00:00:02 Speaker 1: Bloomberg Audio Studios. 00:00:04 Speaker 2: Podcasts Radio News. 00:00:11 Speaker 1: You're listening to Bloomberg. And as always, on the Bloomberg Terminal and the Bloomberg Business Hour. We begin this hour with stocks and bonds edging higher ahead of the Fed's latest decision. Gina Martin-Adams of HB Wealth Management writing, the 10-year is already extremely oversold, but the longer-term drivers of high yields appear unlikely to fade anytime soon. Gina joins us now for more. Gina, welcome back. 00:00:52 Speaker 3: Thank you. 00:00:52 Speaker 1: An old friend of ours and an old friend of this program. It's good to see you in person. You too. 00:00:56 Speaker 3: Great to see you. 00:00:56 Speaker 1: Let's talk about this decision later on this afternoon. What an appearance. What a time for it. 25 basis points. Does it influence your outlook? 00:01:03 Speaker 2: No, not necessarily. 00:01:04 Speaker 4: It's too small to really influence the outlook. And I also think that the market is priced for an emerging Fed tightening cycle. So it looks like what we have is. 00:01:15 Speaker 3: More of the same. Rates are going higher. 00:01:17 Speaker 4: Stocks are probably going higher longer term, but a little bit of volatility in the short run. And frankly, Volatility this time of year is not unexpected. It is September. That's the only month of the year in which you normally see stocks fall. So, we will likely see volatility this fall. But does it really change the longer-term perspective? 00:01:34 Speaker 3: Not necessarily. 00:01:35 Speaker 4: We need to see a downtrend in earnings growth or some sort of change in economic fundamentals to get really different outcomes for portfolios. 00:01:41 Speaker 1: What is it about this backdrop at the moment that's leading to both higher rates and higher equities? 00:01:45 Speaker 3: Higher growth. I mean, it just is simply growth. 00:01:48 Speaker 4: I think that we all talk about inflation and certainly there is some embedded inflation risk, a lot of supply side shocks that we have to contend with. But frankly, the fact that we have extraordinary investment occurring and that is driving GDP growth is what's really driving overall conditions in both the stock and the bond market. If you look at the bond market, Inflation expectations certainly have driven some of the movement in the long end of the curve, but most of this is just real rates going higher because growth prospects are improving, along with enormous capital spending. I mean, we're in the midst of the largest capital spending boom in at least 150 years of U.S. history, and that's creating different outcomes. Core inflation is going higher, not because prices at the pump are going higher, but because we're spending so much on capital investment. 00:02:35 Speaker 3: So there are two potential constraints to this. 00:02:37 Speaker 5: John was mentioning oil prices as they go higher could potentially start to constrain some of this economic activity, particularly among consumers that are more vulnerable. And then higher rates could constrain some of the ambitions of the CapEx. Just starting on the first side, have you started to see any signs that there really has been demand destruction or A slowdown tied to higher oil prices? 00:02:58 Speaker 3: Not yet, but it's early. 00:03:01 Speaker 4: I think that we probably will see demand destruction on the consumer side as a result of higher oil prices. We will see transportation struggle to some degree because diesel prices are at new all-time highs. I mean, we're sitting at diesel prices that are higher than they were in 2022. 00:03:17 Speaker 3: So we will see some demand destruction. 00:03:20 Speaker 4: But that demand destruction is occurring in industries that are already struggling to some degree. The consumer industries have not been big components of growth. The consumer experienced a minor recovery in the first half of this year as a result of tax refunds and tax refunds alone. But what's really powering economic growth is capital spending from the technology companies. It's filtering through the entire system. that's not likely to experience any sort of slowdown unless the Fed does start to tighten policy to force a slowdown in the capital investment to some degree. At the same time, you have the AI companies coming out and saying that they want a slowdown. So we will see a little bit slower growth going forward. But the Fed, in my mind, is playing catch up. I think a lot of this is in 2025, they probably should not have loosened policy to accommodate the doge sort of cuts that were happening at the federal level. Now we need to take back that insurance that was put into the marketplace in 2025 just to normalize policy rates to be in line with what growth and inflation conditions suggest. 00:04:23 Speaker 5: So you don't think that 75 basis points of rate hikes, as a lot of people are currently calling for, would constrain the earnings momentum that you already had penciled in. 00:04:32 Speaker 1: Is that correct? 