Tomorrow's payroll report will be a tricky one to interpret. The hurricanes that hit the Gulf of Mexico area will cause the headline job-gains figure to be far below its fundamental value. But here are a few thoughts on what we should expect and what it means.
Non-Farm Payroll Could Be Any Number, It Doesn't Matter
Seriously. It could even be negative (although I doubt that). It makes no difference. For background, to estimate the non-farm payroll (NFP) number, the Bureau of Labor Statistics (BLS) calls companies and asks how many people actually got paid that week (called the Establishment Survey). So someone could still be working for the firm, but didn't come in that week because of the storm, and it looks the same as a layoff.
Equally important, there are a huge number of hires and fires every month. Recently, both hires and "separations" (fires plus quits) have been running around 6 million. Having that much job churn in the economy and then boiling it down to a net gain of around 200,000 leaves a lot of room for random fluctuation to cause error in the best of circumstances. Throw in two Category 5 hurricanes, and all bets are off.
The ADP report that came out Wednesday was about 70,000 below the prior trend. So as back-of-the-envelope, we could add 70,000 to the reported NFP number. The Household Survey includes a question asking if you did not go into work due to weather. That will also give us a clue, but the error rate remains too high to draw any strong conclusions no matter what the number.
As far as the Fed is concerned, it too will ignore this headline figure. We know the next chance it might hike rates isn't until December anyway, and there will be two more payroll reports by then. The weather-related noise will sort itself out.
Wage Growth Might Also Be Distorted, but Maybe Not as Badly
The wage measure released with this employment report comes from the Establishment Survey. However, unlike the headline payroll figure, the hourly earnings figure is calculated on a per-hour basis. This has been an issue in the past when the total number of hours worked was relatively low. Sometimes average earnings have been overstated. In theory, if you were paid for a full week's work, you should count as having worked a full week, so in theory there shouldn't be much of a distortion. But we have seen weather create these kinds of distortions in the past.
We have some catching up to do if we want to match last year's +2.85% wage number. The last four months of 2016 (i.e., the ones that will be rolling off) came in at a faster pace than more recent figures (2.64% annualized pace vs. 2.38% annualized so far this year). Unfortunately, the seasonal pattern points the other way. The last three years, September-December has come in slower than the rest of the year.
Despite the potential for distortions, the market will very much focus on this wage number. If it comes in a bit higher than expected (which, again, might happen just because of weather-related noise), expect the bond market to sell off quite a bit.
How Might We Get a Better View of Job Creation?
The aforementioned Household Survey, in theory, is a cleaner gauge, since it involves calling people at home and asking if they have a job. Ostensibly, people who consider themselves employed would say yes even if they didn't get any hours that week. Unfortunately, the Household Survey is notoriously volatile. It is near worthless for just one month's estimate.
However, all is not lost. If we look at various ancillary views of job creation, all look quite healthy. The ISM surveys, a measure of business confidence, both show hiring intentions to be quite strong. On the manufacturing side, hiring expectations are at a five-year high. The NFIB survey, a small-business confidence survey, also puts hiring intentions at nearly a five-year high. And while the 3% GDP truth-ers need to calm down, it does seem very likely the economy accelerated a bit in the third quarter and will probably keep that momentum up in the fourth. All that suggests hiring should remain robust.
Fed Consequences and Trade Ideas
Wage growth remains a mystery. It should be stronger given the general employment situation. Although I can't show hard evidence of this, I think the demographics shifts from older workers to younger workers is part of the story. Regardless, I'm comfortable saying that unless wage growth actually decelerates meaningfully, the Fed will stay on track for December.
I have two trades on playing on this specific view. I'm long the dollar through the U.S. Dollar Bullish Fund (USDU) ETF. I'm also short the six-year part of the curve through Treasury futures. I have OK gains in both, but I'm leaving them on at least until a December hike becomes 100% priced in (currently about 70%). I feel both trades retain some upside optionality in the event Kevin Warsh becomes Fed chair. I still think he's the most likely candidate to prevail, although I wouldn't count out either Gary Cohn or Jay Powell. From where we sit, I'm comfortable with all three (or even a reappointment of Janet Yellen) while holding these trades, although if Warsh became more priced-in, I might close out.
(This commentary originally appeared on Real Money Pro at 10 a.m. ET today. Click here to learn about this dynamic market information service for active traders.)