Tuesday, August 16, 2016

Why You Shouldn't Believe the US Jobs Data

Earlier this month, the Bureau of Labor Statistics (BLS) released the latest jobs report data for July, showing that the US economy created 255,000 jobs. This was viewed as great news, and the financial markets and media seemed to breathe a collective sigh of relief.

That's because everyone was afraid it would be a lot worse.

You see, back in May of this year, the jobs report was nothing short of abysmal. After revisions, the number for May came in at just 24,000 jobs. This came on the heels of dramatic market instability in the first quarter of this year and a series of declining jobs reports preceding it:

  • February (still normal): 233,000 jobs increase
  • March (getting weaker): 186,000 jobs increase
  • April (weaker still): 144,000 jobs increase
  • May (falls off a cliff): 24,000 jobs increase
This trend seemed to be going in one direction, and it wasn't a good one. Thus, everyone was ecstatic when jobs shot back up in June by 292,000 (after revision). This positive sign was seemingly confirmed in July when the economy added 255,000 more jobs, beating all the expectations. Additionally, the hourly earnings for July were reported to increase by some 2.6% year-over-year.

Assuming all of this is true, these latest data points are certainly great signs for the US economy. But there's good reason to think the jobs data is just wrong.

Why does this matter?
It may not be obvious why anyone should care about the total number of jobs in the economy, provided your own job is still part of the total. As the old saying goes, it's a recession when your neighbor loses his job. It's a depression when you lose yours.

And in a sane world where the government did not take responsibility for managing the economy, the jobs data would not matter. In the world in which we actually live, however, it is one of the most important drivers of US fiscal and monetary policy. If the jobs data is too low, that means more spending by the US government, lower interest rates from the Federal Reserve, or both. Both types of policies have negative long-run consequences.

The jobs figures are also the primary numbers used by politicians to either celebrate or criticize the state of the US economy at any given moment. For this reason, we can expect them to make an appearance in the upcoming presidential debates. Hillary Clinton, running for a de facto third term of President Obama, will highlight the numbers if they remain strong through November. Meanwhile, we should expect Donald Trump to call them out if they weaken significantly.

So what does the other data say?
There are several other data points that point in the opposite direction of the jobs numbers. Here are a few of the most important ones:

GDP is basically stagnant
Gross Domestic Product (GDP) never makes the list of talking points when people want to praise the economic recovery during President Obama's tenure. That's because it's pretty awful by historical standards. Two charts make this point quite well.

First, here's a chart showing annual GDP growth for all years between 1950-2014, ranked from least to greatest. Years under President Obama's leadership are shown in red.

Second, here's a more zoomed in look at just the last several quarters. As you can see, things have recently been even worse than normal:

Now, as economic statistics go, GDP probably gets more emphasis from economists than it deserves. And it surely involves many complex calculations and assumptions to compute it each quarter. That said, it still conveys some information.

And right now, the GDP data is telling us that the US economy is still technically growing, but at a rate that is very weak by historical standards, just 1.2% in the latest quarter. It's not a recession; it's just not very good.

Corporate profits are falling
Even more telling than the GDP data is the dramatic decline in corporate profits over the last several quarters. US stocks may be near all-time highs, but their performance has had no connection with the performance of the underlying companies.

Here's the year-over-year change in corporate profits since 2000. All the data hasn't come in for Q2 2016, so here's how things looked through Q1:

As you can see, corporate profits have actually been declining (below 0 on the above chart) on a year-over-year basis for the past five quarters. This decline is expected to persist through Q2 2016 based on the earnings information that has been reported so far.

US tax receipts are basically flat compared to last year
Another bad sign comes from the tax receipt data from the US Treasury Department. For the first seven months of the year, total tax receipts are essentially flat compared to last year, down roughly $20 billion.

Meanwhile, if we look at receipts from income tax withheld--the amount pulled directly out of your paycheck by your employer--these figures are up just 1% year-to-date. Those gains were offset by declines in corporate taxes as well as other non-withheld income taxes.

Given that US tax policy has not radically changed since 2015, the slight decline in tax receipts seems to be driven by deteriorating economic activity overall.

Back to jobs
Bringing all of this data back together, we have the following:
  • GDP growth is hovering around 1%, well below what anyone thinks is a normal level of expansion.
  • Corporate profits have declined for 5 going-on 6 consecutive quarters now
  • US tax receipts have actually declined slightly for July YTD, relative to the prior year
Of these, the latter two data points are particularly striking because they don't rely heavily on assumptions. Corporate profits are calculated according to established accounting principles and subject to regular audits, and one assumes (hopes?) that the Treasury Department knows how to count up all the money it receives.

All that is left is to ask some obvious questions:

If corporate profits have been steadily declining, is it likely that they are going to be creating a lot of new jobs in the US?

The number of jobs in the economy is increasing each month at an year-over-year rate greater than 1.5%, and hourly wages reportedly increased at more than 2% in the latest report. So officially, more people are working, and on average they are making more money. Given this, and the fact that we have a progressive income tax, how is it possible that income tax withholding is only up 1%?

Simply put, these are facts that do not fit together. At least one of these numbers seems wrong.

And basically, we have three core data points to indict: corporate profits, tax receipts, or the BLS jobs data. 

If corporate profits are the source of our problems, that would mean many corporations have artificially understated their profits. This seems unlikely.

If tax withholding data is wrong, that would mean, in essence, that the Treasury Department hasn't figured out how to use a summation formula. I try not to put much faith in the competence of government officials, but this seems like too much of a stretch.

Or, perhaps the BLS's assumptions for estimating job creation in the economy were wrong. The BLS samples a relatively small number of businesses and relies on survey data. It then uses complex assumptions to extrapolate the results into an estimated figure for the overall economy. Clearly, this is the data point that has the largest chance of error.

And in fact, they've been wrong before. Just last week, they announced major revisions to past estimates of wage increases, causing substantial declines to previously published results. But the July jobs data remained unchanged.

Based on the analysis above, it's our expectation that there will be more dramatic revisions to come in the future. Until that time, it's probably best to treat any and all optimistic pronouncements on the US economy with a very large grain of salt.

No comments:

Post a Comment