Q&A of the Day – Can You Trust the Government Jobs Report?
Each day I feature a listener question sent by one of these methods.
Email: brianmudd@iheartmedia.com
Social: @brianmuddradio
iHeartRadio: Use the Talkback feature – the microphone button on our station’s page in the iHeart app.
Today’s Entry: @brianmuddradio Do you buy the government jobs #’s? The suddenly great report seems convenient for generating positive headlines in advance of the election.
Bottom Line: You’re no doubt familiar with the saying, garbage in garbage out. Longtime listeners may recall that once upon a time, while serving as a financial analyst, I modeled and projected monthly jobs reports along with overall economic growth. For several years the estimates I provided proved to be highly accurate. This was true heading into the Great Recession, and also as we began to turn the corner coming out of it. But then, starting in late 2011, the dynamics began to change considerably. The methodology I’d used with success for years would wobble from highly accurate one month to way off the next leaving me to wonder where I’d gone wrong. In time I came to learn that it wasn’t that my modeling was off, it was the initial jobs reports, and the initial GDP reports, were often highly inaccurate when first reported. When all revisions to government reporting had come in, the modeling was still accurate, but because that takes a minimum of three months to account for (and up to a year potentially), I gave it up because the government data I would use to assist in modeling had become highly unreliable.
I offered up that backstory for the purpose of answering your question directly. No, I don’t buy the initial reporting by the Bureau of Labor Statistics in their monthly Employment Situation Reports, but then again that’s been true for about 13 years now. What I pay the most attention to is the ADP Private Sector Jobs Report as it’s historically proven to be the most accurate.
Let's start with the known differences between the two reports:
- ADP Report: Measures exclusively private sector jobs
- BLS Report: Measures public & private sector jobs
So, the BLS figures include government jobs that aren't factored into the ADP Report. If all other factors were equal there’d always be a natural variance to a certain extent due to the government report factoring in government job changes not factored into ADP’s reporting. Public sector employment accounts for approximately 15% of all US employment and 13.5% of overall employment in Florida specifically. Therefore, variances in the neighborhood of 15% or so between the two reports could be explainable without additional factors coming into play. But again, that’s obviously not at all the explanation regarding whatever the heck happened with jobs in January. So now let’s look at methodology.
Methodology of the ADP Report:
- Actual, anonymized, ADP data
Methodology of the BLS Report:
- Call out surveys to HR professionals sampling changes that's then projected by BLS economists.
So, what's the difference in these methods? ADP processes the HR and payroll data for over 14% of Americans in the workforce. They’re far and away the leader in payroll processing which is why they produce monthly reporting that’s used by economists to gauge real-time conditions in employment. The starting point for the ADP Report is actual data covering a 6th of the entire US employment pool. Moody's then uses the data, taking into account sector and regional variances, to project the overall job changes monthly. The BLS uses call out surveys that's then used to project the result. It’s for this reason that the initial reporting by the BLS has commonly proved unreliable independent of other factors. Rather than starting with actual data, they’re using estimates provided by HR professionals about changes within their organization. If the HR professionals provide incorrect estimates to the government – you get the garbage in-garbage out effect. The BLS’s methodology is analogous to political polling. It’s a sample, and done well can be highly accurate, however there’s inherently a margin of error, prior to projections, that’s not present within the ADP Report.
The final piece of variance comes down to timing. The ADP employment data that’s used for the reporting is collected the final week of the month. Given the extensive callout surveying that must be done, the BLS begins sampling early in the month. The average date data was collected was the 12th. That’s a difference of about two and a half weeks between the two reports. Some months that doesn’t matter much. That takes us to revisions in data.
Government revisions have been absurd. All that ever gets reported in non-financial news media is the initial number each month and then it's on to the next. That’s a huge disservice, in part because the BLS has routinely been providing a huge disservice with their initial reporting. Not only are the BLS’s monthly jobs numbers revised monthly, but they’re so unreliable they’re also revised annually. It made news in August when the BLS acknowledged that they’d overstated the previous year’s jobs gains by 818,000 jobs over the prior 12 months ending in March. That was an incredible indictment of the agency’s credibility in providing jobs data. Every month had already been revised three times, to theoretically correct the initial reporting, and yet several months later the government came back and said, oh yeah, we were still off by 818,000 jobs or an average of 68,000 per month? That’s complete absurdity.
Numerous studies have been conducted over the years comparing revised BLS jobs reporting for private sector jobs to the ADP Report. What’s been proven time and again is that the two reports end up in near-identical places once the government’s done revising their numbers. This is to say that well before the Biden-Harris Administration the ADP Report has been the far more accurate initial report between the two. This takes us full circle to your question... Is something fishy going on here?
The recent trend had been considerably weaker job growth. The ADP Report for the third consecutive month reflected that trend in their monthly report showing 143,000 jobs added during the month. The government remarkably produced a report that suddenly seemed to reflect a surge in hiring to 254,000 jobs, plus positive revisions to prior months creating a net number of over 300,000 for the month which served to push down the unemployment rate. The government only showed government jobs accounting for 31,000 jobs during the month – meaning that the disparity between the two reports was nearly 150,000 new jobs during the month. History has told us which of these numbers is most likely to be closest to the truth.
Now, does this mean the government was initially attempting to prop up the Biden-Harris administration's economy and generate positive headlines in advance of Election Day? I don’t know. But you don’t have to be a conspiracy theorist not to trust Friday’s jobs report. You only need to be an informed person. The government hasn’t been especially accurate in reporting jobs numbers in well over 13 years.