Quarterly Webinar Replay

Chris Zand: Hello, and thank you all for joining us today for the Osterweis quarterly call where we will discuss recent market activity and trends, our portfolio positioning, and our outlook for the remainder of the year. I'm Chris Zand, and joining me for today's discussion are Chief Investment Officers, John Osterweis, Carl Kaufman, Nael Fakhry, and Greg Hermanski. Before we get started, I'd like to share some exciting leadership news from here at Osterweis. As many of you know, Carl has served as our Co-CEO while continuing to lead the fixed income investment team. Carl, I'll let you share the news.

Carl Kaufman: Thank you, Chris, and it's great to be with everybody today. Effective June 30th, I stepped down from my role as Co-CEO after almost ten years in that role. I think the firm is now on very solid footing. I'd like to devote my time and energy to what I've always loved doing, which is managing fixed income portfolios. I know it doesn't sound exciting to most, but it is to me. Going forward, Cathy Halberstadt will serve as the sole CEO, and this is part of our ongoing leadership transition planning. And I'm very excited for Cathy as she takes the helm. And with that, let's get to today's discussion. Back to you, Chris.

Chris Zand: Thanks, Carl. Exciting indeed. Okay, let's turn now to our main discussion and start with a broader market backdrop. Looking back on the second quarter, we saw a remarkable reversal from the volatility that really dominated the beginning of the year. Investors entered the quarter facing considerable uncertainty from tariffs and inflation concerns to geopolitical tensions and questions about economic growth. Yet by quarter end, equities had recovered sharply. Credit spreads remained tight, and the risk appetite in the investor base that we saw here in the U.S. came back roaring. On the surface, that's a striking change in sentiment. I'm curious, John, whether you think this was primarily a shift in investor psychology, a reflection of fundamentals proving more resilient than many had expected, or some combination of both.

John Osterweis: Thank you, Chris, and welcome everybody. And before I answer the question, I'd just like to thank Carl for his ten years of service as Co-CEO. For those of us who are either stock market or bond market junkies, some of the administrative burdens of running the firm are not the most welcome chore. And Carl has done a superb job, set the firm on a much more focused path than it had been, and as he said, set us up for a lot of success going forward. So Carl, just I'd like to publicly acknowledge what you've done, so thank you.

Carl Kaufman: Thank you, John.

John Osterweis: As far as the market, I think the simplest way to answer it is the market has been torn, as it usually is, between lightness and darkness. And the lightness is the fact that the economy has been and remains very, very strong, to some extent driven by a very massive investment in AI and all that entails. So that's the plus side. The minus side is the obvious turmoil in the Middle East, the disruption of oil supplies, higher oil prices, potentially higher inflation, maybe potential higher interest rates. And so the market has been going back and forth between a focus on either the strong economy or the problems in the Middle East. This is to put it simply. I'd say in the second quarter, the focus shifted back to the economy, and there was a feeling that maybe things were settling down in the Middle East that Trump would finally have his deal.

And so very strong earnings got expressed with much higher stock prices as a result. The other thing that's been going on in the market is the either infatuation or skepticism about the massive AI investment that's going on. At times, you better own AI-related names or you just couldn't perform. And at other times you better avoid the AI, what they call a crowded trade, too many people chasing too much in the way of momentum. And in the second quarter, things began to shift a bit. And recently, after the quarter, there's been a pretty dramatic shift away from the AI trade into a much broader market, which we have always thought was healthy. When markets become too concentrated, they become vulnerable, and when they spread out, it's much, much better. So those are a few things that have contributed to what happened in the second quarter and a little bit beyond.

Chris Zand: Thanks, John. Building on those last few comments, we're already about a month into the third quarter, and after a strong rebound in the second quarter, markets have seemed to have taken a bit of a breather. Has anything meaningfully changed in that period of time from your perspective, or would you say it's just a healthy pause after a powerful rally?

