Moneybeat: Follow the value

Editor's note: The following text is a transcript of a podcast story. To listen to the story, click on the arrow beneath the headline above.

JENNY ROUGH, HOST: Coming up next on The World and Everything in It:

The Monday Moneybeat

NICK EICHER, HOST: Time now to talk business, markets, and the economy with financial analyst and advisor David Bahnsen. David heads up the wealth management firm The Bahnsen Group, and he’s here now. David, good morning to you.

DAVID BAHNSEN: Good morning, Nick. Good to be with you.

EICHER: So, Dividend Cafe this week, you make the point that everyday investors may be far more exposed to the AI boom than maybe they realize, simply because so much of the S&P 500 is now concentrated in a really small number of technology and semiconductor companies.

Now I see in both The New York Times and The Wall Street Journal over the weekend they are reporting that a freely available Chinese AI model appears to have narrowed the gap with leading American systems. According to the Journal, the news has added some fuel to the sell-off in chip stocks.

But the deeper question isn’t simply whether AI succeeds or fails, right? It’s who makes money, where in the chain do they make it, and how durable those profits prove to be.

So what does it actually mean for an investor to be “invested in AI”?

BAHNSEN: There’s so much to unpack there because it’s a multitude of things that I don’t believe very many investors are really thinking through.

Let’s just say that the returns from AI do not come in the way people think, or as quickly, or as much. Then you have some of the companies that need an ROI on their big investment that are vulnerable and heavily exposed.

But let’s say the returns do come, and then it brings in even more competition—whether it’s foreign, global, Chinese, or just within the U.S. You already see some of that. What does that do? It inevitably puts downward pressure on margins. Companies are competing for something that’s really generating a high ROI, so then you end up competing over price. That, of course, puts downward pressure on the expectations people have for a lot of the companies that right now are priced on very fat margins.

I think, Nick, there’s an expression I want to suggest for Moneybeat listeners that we get out of our lexicon forever, which is this very broad and vanilla and totally unhelpful term: “those invested in AI.”

What does it mean to be invested in AI?

A lot of people are going to make money being invested in AI if AI fails, because the investment they made money on was all the people putting money into the cap-ex of AI with the pick-and-shovel companies—the ones making the chips and the compute capacity and cloud infrastructure.

Things could fail, and those companies made money. As long as investors exit before all that blows up, they can make money.

But here’s the problem: that’s not what people are referring to. They’re not referring to making something and then having it just sort of sit there. It’s the ongoing utility of AI. That’s what we’re talking about when we say, “being invested in AI.”

So to me, I look at it like this: How is WORLD going to make money from AI? How is the Bahnsen Group going to make money from AI? How are McDonald’s and Walmart going to make money from AI?

That’s the entire issue here for people who understand economics: a new tool that, once it’s built out, has greater efficiencies, drives greater productivity, and enhances the production of goods and services in the real economy.

Nobody is talking about AI in that sense—the end user. We’re all focused on the bottleneck up at the top: either semiconductors or hyperscalers or AI labs making the large-language models.

I’m sorry. Until people get past that discussion into the real-world discussion, everything else is still very speculative and very uncertain as to who the winners and losers will be.

I’m perfectly fine with someone saying, “I think this whole thing is going to be really successful and transformative and make a lot of money for a lot of people.” That’s fine.

But to say that with any specificity for an investor about which companies, in what sequence, and in what model—I think that’s the part that is just never discussed, and it’s bothering me immensely.

EICHER: Let’s stay on the real-world use of AI.

The fear, of course, is that businesses will deploy it to gain efficiencies by eliminating jobs. But the evidence you cite so far points the other way.

What do those numbers tell us—and why are you not persuaded that this time technology will permanently destroy more jobs than it creates?

BAHNSEN: Yeah, so far. And I’m also saying that there is substantial evidence of the opposite.

For the companies that are the most widely used adopters of AI, they have a 12% increase in entry-level positions and a total increase of about 10%. Those that are considered low or no adoption with AI are somewhere around 0% job growth.

So there is at least something on the front end of AI adoption requiring new hiring. We already know there’s a significant increase in software engineer hires. That’s counterintuitive to some people who are going around saying we’re not going to need software engineers anymore because AI is going to do it all for us, when in fact companies are hiring a lot more software engineers.

There’s going to be adaptation. There’s going to be training. There are going to be workflow shifts, process changes, all of this type of stuff. But there’s also clearly just entry-level jobs that are being hired now.

