Q3: Measuring by the business, not the quote.

The S&P 500 Index returned 2.30% and the S&P Global Broad Market Index returned 1.15% in the quarter ended September 30, 2026.¹

Market Commentary

This quarter continued the trend we’ve discussed ever since ChatGPT was released in November 2022: investors continue to sell old-economy, non-AI stocks to fund their purchases of ever more expensive AI stocks. At the end of September, Goldman Sachs noted that, while strength in AI stocks has kept the market steady, the median S&P 500 stock traded 16% below its 52-week high. Rather ominously, they pointed out that the last time market breadth was this low was during the dot-com bubble.² None of this should be surprising. Today’s market rhymes with the dot-com boom in so many ways, and both rhyme with the many other technology booms and busts that came before them. The pattern almost always goes something like this:

1. A transformative new technology is born.
2. Companies and individuals invest aggressively to profit from the boom.
3. Profits and valuations of the businesses perceived as beneficiaries rise dramatically.
4. The fear of missing out (FOMO) and the pain of underperforming push both companies and investors to shift more of their time and money out of the boring old areas and into the perceived beneficiaries of the new technology, driving prices of the boring areas down and the exciting ones up still further.
5. This widening gap in valuations causes even greater FOMO and psychological pain, pushing still more money from the old areas into the new ones.
6. Eventually, this coordinated global focus, combined with increasingly optimistic projections and the belief that “this time is different,” causes supply to dramatically overshoot demand, crushing profits and causing widespread investor losses.

While we all understand these dynamics in theory, it’s a wholly different thing to live through them. Quite frankly, it’s excruciating. In hindsight, the late 1990s seem to have gone by in the blink of an eye, but that’s not what it feels like when you’re in it. Time stretches out, and you feel mounting internal and external pressure to do something, anything, to fix the seemingly unrelenting underperformance versus the new technology darlings.

Since we’re living through our own difficult period, we’ve found it helpful to reread perspectives from the late ’90s. Perhaps unsurprisingly, the piece that resonated with us most was an exchange between Warren Buffett and a shareholder at the 2000 Berkshire Hathaway annual meeting.³ It took place near the height of the technology boom, after a year in which Berkshire Hathaway’s stock price had declined by approximately 20% while the S&P 500 had increased by over 20%.

AUDIENCE MEMBER: Now, with technology, computers, electronics, and software transforming our entire world — not just here — the world, I must admit that I personally invested in four technology computer software and aggressive growth mutual funds and made up all of my 1999 losses on Berkshire Hathaway. (Laughter and applause)

Are we asking too much as shareholders of Berkshire Hathaway for you men to put your brains to work and possibly speculate a little bit, maybe 10 percent of our money, into the only play in town, which seems to be technology, electronic?

And I read your report, and I understand a lot of your reasoning that it’s difficult — and it is difficult — to project earnings of a lot — people are going to be a little bankrupt. Are they going to go out of business?

But isn’t there enough left in your brainpower to maybe pick a few and — (laughter) — see what’s going on? Because I made over a hundred percent profit in 1999 on my aggressive position in the technology field, so that’s —

WARREN BUFFETT: OK, well —

AUDIENCE MEMBER: — my question.

WARREN BUFFETT: The answer is we will never buy anything we don’t think we understand. And our definition of understanding is thinking that we have a reasonable probability of being able to assess where the business will be in 10 years.

. . . . And if we can’t evaluate a carbon steel company, we don’t buy it. It doesn’t mean it isn’t a good buy. It doesn’t mean it isn’t selling for a fraction of its worth. It just means that we don’t know how to evaluate it.

If we can’t evaluate the sensibilities of putting in a chemical plant or something in Brazil, we don’t do it. If somebody else knows how to do it, you know, more power to them.

. . . . We really look at it as if there wasn’t any quote on the stock. Because we don’t know what the stock is going to do.

If we do — if the business gets worth more at a reasonable rate, the stock will follow, over time. But it won’t necessarily follow week by week, or month by month, or year by year.

We had a lousy year in 1999, but the stock price did not calibrate with that in any perfect, or close to perfect, manner.

And we’ve had good years other times, when the stock price is way overpriced or over-described what happened during the year.

So we really measure all the time by the business. We think of it as a private business, basically, for which there’s a quotation. And if it’s handy to use that quotation, either in buying more stock or something of the sort, we may do it. But it does not govern our ideas of value.