John Osterweis: No, I think again, what's changed is we went from thinking that, well, maybe we've got a deal in the Middle East, to, oh, maybe we don't. And maybe Iran has us exactly where they want us, which is at their mercy. So I'd say, and this shifts daily by the way, but I'd say since the end of the second quarter, there's been a lot more concern about we're not going to get through this unscathed. And, as you know, oil hit $100 a barrel recently. The second thing that's changed is some real skepticism as to whether all of these enormous investments in AI on the part of the hyperscalers will ever actually pay off. And you've seen the hyperscalers go from being massive generators of free cash to now having to borrow money to support these investments. And so that's led to a dramatic change internally within the market where some of the AI-related trades have been under pressure and other parts of the economy and the stock market have been doing well, which I think is healthy. I'd say we remain relatively constructive on the market, but quite cognizant of the risks.

Chris Zand: Thanks, John. Very helpful. I'd like to come back to the AI investment cycle discussion in a few minutes, but before we do, let's broaden the conversation. Equity markets rarely tell the whole story, and fixed income often provides important insights into how investors are thinking about growth, inflation, and policy. Carl, as you look at the bond market today, what do you think it's telling us?

Carl Kaufman: Well, there's a couple of things. As you know, primarily, we tend to invest in high yield bonds, and those generally move in the same direction as equities. And usually for the same reasons that John mentioned, earnings are good, free cash flows are good. We don't do too much of the debt and the hyperscalers clearly because there's not much value there. Our market had a very nice quarter as well, and that was really due to corporate earnings being strong and the oil price coming down in the Middle East. As you know now, oil price is moving back up, so we'll have to see how the quarter progresses. And, pardon the pun, but the situation remains fluid. So we'll just need to keep an eye on that. Interestingly, my team did look into the impact of higher oil prices on the economy. And there's definitely a case to be made that even if oil remains elevated for an extended period of time, it's not as big a concern as it used to be.

Most people don't realize that on an inflation-adjusted basis, gasoline costs about the same as it did 30 years ago. Other goods and services such as health care and higher education have gone up much more starkly. And to put it in more concrete terms, the amount that the households are currently spending on energy is roughly half of what they spent back during the peak of the 1980s. So the economy is much less energy intensive than it was. So we don't worry as much as we did about energy. And also add to the fact that we were a very large importer of oil back then, and we are now fairly energy self-sufficient, which is a big change.

Chris Zand: Fascinating, Carl. Let's come back to inflation for a moment. John touched on a favorable June CPI report, which at first glance was certainly a step in the right direction. But as we know, one month's data is rarely the whole story. What stood out to you beneath the headline numbers, and has it changed your outlook for inflation for the remainder of this year?

Carl Kaufman: Well, the report was good, as you said for sure, but maybe we're a little less sanguine about it because inflation is still above the Fed's target and remains there, stubbornly so. To refresh everybody's memory, the headline CPI for the month of June fell to 3.5% year-over-year, which is a 0.4% decline in monthly prices. There was an impressive move in the right direction, but the headline number includes food and energy prices. Recall that June was when optimism about a long-term resolution to Iran and oil prices coming down happened. So that had an impact on that month. Now oil prices are back up. We don't know whether we have a deal. If we're even talking, we'll just have to wait and see. So core CPI removes energy and food, which are two of the big factors. Unfortunately, we all need energy and food. So the consumer, even though core CPI or core CPIX and other stuff, they'll keep on taking things out till they get to the number they want, clearly is still elevated. And flat is better than higher, but given that CPI is still above the Fed's target, it still seems we have a ways to go before we declare victory.

So basically we're in a wait and see mode. The Fed is meeting this week, today and tomorrow. So we'll have to see what they decide. I mean, there's one camp says they'll hold. A smaller camp says they may raise this week, but most people are thinking they don't raise till September. But I still think a 25 basis point increase could happen later this year.

Chris Zand: Thanks, Carl. Makes a lot of sense. I'd like to come back to AI now. It's certainly one of the defining investment themes of this cycle. Nael, I'd like to turn to you. John mentioned earlier that investors may be starting to question the sheer scale of capital being spent by the hyperscalers. For the past couple of years, the market clearly has rewarded that kind of investment. Why have things started to change from your perspective? Why are investors beginning to scrutinize spending levels more closely now? And what do you think they're trying to determine?

