Thoughts on the Market

• Funding the Next Phase of AI Development

  Recorded at our 2025 Technology, Media and Telecom (TMT) Conference, TMT Credit Research Analyst Lindsay Tyler joins Head of Investment Grade Debt Coverage Michelle Wang to discuss the how the industry is strategically raising capital to fund growth.----- Transcript -----Lindsay Tyler: Welcome to Thoughts on the Market. I'm Lindsay Tyler, Morgan Stanley's Lead Investment Grade TMT Credit Research Analyst, and I'm here with Michelle Wang, Head of Investment Grade Debt Coverage in Global Capital Markets.On this special episode, we're recording at the Morgan Stanley Technology, Media, and Telecom (TMT) Conference, and we will discuss the latest on the technology space from the fixed income perspective.It's Thursday, March 6th at 12 pm in San Francisco.What a week it's been. Last I heard, we had over 350 companies here in attendance.To set the stage for our discussion, technology has grown from about 2 percent of the broader investment grade market – about two decades ago – to almost 10 percent now; though that is still relatively a small percentage, relative to the weightings in the equity market.So, can you address two questions? First, why was tech historically such a small part of investment grade? And then second, what has driven the growth sense?Michelle Wang: Technology is still a relatively young industry, right? I'm in my 40s and well over 90 percent of the companies that I cover were founded well within my lifetime. And if you add to that the fact that investment grade debt is, by definition, a later stage capital raising tool. When the business of these companies reaches sufficient scale and cash generation to be rated investment grade by the rating agencies, you wind up with just a small subset of the overall investment grade universe.The second question on what has been driving the growth? Twofold. Number one the organic maturation of the tech industry results in an increasing number of scaled investment grade companies. And then secondly, the increasing use of debt as a cheap source of capital to fund their growth. This could be to fund R&D or CapEx or, in some cases, M&A.Lindsay Tyler: Right, and I would just add in this context that my view for this year on technology credit is a more neutral one, and that's against a backdrop of being more cautious on the communications and media space.And part of that is just driven by the spread compression and the lack of dispersion that we see in the market. And you mentioned M&A and capital allocation; I do think that financial policy and changes there, whether it's investment, M&A, shareholder returns – that will be the main driver of credit spreads.But let's turn back to the conference and on the – you know, I mentioned investment. Let's talk about investment.AI has dominated the conversation here at the conference the past two years, and this year is no different. Morgan Stanley's research department has four key investment themes. One of those is AI and tech diffusion.But from the fixed income angle, there is that focus on ongoing and upcoming hyperscaler AI CapEx needs.Michelle Wang: Yep.Lindsay Tyler: There are significant cash flows generated by many of these companies, but we just discussed that the investment grade tech space has grown relative to the index in recent history.Can you discuss the scale of the technology CapEx that we're talking about and the related implications from your perspective?Michelle Wang: Let's actually get into some of the numbers. So in the past three years, total hyperscaler CapEx has increased from [$]125 billion three years ago to [$]220 billion today; and is expected to exceed [$]300 billion in 2027.The hyperscalers have all publicly stated that generative AI is key to their future growth aspirations. So, why are they spending all this money? They're investing heavily in the digital infrastructure to propel this growth. These companies, however, as you've pointed out, are some of the most scaled, best capitalized companies in the entire world. They have a combined market cap of [$]9 trillion. Among them, their balance sheet cash ranges from [$]70 to [$]100 billion per company. And their annual free cash flow, so the money that they generate organically, ranges from [$]30 to [$]75 billion.So they can certainly fund some of this CapEx organically. However, the unprecedented amount of spend for GenAI raises the probability that these hyperscalers could choose to raise capital externally.Lindsay Tyler: Got it.Michelle Wang: Now, how this capital is raised is where it gets really interesting. The most straightforward way to raise capital for a lot of these companies is just to do an investment grade bond deal.Lindsay Tyler: Yep.Michelle Wang: However, there are other more customized funding solutions available for them to achieve objectives like more favorable accounting or rating agency treatment, ways for them to offload some of their CapEx to a private credit firm. Even if that means that these occur at a higher cost of capital.Lindsay Tyler: You touched on private credit. I'd love to dig in there. These bespoke capital solutions.Michelle Wang: Right.Lindsay Tyler: I have seen it in the semiconductor space and telecom infrastructure, but can you please just shed some more light, right? How has this trend come to fruition? How are companies assessing the opportunity? And what are