We’d say this is basically our approach to investing as well. In most cases, because of the ever-present risk of competition and disruption, we have low confidence in our ability to project a company’s cash flows far into the future. That’s especially true of businesses caught up in gold rush technology booms, where the forces of competition, innovation, and disruption operate at hyper speed. However, there are a select few businesses that we believe 1) possess network effects; 2) have additional competitive supply constraints such as not-in-my-backyard (NIMBY) restrictions; 3) sell essential services to large, broad, and growing end markets; 4) have untapped pricing power; 5) have conservative balance sheets; 6) have ownership-minded management teams; and 7) will be allowed by regulators to exercise some of this pricing power over time. In these few cases, we believe the businesses are predictable enough for us to estimate their future cash flows within a reasonable band. Then, if we judge the prices to be attractive, we buy them. Over time, we’ve assembled a diverse collection of these businesses that, in our view, essentially operates like a boring, recession-resistant toll taker on global GDP.

We then continuously re-underwrite the pricing power and predictability of our businesses as the world changes. The ultimate test of whether we’re right is not the short- to medium-term stock price but the long-term business results, which is why we felt it was so important last quarter to highlight those results with the chart below.

As you can see, despite their stock prices having stagnated over the past few years, our businesses have continued to grow. As a result, they’ve gotten cheaper relative to the market and, more importantly, relative to their own earnings. While stock prices are hard to predict in the short to medium term, over the long term they follow earnings growth, because public investors or other companies eventually recognize this increasing value.

Portfolio Updates

Because we believe the long-term trajectory of our businesses’ fundamentals is the most important driver of the portfolio, we spend most of our time following these results. To help you understand why we feel comfortable with the current portfolio, we’ve laid out its key categories of investment and, for each, provided an update on business results, whether the thesis is broadly tracking, and where there have been positive and negative surprises. To keep the main body of the letter as concise as possible, we’ve put this detailed review at the end in a section titled Appendix I – Detailed Portfolio Update. As you read it, remember that you may not own all these businesses because of your own unique tax situation and goals. Nevertheless, we think the exercise will still help you understand how we are evaluating your current portfolio.

We also added two new stocks to the portfolio this quarter: Deutsche Bӧrse AG and Ferrovial SE. Deutsche Bӧrse is a financial services group whose premier asset is Eurex, a futures exchange with close to a monopoly on Eurozone government bond futures. It complements our investment in CME Group, whose futures exchange has a near monopoly on U.S. interest rate futures. These businesses provide important diversification for your portfolio because they are among the few businesses in the world that benefit from higher and more volatile interest rates. Ferrovial is a construction and infrastructure firm that owns some of the most attractive toll roads in the world, with a focus on Canada and the U.S. We spend so much time talking about owning toll takers that we thought we should own a company that actually collects tolls. We believe we’ve acquired it at an attractive valuation because of fears that it overpaid for the right to build and operate a new Nashville toll road. The details of the contract are not public, but several factors make this price an attractive entry point in our view: management’s capital allocation track record; its alignment of incentives (the company is controlled and guided by a billionaire Spanish family that thinks in decades); the limited, ring-fenced value at risk, especially compared to the size of the stock price reaction; and the business’s discount to our assessment of its sum-of-the-parts value. We’ve included detailed write-ups of both stocks in a section titled Appendix II – New Positions.

Concluding Thoughts

We know this is a frustrating period for you. It’s frustrating for us as well. In times like these, it’s very tempting to try to change something. However, we believe that would be a mistake. We believe our strategy of owning a diverse collection of toll takers with durable competitive advantages and enduring pricing power is sound. It has produced attractive annualized results since YCG’s inception almost 20 years ago, and the recent underperformance is consistent with how strategies focused mainly on durable, non-technology stocks have typically performed during capital-intensive technology booms, as the chart below shows.

While we don’t know where the lines on this chart will go next, we do know that capital-intensive technology booms don’t last forever, that investment fashions are cyclical, and that defensive, durable businesses likely won’t stay out of favor forever. Most importantly, our businesses’ aggregate operating results remain strong, with pricing, margins, and profits rising over time. Over the long term, stock prices follow earnings, and our businesses’ earnings are growing.

As always, please reach out if you have any questions or concerns. As investors right alongside you, we appreciate the patience this environment requires. We’re grateful to have you as clients and thank you for your trust.

Sincerely,

The YCG Team

Appendix I – Detailed Portfolio Update

A few important notes before we begin. First, business is lumpy, so we focus on the long-term trend and changes in competitive positioning rather than any individual quarter or year. Second, although we care more about long-term business results than quarterly stock prices, we understand that this period has been challenging, so we’ve included each stock’s price performance to help you see which holdings have contributed to and detracted from performance. Lastly, notice that all the investment theses essentially boil down to durably constrained supply of a mission-critical product serving a demand category that is highly likely to grow over time. Again, the most important phrase is “durably constrained.” Many businesses benefit from constrained supply for months or even years, but only a few sell products whose supply is likely to remain constrained for decades.

