Friday’s jobs report surprised everybody, including me. The headline number came in shockingly below the consensus expectation. The BLS counted just 20,000 new jobs in February. The consensus guess of economists was for a gain of 173,000.
I was especially surprised because February withholding tax collections had gone through the roof, with a year to year gain of nearly 10%. That suggested not a downside surprise in the jobs survey, but just the opposite.
Now we all know that the BLS manipulates the number through the magic of seasonal adjustment. But a miss this big? In the opposite direction of what the tax collections told us they should be?
It’s preposterous. So I took a look at the actual raw data and what I found will make you think twice about the conventional wisdom.
When you look at the chart of the raw, unadjusted survey data, what do you see? Railroad tracks. Nothing has changed. The trend is intact.
Digging into the data we see a couple of things that go part way to explaining why what the BLS reported was so far below what the withholding tax collections implied.
The first thing I noticed was that the raw data for January was adjusted upward by 102,000. I have never seen such a large adjustment to the raw data. The BLS apparently missed a whole bunch of jobs in their initial count in January. So it appears that February was terrible because January was so fabulous after the fact.
If we know one thing, it’s that hiring managers follow the stock market, and January was a helluva month for stocks. Apparently, the BLS just missed that in their initial survey of employers in January.
The raw, not seasonally manipulated, data now says that there were 827,000 new jobs in February. That’s a big number, but February always has a big increase.
We can tell if it’s good or bad by comparing this number to past Februarys. It’s not boffo, but it’s not terrible. Since the recovery was in full swing, starting in 2010, 3 years have been materially worse, and 3 were about the same. Only February 2018 and February 2017 were materially better. On the basis of the raw data, this is a nothing burger. The seasonal manipulation factor made it look far worse than it was.
The annual growth rate come down, but that was after the upwardly revised surge in January. The annual growth rate of 1.7% is right in the middle of the growth rate range of the past 2½ years. There’s nothing happening here.
Given the massive bulge in February withholding tax collections, we’d have to guess that maybe, just maybe, this February number will also see a big upward revision come the next release in early April.
Not only is the big jobs slowdown an illusion, there was another data point in Friday’s jobs report that should concern us. That’s wage inflation. Average hourly earnings of production and non-supervisory workers are soaring. The wage inflation rate was 3.5% year over year in February. As you can see, the trend has been accelerating.
And that’s especially bad news because CPI is being underreported. The headline number in January showed an annual inflation rate of just 1.5%. The problem is that when reported CPI inflation has lagged, it has historically always caught up with wage inflation.
The implication is that CPI will catch up with wage inflation. A 3% plus CPI print may be in our not too distant future.
More important than whether the manipulated BLS CPI data catches up with the manipulated BLS wage data is the reality that real market Treasury yields historically have correlated with wage inflation data.
This suggests that the miraculous rally in Treasuries since December may not be long for this world. We should keep an eye on this. If wage inflation is sticky around 3.5%, then it may not be long before we see the 10-year Treasury yield is at a like rate. It’s just another reason why I don’t like the idea of buying or holding long term Treasuries yet.
There are a quarter trillion other reasons here in March, why I especially don’t like Treasuries, or stocks for that matter. The US Treasury is sucking that much cash out of the financial markets this month to pay its bills. It’s a gargantuan figure.
$66 billion of that has already been issued and settled. We have seen stocks come under pressure because there’s no Fed QE to help bond buyers absorb all the new Treasury issuance. They either must borrow the money to pay for them, leveraging up, or sell something else. They’re doing both, and what they are selling is stocks.
Bond buyers have been encouraged by the weak, but likely temporary, “bad” news on jobs. Motivated by that false narrative, most investors and dealers prefer Treasuries to stocks. Treasuries have been well bid and Treasury prices have been relatively firm. Stocks have taken the brunt of the liquidation this month.
Another $60 billion of net new Treasury issuance settles on March 14 and 15, in what will be another huge absorption test for the markets. That still leaves a TBAC forecast $120 billion to settle over the rest of March.
It’s not good news. So regardless of the official and media spin on jobs, the reality of tight liquidity will put relentless pressure on the markets for the remainder of March. I like my chances continuing to hold short term T-bills, where rates should continue to see upward pressure.