| ANNUALIZED SINCE INCEPTION | QUARTERLY | YTD |
|---|---|---|
| - | - | - |
Douglass Winthrop's Q2 2026 letter emphasizes investment discipline amid AI-driven market euphoria. The manager draws parallels to the dotcom bubble, citing historical examples like Sun Microsystems and Cisco Systems where transformative technologies disappointed investors who overpaid, with shares declining 90-96% despite continued business growth. Rather than owning speculative AI infrastructure directly, the portfolio gains exposure through proven cash-generative businesses—Alphabet, Microsoft, and Amazon—which are the largest data center owners powering AI while maintaining multiple earnings streams. The manager highlights that high price-to-revenue stocks have historically delivered half the return of the S&P 500, yet extreme valuations persist with governance shortcuts going unchallenged. S&P Global exemplifies the opportunity set, having become 27% cheaper since November 2022 while the broader market became 20% more expensive. The portfolio demonstrated resilience during brief selloffs in crowded AI trades. Competition is intensifying in semiconductors as Amazon and Alphabet's custom chips reduce Nvidia dependence. The strategy focuses on standards-based moats and businesses with enduring competitive advantages positioned to compound steadily through market cycles.
Douglass Winthrop maintains strict discipline anchored in business quality and intrinsic value during a market dominated by AI speculation, believing that wealth built quietly on enduring businesses with structural competitive advantages outlasts wealth built quickly on fragile ones priced for perfection.
Manager expects continued market discipline to be tested as speculation around AI persists, but believes the portfolio is positioned to compound steadily, protect on the downside, and generate good returns without requiring everything to go right. They anticipate that high quality, cash generative companies currently overlooked by the market will deliver higher returns going forward as valuations have become more attractive. The tone is cautious on market-level speculation but confident in the long-term positioning of quality businesses with enduring competitive advantages.
| Date | Letter | Tickers | Keywords | Pitches | Quick Takes |
|---|---|---|---|---|---|
| Jul 13 2026 | 2026 Q2 | AMZN, GOOGL, MSFT, NVDA, SPGI | AI, Capital markets, Data centers, large cap, Quality, semiconductors, Speculation, valuation | SPGI | Douglass Winthrop sees AI speculation creating a dotcom-style bubble with extreme valuations disconnected from fundamentals. Rather than chasing speculative infrastructure plays, they own proven cash generators—Alphabet, Microsoft, Amazon—gaining AI exposure as an accelerating tailwind to existing franchises. Quality businesses like S&P Global have become 27% cheaper while markets inflated 20%, creating compelling opportunities. Portfolio discipline protects capital when enthusiasm outruns reality. |
| Apr 6 2026 | 2026 Q1 | AMZN, AON, MSFT | AI, Cloud, inflation, large cap, Quality, rates, technology |
MSFT AMZN AON |
Douglass Winthrop sees Q1 2026 market volatility as creating opportunities in quality technology companies. Despite 20-33% declines in holdings like Microsoft, Amazon, and Aon, the manager emphasizes their strong fundamentals and views current weakness as timing reassessment rather than deteriorating prospects. The firm maintains conviction in AI adoption and high-quality businesses with durable competitive advantages. |
| Jan 5 2026 | 2025 Q4 | AMZN, BRK-A, COST, FAST, GOOGL, MA, MSFT, NVDA, ORCL, UBER | AI, inflation, Quality, technology, Trade Policy, value | - | Douglass Winthrop warns that AI valuations mirror historical innovation bubbles where technology succeeded but early investors lost money. They avoid overvalued AI pure-plays like Nvidia, preferring established companies embedding AI into workflows. Despite extended market valuations and policy risks, they see opportunities in undervalued quality stocks like Amazon and Berkshire Hathaway with strong competitive moats. |
| QUARTER | THEMES | TAGS |
|---|---|---|
| 2026 Q2 |
AIManager views AI as genuinely transformative but believes current valuations reflect extreme speculation disconnected from business fundamentals. They actively avoid direct AI infrastructure plays, citing historical parallels to the dotcom bubble where transformative technologies disappointed investors who overpaid. Portfolio exposure is through established cash-generative businesses (Alphabet, Microsoft, Amazon) that monetize AI as an accelerating tailwind rather than the entire investment premise. |
Valuation Speculation Infrastructure Data Centers |
Data CentersManager holds positive exposure to data center demand through Alphabet, Microsoft, and Amazon, described as the largest owners of data centers powering AI. These companies represent compelling AI exposure because they are proven, cash-generative businesses with multiple earnings streams today, with emerging AI businesses as an accelerating tailwind. Amazon's development of AI-optimized chips and investment in Anthropic positions it at the center of advanced AI development. |
Cloud Infrastructure Chips | |