Durable Luxury Goods (Ferrari and Hermès)

Q3 stock price performance

Ferrari: +5.5%

Hermès: -17.7% %

Thesis: Extreme scarcity, combined with global desirability and rising global wealth, equals durable volume growth, pricing power greater than inflation, margin expansion, and rising returns on invested capital.

Ferrari: Sells approximately 14,000 cars per year, compared to 92 million total cars sold globally each year and roughly 60 million millionaires worldwide.

Hermès: Sells approximately 120,000 Birkins and Kellys per year, compared to 1.8 billion total handbags sold globally each year and roughly 60 million millionaires worldwide.

Long-term thesis is tracking, with volumes and prices increasing. However, after a period of supernormal growth, Hermès’s growth has slowed to the mid-single digits because of broad weakness in global luxury and a cyclical, industry-wide shift in luxury spending toward jewelry and away from soft goods like handbags.

Financial Exchanges (CME Group and Deutsche Bӧrse)

Q3 stock price performance

CME Group: +19.2%

Deutsche Bӧrse: +19.0% (new position during the quarter, so the portfolio didn’t capture this whole move)

Thesis: Network effects in trading and clearing, plus rising global wealth, result in more trading on these dominant exchanges over time. Because their costs are largely fixed, this increased trading drives margin expansion and rising returns on invested capital. The exchanges are also a hedge for the rest of the portfolio, since they generally benefit from rising interest rates and volatility.

Rating Agencies (Moody’s, S&P Global – Ratings division, and FICO)

Q3 stock price performance

Moody’s: +0.5%

S&P Global: +2.0% adjusted for the Mobility Global spin-off (-3.1% unadjusted)

FICO: -50.4%

Thesis: Credit ratings are deeply embedded standards that global investors, issuers, and regulators use to communicate about fixed-income securities, making it incredibly difficult for new competitors to enter. This constrained supply combines with rising global wealth and growing capital markets to enable pricing above inflation, increased volume, and durably growing long-term profits.

Thesis is tracking for Moody’s and S&P Global, as their market share remains strong and there are no material new entrants to their markets. Near-term fundamentals are strong, as the debt needs of the AI buildout drive strong growth in bond issuance. This strength is somewhat offset by worries that rising interest rates will make borrowing more expensive for companies and perhaps dampen medium-term demand.

Our FICO thesis experienced a material negative surprise. While Moody’s and S&P Global rate fixed-income securities sold to an extremely fragmented group of global investors, FICO has significantly more customer concentration risk. It assigns credit scores to a wide variety of U.S. consumer loans that are then packaged together and sold to a fragmented group of investors, but in its most important segment, mortgages, there are two big buyers: Freddie Mac and Fannie Mae. Combined, these entities own or guarantee roughly half of all U.S. mortgages.⁴ Importantly, they are both under conservatorship and controlled by the Federal Housing Finance Agency (FHFA).

FICO has operated as a monopoly in U.S. credit scores for many years but kept its pricing constant for most of that time, even as the value of the score to global investors rose dramatically. Over the last few years, FICO has aggressively raised its prices to capture some of that value. Meanwhile, an alternative, VantageScore, has been trying to break into U.S. credit scores for years, but it’s been unsuccessful because a networked and trusted standard provides so much more value than a much less well-known alternative. Recently, however, the director of the FHFA has made it his mission to break FICO’s monopoly. Our view was that, to materially impact that monopoly, he would have to reduce Freddie and Fannie’s credit standards. We thought that was possible but unlikely, given the size of the mortgage market, the implicit government guarantee, and the institutional memory of the mortgage crisis that precipitated the Global Financial Crisis in 2008. Well, not only was it possible, but now it has happened: in September, the FHFA eliminated the 20-point haircut it had applied to VantageScores to account for their tendency to run higher than FICO scores, putting the two on the same pricing grid. In effect, the same borrower can now qualify for better mortgage pricing simply by being scored with VantageScore. As a result, VantageScore is likely to take much more market share than it would have otherwise. Despite this negative outcome, we still think the U.S. credit score market is likely to eventually settle into a duopoly between FICO and VantageScore, an industry structure that should enable price increases and attractive returns over time. Over the next few years, however, both companies are likely to compete aggressively on price and market share until a stable equilibrium emerges.