other key implications that you would flag?Michelle Wang: Yeah, for the benefit of the audience, Lindsay, I think just to touch a little bit…Lindsay Tyler: Some definitions,Michelle Wang: Yes, some definitions around ...Lindsay Tyler: Get some context.Michelle Wang: What we’re talking about.Lindsay Tyler: Yes.So the – I think what you're referring to is investment grade companies doing asset level financing. Usually in conjunction with a private credit firm, and like all financing trends that came before it, all good financing trends, this one also resulted from the serendipitous intersection of supply and demand of capital.On the supply of capital, the private credit pocket of capital driven by large pockets of insurance capital is now north of $2 trillion and it has increased 10x in scale in the past decade. So, the need to deploy these funds is driving these private credit firms to seek out ways to invest in investment grade companies in a yield enhanced manner.Lindsay Tyler: Right. And typically, we're saying 150 to 200 basis points greater than what maybe an IG bond would yield.Michelle Wang: That's exactly right. That's when it starts to get interesting for them, right? And then the demand of capital, the demand for this type of capital, that's always existed in other industries that are more asset-heavy like telcos.However, the new development of late is the demand for capital from tech due to two megatrends that we're seeing in tech. The first is semiconductors. Building these chip factories is an extremely capital-intensive exercise, so creates a demand for capital. And then the second megatrend is what we've seen with the hyperscalers and GenerativeAI needs. Building data centers and digital infrastructure for GenerativeAI is also extremely expensive, and that creates another pocket of demand for capital that private credit conveniently kinda serves a role in.Lindsay Tyler: Right.Michelle Wang: So look, think we've talked about the ways that companies are using these tools. I'm interested to get your view, Lindsay, on the investor perspective.Lindsay Tyler: Sure.Michelle Wang: How do investors think about some of these more bespoke solutions?Lindsay Tyler: I would say that with deals that have this touch of extra complexity, it does feel that investor communication and understanding is all important. And I have found that, some of these points that you're raising – whether it's the spread pickup and the insurance capital at the asset managers and also layering in ratings implications and the deal terms. I think all of that is important for investors to get more comfortable and have a better understanding of these types of deals.The last topic I do want us to address is the macro environment. This has been another key theme with the conference and with this recent earnings season, so whether it's rate moves this year, the talk of M& A, tariffs – what's your sense on how companies are viewing and assessing macro in their decision making?Michelle Wang: There are three components to how they're thinking about it.The first is the rate move. So, the fact that we're 50 to 60 basis points lower in Treasury yields in the past month, that's welcome news for any company looking to issue debt. The second thing I'll say here is about credit spreads. They remain extremely tight. Speaking to the incredible kind of resilience of the investment grade investor base. The last thing I'll talk about is, I think, the uncertainty. [Because] that's what we're hearing a ton about in all the conversations that we've had with companies that have presented here today at the conference.Lindsay Tyler: Yeah. For my perspective, also the regulatory environment around that M&A, whether or not companies will make the move to maybe be more acquisitive with the current new administration.Michelle Wang: Right, so until the dust settles on some of these issues, it's really difficult as a corporate decision maker to do things like big transformative M&A, to make a company public when you don't know what could happen both from a the market environment and, as you point out, regulatory standpoint.The thing that's interesting is that raising debt capital as an investment grade company has some counter cyclical dynamics to it. Because risk-off sentiment usually translates into lower treasury yields and more favorable cost of debt.And then the second point is when companies are risk averse it drives sometimes cash hoarding behavior, right? So, companies will raise what they call, you know, rainy day liquidity and park it on balance sheet – just to feel a little bit better about where their balance sheets are. To make sure they're in good shape…Lindsay Tyler: Yeah, deal with the maturities that they have right here in the near term.Michelle Wang: That's exactly right. So, I think as a consequence of that, you know, we do see some tailwinds for debt issuance volumes in an uncertain environment.Lindsay Tyler: Got it. Well, appreciate all your insights. This has been great. Thank you for taking the time, Michelle, to talk during such a busy week.Michelle Wang: It's great speaking with you, Lindsay.Lindsay Tyler: And thanks to everyone listening in to this special episode recorded at the Morgan Stanley TMT Conference in San Francisco. If you enjoy Thoughts on the Market, please leave us a review wherever you listen and share the podcast with a friend or colleague today.

