You are here

An Alarm Goes Off Threatening The "Strong U.S. Jobs" Myth: Withheld Income Taxes Are Stalling

Of all the indicators that the Fed has presented to justify its rate hike mentality and to validate that the US economy remains on a growth path despite clear recessionary signals from both the manufacturing sector and the dramatic tightening in financial conditions in recent months, Yellen's preferred metric also happens to be the most lagging one: nonfarm payrolls and the unemployment rate, both of which supposedly signal the collapsing slack in the labor market and a jump in wages that has been "just around the corner" for years.

The problem is that when shifting away from lagging indicators of the labor market, to coincident ones, a starkly different picture emerges. The best example of this is when looking at the growth of federal income and employment tax withholdings, the broadest and most timely read on the health of the job market, which as Jed Graham writes, "has been sinking at an alarming rate."

While for most of 2015, tax withholdings rose at a rate of 5% or more from a year ago, on the back of job growth and gains in wages, commissions and other incentive pay, in recent months there has been a substantial dropoff in this key indicator.

As shown in the chart below, revenue inflows to the Treasury Department steadily slowed through the fall, bringing the annual growth rate down to just below 4% by the start of 2016. That’s when growth seemingly collapsed — to just 1.8% over the past five-plus weeks, from Jan. 11 through Feb. 16.

 

The problem with the jobs report is that it relies on statistically interpolated, and seasonally adjusted data from the Bureau of Labor Statistics, which not only has a significant revisionist history aspect being materially revised after a given period, but is also subject to clear political bias and huge "birth/death" assumptions, which correlate the growth in labor with the net creation of new U.S. businesses despite clear indications over the past several years  that there should be no net additions as a result of collapsing "dynamism" as the Brookings institute itself calculated some time ago and as we chronicled in August of 2014 in "4 Million Fewer Jobs: How The BLS Massively Overestimated US Job Creation"

 

On the other hand, the official Treasury tax-receipt data — which don’t come with a margin of error and aren’t subject to revision — are obviously at odds with the much more upbeat numbers reported by the Labor Department. January’s year-over-year payroll increase of 2.665 million, or 1.9%, along with a 2.5% gain in average hourly earnings should yield something in the neighborhood of 4.5% year-over-year growth in tax withholdings — or more than double the actual growth rate in recent weeks.

And yet over the past 10 full weeks, starting Dec. 7, tax withholdings have grown just 3.1% from a year ago. While December and January data can be influenced by the size and timing of year-end bonuses, the pronounced weakness has been sustained for long enough to rule that out as the principal cause.

As Graham notes, "companies that were already facing a tough earnings environment, thanks to a stronger dollar and lackluster economic growth, have seemingly pushed the pause button on hiring amid the financial market tumult that greeted the new year."

It is not just the Treasury tax receipt data that is a major concern: as we previously reported, according to Challenger there was a total of 75,114 layoff notices in January, up 42% from January 2015, and the highest January total since 2009 as Wal-Mart, Macy’s and Yahoo joined oil services firms Halliburton and Schlumberger in workforce restructurings. Wal-Mart on Thursday reported declining earnings and weaker-than-expected same-store sales.

 

In fact, it appears that the only place where the strong jobs myth persists is in official government data, and it's not only in the payrolls report: after rising to 294,000 in the January 16 week, new jobless claims have steadily fallen to 262,000. The four-week average fell 8,000 to 273,250 last week.

Graham further notes that "the benign data are hard to square with a stalling out of growth in federal income and employment taxes, unless the weaker receipts are more related to slower hiring, fewer hours of work and less incentive pay, than to layoffs."

The slower pace of year-over-year gains in tax withholdings has pointed to a significantly slower pace of hiring since September, says TrimTabs Investment Research, which estimates that 820,000 jobs were added from September to January. By contrast, the Labor Department estimates 1.137 million new jobs over that span, or nearly 40% more.

“The deceleration really started back last autumn,” said TrimTabs CEO David Santschi.

Which means that it is likely only a matter of time before either the BLS admits the truth, or the official data turn significantly south.

This also leads to the question whether the Fed have been looking at the wrong jobs data when it decided to raise rates in December for the first time in nearly a decade?

A partial admission was made in the January FOMC statement when the Fed said that "information received since the Federal Open Market Committee met in December suggests that labor market conditions improved further even as economic growth slowed late last year"

However, as noted up top, employment is one of the most lagging indicators, while tax withholdings are a coincident one, and traditionally signal job weakness before the official employment data catch up.

Will this be one of those occasions?

The answer will depend on whether the Fed wants to admit policy error and halt the rate hike process, perhaps with a view to unleashing more stimulus and even the increasingly more discussed "helicopter money."

Or perhaps the Bureau of Labor Statistics will merely take its cue from Australia which months after reporting stellar jobs data, admitted that the 6 and 8-sigma outlier October and November job additions were cooked and the result of "technical issues."

Will the failure of the US recovery likewise be chalked up to a "technical problem"? If so, it is very much unclear just what the market's reaction will be to a government whose data credibility is now the same as China's, even if it means far more stimulus in the near future.