Financial Markets Benchmarks (MSCI and S&P Global – Financial Markets Benchmarks division)

Q3 stock price performance

MSCI: -3.5%

S&P Global: +2.0% adjusted for the Mobility Global spin (-3.1% unadjusted)

Thesis: Financial benchmarks such as the MSCI All Country World Index and the S&P 500 Index are deeply embedded standards that global investors, companies, and regulators use to communicate about investment performance, making it incredibly difficult for new competitors to enter. This constrained supply combines with rising global wealth and growing capital markets to enable pricing above inflation, increased volume, and durably growing long-term profits.

Thesis is tracking for both MSCI and S&P Global, as their market share remains strong and there are no material new entrants to their markets. Because rising markets support their two main revenue sources—subscriptions to their index data and total assets under management benchmarked to their indices—fundamentals continue to be strong. Also of note, S&P Global owns a few unrelated businesses that are good but not great, and it has begun to spin some of them off. The first spin-off, Mobility Global (whose premier asset is Carfax), was completed on July 1^st, creating a new, tiny holding in your portfolio. While Carfax is a good business, it doesn’t meet our stringent criteria for inclusion in your portfolio. However, spin-offs typically trigger a wave of mechanical and sometimes forced selling as funds rid their portfolios of these tiny positions. Research suggests this non-fundamental selling unduly depresses the stock price until it runs its course, typically in six to twelve months. Given these dynamics and our view that the stock appears undervalued, we are likely to wait until next year to sell.

Payment Processing (Mastercard and Visa)

Q3 stock price performance

Mastercard: +7.6%

Visa: +4.9%

Thesis: Payment processors connect merchants, cardholders, merchant banks, cardholders’ banks, and regulators in a five-sided network that is incredibly difficult to replicate. This constrained supply of payment networks, combined with rising global spending, drives increased pricing, volume, and long-term profits.

Thesis is tracking for both Mastercard and Visa as they continue to integrate new payment technologies, such as crypto and buy now/pay later, into their complex digital commerce networks, which increasingly serve as the bridge between rapidly proliferating alternative payment technologies and the banking system.

Insurance Brokers/Analytics (Aon, Marsh, and Verisk)

Q3 stock price performance

Aon: -16.4%

Marsh: +3.0%

Verisk: -6.2%

Thesis: Aon and Marsh have built global, virtually unrivaled two-sided networks that connect companies with insurance providers and offer customers a true one-stop shop. Given the complexity and importance of protecting a large business against an ever-evolving risk landscape, we believe this value proposition is incredibly difficult to replicate. Verisk is a trusted third party to which insurance companies contribute their proprietary data in return for aggregated industry data, which Verisk sells by subscription to them, to regulators, and to other industry participants. Verisk’s products are a very small percentage of insurance companies’ costs and benefit from very strong network effects. As long as Verisk remains a good partner, it will be very hard for a new and unproven competitor to convince all the insurance companies to contribute their proprietary data at the same time. Even then, the alternative product would still be inferior, since it wouldn’t have Verisk’s depth of historical data, including from companies that no longer exist. Combined with steadily rising insurable business activity, we think these strong moats are likely to produce increased pricing, volume, and profits over time.

Thesis is tracking for these companies, though we are disappointed by a recent acquisition announcement from Aon. Aon and Marsh have successfully acquired smaller insurance brokerages and rolled them into their much larger networks, and we agree with this strategy. However, we think Aon overpaid for its most recently announced brokerage acquisition. Because it is a large deal ($17 billion compared to Aon’s roughly $75 billion enterprise value), investors punished the stock. While the price paid is disappointing, the strategic logic is sound, the long-term thesis is intact, and we believe the market has over-penalized the stock. Importantly, these stocks should be decent hedges against inflation and rising interest rates. Inflation increases the cost of insurance claims, causing insurers to raise the premiums they charge. Brokers benefit from this dynamic since they earn a percentage of these premiums, as well as interest on client cash moving through their networks.

Hyperscalers (Microsoft, Amazon, Alphabet, and Meta)

Q3 stock price performance

Microsoft: +37.8%

Amazon: +4.5%

Alphabet: -3.5%

Meta: +28.8%

Thesis: While the hyperscalers’ businesses differ somewhat, they all possess dominant networks with distribution into billions of households and millions of businesses worldwide. Artificial intelligence has created both new threats and new opportunities for these businesses. As a result, both to protect their core and to capture new revenue, the hyperscalers have embarked on aggressive capital spending to build out data center capacity and develop large language model and semiconductor chip expertise. Because of their unrivaled global distribution networks, which constrain competitive supply, and the continued growth in global technology spending, these companies have successfully navigated past technology transitions. We think they will navigate this one as well, leading to durably growing profits over time.