00:04:33 Speaker 4: I don't think so, because the consumer momentum isn't there. And so the 75 basis points of rate hikes, to the degree that it impacts earnings, will probably impact the consumer. To the degree that it impacts the business investment, it's got to be just minor. It removes some of the excess of 2025, is the way that I see it. But I don't think that it has this meaningful impact on growth. If we do have a growth deceleration, it's probably because of other things that are impacting the growth outlook right now. If Fed policy tightens today, I wouldn't anticipate that really impacting earnings until late 2027 or 2028. 00:05:09 Speaker 3: It just has too much of a lagged impact. 00:05:12 Speaker 4: What we're experiencing now is an earnings revival in part because of the eases of last year. 00:05:16 Speaker 1: Let's deal with the other things. There is a company out there, a number of them, making the suggestion that their technology that they're developing could threaten human life. 00:05:25 Speaker 6: Yeah. 00:05:25 Speaker 1: A one in 10 chance of ending humanity in the next several years, which is kind of crazy. 00:05:29 Speaker 5: Yeah. 00:05:30 Speaker 1: Doesn't that lead to a pause or at least a slowdown in the investment here? 00:05:33 Speaker 4: Yeah, it should lead to a slowdown in investment. And that's where I think we are experiencing some friction in the marketplace is it probably does lead to Something of a slowdown, at least in the development pipeline. Does it lead to a slowdown in investment into 2027? Probably not, because you still need the data centers in order to ultimately sustain the development that has already occurred. The deployment of the technology that has already occurred occurs through the data center development that's already in the pipeline into 2027. What a potential slowdown in the development cycle does is potentially slow down the 2028, 2029 investment assumptions, which are still for more than a trillion dollars in spending coming in 2028, 2029. It effectively smooths the cycle. But frankly, the AI train has left the station. We're going to continue to see AI deployed. It's just a matter of the pacing of that deployment and the speed of the development that occurs. And right now, it had all been pushed into a very, very short cycle. 00:06:29 Speaker 2: Yeah. 00:06:30 Speaker 4: That is probably going to be extended over time, but we will still need the data centers, we still need the space, we still need the computing power, even for today's technology. 00:06:39 Speaker 1: Where does it leave the hardware trade in the meantime? 00:06:41 Speaker 4: Yeah, it does leave it a little bit in the dust. And I think that this is probably a little bit underappreciated by the marketplace right now, as we were pricing for that enormous boost in activity in the very, very short run, that maybe creates a little bit of, maybe as we push that out over the cycle, We see actually profitability of the hyperscalers, profitability of the companies that are developing the technology improve in the short run because they can monetize their capabilities faster. They're not producing technologies that will ultimately cannibalize today's technology as quickly. And so you see a monetization emerge for some of the companies that we had punished for spending too much too fast. So I think what you see is this rotation inside technology that's kind of profound. Frankly, everybody's been all in on all of the companies that are going to benefit from spending, leaving the companies that are actually spending behind. Now I see an environment where the companies that are spending money might actually see some monetization of that spending faster. They'd perform reasonably well, while the companies that were the beneficiaries of the spending maybe take a backseat for the short run. 00:07:43 Speaker 1: Stay with us. More Bloomberg Surveillance coming up after this. The Treasury Secretary, Scott Besson, testifying about the future of AI regulation as senior Trump officials meet with Anthropix co-founder and chief compute officer Tom Brown to discuss safety risks with the technology. Stuart Russell, professor of UC Berkeley, participating in the bipartisan pro-human rally this week, writing, we've already sunk trillions of dollars into the current technology path and plan to sink trillions more. But the sunk cost fallacy is just that. If we double down on a mistake, it's still impossible. A mistake. Stuart joins us now for more. Professor, welcome to the program, sir. Stuart, it's good to see you. That mistake, I want to give you some time. What kind of mistake do you think we're making right now? 00:08:31 Speaker 7: So I think this is a long-term mistake. The technology path that the companies chose back around 2018, 2019 was the large language model. as we know, so Chad GPT and Claude. These are gigantic black boxes. They have literally trillions or now even tens of trillions of parameters, and we have essentially no idea how they work. So as they're becoming more capable, the problem is that we need the kinds of safety guarantees that we have for nuclear reactors and airplanes