  --------  

  10:39
• Is There Too Much Focus on Fed’s Moves?

  While central bank policy will always matter for markets, our Head of Corporate Credit Research Andrew Sheets explains why investors should not be worried about the number of Fed cuts in 2025.----- Transcript -----Welcome to Thoughts on the Market. I'm Andrew Sheets, Head of Corporate Credit Research at Morgan Stanley. Today I’m going to talk about why the number of Fed rate cuts this year may matter less than you think.It's Wednesday, March 5th, at 2pm in London.Financial markets spend a lot of time discussing the Federal Reserve. And for good reason. The central bank of the world’s largest economy plays a central role in fighting inflation and setting interest rates. And what they’ll do this year is topical and shifting. At Morgan Stanley, our economists think that US Tariff and Immigration policy will lead the Fed to keep rates somewhat higher, for somewhat longer, than they did at the start of the year.Yet we think there may be just a little bit too much focus on just how much the Fed changes policy over the course of the year. Indeed, we’d go as far as to say that given the choice, investors should be rooting for less change, not more.To start, for all that’s happened in the world since the end of October of 2024, expectations for the Fed’s interest rate path have been remarkably stable. The US 2-year Treasury, which is a decent proxy of where the Fed’s rate will average over the next 24 months, has hovered in a very narrow range. It simply hasn’t been telling us very much; other factors have been moving markets.There’s also a pretty reasonable rule of thumb from history: stability is good. A stable Fed funds rate, almost by definition, implies a stable equilibrium that doesn’t involve overly high inflation pushing rates further up, or overly weak growth pushing them further down. The best growth in recent history, in the mid-1990s, occurred after the Fed reduced interest rates less than one-percent, and then kept them stable, at a pretty elevated rate for a pretty extended period of time.Large changes in rates, on the other hand, in either direction are a different story. Some of the markets worst losses have coincided with the largest declines in the Fed’s target rate – because those large rate cuts usually occur only when there is a large, unexpected slowing in the economy; something markets often don’t like.Meanwhile, we think the Fed also very much wants to avoid a scenario where it has to start raising rates again, given the potential confusion that this could signal after it only recently continued to lower them. And so if over the course of this year, the Fed does need to raise rates, given the very high bar we think they’ve given themselves for action – it probably suggests that something unexpected, and not in a necessarily good way, has occurred.Central bank policy will always matter for markets. But for investors, the question of whether the Fed will cut once, which is the Morgan Stanley base-case, twice, or not at all in 2025 may not matter all that much, at least for credit. Far more important is the performance of the economy, and whether big changes to tariffs or immigration policy drive big changes to growth and inflation. Those big changes, which could drive big changes in Fed policy responses, are the scenario that worries us.Thanks for listening. If you enjoy the show, leave us a review wherever you listen and share Thoughts on the Market with a friend or colleague today.

  --------  

  3:36
• What Will Tariffs Do to the U.S. Dollar?