Thesis is tracking for all four businesses. Each of these mega-cap technology companies has weathered waves of disruption fears. Last year, investors feared ChatGPT and other LLMs would disrupt Alphabet’s search business. This year, investors feared AI agents would disrupt Microsoft’s Office business and that its cloud business was structurally higher cost because Microsoft lags in LLM development and semiconductor chip design. They also feared that Meta’s aggressive spending on its large language models was a waste of money. In all three cases, the companies responded with products and services that assuaged investor fears. Alphabet developed a highly successful semiconductor chip and AI modes for search that have kept revenue and engagement strong. Microsoft has successfully positioned itself as the trusted, Switzerland-like layer on top of the enterprise that can manage and optimize large language model usage, AI agent behavior, and security. And Meta’s large language models have largely caught up to the frontier, while its launch of an AI agent, Muse, showcased its incredible ability to create new and useful products for its billions of users worldwide.

Auto Salvage/Toll Roads (Copart and Ferrovial)

Q3 stock price performance

Copart: -3.2%

Ferrovial: -21.1% (new position during the quarter, so the portfolio didn’t capture this whole move)

Thesis: Copart is the largest two-sided network for totaled and aged vehicles, connecting insurance carriers, rental car companies, and corporate and government fleets to a global buyer base of dismantlers, rebuilders, repair shops, and used car dealers. In addition to this digital auction platform, Copart owns a network of hundreds of strategically located salvage yards that is almost impossible to replicate given NIMBY restrictions. This constrained supply of salvage yards, combined with rising global miles driven and increased total loss frequency (partially offset by a slow decline in accident frequency), enables pricing above inflation, flat to slightly growing volumes, and durably growing long-term profits.

Ferrovial owns some of the most attractive toll road concessions in the world, with a focus on high-traffic-growth areas in North America such as Toronto, Dallas-Fort Worth, Northern Virginia, Charlotte, and Nashville. Through its construction arm, it also bids on new public-private partnerships to develop and operate additional infrastructure assets such as new toll roads and airports. As the areas this infrastructure serves grow denser over time, it becomes harder to add competing infrastructure and demand for the existing infrastructure increases, enabling Ferrovial to grow prices and profits significantly faster than inflation.

Long-term thesis is tracking for Copart, as the structural competitive supply constraints in auto salvage remain as strong as ever. However, the other player in the auto salvage duopoly, IAA, was a weak, poorly managed competitor that ceded market share for many years, leading to a roughly 65%-35% market share split. After a recent change in ownership, IAA’s operations have significantly improved, encouraging insurance companies to split their totaled car inventory more evenly going forward. This dynamic, combined with a cyclical downturn in insurance coverage, has pressured Copart’s volumes. However, pricing remains rational, and the market share readjustment is likely to end once market share gets closer to 50-50. Despite this bounded volume downside, Copart’s stock has declined dramatically. We believe it is quite attractive, trading at a substantial discount to the stock market despite having more constrained long-term supply and more enduring pricing power than the vast majority of businesses in the S&P 500 Index. The one additional negative in the quarter was Copart’s purchase of ACV Auctions, a player in the competitive dealer-to-dealer wholesale vehicle remarketing space. We would have preferred that Copart aggressively buy back stock and focus on its core operations. However, given the relatively small size of the acquisition and the fact that the CEO who made it has been one of the biggest drivers of Copart’s success over the years, the acquisition doesn’t materially impair our long-term thesis.

Ferrovial’s long-term thesis is tracking, with strong price growth at its toll roads. We were able to buy a position this quarter after a Ferrovial-led consortium made a bid for a new project that investors worried was too high. We think investors are overweighting this datapoint because 1) Ferrovial’s exposure is ring-fenced within the project, and even if the bid proves too high, the plausible value lost is small relative to the decline in Ferrovial’s market value; 2) the details of the contract are still not public, so everyone is operating with limited information; 3) Ferrovial’s long-term focus and unmatched expertise in dynamic pricing will likely enable it to make the toll road more profitable than investors expect; and 4) family ownership and a history of savvy capital allocation make it unlikely that this bid is as bad as investors fear or, if it is too high, that it is representative of future decisions.

Waste Collection and Disposal (Waste Management, Republic Services, and Waste Connections)

Q3 stock price performance

Waste Management: -8.1%

Republic Services: -1.2%

Waste Connections: -7.4%

Thesis: These businesses possess unmatched networks of landfills that are almost impossible to replicate because of NIMBY restrictions. As wealth and economic activity increase, demand for waste collection and disposal services is likely to increase as well. This constrained supply will enable pricing above inflation and growing profits. Additionally, because required regulatory and automation investments are becoming increasingly costly over time, these businesses have a long runway to acquire smaller, less sophisticated operators, fold them into their networks, and generate attractive returns.

