Why jobs reports revisions happen and what they mean

Yahoo Finance's Bran Cheung details the context around revisions in monthly jobs reports from the Bureau of Labor Statistics.

Video Transcript


BRIAN SOZZI: All right, now let's get over to our very own Brian Cheung for a deeper look into the jobs report, Brian?

BRIAN CHEUNG: Hey, Brian. Yeah, well, we did get the headline number, 467,000 non-farm payroll adds in the month of January. That was a big surprise from some who even expected a negative print because of the impact of the Omicron variant. Unemployment clocking in at 4%, basically unchanged from the 3.9% we had seen in the month prior. And then wage growth coming in at 5.7% on a year-over-year basis.

But obviously, a lot of interest in just the revisions that were made to previous months as well. I want to highlight that there were two major adjustments that were made. First, they benchmark against data as an update to the quarterly count that they get on the unemployment insurance tax system. This is a way that the BLS can make sure that their data lines up with figuring out when people are leaving or entering the workplace, based off of their UI filings, as well as just regular seasonal fluctuations they do on a yearly basis to kind of smooth it over major holidays.

And you can see that they made revisions to all 12 months over the course of 2021, very dramatic upside but also downside revisions to over the month changes when you consider that some months for example, like July of 2021, as you see ahead of you, actually declining by 402,000 but then some sharp upward revisions like in the later months of last year, November revised up by 398,000. Again, that's in addition to what they had already reported, and then December by 311,000 as well. So there's a lot going on under the hood with that jobs report beyond that headline number, which surprised on the upside today, guys.

JULIE HYMAN: Yeah, and Brian, when you see the size of those revisions, like why do we pay any attention to the monthly numbers when the revisions are that-- I mean, are that huge in a typical year or was this atypical that we saw revisions of that size?

BRIAN CHEUNG: Yeah, this is a very important question. In my past life, I was an econometrics kind of person. I've kind of washed-up since then but basically what you need to know is that there is a large standard of deviation when it comes to the revisions that are made to the data that's reported on a monthly basis. And this isn't by any means any sort of an accusation that the numbers that are reported by the Bureau of Labor Statistics are wrong in any way.

It's just the way that data smoothing happens requires you to after the fact, go back and look at the data that's already been reported to take a look at any sort of unusual fluctuations. Ask the questions, well, why might those fluctuations be happening? And then take a look under the hood and see if you can actually see oh, actually, there was an unusual amount of seasonal hiring because of the holidays or we didn't properly take account for the unemployment insurance numbers that maybe tell us a little bit of a different story and that's the reason why they do this on an annual basis.

But it is indeed the case that even when we look at the singular report that we had gotten for the month of January, again being way above estimates and people are trying to back into narratives right now and there's no right answer as to why this was well above what the Street was estimating, is it because it covered the holiday season? Is it because of workers coming from somewhere else? Is it because of people that are working multiple jobs? Why is all this happening? Well, the jury's still out on that. But there will be econometric ways by which the BLS can look back on this data come January of 2023 to say oh, well, maybe we can revise those numbers.

So again it is indeed the case that you should take the headline number with a little bit of a grain of salt. However, trend-wise it's still telling the same story, which is that the labor market is still continuing to recover. The fact that it is a positive number in the first place still underscores that people are coming back into the labor force even though there was some question about whether or not we had reached maximum employment already.

When it comes to those that are watching the labor market, this is still a net positive. Even the labor force participation rate as we're showing you, ticking up. So all of these showing at least on a macro scale, people going back to work, employers able to hire, all positive things at least for the economic side of things.

BRIAN SOZZI: Brian, past lives? Weren't you born in 1992 you're always telling us? How many lives can you have?

BRIAN CHEUNG: I've done other stuff, Brian, you know.


BRIAN CHEUNG: In my past life maybe I was a real economist, I don't know.

BRIAN SOZZI: All right, fair enough. Get out of here. We'll talk to you later. Brian Cheung.