Prof G Markets - The Jobs Report Is Worse Than It Looks

Episode Date: February 12, 2026

Ed Elson breaks down the January jobs report with Kathryn Anne Edwards, Labor Economist and host of The Optimist Economy podcast. They discuss why 2025 was the worst “non-recession” year for hiring since 2003. Then, Ed is joined by Alex Heath, founder of the Sources newsletter and host of the Access Podcast, to unpack the latest AI models, as well as a viral blog post shaking up the AI conversation. Finally, Ed explains why AI policy is so important, and why it's currently lacking.  Check out Alex’s interview with OpenAI's Fidji Simo Check out our latest Prof G Markets newsletter Follow Prof G Markets on Instagram Follow Ed on Instagram, X and Substack Follow Scott on Instagram Send us your questions or comments by emailing [e-mail] Learn more about your ad choices. Visit podcastchoices.com/adchoices

Support for the show comes from Funrise. For the past seven years, there's been a room in finance most people couldn't enter, a room where you could have invested in some of the biggest names in tech companies like Airbnb and Uber before their multi-billion dollar IPOs. I'm talking about venture capital. Fundrise recently took a sledgehammer to those closed doors by launching a venture capital product that's available to anyone. Their mission is to give everyone the chance to invest in the best tech and AI companies before they go public. You can visit Funrise.com slash ProfG to check out Funrise's venture portfolio and get in early today. All investments involve risk, including the potential loss of principle, past performance is not indicative of future results.

This is a paid advertisement. You might be tempted to let Taco Bell's new Lux Value menu go to your head. Because 10 indulgences for $5 or less makes you feel fancy. Like you might think you need cloth napkins. Well, you don't. Just use the ones that come in the bag. Don't let the Lux go to your head. Today's number eight.

That is the percentage of Americans who say their favorite Winter Olympic sport is curling. That is also roughly the percentage of Americans who have a serious mental illness must be a coincidence. Welcome to Prowdy Markets, I'm Ed Elson. it is February 12th. Let's check in on yesterday's market vitals. The major indices ticked down after employment data was released. More on that in a minute. Meanwhile, Treasury yields spiked as the Jobs Report increased the likelihood that the Fed will hold rates at the next meeting, and finally, Bitcoin sank below $67,000. Okay, what else is happening?

The January Jobs Report came in stronger than expected. The economy added $130,000,000, jobs and the unemployment rate edged lower to 4.3%. But the headline growth marks a lot-sided picture of the job market. Nearly all of the gains came from just two sectors, healthcare and social assistance. Meanwhile, new annual revisions show that the economy added just 181,000 jobs in 2025, which is down from the initial 584,000 that was reported. That makes 2025 the worst non-recession year for hiring since 2003. Okay, here to help us break down this jobs report with speaking with Catherine Ann Edwards,

Labor Economist, Economic Policy Consultant, and host of the Optimist Economy podcast. Catherine, thanks for joining us. Thank you again for having me. So let's get into this jobs report. We added 130,000 jobs, unemployment rate came down. That's kind of the good stuff.

There is some bad stuff as well. Let's just start with your initial reactions. What did you make of the report? I thought it was a report that reminds us that good and bad news is relative. Two years ago, we would have been alarmed at a jobs number this low. Given the tumult of 2025, I think it gave a sense of relief to see even 130,000 jobs added. The big concern for me looking at it was, yes, we added all these jobs, but it's all healthcare jobs. and social assistance jobs.

And this is a point that you've been making on the podcast for a long time. We are adding jobs, but specifically in those categories. And if we didn't have those sectors, then actually job growth would be in decline, is my understanding. Take us through the role that healthcare played in this jobs report and what it tells us about the strength of the labor market right now. In some ways, it's indicative of the last year. the majority of job growth came from health care, education, and social assistance, these are industries that do not reflect the strength of the economy. They don't reflect investments in businesses. They don't reflect opportunities for hiring and expansion. They reflect a kind of

permanent, acyclical need to service the human population. And in that sense, they're adding jobs because of the composition of our, you know, population. And we have a lot of old people. We need to educate young people, but they don't reflect that our economic policies are being successful, that the market is positive and is, you know, enabling economic activity. And in some ways, it's a ballast to a bad economy that is large enough to keep it afloat. But that's, I mean, it's comfort, but I don't know if it's cold comfort or just depressing. It depends really on how you think about it. I mean, I'm glad that we have it. I'm glad that we have this ballast, that there's an anchor to keep us from.