Thesis is tracking for all three companies as they continue to raise prices faster than inflation and make small tuck-in acquisitions. Interestingly, Bloomberg recently published an article highlighting “trash” stocks as possibly one of the best hedges against an AI selloff because of their low obsolescence risk and strong negative correlation to AI stocks.⁵ While we don’t know what the future holds, we’d point out that Bloomberg’s observation is strong evidence for our thesis that investors are selling boring stocks to chase the gold rush in AI. You don’t get much more boring and durable than trash, and the data clearly shows these stocks are some of the most negatively correlated to the AI trade.

Aggregates (Vulcan Materials, Martin Marietta Materials, and CRH)

Q3 stock price performance

Vulcan Materials: -16.9%

Martin Marietta Materials: -16.7%

CRH: -22.3%

Thesis: These businesses possess unmatched networks of rock pits that supply critical, non-substitutable inputs to all construction activity. They are also almost impossible to compete against because of NIMBY restrictions and transportation costs that make non-local options uneconomic. As a result, supply is likely to remain constrained even as wealth and construction activity increase over time, allowing pricing above inflation and growing profits. Additionally, because required regulatory and automation investments are becoming increasingly costly over time, these businesses have a long runway to acquire smaller, less sophisticated operators, fold them into their networks, and generate attractive returns.

Thesis is tracking for all three aggregates companies as they continue to grow their pricing and make tuck-in acquisitions. However, construction is cyclical in the short term and highly sensitive to interest rates, so the increase in rates this quarter pressured the companies’ stock prices (the inverse of the benefit our financial exchange holdings enjoyed). Since essential construction can’t be deferred indefinitely and competitive supply remains constrained, our thesis is unchanged, and we think the stocks are even more attractively priced today.

Railroads (Canadian National and Canadian Pacific Kansas City)

Q3 stock price performance

Canadian National: +0.3%

Canadian Pacific Kansas City: -1.6%

Thesis: The Canadian railroads own dominant, NIMBY-protected transportation networks that connect customers across much of North America. Because of much lower labor and fuel costs, rail transportation has huge cost advantages over trucks, especially for bulk commodities and long hauls. This cost advantage, combined with constrained competitive supply, should enable the Canadian railroads to translate rising industrial and consumer demand into continued pricing above inflation, increased volume, and durably growing long-term profits.

Thesis is tracking for both railroads. After enduring a brutal post-COVID transportation recession, the railroads are seeing improved pricing and volume trends. Additionally, recent structural supply reductions in trucking⁶ and rising labor and fuel costs have widened rail’s already significant cost advantage, further strengthening the railroads’ pricing power. Because higher fuel and labor prices widen this advantage, the railroads also offer some protection against inflation, though not as much as our financial exchange holdings.

Industrial Gases (Linde)

Q3 stock price performance

Linde: -8.2%

Thesis: Industrial gas companies essentially own mini-monopoly distribution networks in each of their service areas because (1) additional industrial gas from an existing plant is cheap to produce; (2) industrial gas is expensive to transport; (3) industrial gas is mission-critical, so customers are willing to pay dense networks more for reliability; (4) industrial gas is typically a small percentage of customer costs; and (5) NIMBY restrictions often make it cost-prohibitive to build pipelines where existing ones are already in the ground. When you combine these competitive supply constraints with the high likelihood that Linde and its smaller competitors will keep increasing volumes, thanks to the broad range of growing end markets they serve, we believe Linde is positioned to grow pricing, profits, and return on capital over time.

Thesis is tracking. Linde’s project backlog, pricing, and profits continue to rise steadily.

Aerospace (TransDigm)

Q3 stock price performance

TransDigm: -18.7%

Thesis: Given the number of unique aircraft designs produced over the last thirty to fifty years, there are likely tens of millions of distinct aircraft parts (Boeing alone has 15 million individual part numbers in its supply catalog⁷). Many of these parts are a tiny percentage of the plane’s total cost (purchase cost plus operating cost) but complex and mission-critical enough that they can’t be substituted without cost-prohibitive recertification. TransDigm has spent the last 30 years building a portfolio of hundreds of thousands of parts with these characteristics. And while each part breaks unpredictably and infrequently, by assembling this vast portfolio, TransDigm now possesses an irreplicable collection of near monopolies that act as an incredibly smooth annuity stream with high pricing power. Moreover, the aerospace industry’s demand growth is arguably the strongest and most predictable in our portfolio. As the world gets wealthier and the cost of necessities declines in real terms, people devote a larger share of their growing wallets to travel. As a result, airline passenger miles have grown almost 5% per year since 1990.⁸ Despite this long period of growth, only 11% of the global population flies each year, only 4% fly internationally, and 80% have never flown at all,⁹ giving us confidence in continued strong industry growth for decades to come.

