---
title: 'Labor Market Fragility: Why the Unemployment Rate Is Lying to You'
source: 'https://youtube.com/watch?v=Rlj9JBrfGCU'
video_id: 'Rlj9JBrfGCU'
date: 2026-09-02
duration_sec: 1094
---

# Labor Market Fragility: Why the Unemployment Rate Is Lying to You

> Source: [Labor Market Fragility: Why the Unemployment Rate Is Lying to You](https://youtube.com/watch?v=Rlj9JBrfGCU)

## Summary

Kevin from Meet Kevin analyzes the latest ADP employment report and the broader labor market, arguing that a declining labor force participation rate is masking underlying job losses and creating fragility that could be exposed by future interest rate hikes. He explains the mechanics of how a falling participation rate can lower the unemployment rate even as jobs are lost, and warns that a normalization of participation could trigger a sharp spike in unemployment and market sell-offs.

### Key Points

- **ADP Report Misses Expectations** [00:00] — ADP employment numbers came in at 38,000, below the average estimate of 57,000 and Deutsche Bank's 65,000. Kevin notes this is consistent with recent weekly gains of $8,000-$10,000, so he doesn't see it as a big deal.
- **Labor Market Fragility** [00:41] — The labor market appears stronger than it is due to an underlying risk: the more the labor market moves in a certain direction, the more fragile it becomes. This is tied to the labor force participation rate.
- **Uber Layoffs 10% of Staff** [01:07] — Uber is laying off 10% of its staff, partly a post-COVID correction after a surge in hiring. Kevin notes some trimming is normal, but combined with other factors, it's worth watching.
- **ADP Choppy Environment** [03:30] — ADP itself describes a choppy environment. AI construction boom and leisure/hospitality spending are positives, but job creation is at the slowest pace since January.
- **False Impression of Labor Market Boom** [04:01] — A hiring boom between March and May was driven by manufacturers building inventory in anticipation of inflation (fear of war with Iran). This gave a false impression of a booming labor market, which has since stabilized at ~10K jobs per week.
- **BLS Forecast for Friday** [05:58] — BLS forecast is 55,000 non-farm payrolls and 50,000 private payrolls. Even 30-40K would be consistent with the thesis. Still above zero, so growth continues.
- **Labor Force Participation Rate Collapse** [06:18] — The labor force participation rate has collapsed this year, falling off a cliff relative to 2023-2024 stability. Small percentage changes are significant because they apply to 157 million people: 1% = 1.5 million people, 0.1% = 150,000.
- **Low Participation Creates Fake Floor Under Unemployment** [09:14] — A declining participation rate lowers the unemployment rate. If participation falls faster than job losses, unemployment can go down even as job losses skyrocket, hiding the true state of the labor market.
- **Rate Hikes Could Normalize Participation** [11:20] — Higher interest rates make life more expensive, potentially driving discouraged workers back into the labor market. If participation normalizes while hidden job losses exist, the unemployment rate could shoot up.
- **Scenario: Participation Normalization** [12:21] — If participation returns to 62.4% (1.5 million more people), and only 100,000 jobs are added, net loss would be 1.4 million jobs, potentially jacking up the unemployment rate by a full percent (4.1% to 5.1%), triggering the Sahm rule and market sell-offs.
- **Rubber Band Analogy** [13:30] — The more labor force participation goes down, the more we're pulling on a rubber band. Eventually it will snap back to normal, creating a big risk for the economy.
- **Corporate Earnings Strong** [14:15] — S&P 500 ex-tech earnings growth is 23%, and overall EPS growth is 44% year-over-year. Companies are making more money from AI, but this doesn't mean labor force participation won't normalize.
- **Wage Growth Plummeting** [15:49] — Average hourly earnings growth has fallen from ~4.5% to 3.1% year-over-year. Worker pricing power is trending down, and after inflation, real wage growth is near zero or negative.
- **Two-Worker Household Squeeze** [17:17] — Higher interest rates squeeze households, potentially forcing a second worker back into the labor force. On a large scale, this could turn the tide and trigger the normalization scenario.
- **Key Number to Watch: Participation Rate** [17:45] — On Friday's BLS report, the only number Kevin cares about is the labor force participation rate. When it skyrockets, that's the bat signal that 'winter is coming'.

