July 2026
Dear clients and fellow shareholders:
After dipping into correction territory (a 10% decline) in the first quarter, the S&P 500 posted one of its strongest quarterly gains in years during the second quarter, bringing the index’s year-to-date return to 9.6%. The technology-heavy Nasdaq Composite experienced even more extreme volatility, falling as much as 15% earlier in the year, before rallying 30% in May alone.
(The stocks in our equity portfolio increased approximately 4% in the first half of the year. Your equity return may differ due to legacy positions, liquidity needs/constraints, and/or rounding.)
Inflation has been no less volatile. The 4.2% increase reported for May marked the largest monthly jump in consumer prices since 2023. Just one month later, CPI fell 0.4% from May—the largest monthly decline since April 2020. Erratic inflation data inevitably leads to erratic interest rate expectations. Six months ago, investors were debating how many times the Fed would cut rates in 2026. Today, markets are contemplating the possibility of a rate increase later this year—or at least they were prior to the June inflation report.
Much of this volatility can be traced, directly or indirectly, to the conflict in Iran. Yet despite the uncertainty, the market’s reaction has ultimately been limited to date. From the start of the conflict on February 27 through the end of the second quarter, the S&P 500 Index gained 9.5%. However, excluding the contribution from AI-related stocks, the index’s return over the period was only 1.6%. This divergence highlights how heavily recent market performance has depended on a relatively narrow group of AI beneficiaries.
While markets can experience sharp short-term reactions to geopolitical events, we believe these fluctuations are rarely meaningful drivers of long-term investment returns. As investors, rather than speculators, we have the luxury of largely ignoring the short-term market swings that accompany wars and military actions. Our core assumptions in this regard are simple: every war eventually ends, and it is impossible to reliably predict when one will begin or when it will end. And if we are ever proven wrong about that first assumption, our primary concerns are unlikely to involve determining the appropriate P/E multiple for the S&P 500.
There’s nothing artificial about AI capital expenditures
Artificial intelligence is powered by two things: massive amounts of electricity, and even more enormous amounts of cash. Data centers, especially those housing the specialized hardware used to train and operate machine learning models, require staggering amounts of power. But while the potential impact of AI on electricity prices has become the most visible public concern, the more significant concern for investors is whether the enormous amount of capital being deployed to build AI infrastructure will ultimately generate adequate returns.
(Although a recent Harvard poll found that two-thirds of Americans believe that data centers in their community cause an increase in local electricity prices, the data on this isn’t as overwhelming. This paper, “Have Data Centers Raised Your Electric Bill?”, is a comprehensive look at changes in electricity prices following data center construction in every state. The findings may surprise you.)
History has shown that even the most transformative technologies can attract levels of investment that eventually exceed their economic value. And the race to build the leading AI platform has a handful of companies investing at a pace for which there is virtually no historical precedent.
These massive expenditures have fundamentally altered the economics of several of the world’s highest-quality businesses, transforming them from capital-light, high cash flow generators into capital-intensive, high fixed-cost businesses that, for the foreseeable future, are expected to generate only a de minimis amount of free cash flow.
During 2026 and 2027, Amazon, Alphabet, Meta and Microsoft are expected to spend roughly $1.8 trillion on AI-related capital projects. For some context, just five years ago those four companies collectively spent only $110 billion on capital expenditures, most of it directed toward assets that generated immediate economic returns. Less than one year ago, these businesses produced enough free cash flow to internally fund all of their AI investments. Today, that free cash flow has largely been exhausted. Carlyle recently estimated that roughly 80% of projected AI capex over the next several years will ultimately require external financing through debt or equity issuance.
(In June, Berkshire Hathaway invested $10 billion in an Alphabet private equity placement, part of an $80 billion equity offering (later raised to $85 billion) Alphabet is using to help fund AI investments.)
History offers a familiar pattern. Whether it was dot-com companies in 2000, the housing market in 2007, the entire stock market in 1929, or Dutch tulip bulbs in the 17th century, speculative bubbles tend to become debt-financed near their peak.
