Those noisy payrolls figures

By Felix Salmon
February 7, 2014
Betsey Stevenson, and helps to show just how noisy the payrolls data really are.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">


The chart of the day comes from Betsey Stevenson, and helps to show just how noisy the payrolls data really are. The big headline figures of the day, 113,000 is ostensibly the increase that we saw, in January, in the number of people on American payrolls. It’s a disappointing number, while a print of say 200,000 would have been decidedly encouraging.

But just look at how we got to that 113,000 figure. We took January’s workforce, of 135,396,000 people, and then subtracted December’s workforce, of 138,266,000 people — for a total decrease of 2,870,000 jobs. But we know that the number of jobs in America always decreases in January — even when the economy is surging. It’s cold out, making outdoor jobs very difficult to do, and the Christmas seasonal jobs are all in the past. So the BLS institutes some seasonal adjustments. In this case, it subtracted 880,000 jobs from the December number, and it added 2,103,000 jobs to the January figure.

All of which means that the 113,000 headline figure is, in fact, 135,396,000 + 2,103,000 – 138,266,000 – 880,000.

You want to trade on that being 70,000 jobs lower than you thought it would be?

But wait: we’re not even close to being done. This month’s payrolls release is much longer than normal — 2,465 words — because it has to explain a lot of changes. As it says in a big box at the very top of the page:

Changes to the Employment Situation Data

Establishment survey data have been revised as a result of the annual benchmarking process and the updating of seasonal adjustment factors. Also, household survey data for January 2014 reflect updated population estimates.

These changes are not small: last month’s preliminary number, for instance, was revised up — on a seasonally adjusted basis — to 137,386,000 workers from 136,877,000. That’s a difference of more than half a million people.

The noisiness of the payrolls report is good news, truth be told. Now that the taper is well under way, there’s very little doubt about the direction of monetary policy for the next year or so. We’ll taper all the way to zero, QE will be over, and then we’ll look at where we are and start wondering whether and when rates might actually start rising. The employment situation when QE is finally over will have almost nothing to do with what happened this month, or next month, or the month after that. Most importantly, it will have to do with the number of people actively looking for work: as the unemployment rate comes down, and the economy continues to grow, will discouraged workers start returning to the workforce, or at least start looking for work again?

There are a lot of unemployed and underemployed workers on the sidelines of the economy, who would work much more if work was available. The Fed’s full-employment mandate means that it’s Janet Yellen’s job to find work for those people. How she’s going to interpret that mandate is something we’re not going to get a real hint of for a long time yet. But one thing’s for sure: we’re not going to be able to guess anything useful by looking at today’s payrolls report.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

So here’s a question: why do people focus so hard on the month to month jobs numbers if they are essentially noise and people in the know should know it? And how is the U2 number derived in that the loss of nearly 3m jobs can equate to a drop in the rate by 0.1%? (Honestly – no sarcasm intended…)

I’ve read BLS reports before and not really seen (or perhaps *gulp* understood) exactly how they derive their numbers. Maybe they should publish more openly these raw figures and let the public at large have a better understanding of what’s going on, because all we’re gonna get is politicized slant (from both sides) and very little explanation/understanding.

Posted by CDN_Rebel | Report as abusive

I think it’s a bad apperception to think “it’s Janet Yellen’s job to find work for those [unemployed & underemployed] people.”

Posted by dedalus | Report as abusive

Felix, great that you’re pointing this out. The press release also discloses the margin of error which tells the same story. It’s sad that the mainstream press is ignoring this.
We really shouldn’t be dumbfounded by why Wall Street trades on the “noise”, it isn’t that surprising. You need variability (i.e. “noise”) to make money. If it’s easy to predict what the number is, then the profits would disappear.

Posted by junkcharts | Report as abusive