## Transcript

As a reminder, we had a payment rail issue with our provider, and so we had to extend the coupon code to Friday, this Friday, September 4th. They say the problem has been solved. We could see the problem solved, which is great.
We wanted to apologize for that. We did have to extend the coupon code for the Meet Kevin membership. And remember, you can still bundle, last chance here to bundle, and get that lifetime access for Reinvest AI. If you're a course member, you could also email us at staffatmeetkevin.com,
get a bundle code, and lock in that lifetime access before it goes away. you'll keep it, but everybody else will have to pay a subscription price to get it going forward. ADP employment numbers came out this morning. They missed expectations, but what's happening with the labor market at a stamp shop today
I think matters less than what's happening with the labor market very broadly because there is an underlying risk occurring in the labor market that is making the labor market seem stronger than it is.
And ironically, the more the labor market moves in this particular direction, the more fragile we actually become. In other words, the more what's currently happening keeps happening,
the more dangerous the labor market actually becomes. And look, today, you know, for example, we heard of news that Uber's laying off 10% of its staff. Okay, some of it is mostly because Uber hasn't really had layoffs since post-COVID.
Obviously, there was a surge of hiring post-COVID, and some companies are still filled up to the point where they're a little overloaded with employees. And so some of the trimming is normal, and that can always obviously seem insensitive to humans.
It's like, oh, let's cut labor to increase profits, but that's going to happen. And so we're just observing that from the outside. What's more interesting to me, once we get, we'll go through the ADP data here, what's more interesting to me is how if you combine the fragility that's actually building in the labor market right now
with potentially higher interest rates, that's where we really want to keep a close eye on what's happening. But there's one data point, so we'll hit that in just a moment. For now, though, let's go to the Meet Kevin app.
So we're going to go to app.meetkevin.com, and let's pull some of these charts that we've got here. Remember, you can download the app on your phone for free at meetkevin.com or, well, sorry, download the app for free in the Apple or Android app store
if you type in Meet Kevin. Okay, so what do we have here? These are the ADP estimates versus reality for the information we got today. So you can see the average and median estimate was right around 57,000.
Deutsche Bank thought we were going to come in at 65,000. What's interesting is their estimate isn't even on here. Sometimes you get estimates like they've got, what do we got? They got 29 estimates here.
Sometimes banks have their own. They don't even make it into these surveys. So know that there are probably more that are likely to distribute along a similar pattern here. You've got someone who's not thinking that the labor market's going to send ADP up at a weekly rate of like $30,000 a week.
We have not seen weekly rates of gains in employment of $30,000 a week since March and April. And there's a potential reason for that we'll touch on. But you can actually track the weekly data every Tuesday.
And, you know, we're at like $8,000 to $10,000. And so we actually ended up getting $38,000 as a read this morning, which is, in my opinion, relatively consistent with that $8,000 to $10,000 move. So I don't think this is really that big of a deal, the ADP report.
You've got a little bit of a mix here where ADP themselves say, hey, we're sort of a choppy environment. It used to be more predictable to kind of see what's going on with the labor market. Things are a little weird right now. Yeah, you've got this AI construction boom.
People are still spending summer vacations, leisure, hospitality. The wealth effect is in full effect. But you've still got the slowest pace of job creation since January. Some of that should be offset, though, by why we have the slowest pace of job creation since January.
I think one of the easiest things to do here is just do a quick little whiteboard explanation. Because it's really important when you're looking at the labor market, you almost have to adjust the way the first quarter. We had a really slow labor market going into the year.
and then in February we started hearing about oh no, we're going to war with Iran and what actually happened is a lot of manufacturers, if you study the weekly ISM or S&P manufacturing reports
you could see it happening a lot of the manufacturers when they say it bluntly were hiring to start building stock assuming that prices would rise So there was basically this you know boogeyman of inflation that companies expected
They're like, well, we may as well build supply now, build up inventory, hire a bunch of people. So you saw this boom in hiring really between March and May that kind of gave this, I would almost call it a false impression that the labor market was on fire and really, really booming.
It actually aligned with all-time highs in the NASDAQ before the SpaceX sucketing in June, right? And so we've kind of seen this, like, what feels like a collapse since then, but it's stabilized.
It's stabilized at around 10K jobs per week. That's what we've seen on the BLS data. That's what we've seen on the ADP data. And in fairness, we do have new BLS data coming out on Friday. So I would say take the information that we have here
and sort of use it to prepare for Friday so you know what single number to really pay attention to. It's not really the headline number that I'm worried about. And it could be really premature to even talk about this,