having, you know, very drastic economic policies pull us quickly into a recession like what we've seen come out of the administration over the past year. But at the same time, it is almost enabling this bad activity because it's propping up the economy. You pointed out the fact that we have a lot of old people in this country and more and more old people in this country. That was my conclusion, too, that it's not that we have a strong labor market. It's that we have an extremely old nation. And therefore, if you have a lot of old people, you need a lot more health care workers. I'm sure that's a little bit of a generalization. But is that kind of the conclusion that we should be drawing here? Well, the other conclusion is that, you know, health care as a sector is not,

I would feel better about what's happening and just say, look, old people demand a lot of health services, you know, people in general are demanding health services. That's fine. What worries me is just how poor performing the health care sector is, you know, when it comes to the job that it's supposed to do. Americans pay a fortune for health insurance, and they don't like it. They don't like what they can access. They don't like how much they pay for it. They don't like how insurance works. I mean, there's just, there's so much dissatisfaction with this industry that from the lens of the macro economy,

health care is adding jobs, and that makes sense because we have an older population and we consume a lot of health care. It's one-sixth of our economy. I think that that type of role that it plays in propping up a market through what would be kind of recessionary periods blinds us to the fact that this industry has severe problems. And I worry that its role of we still added 80,000 jobs because healthcare was added, or health care added 82,000 jobs. So we had a good report is really blinding us to the problems within this sector. Right. Right. We're too afraid to look too closely at just how bad of a sector health care is on some pretty important metrics like making people healthier. Right. And it seems like one of the big complaints with health care or the big problems with health care has been how bloated it has become. This is basically showing us it's getting even more bloated. Maybe it's warranted the number of jobs, but it does seem that the labor market is increasingly reliant on health care, perhaps on bloat in health care, which doesn't seem that sustainable.

The other problem to highlight here is the revision to last year's job numbers. So they revised the annual job tellers from last year to 181,000 jobs added in 2025, and that makes 2025 the worst non-recession year for hiring since 2003. What do you make of the revision? What does it tell us about the state of the economy right now? Yeah, well, two things. One, it's bad and there's no way to spin it. Okay. The number of jobs that we added in 2025 was remarkably low. The second thing I say is it is not too late for 2025 to be a recession year. The way that the United States declares recession is with an incredible amount of lag. So when I, when someone says, when the National Bureau of Economic Research, Business Cycle Dating Committee comes down and says a recession has started, they might declare in June. that a recession started last October. You know, we don't really start a recession until we're deep in it.

There's lots of reasons for that. Mostly good. We don't want to instill panic. We want to give a chance for policy to make a difference. But this report and this revision keeps in play that we could be in a recession where the declaration was to have started last fall or last spring. It's not out of play yet. So the fact that it was such a low job-producing year means that.

that we're still at risk of, you know, one really bad jobs report, a couple really bad rounds of layoffs in an aggregate number, and we'll have decided that we've been in a recession for six months. That's not unheard of. A lot of recessions are small and weak, and a report like this could push us into one. I don't think that that's what's going to happen, but it doesn't take it off the table.

What does this tell us about the current data? Because it seems like what keeps on happening is we pay a lot of attention to say the January numbers, and then we learn in that same report that, by the way, all of the data that you looked at over the past year is completely wrong, and actually the numbers are way worse than we thought. And to be fair, this is something that keeps on happening

where we keep on revising the numbers, generally speaking, the numbers are being revised to show a slightly darker picture of what is really going on, which makes me question, And how seriously should we even be taking the monthly data? Like, should we even be drawing any conclusions about how many jobs were added or lost in January? Well, there is a little bit of when you buy a new car, you suddenly see it everywhere on the road.

