February 6, 2017

AP’s Rugaber Weakly Claims That Trump Has Inherited a ‘Healthy Job Market’

As far as the Associated Press’s Friday headline following the government’s January employment report is concerned, the matter is settled: “President Donald Trump Inherits a Healthy Job Market, with Solid Hiring.”

There are two elements to the headline’s claim, both of which are suspect. While the obvious one is the belief that the U.S. job market is genuinely healthy, the AP also effectively and misleadingly asserted that Trump has had no influence to this point.

This post will demonstrate that job losses — that’s right, losses — during the past four months have been significantly greater than usual. Additionally, readers will see, despite what the mostly lackluster to miserable payroll data showed, that the labor market, according to the government’s Household Survey, finally began to wake up after years of slumber in January.

Outlets on the left and right have missed the mark in covering January’s jobs report. Both the AP’s Christopher Rugaber and conservative outlets like Investor’s Business Daily gave too much respect to January’s seasonally adjusted payroll job additions. Rugaber, while he recognized a positive January impact that many others missed, failed to directly give Trump any credit for its occurrence.

In that regard, Rugaber’s report moved into flat-out contradiction in Paragraph 4 (bolds are mine throughout this post):

President Donald Trump has inherited a healthy-looking job market from his predecessor, with the U.S. economy registering a burst of hiring in January and an influx of Americans looking for work

U.S. employers added 227,000 jobs last month, the Labor Department said Friday . That’s the biggest gain since September, and it exceeded last year’s monthly average of 187,000.

Unemployment ticked up to a still-low 4.8 percent from 4.7 percent in December. But it rose for an encouraging reason: More Americans started looking for work last month.

The unemployment rate counts only those people who are actually trying to find a job. All told, more than a half-million Americans began looking in January, and the vast majority landed a job.

Rugaber should be congratulated for separating the BLS’s incorporation of the Census Bureau’s December 2016 population control adjustments from what actually happened in January per the government’s Household Survey, as seen here (red box added by me):


The “2017 population control effect” in the second column is really the result of the Census Bureau transmitting to BLS its 2016 findings about the U.S. population, most importantly that the nation had over 800,000 fewer adults than it had previously estimated. The third column represents something we haven’t seen in quite some time, namely that a lot of new people entered the workforce and found jobs. Though the figures are seasonally adjusted and have to be treated with some skepticism, as will be explained shortly, those are pretty impressive numbers.

But Rugaber was clearly reluctant to personally give credit to Barack Obama’s successor. He noted, and quoted, “Economists (who) said optimism may have played some role in lifting hiring last month.” One of them even credited “optimism related to the election.” This clearly indicates, contrary to AP’s headline and Rugaber’s opening paragraph, that at the very least Donald Trump’s election, along indications that he intends to keep the promises he made during the presidential campaign, had a positive effect on January’s numbers. Trump has obviously inherited a lot from the Obama administration (mostly baggage), but he also quite obviously influenced January’s Household Survey improvements. If Obama is to be credited with a positive influence, it would be that people knew that he would very soon be positively out of office.

Now let’s get to the AP’s alleged January “burst of hiring” as reported by employers, something which IBD and others also bought into. Contrary to what the Household Survey numbers tell us (which makes this exercise quite frustrating at times), the seasonally adjusted 227,000 figure for January payroll additions is clearly inflated.

To understand why outlets on both the left and right could be fooled, it’s important to explain how the BLS comes up with its estimates of payroll job changes every month. Each month’s headline number for job additions on company payrolls is the result of seasonally adjusting the actual amount of jobs added or lost in the context of actual (i.e., not seasonally adjusted) monthly results seen during the past five years.

Readers may be surprised to learn that certain individual months during the year typically have raw job gains of over 1 million, while others, such as January, routinely show raw job losses of well over 2 million. The BLS’s seasonally adjusted figures are supposed to indicate whether a given month’s underlying (and rarely highlighted) result was better or worse than others in the context of the past five years. Sometime those results accurately reflect the underlying raw data. Often, particularly during the past four months, they don’t.

The following table shows readers how weak three of the past four months have really been:


Looking at the four previous months:

  • The figures for actual jobs added in October 2016 (885,000 overall and 465,000 in the private sector), compared to previous Octobers, especially 2015 and 2014, were weak. My reviews of previous years’ seasonally adjusted values (not presented, but seen here) indicates that BLS was overly generous in its seasonal conversions, and that figures well below what the government reported, i.e., 80,000 jobs added overall and 90,000 in the private sector, represent the genuine strength of that month’s job market. October’s genuine strength was therefore pretty weak.
  • November 2016, on the other hand, was a relatively strong month, especially considering that previous years’ seasonally adjusted figures have been relatively high. BLS’s seasonally adjusted figures significantly understated that month’s strength, and I have estimated that the genuine strength was about 280,000 in each category.
  • December 2016 saw a significant return to weakness compared to four of the previous five years. The government’s official seasonally adjusted figures overstated November’s strength, which I have estimated as only 70,000 for total nonfarm and 90,000 in the private sector.
  • Job losses in January 2017 were extraordinarily high compared to most previous Januarys. Both nonfarm and private-sector losses were the second-worst since 2009, when the nation was still in the midst of the Obama administration’s “jobs recession” (the economy did not stop losing jobs until February 2010, eight months after the recession’s officially declared end). January’s seasonal conversions were ridiculously generous, as seen below:


    To cite just one comparative example, 132,000 more nonfarm jobs were actually lost in January 2017 than in January 2015. It’s absurd to pretend that January 2017, after seasonal adjustments, was actually a bit stronger (227,000 compared to 221,000). My review indicates that January 2017′s genuine strength was 130,000 overall, and 140,000 in the private sector.

Having explained each month, we can see in the two columns on the far right of the “Job Gains/Losses” table above that the final two months and four months of the Obama administration saw the worst actual (not seasonally adjusted) payroll jobs results in the past six to seven years.

Donald Trump has not inherited a healthy job market. The Household Survey results cited near the beginning of this post indicate that Trump’s election, his economy-related conduct during the presidential transition, and his imminent inauguration had a positive influence on January’s job market that hasn’t yet been reflected in reported payroll jobs.

These facts are contrary to what AP’s headline and Christopher Rugaber’s opening paragraph claimed. Sadly, many news consumers will only see the wire service’s false opening presentation, and will believe it.


No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.