, from 2.5% to 2.6%, a positive sign that some of the benefits of the ongoing recovery are finally reaching workers’ paychecks. As shown below, this pace of wage growth remains well below Fed chair Yellen’s benchmark target of 3.5%. In other words, this trend should be very much welcomed, not feared! It’s what’s supposed to happen as the job market improves and working people get a little more bargaining power.

The punchlines are thus as follows: May’s dismal report was an outlier; the US recovery proceeds apace; Brexit hasn’t shown up in the jobs numbers; wage growth is slowly picking up a bit of speed, as I’d expect; and the job growth engine has downshifted from around 200K/month to around 150K/month, once you smooth out the monthly noise. That’s also to be expected as we get closer to full employment, though given that we’re not there yet, both monetary and fiscal policy needs to continue to be as pro-growth as possible. This policy stance is underscored by the absence of inflationary pressures.

JB’s patented monthly smoother is particularly important this month. The monthly trend job gains over the past 3 months is about 150K, 6 months: ~170K; 12 months: ~200K. There’s the downshift noted above.

Source: my analysis of BLS data.

Source: my analysis of BLS data.

A few other highlights from the report:

–The underemployment rate ticked down from 9.7% to 9.6%, but this more comprehensive measure of labor market slack (it includes part-time workers who’d rather be working full-time and those “marginally attached” to the labor force) shows we’re not yet at full employment, which calls for an underemployment rate about a full point lower.

–As noted, there’s a positive trend in average hourly earnings (see figure below). But note also Chair Yellen’s wage benchmark of 3.5% shown in the figure. There’s considerable room for wages to continue to accelerate. Those calling on the Fed to raise rates and thwart this trend are both wrong on the substance—wage-growth is not pushing up price growth—and implicitly suggesting that those who’ve gained the least from the recovery thus far need to take a hit. This, in my view, is the monetary policy version of “the system is rigged.”

Source: BLS

Source: BLS

–After losing 16,000 jobs in May, manufacturing rebounded in June, adding 14,000 factory jobs.  Smoothing over recent months, employment in the sector is down 4,000 jobs/month this year, compared to +10K/month in 2014-15. This reversal is partly a function of the stronger dollar, which makes our manufactured exports less competitive, leading to larger trade deficits. More broadly, it reinforces the need for better rules of the road in our trade deals, a topic that’s become highly elevated in the presidential campaign.

Federal Reserve economists are just as good at taking averages as I am, and thus I’m very confident the Fed won’t overreact to the big June jobs number and resume their rate normalization campaign. Brexit uncertainty, the risk of further strengthening of the dollar, and labor market volatility all push against an interest rate increase. Once you average over the past few months, choppy waters smooth out a bit, suggesting a generally solid, ongoing labor market recovery.

Print Friendly, PDF & Email

5 comments in reply to "Jobs Report: June’s big number reveals welcome bounceback, but you’ve gotta smooth these noisy data."

  1. Smith says:

    I thought the labor force is presently about 150 million but just checking it’s 130 million full time and 26 million part time. If the labor force grew as expected, about 1% a year, you’d need 1,500,000 new jobs to break even (no increased unemployment), and the latest 3 month trend of 150,000/month brings us 1,800,000. The 300,000 extra would drop unemployment a whole .2 percent. But more troubling is the data here.
    http://www.bls.gov/news.release/empsit.t01.htm
    The figures are for June 2015 and June 2016
    Civilian noninstitutional population 250,663 253,397 (16 years of age and older)
    Civilian labor force 158,283 158,880
    Participation rate 63.1 62.7

    The population increased 1% but the labor force increased .38% Why? The participation rate dropped .4% which is .4% of the 253,397 population, not the 158,880. That accounts almost entirely for the missing .62% expected labor force increase, which is a million people by the way. Thus, this month 287,000 jobs were added but the unemployment rate rose, as more people jumped back into the labor market. This makes the trend of the trend troubling, as you go 12 month, 6 months, 3 months, the smoothed average drops. 150,000/month will just about take care of new population, and won’t do anything to reduce underemployment (beyond the .2% 300,000 surplus mentioned). It’s surprising to consider that in Obama’s last year in office, a million more people dropped out of the labor market. This is why running for the third term and continuity may not be how to sell voters on an economic program.

    The participation rate drop off is not demographics because table 1 https://www.census.gov/prod/cen2010/briefs/c2010br-09.pdf shows a half point increase in percent population over 65, growing .5% 2000 to 2010 which is .05% a year. A larger increase in this decade due to peak boomer retirement isn’t going to significantly affect the numbers.


    • Smith says:

      Civilian noninstitutional population 250,663 253,397 (16 years of age and older)
      Civilian labor force 158,283 158,880

      These figures above from BLS site are reported in 1,000s so the actual numbers are
      Civilian noninstitutional population 250,663,000 253,397,000 (16 years of age and older)
      Civilian labor force 158,283,000 158,880,000


    • Smith says:

      Error. Correction of my data, and half the severity labor withdrawal, the demographics, the aging of the U.S. workforce is quite significant and may easily account for half the missing million workers (lower workforce participation rate over last 12 months). Data shows age 65+ going from 13 to 16% 2010 to 2020 or about .3 percent per year, and participation rate falls for over 54 – 64 years old at 64% (2014) 65 – 74 years old 26% (65+ is 19%) so a little more than 2/3 of .3 or .2% dropping out due to aging workforce.

      So at 150,000 in four months you’ve got the new .4% increase in workforce covered. Then another two months takes care of the .2% that dropped out last year. Then six months of gravy, adds almost a million more jobs. But fear not because in 1994 workforce participation was 66% compared to 61%, and population over 65 years was still 12.5% compared to 15% today. (66-61) – (15-12.5) = 2.5%
      But again that 2.5% at a minimum applies to total working age population of 253 million or 5 million. Is that correct? So you need another nearly 4 years with no recession to get back to where you were in 1994. Yet the unemployment rate in 1994 was 6% while the workforce participation rate was 66% and the population over 65 years was already 12.5%. What am I missing? Why was workforce participation so high while unemployment was 6%? (Workers beaten and downsized, globalized, rust belted, outsourced, and prematurely disabled?)

      Workforce participation by age.
      http://www.bls.gov/emp/ep_table_303.htm
      Percent of population over 65
      http://www.aoa.acl.gov/aging_statistics/future_growth/docs/By_Age_Total_Population.xls


  2. Bystander says:

    I don’t dislike Krugman. But I’m disillusioned by his political stance and by his willful ignorance of modern globalization. I don’t mean to hurt the man. I like him generally.

    Strangely I’ve never met him and I don’t expect to ever meet him or any of you. This is the kind of alienation that only the internet can provide.

    Is the Internet good or bad? So far we haven’t even scratched the surface of how it can be used effectively. So far it is almost pure entertainment.


  3. Bystander says:

    Good Economist: Joseph Stieglitz.

    Good national security advisor: Richard Clarke.

    What did they have in common?

    A tendency to smile when no smile was warranted. It is indicative of their willingness to contemplate and understand things that would make most people frown.

    That smiling tick is not nothing, rather it is something important…


Leave a Reply

Your email address will not be published.