Like, we have been in a tenuous economic position for more than two years. We've had very high inflation. We've had interest rates gutting people's ability to purchase things. And now we've been met with tumultuous administration, policy. So we are in this, like, in the new car looking for the car that looks like ours, which is maybe the economy being weaker than we thought it was. That's happening to a certain degree because this happens every downturn. There are big revisions to the numbers. The growth wasn't as high as we thought. We just don't normally get held in kind of this, are we in a recession purgatory for three years? It's a very unique time in American economic history for that reason. I mean, normally it's like recession in, recession out, and we know where we are you. are. I think that's part of it. But yes, you should believe the monthly numbers, because even when a revision comes in, it never, like, dramatically changes the direction. It's not as if we replace every month of last year with a whole new number that was unrelated to the one that came before. They get, they get

revised in degrees. Yes. And not, and not it was, actually, we thought we added all these jobs and they were, you know, revised downward and we actually lost all these jobs. I mean, for the most part, It was a weak year of job growth, weaker in the fall than in the spring. And that's exactly what the revision showed, too. It just took it and down a little bit, you know. But I would be remiss if I didn't say any reliance on one single number produced by the data is not the right way to interpret it. The economy always moves in directions and not with the number. So that it's a lesson to not focus too much on the number, more so the relative position of the economy.

report today is a great example, is 130,000 good or bad? Well, would your opinion change if it went from 110 to 150? You know, the range of a typical revision? No, you would still think this is a pretty middling job report and the U.S. economy should be adding more, but you're going to be grateful that it wasn't a really low number because we've had such a weak year. So I think about it that way of just take whatever is the jobs number, add 25,000 to each side and ask if it would really change your opinion of the report. Yes. Yeah, that is a helpful framework. Final question. unemployment rate for young people, the 16 to 24 year olds fell again. It's now down to 9%. Also, the starting salary for college graduates is down 8% year over year. It's the lowest it's been in six years.

I mean, overall, the job market for young people has been at least worse than it has been for middle-aged people, older people, etc. I mean, it seems like whatever tightness, we're seeing in the labor market is being acutely felt among young people especially. And this seems to keep on happening every time we get these reports. I just wanted to end with your conclusions on that. What is this say about the job market for young people right now? Young people are always doing worse than the job market because they have less experience and they have less financial responsibilities.

So they're also able to kind of ride out a bad labor market for longer because they are not, you know, most 21-year-olds don't have mortgages. They don't have children. They don't have child care payments and so on. So they are both weaker in how they're treated by the labor market and weaker and how they respond to that. This labor market in particular, this low hiring, almost frozen labor market that we see is particularly damaging to young people who are not getting their foot in the door. And again, you know, it's dropping to nine, being up at 11 doesn't change the story on the ground, which is that young people are not finding jobs. And they're not finding them at the degree we want or the pace that we want.

I welcome this slight dip. I didn't love that the 25 to 34-year-old rate jumped up a little bit. So you do wonder if a bunch of people just got a little older. And that might have been the reason for the fall is the composition around the age. cut off, but again, it's a feature of economic business cycles that young people tend to be almost a leading indicator, as well as the slowest to recover, given their position in the labor market. But we're not normally held here for so long to be kind of wallowing in this labor market purgatory. It's not supposed to be multiple years long. And so they're in some ways, they're kept in the worst position for a long period of time when workers who have kept their job are okay. I don't know if that answer is satisfying at all because there's no, Short of, let's make the economy stronger, there's not really a solution for them. All right, Catherine Ann Edwards, Labor Economist, Economic Policy Consultant, host of the Optimist Economy podcast, Catherine, appreciate your time. Thank you. Bye, y'all.

After the break, AI reaches a turning point. And for even more markets insights, you can subscribe to my weekly newsletter, simply put at edwardelson.substack.com. We're back with Profji Market. OpenAI and Anthropics' AI releases are always interesting, but the impacts of their latest tools have been profound. Last week, Arthur Anthropic released Opus 4.6, and Open AI released Codex 5.3. Software stocks lost $2 trillion in value, and we discussed that on our Monday episode. This week, tech firm Altruist released an AI tool, which helps you do taxes, and financial stocks such as Charles Schwab and Raymond James lost around.

around 10% of their value. And then yesterday, a blog post that discussed these new AI tools didn't just go viral. It went nuclear. The post reached over 50 million people in 24 hours. That is equal to roughly one-third of this year's Super Bowl audience.