  Our U.S. Public Policy and Currency analysts, Ariana Salvatore and Andrew Watrous, discuss why the dollar fell at the beginning of the first Trump administration and whether it could happen again this year. ----- Transcript ----Ariana Salvatore: Welcome to Thoughts on the Market. I'm Ariana Salvatore, Morgan Stanley's U.S. Public Policy Strategist.Andrew Watrous: And I'm Andrew Watrous, G10 FX Strategist here at Morgan Stanley.Ariana Salvatore: Today, we'll focus on the U.S. dollar and how it might fare in global markets during the first year of the new Trump administration.It's Tuesday, March 4th at 10am in New York.So, Andrew, a few weeks ago, James Lord came on to talk about the foreign exchange volatility. Since then, tariffs and trade policy have been in the news. Last night at midnight, 25 percent tariffs on Mexico and Canada went into effect, in addition to 10 percent on China. So, let's set the scene for today's conversation. Is the dollar still dominant in global currency markets?Andrew Watrous: Yes, it is. The U.S. dollar is used in about $7 trillion worth of daily FX transactions. And the dollar's share of all currency transactions has been pretty stable over the last few decades. And something like 80 percent of all trade finance is invoiced in dollars, and that share has been pretty stable too.A big part of that dollar dominance is because of the depth and safety of the Treasury security market.Ariana Salvatore: That makes sense. And the dollar fell in 2017, the first year of the Trump administration. Why did that happen?Andrew Watrous: Yeah, so 2017 gets a lot of client attention because the Fed was hiking, there was a lot of uncertainty about would happen in NAFTA, and the U.S. passed a fiscally expansionary budget bill that year.So, people have asked us, ‘Why the U.S. dollar went down despite all those factors?’ And I think there are three reasons. One is that even though the possibility that the U.S. could leave NAFTA was all over the headlines that year, U.S. tariffs didn't actually go up. Another factor is that global growth turned out to be really strong in 2017, and that was helped in part by fiscal policy in China and Europe. And finally, there were some political risks in Europe that didn't end up materializing.So, investors took a sigh of relief about the possibility that I think had been priced in a bit that the Eurozone might break up. And then a lot of those factors went into reverse in 2018 and the U.S. dollar went up.Ariana Salvatore: So, applying that framework with those factors to today, is it possible that we see a repeat of 2017 in terms of the U.S. dollar decline?Andrew Watrous: Yeah, I think it's likely that the U.S. dollar continues to go lower for some of the same reasons as we saw in 2017. So, I think that compared to 2017, there's a lot more U.S. dollar positive risk premium around trade policy. So, the bar is higher for the U.S. dollar to go up just from trade headlines alone.And just like in 2017, European policy developments could be a tailwind to the euro. We've been highlighting the potential for German fiscal expansion as European defense policy comes into focus. And unlike in 2017, when the Fed was raising rates, now the Fed is probably going to cut more this year. So that's a headwind to the dollar that didn't exist back in 2017.So, on trade, Ariana. What developments do you expect? Do you think that Trump's new policies will make 2025 different in any way from 2017?Ariana Salvatore: So, taking a step back and looking at this from a very high level, a few things are different in spite of the fact that we're actually talking about a lot of similar policies. Tariffs and tax policy were a big focus in 2017 to 2019, and to be sure, this time around, they are too, but in a slightly different way.So, for example, on tax cuts, we're not talking about bringing rates lower on the individual and corporate side. We're talking about extending current policy. And on tariffs and trade policy, this round I would characterize as much broader, right? So, Trump has scoped in a broader range of trading partners into the discussion like Mexico and Canada; and is talking about a starting point that level-wise is much higher than what we saw in the whole 2018 2019 trade friction period.The highest rate back then we ever saw was 25 percent, and that was on the final batch of Chinese goods, that list four. Whereas this time, we're talking about 25 percent as a starting point for Mexico and Canada.I think sequencing is also a really important distinction. In 2017, we saw the tax cuts through the Tax Cuts and Jobs Act (TCJA) come first, followed by trade tensions in 2018 to 2019. This time around, it's really the inverse. Republicans just passed their budget resolution in the House. That lays the groundwork for the tax cut extensions.But in the meantime, Trump has been talking about tariff implementation since before he was even elected. And we've already had a number of really key trade related catalysts in the just six weeks or so that he's been in office.Andrew Watrous: So, you mentioned expectations for fiscal policy. What are recent developments there, and what do you think will happen with U.S. fiscal?Ariana Salvatore: I mentioned the budget resolution in the house that was passed last week. And you can