and so on, but we just can't get those guarantees because we don't understand... how the technology works. And we have no way of stopping these systems from doing things that we don't want them to do. So the only guarantee we can get is that they are not safe. And so my argument is that the technology path itself was a mistake. and it's not going to get better by pouring more money into it. 00:09:33 Speaker 1: Stuart, let's assume that they do just pour money into it, which is what most people assume. That's the direction of travel now. We're on this path. The Treasury Secretary made the argument, and we played out some sound just there, and I'll read it verbatim. The best way to guarantee safety is that the creators are liable for what they build. Do you see that liability, those guardrails, as being sufficient? 00:09:54 Speaker 4: No. 00:09:55 Speaker 7: And there are really two reasons. So, the, most obvious reason is that the risk that is uppermost on the mind of, for example, Dario Amadei, the CEO of Anthropic, is the risk of loss of control. which really means that the systems have become intelligent enough that they end up taking control of the world, and there's nothing we can do to get that control back. And after that, we really have no say in what happens at all. And obviously, liability is not going to deter a company from incurring that risk, because if it does come to pass, there will be nobody around to pay the money and nobody around to receive the money. 00:10:40 Speaker 2: So that's the first obvious point. 00:10:41 Speaker 7: But the second more important point, from a practical point of view right now, is that we need more than just liability. 00:10:50 Speaker 2: So think about airplanes. 00:10:51 Speaker 7: Of course, if there's a defect in an airplane and it results in a crash, then the company is liable. And we saw that with the Boeing 737 MAX 8. But that doesn't mean we don't also have safety requirements for the airplanes. And to get an airworthiness certification, The company that develops the aircraft has to do enormous amounts of flight testing and structural integrity testing and mathematical analysis of the design of the aircraft and software that runs the aircraft and so on. 00:11:25 Speaker 2: Before they get that certification. And we do that for obvious reasons. 00:11:30 Speaker 7: And exactly what happened when Boeing asked for basically a bypass on that certification process was that they ended up killing 340-something people. They lost about $ 80 billion, and they ceded the American lead in civil aviation to Airbus, which had a much stronger focus on safety. 00:11:52 Speaker 2: So we need both. 00:11:54 Speaker 7: Requirements before a developer has market access, as well as liability after the fact. 00:12:00 Speaker 5: So, Stuart, there's a big difference, though, between ensuring that a plane's not going to crash and regulating AI. On one hand, it's very clear what the goal is when it's going into an airplane, making sure it doesn't crash. When it comes to AI, what's the ultimate goal? To prevent artificial general intelligence or to just make sure that some of these companies are keeping the models in a box? 00:12:24 Speaker 7: So the answer to that question, which is a very good question, and you're right to say that the range of things that an AI system can do wrong is much broader than an airplane crashing. So the answer, we have this concept of behavioral red lines. So red lines just demarcate things that we absolutely do not accept that AI systems will do. So that includes AI systems breaking into other computer systems. 00:12:53 Speaker 2: That's a very clearly defined event. It's already a. 00:12:58 Speaker 7: Crime if a human being does it. So we have clear definitions from the legal process that applies to humans. 00:13:05 Speaker 2: We don't want AI systems. 00:13:06 Speaker 7: That improve their own capabilities without human authorization and oversight. We don't want AI systems that advise terrorists on how to build biological weapons. 00:13:17 Speaker 2: So these are a few examples. And it doesn't have to be exhaustive. 00:13:21 Speaker 7: The point is that as the companies develop the understanding of their own systems and the safety engineering technology to make sure that they can provide evidence of safety with respect to these red lines, then that process will lead them to prevent all the other kinds of harms that we haven't yet enumerated. 00:13:42 Speaker 1: Stay with us. More Bloomberg surveillance coming up after this. Let's stick with AI. Investment firms increasingly backing the technology, including the infrastructure needed to fuel the boom. Apollo Global Management President Jim Zauter saying in the company's most recent earnings call, quote, the sheer size of the AI infrastructure build-out is unprecedented. We see an enormous opportunity for private capital to finance a portion of this, along with public capital. Jim joins us now for more. Jim, welcome. It's good to see you. Good to see you. No days off for Apollo. What an August you just had. 00:14:21 Speaker 6: We didn't have an August break. It was pretty much pedal to the metal, a bunch of announcements, a bunch of travel, and a bunch of financing. So the firm did an amazing job. 00:14:32 Speaker 1: Walked in with the Yankees cap as well, distributed some merch. Brammo, how do you feel about that? Going to put that on? 00:14:37 Speaker 5: Well, I'm going to just amend and embroider a couple of lines around the Y, and I think it'll be fine. 00:14:42 Speaker 1: Can we pair the two stories? 