Thesis is tracking for TransDigm. Its volumes, pricing, and profits continue to grow at a brisk pace. The stock likely suffered this quarter because of the rapid rise in interest rates. TransDigm’s cash flows are so consistent that management has, intelligently in our view, always kept its balance sheet levered to further magnify its already high returns. While we believe TransDigm’s low capital needs, high margins, and strong pricing power make it an inflation-protected asset over the long term, its steady, long-duration cash flows and material debt load cause its valuation multiple to compress during periods of rising interest rates.

Consumer Staples (Procter & Gamble and Colgate)

Q3 stock price performance

Procter & Gamble: -0.2%

Colgate: -6.4%

Thesis: Procter & Gamble and Colgate dominate many household, personal care, and pet product categories with brands such as Tide, Pampers, Gillette, Crest, Colgate, and Hill’s Science Diet. Psychological switching costs such as habit and risk aversion, combined with low price points and unmatched global distribution, make these businesses hard to disrupt. As global wealth rises, we believe these companies are likely to grow their volumes, prices, and profits over time.

While private label alternatives from retail behemoths such as Costco, Walmart, and Amazon continue to slowly gain share, the thesis is tracking for both businesses, which continue to grow volume, pricing, and profits in most of their key product categories.

Appendix II – New Positions

Deutsche Bӧrse Group

Deutsche Bӧrse Group owns Eurex, the Frankfurt Stock Exchange, Xetra, the European Energy Exchange, and 360T, making it Europe’s leading futures, options, and equities exchange operator, with a near monopoly on Eurozone government bond futures trading. The company benefits from two network effects. The first occurs at the exchanges themselves, which are global, two-sided networks that connect buyers and sellers of futures and options contracts and provide them with unrivaled liquidity. The nature of futures contracts strengthens this two-sided network effect even further. For most assets, traders have no restrictions on where they can buy and sell. For instance, if a trader buys Microsoft stock on the Nasdaq, he or she can subsequently sell it on any number of exchanges or on increasingly popular alternative trading systems (ATS) such as dark pools and electronic communication networks (ECN). Futures contracts are different. A futures contract opened on one exchange cannot be closed on another, making it much harder for competitors to develop alternative sources of liquidity for futures buyers and sellers.

Deutsche Bӧrse also owns the clearinghouse that validates and finalizes all the futures transactions on its exchanges, further strengthening its competitive position through a second, overlapping network effect. Clearinghouses, which serve as third-party intermediaries to ensure that both buyers and sellers honor their contractual obligations, benefit from network effects through something called position netting. Position netting occurs when a customer places two or more trades through the same clearinghouse that fully or partially offset each other. In that case, the clearinghouse requires the customer to put up only enough money to cover the net position, allowing the customer to earn the same dollar return with less capital invested and raising its return on investment. For example, if a trader bought $100 of Microsoft stock through one clearinghouse and sold $90 short through another, and each clearinghouse required the trader to post 100% of the capital at risk, then he or she would have to post $190. However, if the trader bought $100 of Microsoft and sold $90 short through the same clearinghouse, he or she would only need to post $10 as collateral. As you can see, this netting effect creates a huge incentive for customers to use as few clearinghouses as possible. Importantly, Deutsche Bӧrse’s ownership of both the exchanges and their clearinghouse is a huge and relatively uncommon advantage. For instance, most stock and bond transactions in the United States are cleared through the Depository Trust & Clearing Corporation (DTCC), a private company jointly owned by its member banks and brokers, which makes it easier for alternative sources of liquidity to crop up and steal market share from stock exchanges like the NYSE and Nasdaq.