really think of that as the starting point for the reconciliation process to kick off. And consequently, the extension of the Tax Cuts and Jobs Act.To be clear, we think that House Republicans will be able to align behind extending most of the expiring Tax Cuts and Jobs Act, but that's still in the books until the end of 2025. So, we see many months needed to kind of build this consensus among cohorts of the Republican caucus in Congress, and we already know there's some key sticking points in the discussion.What happens with the SALT [State and Local Tax] cap? What sort of clawbacks occur with the Inflation Reduction Act? All these are disagreements that right now are going to need time to work their way through Congress. So not a lot of alignment just yet. We think it's going to take most of the year to get there.But ultimately, we do see an extension of most of the TCJA, which is like I said, current law until the end of 2025.But Andrew from what I understand when it comes to fiscal policy, there are really two stages in terms of the market impact that we saw in the last administration. Can you walk us through those?Andrew Watrous: Yeah, so one lesson from 2016 to 2018 is that there were really two stages of when fiscal developments boosted the dollar. The first was right after the U.S. election in 2016, and the second was much later after the Tax Cuts and Jobs Act passed. So right after the 2016 election, within a couple of weeks, the dollar index rallied from 98 up to 103, and 10-year Treasury yields rose as well.And then things sort of moved sideways in between these two stages. Ten-year Treasury yield just moved sideways. Fiscal wasn't as supportive to the U.S. dollar. And as we know, the dollar went down. And then we had the second stage more than a year later. So, the TCJA was passed in December 2017. And then the dollar rallied after that along with the rise in Treasury yield.So, we think that now, what we've seen is actually very similar to what happened in 2017, where the dollar and yields moved a lot after the 2024 election; but now the budget reconciliation process probably won't be a tailwind to the dollar until after a tax cuts extension passes Congress. And as you mentioned, that's not going to be for many, many months. So, in the interim, we think there's a lot of room for the dollar to go down.Ariana Salvatore: And just to level set our expectations there to your point, it is probably going to be later this year. House Republicans have to align on a number of key sticking points. So, we have passage somewhere on the third or fourth quarter of 2025.But when we think about the fiscal picture, aside from the deficit and the macro impacts, a really key component is going to be what these tax changes mean for the equity market. The extension of certain tax policies will matter more for certain sectors versus others. For example, we know that extending some of the corporate provisions, aside from the lower rate, will have an impact across domestically oriented industries like industrials, healthcare, and telecom.But Andrew, to bring it back to this discussion, I want to think a little bit more about how we can loop in our expectations for the equity market and map that to certain dollar outcomes. How do you think that this as a barometer has changed, if at all, from Trump's first term?Andrew Watrous: Yeah, currency strategists like me love talking about yield differentials. But from 2016 to 2018, the U.S. dollar did not trade in line with yield differentials. Instead, in the initial years of President Trump's first term, equities were a much better barometer than interest rates for where the U.S. dollar would go.After President Trump was elected in 2016, U.S. stocks really outperformed stocks in the rest of the world, and the U.S. dollar went up. Then in 2017, stocks outside the U.S. caught up to the move in U.S. stocks, and the U.S. dollar fell. Then in 2018, all that went into reverse, and U.S. stocks started outperforming again, and the U.S. dollar went up.So, what we've been seeing in stocks today really echoes 2017, not 2018. Stocks outside the U.S. have caught up to the post election rise in U.S. stocks. And so, just like it did in 2017, we think that the U.S. dollar will decline to catch up to that move in relative stock indices.Ariana Salvatore: Finally, Andrew, we already discussed the U.S. dollar negative drivers from 2017. But what happened to these drivers the following year in 2018? And is that any indication for what might happen in 2026?Andrew Watrous: So 2018, as you mentioned, does offer a blueprint for how the U.S. dollar could go up. So, for example, if trade tensions evolve in a direction where our economists would have to significantly downwardly revise their global growth forecasts, then the U.S. dollar could start to look more attractive as a safe haven. And in 2018, there was a big rise in long-end Treasury yields. That's not what we're calling for; but if that were to happen, then the U.S. dollar could catch a bid.Ariana Salvatore: Andrew, thanks for taking the time to talk.Andrew Watrous: Great speaking with you, Ariana.Ariana Salvatore: And thanks for listening. If you enjoy Thoughts on the Market, please leave us a review wherever you listen and share the podcast with a friend or colleague today.

