The S&P 500 Index returned 2.41% and the S&P Global Broad Market Index returned -1.16% in the quarter ended December 31, 2024.1
During the fourth quarter, we continued the rebalancing discussed in our last two letters. First, we trimmed our position in CBRE Group to buy more FICO. While both stocks outperformed in 2024 due to strong idiosyncratic business momentum as well as interest-rate driven improvements in commercial and residential transaction activity levels, our continued research on FICO’s untapped pricing power and capital-lite economic model led us to reappraise our relative weightings in the stocks. Second, we sold our positions in Nike and Estee Lauder to buy more Hèrmes and LVMH. As we mentioned in our last letter, we’ve gained more and more conviction that the most favorable long-term economics accrue to the brands that become the most desirable status symbol in their category through a combination of heritage, story-telling, and controlled supply. While Hèrmes and Ferrari are the two luxury companies that most epitomize these characteristics, LVMH has amassed an incredible collection of luxury assets since Bernard Arnault founded the company in 1987. In fact, LVMH is the largest luxury goods company in the world with over 75 brands and 6,097 stores,2 giving it tremendous bargaining power with retailers, celebrity endorsers, and owners of prime real estate on fashionable streets such as Fifth Avenue, New York. Lastly, since the time of our incremental purchases of Hèrmes and LVMH roughly two months ago, both stocks have significantly outperformed the index while our holding in FICO has significantly underperformed the index. Given our desire to opportunistically increase the size of our holding in FICO, we took advantage of this big relative performance gap by selling some of our LVMH position during January 2025 to buy more FICO.
Late 90s Tech and Telecom Bubble Redux?
Since the introduction of ChatGPT in November 2022,3 stocks powering or utilizing artificial intelligence (“AI”) have dominated best performer lists. 2024 was no exception. A disproportionate number of the top 10 best performing stocks of 2024 were the semiconductor, software, energy, and utility companies driving the massive datacenter buildout that is required to advance the capabilities of ChatGPT and its competitors.4 While this dominance of the best performers by the hot new thing is not that unusual, what is unusual is that the top eight stocks in the S&P 500 as of December 31, 2024, accounting for over 35% of the index,5 are all perceived to be beneficiaries of the AI boom and thus have massively outperformed the market as well. They are, in order of market cap, Apple, Nvidia, Microsoft, Amazon, Alphabet, Meta, Tesla, and Broadcom. The last time we can remember seeing similar enthusiasm around a new capital-intensive technology was during the late 90s buildout of the internet. However, even then, the market was arguably much less concentrated into one thematic area, with Microsoft, Cisco, Intel, IBM, and Oracle accounting for only 5 of the top 10 companies and 13% of the S&P 500 Index. 6
Given this analogy, the obvious question is, “Okay, if the AI boom reminds you of the dotcom bubble, does that mean it’s also a bubble, and will it have a bust as painful as previous capital-investment-driven technology bubbles?” To answer this question, let’s first look at some of the previous bubbles in more detail. The two most famous examples of these types of bubbles are the railroad bubble of the 1840s and the dotcom bubble of the late 1990s. Both these bubbles followed a similar script. Exciting new technologies were created that had tremendous potential to drive massive productivity improvements but that required tremendous capital investment. Investors and promoters told compelling, plausible stories about the revenue and profits that could be captured by funding these innovations, leading to a flood of capital investment across multiple industries. Over time, as the buildout continued, the incentives of the promoters and animal spirits of the investors resulted in more and more aggressive short- and medium-term predictions. Eventually, while demand continued to grow rapidly for both railroad networks and the internet, it ended up falling far short of the demand growth expected by the stock and bond markets. Simultaneously, the builders of the railroads discovered ways to ship far more goods per railroad track than they anticipated, and the builders of the internet created new technologies that allowed far more data to be transmitted on each fiberoptic cable than expected. Furthermore, these supply-side innovations occurred after huge amounts of capital had already been invested in projects that were midway through completion and thus could not easily be reversed. As a result, the supply of goods, services, and infrastructure built during the booms ended up being far greater than the short- and medium-term demand, crushing prices, revenues, and profits relative to expectations and causing huge stockholder losses. Even worse, a large proportion of these projects were funded with debt, further magnifying stockholder losses and creating dangerous distress in the debt markets as well.