00:14:43 Speaker 2: Sure. 00:14:43 Speaker 1: The boom in Apollo Sports Capital with what's happening in AI. Is that one of the reasons why sport is becoming such a big focus for investment firms? 00:14:52 Speaker 6: Well, I do think there's been a few themes as I've been on this show for the last couple of years. And there's no doubt there's a broader theme with society as the AI technology plays a larger and larger role. What are industries, what are activities, what are businesses that will have a very low chance of obsolescence? And I think it's as simple as that. The last 50, 100 years, there's been more and more leisure time, leisure time activities, whether it's the whole travel, entertainment business, and other activities like that. And I think people see the purview of sports, especially mainstream sports, and the aggregate followership, if you look at the top 100 shows in a year, you know, 95, 97 are sporting events, and that's probably unlikely to get disrupted. So I think this was a thematic view that we had over the last three or four years that about putting our firm in the limelight of those activities. The other thing I would say is we've been much more of a, within the arena of sports and entertainment, we've been much more of a debt financing partner. Certainly there's transactions where there's equity along as well. I think a lot of folks have put their stake in just the equity side of the business, but we've done the equity as well as the debt financing and the capex expansion. 00:16:15 Speaker 1: Why is the debt slightly more interesting to you? 00:16:17 Speaker 6: Well, when you think about, you know, a franchise like the Yankees, there has not been historically a variety of financing alternatives of companies of that, of teams. And the loan to value is exceptionally low that you can make a loan at versus other comparative industries. You're loaning 10, 15, 20 cents on the dollar of value versus other industries, 50, 60, 70. So you have a great margin of safety, of protection on what we would say is world-class, unique assets that we want to be associated with for decades. 00:16:55 Speaker 5: How much is that applicable to the broader investment universe? This has been a debate on Wall Street for the better part of 18 months now. Would you rather be on the debt side of the industrial build-out or the equity side to participate in the upside of what could potentially come? 00:17:11 Speaker 6: Well, it depends on the capital you have. I mean, as I've spoken on this show many times, you know, half of our capital comes from regulated balance sheets. And on those regulated balance sheets, our objective is to make plus or minus 7%, depending on the rate environment. And so when you see the short end right now in three and five years where they are in their yields right now, I'm focused in, we are focused on high quality spread in that short duration so you don't have to take that equity risk. There's no doubt that what we bring, and you had the quote up as the show started, The message that we've been saying is the scale of this CapEx cycle is unprecedented. It's going to take any and all precincts, equity, public equity, private equity, private capital, investment grade debt, and everything else in between. And so for us, for our regulated balance sheets, we're really a debt financing provider. And hence, the Intel financing, the Broadcom financing, the NVIDIA financing. On the other side of our business, transactions like what we announced for the Yankees, We announced a very interesting transaction for Atlantic Aviation, which is arguably another business that I don't think is going to get disrupted. This is an airport fixed-based operator of private aviation. Those are businesses that we think have amazing resilience. And so that's how we're trying to make sure that we are thoughtful on the debt side and on the equity side being very, very thoughtful about disruption. I'm often fond of saying we don't want to take equity risk for a fixed coupon. And I think when you think about what's going on right now in a lot of the build-out across the whole spectrum from the large language models to the racks and the chips, It's interesting to see today the gross margin is highest away from the models. And so companies like Broadcom and many others are doing very, very well, notwithstanding who's the winner of the LLM race. 00:19:17 Speaker 5: So in other words, if there's a huge debt issuance slate from the likes of OpenAI and Anthropic, wouldn't be as interested? 00:19:23 Speaker 2: No, I think there's opportunities. 