It's also more views than the Wall Street Journal website gets in a month. The premise of the post is simple. These new tools aren't about to change everything. They already have changed everything. The author Matt Schumer writes, quote, The experience that tech workers have had over the past year of watching AI go from helpful tool to does my job better than I do is the experience everyone else is about to have.

Leaders from multiple industries praised the piece. Alexis O'Hanian said he strongly agrees. Medi Hassan said that it is, quote, the most important piece you read today this week this month. Clearly AC change is happening in AI right now, one that is pushing the imaginations of not just coders, but everyone who works with a computer. We wanted to dig into this, the new AI tools, as well as the blog post. So joining us to discuss it. We're speaking with Alex Heath, founder of the Sources newsletter and host of the Access Podcast. Alex, good to see you. Good to see you, Ed.

So it's been a while since we had you on. So much has happened in tech and in AI, which I want to get into with you. I guess let's just start with these new AI tools that we've seen released Opus 4. 4.6, 5.3 Codex, all of these plug-in tools. We saw this just evisceration of software stocks last week. Now we're seeing financial stocks are in decline. What do you make of the new AI tools? What do you make of how the market is reacting to these new AI tools? Well, I think out here on the West Coast, everyone's obsessed with Claudebot, right? this automation agenic software that you have to give a Mac Mini to or some kind of cloud

instance and that it can control everything on a computer for you and you can route it through iMessage or telegram or whatever. That's definitely taken, I would say, the tinkers and the early adopters in Silicon Valley by storm. And then the new models are good. I mean, the models continue to improve, especially on tool use, the ability to control and manipulate software. You're seeing that with Codex, opening eyes, seeing a ton of traction there. I think stealing market share a little bit from Claude and Anthropic. And then Claude continues to ship and be at the frontier with Opus 4.6, like you said. So the tools are progressing, it seems, at a steady pace.

And I think everyone is looking for a story. Everyone is looking for a narrative to hang their jitters on and the market's jitters on. and I'm not a macro expert, but I would venture to say that this much sell-off maybe doesn't have to do with Opus 4.6 and Claudebot. You know, that just seems a little overblown, but again, I'm a tech guy, not a macro guy. Well, it's interesting because your read on what's happening with AI sounds to me as a lot more sober than what I'm hearing from other people. And this blog post, which I want to get into with you, that went absolutely viral yesterday, more than 50 million views just on X, and it's on other platforms too. So it's large than that. The story of that article isn't that these new tools represent sort of a steady progression, as you're portraying, but an exponential progression, and that suddenly everything is different.

You know, these AI tools have been getting better and better and better, but now these new tools, ones represent an inflection point. And from this point forward, everything is about to change. That's not what I'm hearing from you. I think what it's doing is it's rapidly changing coding. It's changing the job of a software engineer. You saw that in that post you were talking about. He was talking about engineering. And it's true that it is abstracting away much of what was traditionally considered software engineering. It's doing it incredibly quickly. people who build these systems, the leaders of these AI labs, they don't really expect the models to generalize to new domains and the way that they have to coding, at least not as quickly as coding has taken off. There's a special thing to coding. It's deterministic. You can know for sure that your code works. It's math. And AI and LLMs really excel in that kind of environment. And sure, you know, we saw the sell off with legal stocks.

for example, when Claude added the legal AI and the contract review and all of that. So there is a threat of, you know, these AI assistants eating whole sectors as they mature. But, you know, Sam Alman pointed this out around the Super Bowl, you know, chat GPT has more users in Texas than Claude has globally. These are still not widely used tools. And so this hype that you see, and I was joking on my ex account earlier today, like, it's like neo-dodging bolts in the Matrix, trying not to like get sucked into all these, you know, the world is never going to be the same essays on X, right? All these think boys and these AI marketers that are just going crazy on X about all this stuff. You know, I'm of two minds. I do appreciate what they're experiencing, which is, you know,

if they're engineers, a key part of their lives quickly changing before their very eyes in a very fast clip. And I do think there is a bit of a gap between the rest of the world that is not experiencing that and is just experiencing chat chitpit which you know will still add an extra finger to a hand or something like that and then these coding systems building apps and software you know one shoted so will that generalizes the debate how quickly will it generalize to other domains um i think um if if one of these labs comes out and says we've solved you know it's called continual learning but this idea of like the lm like continues to learn and build on itself that's probably another C-Change moment.