As we look at the artificial intelligence boom today, there are obvious similarities. Just like in previous booms, investors and promoters are telling plausible stories about the ways in which artificial intelligence could drive huge productivity improvements. AI agents, robots, self-driving automobiles, and even artificial general intelligence all seem within reach, and it certainly seems reasonable that the companies that own or power these technologies could generate large and growing revenues and profits. As a result, trillions of dollars are being invested into manufacturers, energy producers, utilities, and technology companies to drive the technical solutions that will make these possibilities a reality. However, just like in previous booms, it’s incredibly difficult to have confidence in the timing of the technical solutions, the investment required to develop these solutions, and, lastly, the monetizability of the solutions once they’re developed. For instance, it’s plausible that Tesla solves self-driving next year. It’s also plausible it takes the company ten years or even much longer to solve it. In each of these scenarios, the amount of capital required to develop the solution and the present value of the profits resulting from the solution are likely to be wildly divergent. Furthermore, it’s plausible that only Tesla solves self-driving, that Tesla and five other companies solve it at the same time, that another company solves it instead, or that one of the many other possible permutations occurs. Just like in the timing question, the revenue and profit outlook for Tesla in particular and the industry as a whole are likely to be dramatically different depending on which scenario occurs. In some of these scenarios, Tesla or a competitor could develop a networked monopoly, and, in others, self-driving could become completely commoditized and the consumer could capture all the benefit of the new technology.
Now that we’ve looked at the similarities, let’s look at some of the differences. Perhaps the most important difference is that the boom today is largely being funded by equity and not debt. More specifically, it is being funded by four of the top ten companies in the S&P 500 (Microsoft, Amazon, Alphabet, and Meta) using their massive cash balances and prodigious cash flows as sources of funds. Another important difference is that we believe the eight technology businesses at the top of the S&P 500 are, as a group, much better businesses than their counterparts in both the railroad era and the dotcom bubble. Most of them are globally networked businesses with high margins, strong returns on invested capital, and/or significant untapped pricing power. Furthermore, as a group, they trade at significantly cheaper valuation multiples than their counterparts did in the tech and telecom bubble,7 though they will still turn out to be quite expensive if their hundreds of billions of dollars of AI investments ultimately end up generating poor returns.
Concluding Thoughts
We are currently in the middle of a historic capital-investment-driven technology boom that promises huge advancements in artificial intelligence and productivity. While there are some important business quality, valuation, and capital structure differences that somewhat mitigate the risk of the AI boom as compared to previous capital-investment-driven technology booms, the fundamental uncertainty created by the boom is the same. The speed of innovation combined with the huge amounts of capital being invested in multi-year projects make it especially difficult to predict how fast the demand and supply of useful services will grow, and, therefore, how profitable all this spending will turn out to be. Furthermore, relative to history, the market is uniquely concentrated, causing, in our view, a worrisome degree of exposure to this fast-changing and unpredictable boom. Given these facts, we prefer our index-agnostic approach where we first try and identify the most durable, predictable businesses in the world, and then, so long as we believe they are attractively priced, diversify among these businesses as much as possible so that your portfolio isn’t too reliant on a single investment theme, industry, macroeconomic factor, or geography. Our hope is that this process has resulted in a portfolio that can grow your purchasing power over the long term in a wide range of future scenarios by effectively operating as a diversified, recession-resistant toll collector on global GDP. See the below infographic for a visualization of this concept. 8
Sectors and Corresponding Oligopolistic Toll Collectors
As you look at the above portfolio, we’d like to highlight a few of its key characteristics. First, observe that it’s full of dominant, oligopolistic global companies that have successfully grown for decades or even centuries despite the many economic and geopolitical challenges that have occurred over their lifespans. Second, note that it does include some of the mega-cap tech companies (Microsoft, Amazon, Apple, and Alphabet) that are involved in the AI boom. Our reason for owning these companies despite the significant uncertainty created by the AI boom is that we believe they, as well as all the other companies in the portfolio, continue to possess the six characteristics that we’ve found make businesses durable and predictable. In our view, they 1) own dominant networks; 2) possess other checks on competing supply such as not-in-my-backyard (NIMBY) zoning restrictions, institutional risk aversion, and switching costs; 3) possess significant untapped pricing power; 4) operate in categories that we believe will grow at least as fast as GDP; 5) have conservative balance sheets; and 6) are run by ownership-minded management teams. Lastly, observe that, while the portfolio does have some exposure to the AI theme, it also has large weightings in industries and companies with, at most, a tangential relationship to the AI boom. Specifically, we believe the long-term business prospects of our holdings in payments, luxury, railroads, insurance brokerage, capital markets, waste management, and commercial real estate are unlikely to be significantly impacted by artificial intelligence. As a result, no matter what path the AI boom takes, we believe your portfolio is prepared.