but when it is, it's going to suck. The DLS forecast, just so we can get it out of the way, we are looking for 55,000 jobs on non-farm payroll, 50 on private payroll. So even if we get anywhere close to that, now 30, 40, it's going to be relatively consistent with this thesis here.
And the good news is this is still above that zero line where we're shrinking. So we still have some growth happening. There's a lot of rejiggering happening in the regular market, of course. The big problem that we have, though, is something that actually makes us more sensitive to interest rate hikes in the future.
Interest rate hikes make things more expensive. We know that. But they also mean people who don't have a job might end up getting convinced that, ah, crap, we're going to have to jump back to work sometime soon.
That issue relates to the labor force participation rate, and we've talked quite a bit about this, but I want you to see how extreme it's getting. This chart right here, this lower one, the labor force participation rate, It indicates, while these percentage changes are very, very small, they're actually a big deal because you're applying those percentages to 157 million people.
So when you get a, you know, 0.10%, so like a 10 basis point move, 0.1% move, you're talking about lots of people, right? So, you know, every percent is one and a half million.
Every 0.1% is 150,000 people. It's like two Taylor Swift stadiums, right? There are a lot of freaking people. The labor force participation rate has really collapsed this year.
It was already falling in 2025, and our ice rates kind of got us a little temporary blow over here in the middle of 2025, maybe post-terrace or whatever.
But it has fallen off a cliff relative to the stability that we had in 2023 and 2024 2024 on the participation rate, which is ironic because in 2023 and 2024, we were contending
with this. These are the number of payrolls right here. This was before Trump. Look right here where this mouse is. You see this collapse in payrolls right here? Right there, where we went to basically zero on jobs.
Right there, that's where people are like, oh, Jerome Powell cut interest rates because of the election. remember september of 2024 he cuts 50 basis points to be like that a-hole just wants to try to get
biden in or you know kamala in at the time right and the reality was and i'm not saying there's not a political bias okay i personally think kevin warsh is a hack and you know he's gonna do
whatever he can to to show the dollars up so don't get me wrong i got political jade as well I just think, looking at the data then, maybe there was 10% political rationale.
I think 90% of it was because of that collapse in the labor market we saw. And we were watching it, and we were like, holy smokes, this is not good. And then, of course, you had the Japanese carry trade. You know, that was the ice cream of the state. We had a recovery after that, though.
Just like this magical recovery. I thought some of it could have been because of enthusiasm around actually Trump winning the election. Because of the data collection. Who knows, right? But the point is all of that drama that happened in 2024 sorry to make you relive that drama All of that drama did not affect the labor force participation rate Okay labor force participation rate has literally
collapsed since then. And here's what you need to know about a low labor force participation rate. Increase a fake floor of glass under the unemployment rate. And this is the nasty
part that doesn't make sense to a lot of folks, because, you know, you've got even the Federal Reserve, they're like, oh, unemployment's fine, look how stable the unemployment rate is, everything's fine, there's nothing to worry about. You get that talk all the frickin'
time, you know, the unemployment rate, oh, it's 4.1, 4.2%, whatever, somewhere in this range, somewhere between 4.1 to 4.3, that's frickin' low, people are like, oh, that's maximum employment, whatever. But what happens with the math is that every time the participation
rate goes down, you actually lower the unemployment rate. And you can actually see a decline in the participation rate that accelerates faster than the job loss you're seeing. So in other words,
you could literally see job loss skyrocketing. If the participation rate is falling faster, job loss can go up and the unemployment rate can go down.
So in other words, you could literally have the labor market collapsing from a labor force point of view in terms of actual job loss, how many people are losing their jobs. But the participation rate or the way the analysts tell you
what percentage of people are actually involved and actively looking for a job, they're not discouraged workers, You know, they haven't given up because they can't find a job with their skills, or they haven't retired early, or whatever.
Or they're employed part-time, and they wish they could be working full-time, right? All those metrics go into this. They tell you, ah, labor force participation rate is plummeting. The more it plummets, the more it actually hides this.
So the market doesn't see job losses skyrocketing. It only sees the unemployment rate going down. That's all it sees. But the problem is, not that this is going down.
The real problem, and this is the scary part, is when interest rates go up and the Fed hikes, you make life more expensive, and you drive people who otherwise may have given up trying to look for work, you potentially drive them back into the labor market.
Now, if you have an increasing labor force participation rate, or a normalization of the labor force participation rate, and you have those hidden job losses, you now have two forces that push up the unemployment rate.
Like, if you had stable jobs, no job losses, no job creation, and the participation rate went up, the unemployment rate would shoot up. If you have participation rate going up, and you have job losses,