People think that's maybe next year, not this year. But, you know, a lot of these stocks that are getting sold, they have, you know, long-term enterprise contracts. You know, Figma, I think, is an early interesting example, right? Like, Figma does giant Fortune 500, you know, many, many seat deployments, has a huge network effect of, like, people using it together on teams. I don't see tools like that getting completely destroyed in the near term, maybe in the long term. But I think everyone's jittery and everyone's looking for a reason to sell. So any new thing gives you that opportunity. It is remarkable the level of hysteria that I'm seeing across the internet right now and across the markets.

And you make the point that we're seeing a lot of think boys saying, everything has changed. there was another viral tweet that I want to point your attention to, which I think is representative of what's happening. There was a safety researcher at Anthropic who resigned, and then he posted his resignation letter on Twitter. It got 12 million views, and in the letter, he says, quote, the world is in peril. He doesn't really explain what actually happened, but it's this abstract idea that I have to quit. Everything's going to shit. the world is over. Why is this happening and why is it, why is it catching on right now? Is this just content? Is this just the story of media today? It is. I think also if you're an AI researcher

who has been relatively invisible inside of one of these companies, and I don't want to suggest that these people don't mean well or they don't believe what they're saying, but your departure note is your claim to fame. It is the way you can pay, seriously, I mean, it's the way you can plant your flag and hopefully maybe use it to raise money if you go start a company, right? Like there's a reason everyone is doing this. It's kind of crazy, especially this week, the amount of departure notes like this, like the one you mentioned. Yeah. And it's all these vagaries and it gets a ton of views because it's the same feeling. The feeling behind those views is the same feeling behind, you know, all legal stocks selling off when Anthropic does something, right?

everyone is looking for what is actually happening because they feel that the world around them is changing and they feel that it's changing in a way they can't control. And so these, you know, hollowed places that these people work, you know, they come out and they get to like maybe say something. And it's like, yeah, man, I mean, actually tell me what's going on. That'd be nice. Yeah, this is how I think will end because you wrote in your newsletter, you said, off to Davos, you went to, to Davos, you got the sense of everything, and then you said, quote, the industry has collectively decided to gang up on Open AI. I found that really notable. I feel that that is happening in the consumer world, too. I think the ad, the anthropic ad that went viral was really the illustration of that dynamic. And by the way, you said that long before the ad came out. And then the ad comes out and it's like, yeah, we all decide that we don't like Open AI. But it seems

very important the fact that actually the industry is ganging up on Open AI. Yes. Say more about that and what that means for Open AI going forward. Well, they're the market leader. They have 800 plus million weekly users. They're the definitely household name, right? They're the Kleenex of AI. And I think everyone wants to swing at that.

And I think people also perceive a moment of weakness, right? They did the code red when Google's Gemini really got better at the end of last year. By the way, Fiji told me that they hoped to lift that code red when their next big model comes out, which I think could be as soon as, you know, a month or two. So they were on their back foot, I think, for the first time since ChatsbyT really launched towards the back half of last year. And I think Anthropic Google, Meta, everyone else sees that opening of weakness and goes, let's pounce. Now, I think Codex is putting them back in the conversation on coding. It's growing very much, and I think they're happy with it. And I think Chat, Chat, Chat, ChiPT has started to reaccelerate its growth.

So we'll see if that shifts. We'll see if the rhetoric around them shifts or not. I mean, I do think those ads kind of fell flat at the Super Bowl. There were some reporting on, you know, the brand sentiment and what people thought of them. And I don't think it really got the message across in the way that Anthropic hoped. But, you know, the road is long here, and everyone's trying to go public this year. And that also increases the stakes because you have to be the winner. You have to be seen as the market leader.