As always, thank you so much for your trust, know that we continue to be invested right alongside you, and please always reach out to us if you have any questions or concerns. We’re here to help!
Sincerely,
The YCG Team
Disclaimer: The specific securities identified and discussed should not be considered a recommendation to purchase or sell any particular security nor were they selected based on profitability. Rather, this commentary is presented solely for the purpose of illustrating YCG’s investment approach. These commentaries contain our views and opinions at the time such commentaries were written and are subject to change thereafter. The securities discussed do not necessarily reflect current recommendations nor do they represent an account’s entire portfolio and, in the aggregate, may represent only a small percentage of an account’s portfolio holdings. A complete list of all securities recommended for the immediately preceding year is available upon request. These commentaries may include “forward looking statements” which may or may not be accurate in the long-term. It should not be assumed that any of the securities transactions or holdings discussed were or will prove to be profitable. S&P stands for Standard & Poor’s. All S&P data is provided “as is.” In no event, shall S&P, its affiliates or any S&P data provider have any liability of any kind in connection with the S&P data. MSCI stands for Morgan Stanley Capital International. All MSCI data is provided “as is.” In no event, shall MSCI, its affiliates or any MSCI data provider have any liability of any kind in connection with the MSCI data. Past performance is no guarantee of future results.
1 For information on the performance of our separate account composite strategies, please visit www.ycginvestments.com/performance. For information about your specific account performance, please contact us at (512) 505-2347 or email [email protected]. All returns are in USD unless otherwise stated.
2 See https://www.lvmh.com/en/investors.
3 See https://en.wikipedia.org/wiki/ChatGPT.
4 See https://markets.businessinsider.com/news/stocks/the-10-best-performing-s-and-p-500-stocks-in-2024-include-nvidia-tesla-2024-12#2-vistra-9.
5 See https://www.ssga.com/library-content/products/factsheets/etfs/us/factsheet-us-en-spy.pdf.
6 See https://www.finhacker.cz/top-20-sp-500-companies-by-market-cap/#1999.
7 As of January 22, 2025, according to Refinitiv data, the top eight technology stocks in the S&P 500 traded at a market cap weighted next-twelve-months earnings multiple of 36.9x and a simple average next-twelve-months earnings multiple of 42.2, both of which compare favorably to the 55.3x and 92.4x multiples of the top ten tech stocks in March 2000. See https://x.com/GavinSBaker/status/1526615551480582145 for data from March 2000. See https://x.com/GavinSBaker/status/1815421623941341303 and https://x.com/lhamtil/status/1819093619539615989 for additional valuation comparisons between then and now.
8 While they are great businesses with powerful competitive advantages that we believe will persist over time, our small holdings in Pepsi, Procter & Gamble, Colgate, and Progressive don’t fit quite as neatly into our toll collector framework, so we’ve excluded them from the graphic.