I don’t want to get ahead of myself, Nick. I’m totally open to the idea that there will end up being job loss. Already, over the years—even pre-AI—automation led to certain job loss on a gross basis, meaning some fewer jobs. But then on a net basis, it opened up new opportunities.

That’s always been my base case here because there’s no precedent where that hasn’t happened in history.

There are some who say this time it’s different—that this time AI is not only going to do more things, but then it will not allow for new hiring as new needs come about. I have a very hard time getting there because of what I believe about human nature.

I think human beings are constantly growing in their needs and wants. Therefore, there’s an insatiable need for work to fill those needs and wants.

EICHER: I want to leave AI but stay on the jobs market. Because I read in The Wall Street Journal that nearly 2 million Americans have been looking without success for work for at least six months, and that figure is near the highest level since late 2021.

That sounds ominous. But at the same time, overall unemployment is still relatively low, so how much weight should we put on that number—and what may explain it?

BAHNSEN: Well, it’s still a very low number in the context of total unemployment.

There are more people who are long-term unemployed, and I’ll make a comment in a moment as to where that usually comes from. But that number becomes very problematic when it’s in the context of a rising unemployment rate, because then you have a lack of jobs and a lot of people looking for work at a time when there’s a lot of people who have been looking for work for a long time and not finding it.

You don’t see all of those things together. That’s a recessionary condition.

But here it’s more an outlier—a subset within what’s a reasonably low unemployment number—that there is a slightly higher amount of people who have been unemployed for some time.

It’s tough for economists to unpack because you don’t know what those who say they’ve been looking for work for a long time have done or are willing to do. When the weekly unemployment claims are not going higher, it isn’t because there’s a lot of firing going on.

So is there a segment of the population that is maybe stuck with obsolete skills? Or in an age stage that is just more difficult on a micro level—the individuals who find themselves in those situations?

I think back to things I’ve studied from the early ’90s, when the defense and aerospace industry had substantial layoffs. There were people who, for 30 years, were quite skilled in certain types of work, and the work they did was just not needed anymore and wasn’t coming back.

To go tell them at age 59 to go learn a new skilled trade—I can’t even imagine what that must have been like.

I don’t think we have a ton of that, but I’m positive we have some.

From a macroeconomic standpoint, though, that becomes a much bigger concern when it’s in the context of rising unemployment, which right now it’s not.

EICHER: Let’s end on an up-note, the apparent boom in new-business formation.

The New York Times reports a record 5.7 million applications to start businesses last year, with roughly half a million new firms actually formed.

Many may remain tiny and never create big numbers of jobs, or fail. Is that enough to discount the trend—or does the rise in entrepreneurship genuinely cheer you?

BAHNSEN: No, it does cheer me.

I wrote about it in Dividend Cafe this Friday. A 30% increase in business formation—that is largely... there are an awful lot of AI startups or AI-adjacent startups in there. I’ve written about it in recent weeks in the small-business data because, on one hand, I love seeing new business formation.

There are an awful lot that don’t make it, never get off the ground, and are sort of hobby-type businesses. There’s always been that, though. So the numbers are relevant in the sense that they’re relative to one another—that there’s a sequential growth in this that has been substantial.

Now, I don’t get particularly excited about all of them because somebody starting a little business to incubate themselves by walking around putting up Instagram reels of themselves eating lunch doesn’t excite me very much.

But that other people are trying to give it a shot in the small-business world—that does.

You’re right. It may not be a big hiring phenomenon, but I feel very strongly about entrepreneurialism. I do think that in the AI and tech sector, this is allowing certain people who aren’t going to get hired at big tech companies to go out and try something, whether it’s in coding or programming or the enterprise or facilitating services.

There’s a lot of need out there.

So the data is encouraging. That doesn’t mean everyone is going to make it. That’s part of celebrating entrepreneurialism—is celebrating risk-taking.

Once you get through some of the froth in the data, you still get a number that is on the rise.

That’s a good thing.

EICHER: All right. David Bahnsen is founder, managing partner, and chief investment officer of The Bahnsen Group.

You can read him every week at Dividend Cafe. You can sign up for it, read it, or both at dividendcafe.com.

David, thanks so much. We’ll see you next week.

BAHNSEN: Thanks so much, Nick.

WORLD Radio transcripts are created on a rush deadline. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of WORLD Radio programming is the audio record.