Nael Fakhry: Yeah, good to see you Chris, and good to be with everyone. I think the best way to approach the question is to think about how AI works. It's obviously a revolutionary technology, but it's also simple in that every time you interact with AI in terms of a chatbot or if you're using some sort of a coding program or whatever it might be, the AI tool basically pings a data center. And what's unique is that that process is very compute-intensive. And because AI is so ubiquitous, adoption is rapidly growing, demand for compute is just skyrocketing. So what does that mean? Well, chips, the semiconductors, the servers, the actual data centers, the power, the cooling, everything that goes into it is really expensive and getting more expensive. And most companies just don't have the capital, the resources to handle all of that. And that's where the hyperscalers fit in.

So companies like Amazon, Alphabet, Microsoft, Meta, Oracle, those five companies are what are called hyperscalers, and they are basically building the AI infrastructure and capacity to handle all of that. In aggregate, they're spending this year over $740 billion in capital expenditures to build out AI infrastructure. That's an astonishing number. And longer term, that is expected to accumulate to something on the order of $5 trillion by 2030. So we're talking about really big numbers and a constant resetting higher of the bar in terms of capital spending that these companies are doing.

Chris Zand: Thanks, Nael. That is remarkable in terms of how much spending is happening and the fact that these companies have been able to evolve from asset light software businesses to now some of the world's largest infrastructure investors. Greg, I'd like to turn to you and bring this back to our portfolios. How has this evolution influenced how we're positioned today?

Greg Hermanski: Sure. Thanks, Chris. As you've heard us say many times before, we're looking for companies that fit our quality growth framework. And one of the key elements of a quality growth company is increasing levels of cash flow. So as we consider the hyperscalers that Nael was just talking about, Alphabet, Amazon, Microsoft, we've been looking at metrics that include revenue growth, increasing margins, increasing operating cash flow, as well as how their balance sheet is, the absolute level and strength of their balance sheet as key indicators of whether or not incremental capital investments make business sense. And so far as we look at these numbers, all of those metrics have been growing and they're positive indicators. So one of the things that we love with a quality growth company is it gives them the choice with the strong free cash flow generation to either to invest back into their businesses when they see growth and they see wonderful opportunities.

And really not all companies have that capability or choice. So right now, the hyperscaler management teams, as they look at their business and opportunities in front of them, they're investing because they see these wonderful opportunities. But we like the fact that they're in the position of choosing, if things warrant, to slow down their investments or even stop their investments when things change, and then they'll be in the position to generate significant free cash flow to get back to shareholders.

Chris Zand: Thanks, Greg. One of the things we've focused on so far has been the enormous amount of AI investments, but the opportunity set is clearly much broader than that. Can you talk a bit about where else we're finding attractive opportunities in the AI ecosystem?

Greg Hermanski: Yeah, for sure. We've talked about a lot of these opportunities in the past. They include semiconductor companies that are supplying the key tools for the data center builds. But there's also companies like EDA companies, Synopsys is an example, that make the software to design the semiconductors. And then there's companies like Applied Materials that make the equipment to manufacture the semiconductors. And then also companies like Keysight, which make testing equipment to test semiconductors. They test the data networks and they test other electronic products that go into the AI ecosystem. And then also lesser-known would be companies like the industrial gas companies that provide critical gases that are used to manufacture semiconductors and other electronics. All of these industries are benefiting from the AI theme. And that having been said, we think it's important to recognize that because they're all tied together by the AI theme, that it creates something that we call "stacked risk."

So this year, as we've seen valuations go up and expectations go up, we've been careful to reduce our exposure to many of those industries.

Chris Zand: Greg, I have one more question for you. In our last Reading the Tea Leaves episode, we discussed SpaceX. Can you briefly recap what we covered and share any new thoughts you may have about the company?

Greg Hermanski: Sure. SpaceX is an amazing company, first of all. It's used its pace and innovation to push the known technology limits. And as a result, it's created amazing business opportunities for shareholders. Some of the notable innovation that they've had, they've created reusable rockets, which has significantly driven down the cost of putting things in space, as well as low orbit satellite-based data networks that reduce latency and drive down the cost of communication around the world. SpaceX's innovation is enabling some very large long-term market opportunities, not just in communications and launch, but also opportunities like data centers in space. So while we think that SpaceX is truly a remarkable company, we also see a lot of technology risk to get there and very significant free cash flow burn, and that'll be necessary to fulfill the potential of all their opportunities. So all in all, it's an amazing story, and we'll closely follow it to see if it becomes an amazing stock opportunity as it is an amazing innovation story.