00:19:25 Speaker 6: I mean, the big challenge and another theme that we've talked about is the scale of this. If you look at the largest company on the planet, NVIDIA, their four or five top 10 investors have upwards of anywhere from 2%, 3% of the equity to almost 9% of the equity. That's a $ 500 billion exposure. The biggest companies in the globe are not going to have people provide that scale of debt. So it's a scale and capacity issue. And so there's certainly financing opportunities for OpenAI and Anthropic that we want to certainly be front and center in, and we have been, and we will continue to do so. But it's all about sizing and diversification. And that's the difference between funding debt and equity in this industry today. The equity, you have massive convexity. You don't mind getting very, very concentrated because the nature of debt and just getting paid back par, it's much more of a diversification game. 00:20:25 Speaker 1: Do you sense that investors are becoming more discerning about the available opportunities? We've talked about NVIDIA a few times attempting to anchor borrowing costs for the ecosystem. You wouldn't think they'd have to do that if demand was there for the amount of debt that needs to be raised. What do you think is happening there? 00:20:39 Speaker 6: Well, I think that's a complex question. I think there's a lot of reasons why the reality is the IG market has seen a tremendous amount of issuance this year, the public investment grade market. But now you're seeing the broad ecosystem of AI. People are predicting that that system will be 10% of the IG market. So, I think those companies wisely are saying we have to count on a variety of precincts to raise this capital. And whether it was what Alphabet did earlier this year with the debt and the mandatory, you need to hit a variety of sectors. So there's no one asset class that's large enough to fulfill the aspirations of all these companies. They're going to need any and all, and I call it calling all precincts. 00:21:28 Speaker 1: Yeah, they're tapping equity, they're tapping debt, they're tapping different currencies. We can see it from Amazon, from Alphabet. That competition for capital, the crowding out forces that we've been talking about all year. Who's the victim? Who's getting left behind? 00:21:40 Speaker 6: Well, I think this is certainly, you know, we had a, the private capital and the private equity industry was one where for 20 years it was about asset light, go back to software, distribution businesses, and now we have transcended into an asset-heavy industry. So there's no doubt we are in a winner-take-most environment in a lot of businesses, and there's no doubt what you've just pulled the thread at is Having access to capital is a competitive advantage. And whether you're seeing it in the industry that I happen to operate in day in, day out, or the industry that we're focusing on, there's the winners and the haves and the have-nots. And having access to capital in scale is certainly a competitive advantage. 00:22:27 Speaker 5: It's one thing for an AI company with the prospect of 30% growth to borrow at a 6%, 7% rate. It's another thing for a mom-and-pop company that's been grappling with. 00:22:38 Speaker 1: Supply shock after supply shock. 00:22:40 Speaker 5: How feasible is it for corporate America at large to handle almost 6% average yields on the investment-grade index going forward, especially with the refinancings coming to the fore? 00:22:54 Speaker 6: No doubt, if you also look where mortgage rates are right now, that's having an impact. If I was here, I did look up. I've been here 15 times since last January. Thank you very much. 00:23:02 Speaker 1: You've been counting. 00:23:03 Speaker 6: Because I wanted to talk about how many times I've talked about higher for longer. And if I was sitting here six months ago and the short end and the long end in rates and oil would have been at these levels, I think we all would have been surprised at the level of the equity market. This economy and market have been resilient. 00:23:21 Speaker 2: Yeah. 00:23:21 Speaker 6: And I suspect that when you think about where we are on the rate curve right now, it is a combination of some inflation issues, some fiscal concerns, as well as a massive amount of supply hitting the market. And it's a combination of all three of those. But you're asking the good question about at what point does the economy say no mas and the actual rate of cost of capital is going to slow down things. We're not seeing it right now. Now, we're right in the middle of this massive capex cycle. You do worry about the breadth of that capex cycle being narrowed a little bit because of the financing costs. But it's certainly a macro concern that investors have to think about. 00:24:02 Speaker 1: This is the Bloomberg Surveillance Podcast, bringing you the best in markets, economics and geopolitics. You can watch the show live on Bloomberg TV weekday mornings from 6am to 9am Eastern. Subscribe to the podcast on Apple, Spotify or anywhere else you listen. And as always on the Bloomberg Terminal and the Bloomberg Business App.