Are The Jobs Numbers Fake? Trump Fires Head of BLS

President Trump called into question the integrity of our economic data when he fired the head of the Bureau of Labor Statistics after the latest jobs report in the beginning of August. In today's video, we're going to look at how we can tell if the employment data has really been rigged. Let's get into it. Welcome to Real Estate Watch, the channel that focuses on forward-looking real estate and economic analysis. My name is Eric, professional economist and licensed real estate agent. For my regular viewers, if you're wondering what happened to all my hair, I cut it off. Just as I rely on data to inform my analysis in these videos, my thinning hair as I grow older was a strong indicator that I should stop pretending I have a full head of hair and do something about Speaking of data, that's what we're talking about in today's video. After massive downward revisions to previous month's job growth numbers, President Trump fired the head of the Bureau of Labor Statistics, or BLS, suggesting that the new job numbers were not accurate. Make no mistake about it, the downward revisions presented in this latest jobs report were quite concerning, if accurate. Last month, the BLS reported that non-farm job growth came in at 144,000 in May and 147,000 jobs in June. In the report released in August, those numbers were revised down to just 19,000 in May and 14,000 in June. In a labor market the size of the U.S. is, that's practically no job growth at all. And it wasn't just a one-off event, it was two months in a row. And the July job growth figure was just 73,000, which is very weak job growth. There are two possible explanations for these numbers. The first is that the numbers are just highly inaccurate, as President Trump has suggested. The second is that something is happening in the economy that is causing a massive slowdown in hiring. In an upcoming video, I'll be doing a deep dive on the latest economic data, not just the job number, so I'll deal with that possibility then. Right now we're considering the possibility that the job numbers are highly inaccurate, maybe even rigged, and how we can know for sure. There are actually two different data sources we get from the BLS on employment growth, and one is much more reliably accurate than the other. The monthly job growth numbers that we hear about in the media each month is from what's called the Current Employment Survey, or CES. The BLS surveys about 121,000 business establishments each month in order to produce these job numbers we hear about. Now because it is a survey, it is subject to a margin of error. The BLS attempts to minimize estimation errors from the survey by making revisions each month after they have more data. This is where things got a bit crazy this month. The other employment data source, which is much more accurate, is called the Quarterly Census of Employment and Wages, or QCEW. Like the name implies, it's a census and not a survey. The QCEW uses unemployment insurance data reported by employers, which covers 95% of all jobs in the U.S. It's the closest thing we have to hard data on how many jobs there actually are. Now you might be wondering, why not just use that and get rid of the CES altogether? Well the problem with the QCEW is it's very lagged. As of the end of August, we only have QCEW data through the fourth quarter of 2024, so it's not as useful for gauging the current state of the economy. That is why we have the CES and the monthly job numbers that try to give a more current snapshot of the labor market. But we can still use the QCEW data to get a sense of how reliable the CES data is. Let me show you what I mean. Here's a graph of non-farm employment from the QCEW data and what we get from the monthly CES jobs report. The first thing you'll notice here is that the CES non-farm data from the monthly jobs report is consistently higher than the QCEW data. This is due to what I mentioned before that the QCEW data covers 95% of all jobs, whereas the CES data is intended to estimate all payroll employment in the U.S. Now just looking at the level of jobs doesn't help us that much with assessing the current accuracy of the CES data. So let's look at it in terms of year-over-year job growth. Here are the two series over the last five years. Looks pretty much the same, right? The reason for this is that once a year the BLS revises the CES data based on what is happening with the QCEW data. This is why the job growth is practically identical in their history. The next annual revision will be on September 9th, so it's been almost a year since they've done this. If we zoom in on what has been happening since their last annual revision, we find something interesting. Since December of 2023, we see that the annual job growth from the QCEW has been consistently lower than the CES. This tells us that the CES, for some reason, has been overstating job growth relative to what is actually happening. This is why we're seeing downward revisions to the CES data, and it suggests that when the BLS does their annual revision to the CES data, we are likely to see some big downward revisions for several months over the last year or so. What this all means is that the big downward revisions we saw for the May and June jobs numbers that got the head of the BLS fired were probably not rigged. The question remains, however, is whether they should have been revised down so sharply to show practically no job growth. The definitive answer to that will come on December 3rd. That is when the QCEW data for the second quarter of 2025 will be released. It's a long time to wait, but we will get a clear picture on what actually happened in the labor market for May and June at that time. If you like this kind of data analysis and want to see more of my videos, make sure to do all the YouTube stuff. Subscribe, click the notification bell, and hit that like button down below. Thanks for stopping by. I'll see you in the next video.