And I do think there's also just this philosophical alignment that Google and Anthropic feel. I noticed this in Davos, particularly with the heads of those companies, Demas Asabas from Google Deep Mind and Dario Amadeh, from Anthropic. They were on a panel together. I was at a dinner where they, you know, I heard Demas call Dario a good friend during Davos. And I think, you know, Sam is pretty controversial. And I think people want to band together to try to take on Sam's leadership position. All right.

Alex Heath, founder of the Sources newsletter and host of the Access podcast. Appreciate it, Alex. Thank you. Thanks, Ed. So this AI blog post has obviously gone viral. It's gotten more than 50 million views on X. Who knows how many views it's gotten on other platforms, possibly even more. Everyone is talking about it because of what it says,

specifically that AI is changing everything. Now, I want to be clear about where I stand on this blog post. I don't find it particularly groundbreaking. It walks us through how powerful these AI tools have become, something I thought most of us already understood, pushes us to adopt these tools to keep up with the times, again, something I thought we all understood. I think what is striking about it isn't necessarily what it communicates, but rather how it is communicated. It is a really well-written piece, and that to me is simply a testament to the power of great writing.

Having said that, there is one point in this blog on which I wholeheartedly agree. And that is, despite what all these tech leaders told us about AI, that AI won't take your job, that AI will be your thought partner, that it will enhance employment. The truth is actually, no, AI will take your job. The writer says, quote, if AI is smarter than most PhDs, do you really think it can't do most office jobs? That is the salient point. And in fact, we can look at existing data, which indeed tells us that AI is already taking people's jobs. We saw mass layoffs across multiple different industries this month.

Pinterest laid off nearly 1,000 workers. Dow Chemical laid off 4.5,000. Heineken laid off 6,000. Amazon laid off 16,000. In each of these cases, the companies are literally telling us that the reason they're letting these people go is because of AI. In fact, we can look at this on the national level too. A recent report found that AI was responsible for 5% of all layoffs,

last year. That might not sound like a big number until you realize that 1.2 million people were laid off last year, which is the highest number we've seen since the pandemic. Put another way, AI has already taken tens of thousands of jobs. Plus, we're only a couple of years into this. So if we're being realistic, this is just the beginning. Now, this isn't a new predicament for America. We have had big labor market disruptions before. But each time that has happened, we have figured out a way to mitigate the externalities with something that we call policy or laws. During the Industrial Revolution, for example, we came up with child labor protections.

As factories started to reshape the nation, we created the Fair Labor Standards Act, aka minimum wage. During the Great Depression, we made Social Security after World War. World Two, we created the GI Bill. We've dealt with this kind of thing before. And our response has invariably been the same, and that is right laws, right policies that make the situation better. But what is particularly scary about this moment, about AI, is that as of today,

our government has decided that the policy that they want to pursue is no policy. They want to sit back and do nothing. And this isn't hyperbole. Trump has literally ordered a moratorium on all state-level legislation. The plan is to make it illegal to regulate AI. Now, the reason they're doing this is because they have it in their heads over at the White House that any form of policy, any form of regulation is stifling innovation. If you regulate AI, you're getting in the way of AI.

This is what they genuinely believe. Meanwhile, over in China, they actually have figured out real AI legislation. They have made it a priority, which would explain why nearly 90% of Chinese citizens are actually excited about AI. Compare that to America, where the number is 40%. We know where this is headed in America because we know we're doing nothing about it. So if the upshot of this viral blog post is that people now recognize what is happening to jobs, that AI is taking jobs, then great. But as I wrote in my own newsletter,

the next step is to take action. What exactly do we do about this? Well, you can go read my newsletter where I lay out some proposals, but one thing is very clear to me when it comes to AI and when it comes to jobs. There is no worse policy we could pursue than no policy at all.

And yet that is where we are at today. we have decided to do nothing. Okay, that's it for today. This episode was produced by Claire Miller and Alison Weiss, edited by Joel Patterson, and engineered by Benjamin Spencer. Our research team is Dan Shillan, Isabella Kinsel, Chris Nodonoghue, and Mia Silverio. Thanks for listening to Proffty Markets from Proffery Media.

If you like what you heard, give us a follow. I'm Ed Elson. Tune in tomorrow for our conversation with Ishwa Prasad.