Chris Zand: Thanks, Greg. Clearly, whether we're talking about the hyperscalers, the broader set of opportunities related to AI, or even space companies, there's massive amounts of spending happening. Nael, I want to come back to you and ask a really simple question. From your perspective, how sustainable can all this spending truly be?

Nael Fakhry: Yeah, that's I think exactly the right question, and we've touched on this a bit. I would make a couple points. I mean, first of all, as John alluded to earlier, the hyperscalers before AI were massive generators of free cash flow, free cash flow just being another measure of profitability. And they were able to, after generating the operating cash flow of the business, investing in the business, they were able to pay out dividends in many cases, buy back enormous amounts of their own stock, and do acquisitions. And so they were really harvesting the fruits of all these investments they had made over the past 20, 25 years. And with these new data center commitments, the debt and equity markets that they're tapping, things are changing. These are companies that are now spending so much that they're absorbing all of that operating cash flow and investing it in CapEx, then looking for outside sources of capital in order to fund all of that.

And the important thing is, Greg alluded to this as well, the revenues are accelerating, margins are improving for these businesses, and so we're seeing returns on capital, but these companies now have to manage their operating expenses much more carefully. You hear about headcount reductions now and much more prudent growth in headcount. They're much more vigilant about managing their balance sheets, and they have to be. And this is a really radical transformation from companies that were almost in many ways mature. They were returning capital to shareholders, to now they're returning to being almost like startups where they're having to attack outside capital in order to grow and attack what is a massive, massive growth opportunity. I think the key is that we don't think all hyperscalers are made equal. Those that have a long history of going after new areas of growth, investing in them and are disciplined about it, those that have good balance sheets and are not overextended, those are the ones that I think are interesting because at the end of the day, this is very simple.

If they're willing to invest today to capture a big growth opportunity in the future, then the free cash flow in the future post-AI and post this cycle will be higher than what it was before. And we've made that investment many times across a bunch of different industries. And so as long as the payoff is there, we are happy to make that investment, but we're monitoring it very closely because it's a huge shift that you can't really overstate.

Chris Zand: Thanks, Nael. Very helpful. Let's change gears now and touch on one more topic that has received a lot of attention this year, and that's private credit. Carl, we've seen plenty of headlines surrounding questions about the health of this market. What do you think investors should understand about what's happening today and what's your current assessment of the asset class overall?

Carl Kaufman: Sure. Our view on it has not changed much for a while now. We have been skeptical of the wonderful returns that they're able to provide with no risk. I think that what is happening is we are starting to see cracks in the system. Remember, this is a relatively new sector that received a huge rush of money into various funds and sponsors who had to put it to work. And I would say that the credit work that they did and the due diligence they did was probably not as strong as it should have been. So we have been seeing some very high profile defaults.

Some of them have been large. A recent company, Medallia, failed, and it wiped out $5 billion of equity in one fell swoop. So what that's caused is it's caused a demand for redemptions that is much larger than the gates that these people, the sponsors have put up. And so I think it's a case of investors not having read the fine print when they invested. And keep in mind that this is the area where companies that were unable to raise money efficiently in the public markets migrated to both the loan market and private credit. So this consists mostly of smaller or higher leveraged companies, which are more risky. Now, we have a very strong economy underpinning them right now. And despite that, we've had a number of high profile bankruptcies. So I think that it's growing pains. I don't think it's systemic. I mean, keep in mind that in our market, the public markets, nearly 60% of that is rated BB, which is the highest rating in high yield versus 40% in 2007.

While, the CCC component is now 8.5% and it was 18% in 2007. Now they got up to 20% in 2008. So private credit, while it's been pretty good for companies to access that market, it is creating some problems for investors who are exposed to that asset class right now. So we'll have to see how this plays out eventually, but I think we have to find a level and hopefully their due diligence and their infrastructure of these funds has improved to the point where they can start parsing their risk a little better than they have.