  --------  

  10:06
• Will GenAI Turn a Profit in 2025?

  Our Semiconductors and Software analysts Joe Moore and Keith Weiss dive into the biggest market debate around AI and why it’s likely to shape conversations at Morgan Stanley’s Technology, Media and Telecom (TMT) Conference in San Francisco. ----- Transcript -----Joe Moore: Welcome to Thoughts on the Market. I'm Joe Moore, Morgan Stanley's Head of U.S. Semiconductors.Keith Weiss: And I'm Keith Weiss, Head of U.S. Software.Joe Moore: Today on the show, one of the biggest market debates in the tech sector has been around AI and the Return On Investment, or ROI. In fact, we think this will be the number one topic of conversation at Morgan Stanley's annual Technology, Media and Telecom (TMT) conference in San Francisco.And that's precisely where we're bringing you this episode from.It's Monday, March 3rd, 7am in San Francisco.So, let's get right into it. ChatGPT was released November 2022. Since then, the biggest tech players have gained more than $9 trillion in combined market capitalization. They're up more than double the amount of the S&P 500 index. And there's a lot of investor expectation for a new technology cycle centered around AI. And that's what's driving a lot of this momentum.You know, that said, there's also a significant investor concern around this topic of ROI, especially given the unprecedented level of investment that we've seen and sparse data points still on the returns.So where are we now? Is 2025 going to be a year when the ROI and GenAI finally turns positive?Keith Weiss: If we take a step back and think about the staging of how innovation cycles tend to play out, I think it's a helpful context.And it starts with research. I would say the period up until When ChatGPT was released – up until that November 2022 – was a period of where the fundamental research was being done on the transformer models; utilizing, machine learning. And what fundamental research is, is trying to figure out if these fundamental capabilities are realistic. If we can do this in software, if you will.And with the release of ChatGPT, it was a very strong, uh, stamp of approval of ‘Yes, like these transformer models can work.’Then you start stage two. And I think that's basically November 22 through where are today of, where you have two tracks going on. One is development. So these large language models, they can do natural language processing well.They can contextually understand unstructured and semi structured data. They can generate content. They could create text; they could create images and videos.So, there's these fundamental capabilities. But you have to develop a product to get work done. How are we going to utilize those capabilities? So, we've been working on development of product over the past two years. And at the same time, we've been scaling out the infrastructure for that product development.And now, heading into 2025, I think we're ready to go into the next stage of the innovation cycle, which will be market uptake.And that's when revenue starts to flow to the software companies that are trying to automate business processes. We definitely think that monetization starts to ramp in 2025, which should prove out a better ROI or start to prove out the ROI of all this investment that we've been making.Joe Moore: Morgan Stanley Research projects that GenAI can potentially drive a $1.1 trillion dollar revenue opportunity in 2028, up from $45 billion in 2024. Can you break this down for our listeners?Keith Weiss: We recently put out a report where we tried to size kind of what the revenue generation capability is from GenerativeAI, because that's an important part of this ROI equation. You have the return on the top of where you could actually monetize this. On the bottom, obviously, investment. And we took a look at all the investment needed to serve this type of functionality.The [$]1.1 trillion, if you will, it breaks down into two big components. Um, One side of the equation is in my backyard, and that's the enterprise software side of the equation. It's about a third of that number. And what we see occurring is the automation of more and more of the work being done by information workers; for people in overall.And what we see is about 25 percent, of overall labor being impacted today. And we see that growing to over 45 percent over the next three years.So, what that's going to look like from a software perspective is a[n] opportunity ramping up to about, just about $400 billion of software opportunity by 2028. At that point, GenerativeAI will represent about 22 percent of overall software spending. At that point, the overall software market we expect to be about a $1.8 trillion market.The other side of the equation, the bigger side of the equation, is actually the consumer platforms. And that kind of makes sense if you think about the broader economy, it's basically one-third B2B, two-thirds B2C. The automation is relatively equivalent on both sides of the equation.Joe Moore: So, let's drill further into your outlook for software. What are the biggest catalysts you expect to see this year, and then over the coming three years?Keith Weiss: The key catalyst for this year is proving out the efficacy of these solutions, right?Proving out that they're going to drive productivity gains and yield real hard dollar ROI for the end customer. And I think where we'll see that is from labor savings.Once that occurs, and I think it's going to be over the next 12 to 18 months, then we go into the period of mainstream adoption. You need to start utilizing these technologies to drive the efficiencies within your businesses to be able to keep up with your competitors. So, that's the main thing that we're looking for in the near term.Over the next three years, what you're looking for is the breakthrough technologies. Where can we find opportunities not just to create efficiencies within existing processes, but to completely rewrite the business process.That's where you see new big companies emerge within the software opportunity – is the people that really fundamentally change the equation around some of these processes.So, Joe, turning it over to you, hardware remains a bottleneck for AI innovation. Why is that the case? And what are the biggest hurdles in the semiconductor space right now?Joe Moore: Well, this has proven to be an extremely computationally intensive application, and I think it started with training – where you started seeing tens of thousands of GPUs or XPUS clustered together to train these big models, these Large Language Models. And you started hearing comments two years ago around the development of ChatGPT that, you know, the scaling laws are tricky.You might need five times as much hardware to make a model that's 10 percent smarter. But the challenge of making a model that's 10 percent smarter, the table stakes of that are very significant. And so, you see, you know, those investments continuing to scale up. And that's been a big debate for the market.But we've heard from most of the big spenders in the market that we are continuing to scale up training. And then after that happened, we started seeing inference