Sometimes a new technology emerges whose eventual usefulness is difficult to appreciate. Not long ago, many of us struggled to understand why anyone would want a phone with a camera. Today, we use our phone cameras to do things most of us never imagined when they were first introduced, from reading menus with tiny fonts to scanning bar codes at Home Depot.
Artificial intelligence is not one of those technologies. AI has already become an integral part of our work process at MCM, and we are certain that we have barely begun to realize its potential. While AI large language models may seem like little more than a sophisticated search engine to many casual users, they are already transforming how many businesses operate, increasing both the speed and volume of certain types of work.
The question is not whether AI will prove valuable for consumers, but whether companies will generate meaningful returns from these massive initial investments. Can they ultimately earn enough incremental profit to generate returns above their cost of capital on the trillions of dollars now being deployed?
This uncertainty is one of the primary reasons we have been reluctant to fully embrace the AI investment theme. We currently own some Alphabet and have previously owned both Meta and Microsoft. But the extraordinary capital commitments these companies are making in an effort to “win the AI race” have fundamentally changed the economics of these businesses compared with just a few years ago.
At the end of June, approximately 10% of our stock portfolio was invested in technology sector companies (Alphabet, GoDaddy, and SS&C Technologies), compared with roughly 39% for the S&P 500. This is not the result of a top-down decision to underweight the sector, but simply reflects the fact that we have found relatively few opportunities in the sector that meet our investment criteria at current valuations.
Even with this modest tech exposure, we have outperformed the S&P 500 for most of the past three years—and done so while, in our estimation, taking considerably less risk than the overall market. That was not the case this past quarter. And while we aren’t managing our stock portfolio to maximize relative three-month returns, if we are going to trail the S&P 500, we would prefer to do so in a quarter in which our stocks return 10%.
Portfolio sale: DaVita (DVA)
During the second quarter, we exited our position in the kidney dialysis services company DaVita. While DaVita remains an exceptional operator with a well-established competitive advantage, we concluded that the future opportunity had become less attractive following the significant appreciation in the stock. Over our 3.5-year holding period, we generated a total return of approximately 174%, or roughly 35% annualized.
We originally purchased shares at approximately $72 per share following a selloff triggered by the company’s reduction in its 2023 guidance. At the time, the market was focused primarily on near-term volume concerns, while we believed investors were underestimating the durability of DaVita’s cash generation. Based on the company’s EBITDA outlook, we estimated free cash flow would exceed $1 billion annually, allowing us to purchase the business for less than 7x free cash flow, roughly half of its historical valuation.
A key part of our thesis revolved around DaVita’s ability to use its depressed valuation to create significant per-share value through share repurchases. Although buybacks were temporarily paused while the company prioritized debt reduction, we believed they would eventually resume.
This thesis played out well. While free cash flow expectations increased only modestly, DaVita resumed repurchases and retired a substantial amount of stock. The share count declined from approximately 93 million to 64 million during our ownership period, increasing earnings power from roughly $8.50 per share to more than $14 per share.
However, the same factors that drove our strong return have reduced the future opportunity. DaVita is expected to generate approximately $1.1 billion of free cash flow this year against a market capitalization of roughly $14.9 billion. While the valuation remains reasonable, we believe fair value is likely closer to 12x free cash flow given the company’s growth profile. More importantly, the normalization of the valuation multiple has diminished one of DaVita’s most attractive historical advantages. At a higher share price, each dollar of buybacks purchases a much smaller ownership stake, shifting future returns increasingly toward business growth rather than valuation-driven capital allocation.
Unfortunately, DaVita’s growth algorithm has become more challenging. After decades of expansion, total dialysis treatment volume has been roughly flat for the past six years. Excluding reimbursement increases, this is effectively a zero-growth business. The company has also acknowledged that reimbursement growth continues to lag inflationary pressures affecting its cost structure.
After the stock’s substantial price appreciation and the reduced benefit of valuation-driven buybacks, we believe future returns will depend primarily on business growth, a more difficult path given the company’s underlying volume trend. We therefore chose to redeploy the capital elsewhere.