Chris Zand: Great, Carl, thanks for the insights there. As we wrap up today's session, we've covered a lot of ground today from the resilience in the economy and inflation to AI, capital spending, and portfolio positioning. Stepping back from it all, John, what's your outlook for the next six to 12 months? And where do you think investors should be focusing their attention?

John Osterweis: Well, I wish I had something new and exciting to say. I think the market is going to continue to waffle between a focus on a very strong economy on the one hand and these exogenous risks coming from the conflict in the Middle East on the other hand. And within the market, you'll have this tension between the AI trade and the non-AI trade. And so I think probably the safest bets will be in finding companies that may have some temporary problem or setback that are selling cheap, that are fundamentally sound and growing, and will do so in pretty much any market. And I've probably said that on every single call I've ever been on, but I think finding these situations that are not part of some major mania, but are great companies, but under a cloud, hopefully a temporary cloud, is the place to hunt. So that would be my best advice.

Chris Zand: Thanks, John. Very helpful. I think one of the key takeaways from today's discussion is that while the headlines continue to evolve, our investment framework remains very consistent. The economy is resilient. There are attractive opportunities across the AI ecosystem and reasons for cautious optimism, but a continued focus on valuation, risk management, and disciplined portfolio construction remain key to our process. So with that, I'd like to open things up to our audience now. As always, you can raise your hand for asking a live question or enter it into the Q&A box. We'll start with one that we received by email prior to today's question. Carl, this one's for you. Nael mentioned that the hyperscalers are borrowing a lot of money that they need for their massive data center buildouts. How is that impacting fixed income markets?

Carl Kaufman: Thanks. It's a good question. Heavy issuance that clearly drive up rates and it does increase aggregate credit risk as well. We've been avoiding hyperscaler debt thus far. I mean, there's some big issues by Meta earlier in the year, $30 billion that was really tied to the data center's success, not recourse to Meta. Oracle has been a huge issuer of debt, and they are the exception among the, they're a mini hyperscaler, I guess you'd call them, in that they don't have the free cash flows that the other guys do to support it. And that debt has traded down quite a bit since issuance. So we've been avoiding it and waiting for the dust to settle.

Chris Zand: Great. Thanks, Carl. Another question that's come through. There's a lot of discussion lately about moving to open source AI models rather than proprietary models. Does that impact what the endpoint of AI revenue will be and, as a result, might reduce expected AI spending due to lower potential returns?

Carl Kaufman: Quite possibly. There are two million AI models out there, so there's a lot to choose from. Not every one is going to be around in ten years. And I think that what we've seen so far is that you've gone from firms telling their people to token max; you know that tokens are how they charge for AI to, no, no, wait a minute, the price of tokens has gone up, so stop that. And I've heard stories about people going to, we're paying a couple of grand a month as a small firm for AI for a specific model, to being put on the meter and jumping to 50,000 a month. And they've basically gone to an OpenAI model. So you're seeing some of that already, and we're still very early in this development of AI. So I think there's going to be winners and losers. And I think for many, many applications, it's going to be the open source models.

And for highly specialized uses that require, like drug discovery and things like that, they'll still pay the price because it's worth it. They're working on billion dollar projects. But I think there's going to be a shakeout coming and we just don't know who the winners and losers are going to be or what the pricing model is going to look like.

Greg Hermanski: Adding to what Carl said, it's unclear at this point. There is also some issues. Some of the open source models look like they're cheaper, but they're actually less efficient in processing tokens and processing cash. And so it's unclear how much cheaper they are. And then the other element is, and I do think we're going to have some more efficient types of models to use. Does that drive demand overall, and does that positively impact the spending on AI and the development of AI? So it's a little bit unclear, but we're in the early innings and things are going to develop and things are going to change over the next few years, exactly how it's going to work and how it's going to look.

Chris Zand: Okay. Well, that looks like all the questions we've had in the Q&A. So I think we'll wrap up today's session. I'd like to thank all our panelists for their thoughtful insights and all of you for joining us today. We'll be back for the September roundtable and wish you all a great summer.