suddenly as a big user of advanced processors, GPUs, in a way that they hadn't before. And that was sort of simple conversational types of AI.Now as you start migrating into more of a reasoning AI, a multi pass approach, you're looking at a really dramatic scaling in the amount of hardware, that's required from both GPUs and XPUs.And at the same time the hardware companies are focused a lot on how do we deliver that – so that it doesn't become prohibitively expensive; which it is very expensive. But there's a lot of improvement. And that's where you're sort of seeing this tug of war in the stocks; that when you see something that's deflationary, uh, it becomes a big negative. But the reality is the hardware is designed to be deflationary because the workloads themselves are inflationary.And so I think there's a lot of growth still ahead of us. A lot of investment, and a lot of rich debate in the market about this.Keith Weiss: Let's pull on that thread a little bit. You talked initially about the scaling of the GPU clusters to support training. Over the past year, we've gotten a little bit more pushback on the ideas or the efficacy of those scaling laws.They've come more under question. And at the same time, we've seen the availability of some lower cost, but still very high-performance models. Is this going to reshape the investments from the large semiconductor players in terms of how they're looking to address the market?Joe Moore: I think we have to assess that over time. Right now, there are very clear comments from everybody who's in charge of scaling large models that they intend to continue to scale.I think there is a benefit to doing so from the standpoint of creating a richer model, but is the ROI there? You know, and that's where I think, you know, your numbers do a very good job of justifying our model for our core companies – where we can say, okay, this is not a bubble. This is investment that's driven by these areas of economic benefit that our software and internet teams are seeing.And I think there is a bit of an arms race at the high end of the market where people just want to have the biggest cluster. And that's, we think that's about 30 percent of the revenue right now in hardware – is supporting those really big models. But we're also seeing, to your point, a very rich hardware configuration on the inference side post training model customization. Nvidia said on their on their earnings call recently that they see several orders of magnitude more compute required for those applications than for that pre-training. So, I think over time that's where the growth is going to come from.But you know, right now we're seeing growth really from all aspects of the market.Keith Weiss: Got it. So, a lot of really big opportunities out there utilizing these GPUs and ASICs, but also a lot of unknowns and potential risks. So, what are the key catalysts that you're looking for in the semiconductor space over the course of this year and maybe over the next three years?Joe Moore: Well, 2025 is, is a year that is really mostly about supply.You know, we're ramping up, new hardware But also, several companies doing custom silicon. We have to ramp all that hardware up and it's very complicated.It uses every kind of trick and technique that semiconductors use to do advanced packaging and things like that. And so, it's a very challenging supply chain and it has been for two years. And fortunately, it's happened in a time when there's plenty of semiconductor capacity out there.But I think, you know, we're ramping very quickly. And I think what you're seeing is the things that matter this year are gonna be more about how quickly we can get that supply, what are the gross margins on hardware, things like that.I think beyond that, we have to really get a sense of, you know, these ROI questions are really important beyond 2025. Because again, this is not a bubble. But hardware is cyclical and there; it doesn't slow gracefully. So, there will be periods where investment may fall off and it'll be a difficult time to own the stocks. And that's, you know, we do think that over time, the value sort of transitions from hardware to software.But we model for 2026 to be a year where it starts to slow down a little bit. We start to see some consolidation in these investments.Now, 12 months ago, I thought that about 2025. So, the timeframe keeps getting pushed out. It remains very robust. But I think at some point it will plateau a little bit and we'll start to see some fragmentation; and we'll start to see markets like, you know, reasoning models, inference models becoming more and more critical. But that's where when I hear you and Brian Nowak talking about sort of the early stage that we are of actually implementing this stuff, that inference has a long way to go in terms of growth.So, we're optimistic around the whole AI space for semiconductors. Obviously, the market is as well. So, there's expectations, challenges there. But there's still a lot of growth ahead of us.So Keith, looking towards the future, as AI expands the functionality of software, how will that transform the business models of your companies?Keith Weiss: We're also fundamentally optimistic about software and what GenerativeAI means for the overall software industry.If we look at software companies today, particularly application companies, a lot of what you're trying to do is make information workers more productive. So, it made a lot of sense to price based upon the number of people who are using your software. Or you've got a lot of seat-based models.Now we're talking about completely automating some of those processes, taking people out of the loop altogether. You have to price differently. You have to price based upon the number of transactions you're running, or some type of consumptive element of the amount of work that you're getting done. I think the other thing that we're going to see is the market opportunity expanding well beyond information workers.So, the way that we count the value, the way that we accrue the value might change a little bit. But the underlying value proposition remains the same. It's about automating, creating productivity in those business processes, and then the software companies pricing for their fair share of that productivity.Joe Moore: Great. Well, let me just say this has been a really useful process for me. The collaboration between our teams is really helpful because as a semiconductor analyst, you can see the data points, you can see the hardware being built. And I know the enthusiasm that people have on a tactical level. But understanding where the returns are going to come from and what milestones we need to watch to see any potential course correction is very valuable.So on that note, it's time for us to get to the exciting panels at the Morgan Stanley TMT conference. Uh, And we'll have more from the conference on the show later this week. Keith, thanks for taking the time to talk.Keith Weiss: Great speaking with you, Joe.Joe Moore: And thanks for listening. If you enjoy Thoughts on the Market, please leave us a review wherever you listen and share the podcast with a friend or colleague today.