Interestingly, after the DTCC clears U.S. stocks and bonds, it transfers and stores them in its central securities depository, acting as the digital vault for approximately 90% of all U.S. securities, representing a value of $114 trillion. Because securities trade so frequently, the DTCC processes transactions worth many times that value. In 2025, for instance, it processed $4.7 quadrillion of transactions.¹⁰ It earns fees for this service, as well as interest on the buyers’ and sellers’ cash that flows through its platform to settle trades. While this fantastic business is a user-owned, not-for-profit cooperative in the United States, things work differently in Europe. Every European country maintains its own domestic central securities depository (CSD), and, to handle cross-border trading, Europe relies on a duopoly of international central securities depositories (ICSDs) that plug into all the national CSDs. More importantly, these two ICSDs serve another critical capital markets function: combined, they house almost all international debt issued by corporations and governments around the world. Whether it’s Tencent, Alphabet, Airbus, or the Malaysian government, if they issue international debt, it is almost certainly cleared and custodied by Clearstream or Euroclear. And while Euroclear is owned by its user banks, various national stock exchanges, and sovereign wealth funds, Clearstream is wholly owned by Deutsche Bӧrse. Thus, Deutsche Bӧrse acts not only as a tollbooth on Eurozone debt growth through its near-monopoly futures exchange but also, through Clearstream, as a toll collector on international debt issuance.

Beyond these crown jewels, Deutsche Bӧrse also owns index benchmarks such as the DAX and STOXX, the governance and ESG data provider Institutional Shareholder Services (ISS), the investment management software platform SimCorp, and various other smaller financial services businesses.

Lastly, like CME Group and our insurance brokers, Deutsche Bӧrse serves another important function in the portfolio: it is an excellent risk diversifier. For most businesses, earnings and valuations suffer when inflation rises or systemic risks such as geopolitical conflict and natural disasters materialize. In contrast, these companies’ earnings and business prospects generally improve during these periods, providing much-needed ballast for the portfolio.

Ferrovial

In our letters, we’ve often discussed looking for toll takers on big and growing economic categories. Usually, these businesses aren’t literally taking a toll but selling a mission-critical service that is a small part of their customers’ budgets. Our newest holding, on the other hand, is literally a toll collector. Ferrovial owns some of the most attractive toll road concessions in the world, with a focus on high-traffic-growth areas in North America such as Toronto, Dallas-Fort Worth, Northern Virginia, Charlotte, and Nashville. The company’s premier asset, the 407 ETR in Toronto, demonstrates why toll road concessions can be so attractive. It possesses three key characteristics. First, it is long-lived, with more than 70 years remaining on its contract with Ontario. Second, it runs through a congested, growing area, so its time-saving value to drivers keeps rising. Lastly, its contract with Ontario has no rate cap, enabling Ferrovial and the other owners (mostly Canadian pension funds) to raise prices as fast as the value to drivers increases. This rate freedom also allows Ferrovial to use sophisticated pricing models that dynamically adjust tolls based on congestion levels. Combined, these characteristics have enabled the 407 ETR to raise pricing from 11 cents per km in 1999¹¹ to $0.50 per km during off-peak times and $1.19 per km during peak times.¹² That’s a compound annual growth rate of roughly 6% for off-peak times and 9% for peak times, trouncing Canadian inflation of just over 2% per year over the same period. And these numbers don’t include added charges such as a $1.00 trip fee, a $5.30 camera charge per trip, and a $5.00 monthly account fee.¹³

While most of Ferrovial’s U.S. managed lanes don’t have quite as much pricing freedom as the 407 ETR, we believe their revenues and profits are still highly likely to beat inflation over time. Their concession contracts typically include an inflation-adjusted cap on non-congested prices, but they allow Ferrovial to raise congested prices by as much as necessary to keep the traffic flowing. Given that these managed lanes are located in some of the fastest-growing areas of the country, we believe this contract structure will translate into inflation-beating price increases over time.

Lastly, Ferrovial reinvests the cash flow from these toll roads into new public-private infrastructure partnerships, with a focus on building out its U.S. toll road network, though it will also partner on other attractive projects such as the new JFK airport terminal. Because it began as a construction firm and still has design and construction expertise, Ferrovial is one of the few firms that can offer its government partners a one-stop shop, handling the design, construction, and long-term operation of a project and making government officials’ lives easier. And since Ferrovial will be the long-term operator and beneficiary of the concession cash flows, officials can feel confident that it won’t cut construction corners that could create public relations problems later.

Ferrovial’s ownership structure gives it yet another edge. With the Spanish billionaire del Pino family controlling about 30% of the outstanding shares, we believe Ferrovial is in a unique position compared to private equity and other infrastructure bidders. Private equity bids are funded by outside investor cash and optimized for value over a 5- or 10-year holding period. Ferrovial’s bids, by contrast, are funded by virtually perpetual toll road cash flows and optimized for value over the lifetime of the project. This enables patience and the ability to pay more in certain situations, such as when private equity bidders undervalue the long-term tail of cash flows, when Ferrovial can gain extra synergies by plugging the project into its existing network of toll roads, or when its proprietary dynamic pricing algorithms and deep experience with managed lanes allow it to model revenue more accurately than other bidders.