SemiconductorsManager views semiconductor competition as intensifying, with well-financed players loosening Nvidia's grip on the market. Chips designed by Amazon and Alphabet have matured to the point where they are significantly reducing dependence on Nvidia. While Nvidia is acknowledged as still the king of chip design, the manager owns the competition through their holdings in Amazon and Alphabet rather than Nvidia directly. |
Competition Nvidia Custom Chips | |
QualityManager emphasizes discipline anchored in business quality and intrinsic value as protection when enthusiasm outruns reality. They highlight that high quality, cash generative companies have been overlooked as capital flows to AI speculation, creating opportunities. S&P Global is cited as an example where the forward earnings multiple has gotten 27% cheaper since November 2022 while the broader market has gotten 20% more expensive, making it a better opportunity. |
Valuation Cash Flow Discipline | |
Capital MarketsManager holds S&P Global as a core position, describing it as the independent referee and standard by which global capital markets price risk and benchmark performance. Its credit ratings, indices, and commodity benchmarks are embedded directly into bond covenants, investment mandates, and derivatives settlements. Standards-based moats are viewed as more permanent than product or technology moats because displacing them requires coordinated, simultaneous action across an entire industry's existing contracts and institutional processes. |
Benchmarks Ratings Moats | |
Risk AppetiteManager observes that euphoria is enabling extreme valuations and governance shortcuts without meaningful resistance from institutional shareholders, boards, or the broader culture. They cite the SpaceX/xAI merger as an example where a $1.25 trillion transaction with significant conflicts of interest proceeded without a fairness opinion. The market has become a singular bet on artificial intelligence, with a broader swath of participants starting to question that unbridled optimism. |
Speculation Governance Euphoria | |
| 2026 Q1 |
AIThe letter extensively discusses artificial intelligence through the lens of the Hype Cycle framework, noting that AI appears to be moving from inflated expectations toward more measured adoption. The manager views recent market volatility in AI-heavy companies like Microsoft and Amazon as a reassessment of timing rather than deterioration in long-term value, creating potential opportunities. |
Technology Innovation Infrastructure Investment Adoption |
CloudCloud infrastructure is highlighted as a key growth driver, particularly through Microsoft Azure's 25%+ year-over-year growth and Amazon Web Services' dominant market position. The manager emphasizes AWS as a primary toll road for AI deployment, generating 30-35% operating margins despite representing a minority of Amazon's total revenue. |
Infrastructure Growth Margins Deployment Services | |
Data CentersMassive data center investments are discussed as a key theme, with Microsoft and Amazon committing tens of billions annually to AI-related infrastructure. Microsoft's capital expenditures are expected to exceed $100 billion this year, while Amazon's are projected even higher, driven by continued AWS and AI infrastructure expansion. |
Infrastructure Investment Capacity AI Expansion | |
| 2025 Q4 |
Small CapsSmall-cap equities ended 2025 on a positive but volatile note with mixed market conditions throughout Q4. The manager expects the outlook for small-cap equities entering 2026 to be increasingly constructive, particularly within value-oriented segments, driven by Federal Reserve monetary easing and improving earnings momentum. |
Small Cap Value Russell 2000 |
RatesThe Federal Reserve's shift toward monetary easing represents an important inflection point for smaller companies, which tend to be more sensitive to changes in interest rates and credit conditions. Lower borrowing costs should support refinancing activity, capital investment, and margin recovery. |
Federal Reserve Interest Rates Monetary Policy | |
EarningsConsensus expectations point to a meaningful acceleration in small-cap earnings in 2026, with growth projected in the low-to-mid teens and exceeding that of large-cap companies. This anticipated rebound reflects easier year-over-year comparisons, improving operating leverage, and broadening demand across cyclical sectors. |
Earnings Growth Operating Leverage Cyclical | |
ValueSmall-cap stocks trade at approximately 16x forward earnings, compared with more than 21x for large-cap equities, leaving small caps valued at roughly a 30% discount on a relative basis. Within small caps, value-oriented stocks remain attractively positioned as growth stocks continue to trade at a meaningful premium. |
Valuation Forward P/E Discount | |
EnergyStrong stock selection and favorable allocation within Energy made the sector the second-largest contributor to relative performance. Helmerich & Payne led gains, benefiting from improving drilling activity, disciplined capital allocation, and strong operational execution. |
Drilling Oil Services Capital Allocation | |
UtilitiesStrong stock selection and favorable allocation made Utilities the largest contributor to relative performance for the quarter. Portland General Electric led gains, supported by regulatory clarity, steady rate base growth, and defensive characteristics that were rewarded as market volatility increased. |
Regulated Utilities Rate Base Defensive |
| Date | Pitch Type | Author | Ticker | Company | Industry | Sub Industry | Bull / Bear | Exchange | Keywords | Action |