this will go up even faster. And rate hikes make this even more likely. And so that's the fragile nature of the labor market because if the labor market all of a sudden, imagine we get the labor force participation rate normalizes, right?
And we go back to a participation rate that is, you know, 62.4%. So now add 1% of people that are now deemed to be participating. That's one and a half million.
Just pick numbers here for a moment, okay? you go back to a normalized labor force participation rate. That's 1.5 million more people need to have gotten a job. If we only added 100,000 jobs that month, plus this,
then we have now net lost 1.4 million jobs per the statistical changes, which all of a sudden takes this unemployment rate and possibly jacks it up a whole percent.
And so you can instantly go from 4.1% to 5.1%. And then you start getting talk about the Claudia Song rule is getting triggered. Oh my gosh, what's happening? Market sells off.
And then you potentially get this self-fulfilling, you know, CapEx slowdown, artificial intelligence spend slowdown. These people are like, oh my gosh, labor market's collapsing. We must be going into a recession. And it all comes simply because of that one stupid line.
Now, I'm not saying that's what's going to happen. this number could keep going down. They could also be rigging that number down And you know that also kind of weird right This idea that this number could be getting rigged down But what you need to know is the more this number goes down the more we
pulling on a rubber band. Simple analogy. The more labor force participation goes down, the more we're pulling on this rubber band. And eventually it's going to snap. Eventually it's going to snap back to normal.
And so these discouraged people or people gave up or whatever, they end up coming back to work or the statistics, the way they analyze it, the numbers, whatever, change, that's a big risk for the economy right now. And I think not a lot of people are paying attention to that.
Let's look at some of the other numbers on the labor market, though, and the overall economy. Overall, if we look at earnings for companies, they are doing well. You've got the S&P 500, ex-tech, seeing growth.
growth outside of technology saw 23% here at the end of 2026. This is really good. So, you know, people say, oh, but Kevin, you know, growth rates are so good at companies. Yes, yes they are, but both should be true.
Growth rates could be excellent at companies, but they could also be firing people at the same time or hiring fewer people. What I've seen in earnings calls, and I see this at other businesses as well,
but what I see is this, you know, wow, this AI revolution, let's hire some people, Let's acquire companies to help us with artificial intelligence. Let's get our own AI going. At some point, that investment in the human capital stops.
Got the AI team. You don't have to hire anymore. So you sort of temporary up front boom. What happens when you start cutting? You know, this sort of adds to the problem. But ironically, even if that happens, company profits can continue to grow.
And that's the big thing here. Now you look at the S&P 500 year-over-year growth rate for earning per share. We're looking at 44% here as an estimate for the last quarter. That's great. It means companies are making more money from artificial intelligence,
but that doesn't mean labor force participation isn't going to skyrocket and normalize. Here's another thing to look at in terms of a weakening labor market. Average hourly earnings, year-over-year wage growth.
If worker wages were a corporation, I would argue workers have on-net lower pricing power. In other words, the rate of change in people's ability to demand more money for their services,
for their human capital, has plummeted. We've gone from somewhere around getting average hourly wage growth of around 4.5% per year down to 3.1%. That actually means, I'd have to double check if these are real numbers.
These might already be the real numbers. But I was going to say, if they're not the real numbers, if these are just nominal wage growth numbers, you're actually losing money every year to inflation. Your wages are now net negative. Should have checked that before I didn't.
Sorry. But anyway, the trend doesn't matter. If I pull 4% off inflation off of this and 4% on inflation off of this or 3%, whatever level you want to use, the whole bar is just going to move down close to zero.
So, point is, worker pricing power is still trending down. All of that to me suggests we're in a weird place in the economy where corporations can boom in profits, corporations can boom on investment spending.
At the same time, you could be squeezing people out of the labor market that are okay for now, but eventually if these folks, the same percentage of people who are usually in the labor market, come back because rates got higher, corporations are still getting richer, but people aren't getting richer because wage growth is down.
an easy way to think about this is like a two-family household, two-worker household, mom and dad or whatever, one person, you know, I'm just going to stay home, I'll go study or whatever. You get squeezed by higher interest rates. Eventually, you might have to go back to the labor force.
That happening on scale is exactly what sort of turns the whistle, if that's even a good thing. On Friday, we will have a BLS job data. The only number I really care about is that labor force participation rate.
I don't think it'll skyrocket anytime soon, but when it does, when it does, that is your bat signal that winter is coming. Okay, cool. So that gives us... I don't know how to advertise these things each other here.
I feel like nobody else knows about this. We'll try a little advertising and see how it goes. Congratulations, man. You have done so much. People love you. People look up to you. Kevin, pass right there. Finance and Analytics and YouTube meet Kevin. Always great to get your take.