This past week, the Bureau of Labor Statistics (BLS) issued a revision to the previously announced, and previously revised, change in nonfarm payroll numbers from April 2024 through March 2025. It seems, there were actually 911K fewer than it originally reported.
Obviously, that isn’t an insignificant number. In fact, it was the highest annual revision in history. However, in and of itself, I would argue this wasn’t as surprising as it might have seemed. Why? Well, the BLS makes a lot of revisions to its labor data. And to be honest, I’ve had my doubts about the monthly Employment Situation reports for quite some time.
As such, in a strange way, I felt the BLS’s admission of error, if we can call it that, to be a small measure of validation. Without getting into the weeds, the monthly data consistently didn’t make sense to me. Regional purchasing manager reports, business confidence surveys and even the vagaries between the BLS’s own Household and Establishment data sets all strongly suggested the labor markets were weaker than the headline numbers.
Clearly, this is important because I believe the jobs data impacts the future cost of capital in the United States, possibly more than other data sets. After all, the U.S. economy is consumer driven. When the economy creates jobs, it creates paychecks. When it creates paychecks, voila, it creates consumers. Obviously, the inverse is also true.
So, the need for accurate, or at least somewhat accurate, labor market data is crucial. It is the keystone to forecasting economic growth, inflation and, therefore, future changes to U.S. monetary policy. If the data the BLS provides is faulty, so too will be the decisions investors make regarding the cost of money in the U.S. financial system.
In essence, if the BLS doesn’t do its job reasonably correctly, the markets could misprice capital. Intuitively, this could lead to the misallocation of precious resources in the economy, and negatively impact the flow of money to its highest and best use.
No, I am not being hyperbolic.
- In its annual ‘big’ revision in 2024, the BLS expunged 818K announced payroll jobs. Poof! They didn’t really exist.
- The previous year, 2023, the number was 306K.
- As such, over the last 3 years, 2,035K previously announced payroll jobs have dissipated into the ether.
Meanwhile, over the last 12-months, the official “labor force participation rate” has fallen from 62.67% in August 2024 to 62.33% in August 2025. Since the BLS doesn’t count you as unemployed if you aren’t participating in the labor force, this is extremely important. HAD the participation rate been 62.67% in August 2025, and the numbers of jobs held constant, the official Unemployment Rate would have been 4.85% instead of 4.32%.
Who knows? Would the Federal Reserve have already cut the overnight rate at least once if the markets had had the correct jobs data? If so, how would that have impacted corporate profitability, capex decision making and asset class returns?
Then there is the whole inflation thing. You know the Consumer Price Index (CPI)? The most commonly used inflation gauge—even if it isn’t necessarily the Fed’s favorite one? Well, for some reason, the BLS calculates the CPI as well.
So, the Fed has a dual mandate: price stability and full employment. The BLS is responsible for calculating both. What could possibly go wrong?
As for the CPI itself, the BLS announced this past Thursday that inflation increased 0.4% in August. Over the 12-months ending on 8/31/2025, it estimated prices climbed 2.9% in aggregate.
The question remains: Did they?
Well, jewelry shot up 6.8% for the month. Did you spend 0.143% of your budget on such things? Conversely, linens were down 6.7%, and they make up 0.133% of the index. Hey, don’t forget your smokes, which make up 0.488% of your outlays. They apparently set you back an additional 1.0% in August.
I think you get the picture.
There is absolutely no way possible for the BLS to accurately calculate inflation in the U.S. economy. None. There are simply too many products and services. They are too many retail, wholesale and e-tail distribution channels. They are too many regional differences.
How does the BLS account for the price differential between premium, discount and generic brands & products? BOGOs, manufacturers’ coupons and other specials? Daily price changes?
At best, the BLS can only be somewhat directionally accurate. 0.4% in August? That might be somewhat close to the truth in August 2025, perhaps, in aggregate. But what if you didn’t buy, say, a diamond ring, cigarettes or a new sewing machine? Perhaps you didn’t stay in a hotel or have a mechanic work on your car. Maybe your family doesn’t eat beef, tomatoes or drink instant coffee.
If that is the case, you might have even had some deflation in your household budget. But, if you DID buy all of those things, as well as some “men’s pants,” brother, you bled some red ink last month. At least that is what the BLS would have you believe.
Do you think it really captured all of that data correctly? From Miami to Anchorage? From Honolulu to Boston? And all points in between? Is the CPI the same in Scarsdale as it is in Clio, Alabama? Or how about Beverly Hills and Welch, West Virginia? Shoot, do they even consume the same stuff in those areas?
I mean, I would be willing to bet a plug nickel that grocers sell more cans of potted meat in Clio than they do in Scarsdale. On the flipside, I doubt the demand for, say, a 2018 Silver Oak Napa Valley Cabernet Sauvignon is as great in Welch as it would be in Beverly Hills.
In fact, I suspect the demand curve for that bottle of wine in Welch would be almost all the way over to the Y-axis, if not all the way. A little graphical humor for you, pun intended.
So at best, the most the BLS can hope for in calculating inflation is to be directionally accurate. There is absolutely no way it can be precise. Counting all the jobs and all the prices in an economy as large and dynamic as ours? That is about as impossible as it gets.
Yet, is that how the Fed sets the overnight rate? How the markets price capital in the economy? With inaccurate data? It seems kind of crazy when I put it like that, doesn’t it?
The thing is, things have always been this way. The data didn’t just start to be wrong overnight. The methodology was never meant to be 100% precise. Everyone understood the numbers were “good faith” estimates. Folks got the enormity of the situation, and didn’t expect perfection. The data just had to be “close enough for government work,” if you catch my drift.
So, what changed? Why the cynicism these days over the releases? Why the hue & cry over the veracity of the labor market data, or the lack thereof? Fair enough.
Two things:
- First, according to Pew Research, the public’s faith in public institutions is near an all-time low. In an article released on October 17, 2024, Pew reported: “Only 22% of U.S. adults said they trust the federal government to do the right thing just about always or most of the time.” Guess what? The BLS is part of the federal government. Ta-da.
- Second, there is the simple fact that the dollars are enormous in aggregate. I am talking about the movement of massive amounts of cash across the financial system and economy. The potential for gain or loss is simply enormous. This isn’t a game, and ‘close enough’ only counts in horseshoes and hand grenades.
As such, I strongly suspect the administration will put enormous pressure on the BLS and other agencies who disseminate economic data to improve their methodologies. The types of revisions we have seen recently suggest there is a need for change in how they collect and calculate their data.
This stuff is simply too important.
Thank you for your continued support. As always, I hope this newsletter finds you and your family well. May your blessings outweigh your sorrows on this and every day. Also, please be sure to tune into our podcast, Trading Perspectives, which is available on every platform.
John Norris
Chief Economist
Please note, nothing in this newsletter should be considered or otherwise construed as an offer to buy or sell investment services or securities of any type. Any individual action you might take from reading this newsletter is at your own risk. My opinion, as well as those of our Investment Committee, is subject to change without notice. Finally, the opinions expressed herein are not necessarily those of the rest of the associates and/or shareholders of Oakworth Capital Bank or the official position of the company itself.