|---|---|---|---|---|---|---|---|---|---|---|
| Jul 13, 2026 | Fund Letters | Douglass Winthrop | SPGI | S&P Global Inc. | Financial Data & Stock Exchanges | Financial Exchanges & Data | Bull | New York Stock Exchange | Benchmarks, Cash generative, credit ratings, defensive, Equity, financial data, indices, Standards-based Moat, valuation dislocation, Value | Login |
| Apr 6, 2026 | Fund Letters | Douglass Winthrop | MSFT | Microsoft Corporation | Software - Infrastructure | Systems Software | Bull | NASDAQ | Artificial Intelligence, Azure, Cloud computing, data centers, Enterprise software, operating leverage, SaaS, technology infrastructure | Login |
| Apr 6, 2026 | Fund Letters | Douglass Winthrop | AMZN | Amazon.com Inc | Internet Retail | Internet & Direct Marketing Retail | Bull | NASDAQ | Artificial Intelligence, AWS, Cloud computing, data centers, e-commerce, infrastructure, market leadership, operating leverage | Login |
| Apr 6, 2026 | Fund Letters | Douglass Winthrop | AON | Aon plc | Insurance Brokers | Insurance Brokers | Bull | New York Stock Exchange | Cyber security, enterprise clients, financial services, insurance brokerage, professional services, recurring revenue, regulatory compliance, risk management | Login |
| TICKER | COMMENTARY |
|---|---|
| GOOGL | Alphabet, Microsoft, and Amazon, all three of which are core DWA investments, are the largest owners of data centers that power AI. These companies represent AI exposure we find compelling because they each are businesses with multiple earnings streams today, and with emerging AI businesses as an accelerating tailwind rather than the entire investment premise. Chips designed by Amazon and Alphabet have matured to the point where they are significantly reducing their dependence on Nvidia. |
| MSFT | Alphabet, Microsoft, and Amazon, all three of which are core DWA investments, are the largest owners of data centers that power AI. These companies represent AI exposure we find compelling because they each are businesses with multiple earnings streams today, and with emerging AI businesses as an accelerating tailwind rather than the entire investment premise. |
| AMZN | Alphabet, Microsoft, and Amazon, all three of which are core DWA investments, are the largest owners of data centers that power AI. These companies represent AI exposure we find compelling because they each are businesses with multiple earnings streams today, and with emerging AI businesses as an accelerating tailwind rather than the entire investment premise. Amazon, for example, has dominant franchises in commerce and advertising, in addition to its leading data center footprint, which itself has many demand drivers in addition to AI. Amazon develops its own AI-optimized semiconductor chips, and its early investment in Anthropic has positioned it at the center of advanced AI development. Anthropic has emerged this year as the leading AI developer, doing so on Amazon's chips. Three years ago, it would have been difficult to imagine that the world's leading AI lab would train and serve frontier models on Amazon-designed chips rather than Nvidia's, yet that is where we are today. Chips designed by Amazon and Alphabet have matured to the point where they are significantly reducing their dependence on Nvidia. |
| NVDA | Three years ago, it would have been difficult to imagine that the world's leading AI lab would train and serve frontier models on Amazon-designed chips rather than Nvidia's, yet that is where we are today. Chips designed by Amazon and Alphabet have matured to the point where they are significantly reducing their dependence on Nvidia. While Nvidia is still the king of chip design, well-financed competition is loosening Nvidia's grip, and we own that competition. |
| SPGI | One stark example in our portfolio is S&P Global, which exists to be the independent referee that was absent from the SpaceX/xAI transaction. S&P Global is the standard by which global capital markets price risk and benchmark performance. Its credit ratings, indices, and commodity benchmarks are the defined reference points embedded directly into bond covenants, investment mandates, and derivatives settlements worldwide. Its indices underpin $17 trillion in passive investments, and its data is the input on which many models run. Standards-based moats tend to be more permanent than product or technology moats because displacing them requires coordinated, simultaneous action across an entire industry's existing contracts and institutional processes. Its businesses are so deeply woven into financial markets that disruption is nearly unimaginable. The investment opportunity at this moment is notable. Since the November 2022 launch of ChatGPT that sparked the frenzy in artificial intelligence, the forward earnings multiple for the broader market has gotten 20% more expensive, while that of S&P Global has gotten 27% cheaper. S&P Global has become a better opportunity as it has gotten cheaper, while the major indices, inflated with speculation, have become riskier. |
| Ticker | Put/Call | Amount Bought | Shares Bought | % Change | Weight % |
|---|---|---|---|---|---|
| No Recent Buys Data | |||||
| Ticker | Put/Call | Amount Sold | Shares Sold | % Change | Weight % | Status |
|---|---|---|---|---|---|---|
| No Recent Sells Data | ||||||
| Industry | Prev Quarter % | Current Quarter % | Change |
|---|---|---|---|
| No industry data available | |||