  --------  

  12:49
• Searching for Signals in U.S. Policy Noise

  Our Global Head of Fixed Income Research and Public Policy Strategy explains why conflicting news on tariffs and government spending may point to a case for bonds.----- Transcript -----Welcome to Thoughts on the Market. I’m Michael Zezas, Morgan Stanley’s Global Head of Fixed Income Research and Public Policy Strategy. Today we’ll be discussing recent U.S. public policy headline noise and the signal within that for investors.It’s Friday, February 28th, at 12:30 pm in New York.For investors paying attention to events in Washington, D.C., the past few weeks have been disorienting. Tariff announcements have continued, but with shifting details on timing and magnitude. And Congress passed a bill to enable substantial spending cuts, but subsequent media reports made clear the votes to actually enact these cuts later this year may not be there. Our recent client conversations have revealed that investors’ confusion has reached new heights, and there’s little consensus, or conviction, about whether U.S. policy choices are set to help or hurt the economy and markets. Net-net, it's a lot of policy noise, and very little signal. That said, here’s what we think investors can anchor to. For all the headlines on potential new tariffs for China, Mexico, Canada and on products like copper, actual tariff actions have followed a graduated pace, in line with our base case of ‘fast announcement, slow implementation’ – where tariffs on China start and continue to climb, but tariffs on the rest of world move slowly and are more subject to negotiation. Tariffs on Mexico and Canada appear, in our view, likely to be pushed out once again given progress in negotiation on harmonizing trade policy and progress in reduced border crossings. On the other hand, tariffs on China, already raised an incremental 10 percent a few weeks back, seem likely to step up again as there are much bigger disagreements that the two nations don’t appear close to resolving. But even if tariffs move according to the pace that we expect, that doesn’t mean they come without cost. The U.S.’s goal is to bring more investment onshore, with an aim toward increasing goods production, thereby reducing trade deficits, securing important supply chains, and growing industrial jobs. The theory is that higher tariff barriers might incentivize more direct investment into the U.S., as companies build supply chains in the U.S. to avoid the higher tariff costs. But even if that theory plays out, there’s a cost to that transition. In a recent blue paper, my colleague Rajeev Sibal led a team through an analysis demonstrating that the next phase of supply chain realignment would be considerably costlier to companies, given the complexity of production that must be shifted. So either way, companies take on new costs – tariffs, CapEx, or both. That challenges corporate margins, and economic growth, at least for a time. And there’s plenty of execution risk along the way. So what’s an investor to do? Our cross asset and interest rate strategy teams think it's time to lean more heavily into bonds. Equity markets may do just fine here, with investors looking through these near term costs, but the risk of something going wrong with, for example, tariffs escalation or broader geopolitical conflict, may keep a ceiling on investors’ risk appetite. Conversely, a growth slowdown presents a clearer case for owning bonds, particularly since it wasn’t that long ago that better economic data helped the Treasury market price out most of the expected monetary policy cuts for 2025. Thanks for listening. If you enjoy the show, please leave us a review wherever you listen and share Thoughts on the Market with a friend or colleague today.

  --------  

  3:44

Más podcasts de Economía y empresa

Podcasts a la moda de Economía y empresa

Acerca de Thoughts on the Market

Thoughts on the Market: Podcasts del grupo