Portfolio addition: Zoetis (ZTS)
We added one new stock to the portfolio last quarter, Zoetis. (Our average purchase price was about $73.20/share.) Zoetis is the world’s largest pure-play animal health company. We have long viewed Zoetis as one of the highest-quality businesses in healthcare (albeit for mostly four-legged patients), and the company has generally been valued accordingly. However, temporary concerns around slower companion animal growth and increased competitive pressures created an opportunity to purchase the company at an attractive price. After trading above 30 times earnings for much of the past decade, Zoetis now has a P/E of less than 11x, a valuation we believe significantly understates the quality, durability, and long-term earnings power of the business.
The animal health industry is attractive as it combines the defensive characteristics of healthcare investing with a more favorable industry structure than traditional pharmaceuticals. In contrast to human healthcare, a significant portion of animal health spending is paid directly by owners, which reduces reimbursement pressure. The industry also benefits from lower research and development costs, shorter development timelines, and strong relationships with veterinarians. These relationships, combined with regulatory barriers and specialized technical expertise, create meaningful switching costs and contribute to long product lifecycles.
The industry itself is highly consolidated, with Zoetis, Boehringer Ingelheim, Merck Animal Health, and Elanco representing the largest participants. This favorable competitive structure has historically supported pricing power, attractive margins, and strong returns on invested capital. Within this landscape, Zoetis has established one of the strongest competitive positions in the industry. The company holds leading positions in major therapeutic categories, with a broad portfolio spanning hundreds of products across companion animals and livestock, along with a global commercial infrastructure that would be extremely difficult for competitors to replicate.
The company’s recent challenges appear to be primarily related to product cycles rather than a deterioration of the underlying franchise. Zoetis has experienced slower growth in several companion animal categories as certain products mature, competitive intensity increases, and newer products take longer to reach their full potential. We believe investors have extrapolated these near-term headwinds too aggressively and are underappreciating the durability of Zoetis’ portfolio.
Successful animal health products often enjoy longer and more stable lifecycles than many human pharmaceutical products. While patent expirations can lead to rapid revenue declines in human pharmaceuticals, animal health products often experience more gradual erosion due to veterinarian relationships, brand loyalty, a fragmented customer base, and limited generic competition in many categories. Zoetis has consistently extended the life of its key franchises through innovation, including new formulations, expanded indications, and geographic expansion.
The company’s leading franchises in parasiticides, dermatology, and pain management provide a strong foundation for future growth. Longer-term industry trends remain favorable as pet ownership increases, owners continue to spend more on animal health, and advances in veterinary medicine allow more conditions to be treated. Companion animal healthcare, which represents the majority of Zoetis’ revenue and a significant portion of its profitability, benefits from these secular trends and remains one of the most attractive areas within animal health.
Zoetis also maintains one of the strongest innovation engines in animal health. The company has a proven ability to leverage its scientific capabilities and commercial relationships to successfully launch new products and build new franchises. We believe the market is assigning too little value to this pipeline and is placing too much emphasis on recent product-cycle challenges.
At our purchase price, we were able to acquire a business with industry-leading margins, strong free cash flow generation, and a dominant competitive position for less than 11x earnings. We believe this valuation reflects an overly pessimistic view of the company’s long-term growth prospects and fails to recognize the strength and durability of the underlying franchise. Additionally, Zoetis’ remaining $1.8 billion share repurchase authorization provides an attractive opportunity to significantly enhance per-share value at a time when the stock is trading at a depressed valuation.
The investment is not without risks. Competitive pressures may persist longer than anticipated, certain pipeline opportunities may not deliver as expected, and the recovery in companion animal growth could take time. However, we believe these risks are more than reflected in the current share price and that the market is assigning too much weight to near-term challenges.
Great businesses occasionally become undervalued when investors focus too heavily on temporary setbacks and lose sight of long-term fundamentals. (DaVita was a great example of this.) We believe Zoetis represents one of those opportunities: a high-quality company experiencing a period of slower growth rather than a permanent impairment of its competitive position. We were pleased to become shareholders at a valuation that we believe offers an exceptionally attractive risk-reward opportunity.