If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

📊 Market Lens | Two-Week Snapshot
Period: August 18 to August 28, 2026

• VIX: 14.43 (-0.59 pts over two weeks, intraday spike on Jackson Hole day)

• 10Y Treasury: 4.722% (+5.1 bps over two weeks)

• 30Y Treasury: 5.310% (+10.9 bps, highest since June 2007, 19-year record)

• WTI Crude: $83.40 (+1.2% over two weeks)

• Brent: $89.31 (+2.1% over two weeks)

• Gold (XAU): $4,529.90 (+1.0% over two weeks)

• S&P 500 (SPX): ~7,712 (-0.9% over two weeks)

• USD Index (DXY): ~100.50 (-0.9% over two weeks)

Share

The most important two sentences spoken in global financial markets over the past two weeks were delivered on a Thursday morning in Wyoming. Kevin Warsh, speaking at the Jackson Hole Economic Symposium on August 28 in his first appearance as Federal Reserve Chairman at the annual gathering, said: “We would have work to do if policymakers were not confident inflation was returning to its 2% target.” He then confirmed that the September FOMC meeting remains live for a rate increase, refused to provide forward guidance, and declined to indicate any threshold at which the Fed would move to ease. The speech lasted less than an hour. Gold fell $134 in the same afternoon. Short-term Treasury yields spiked immediately. The market repriced September hike probability higher in real time.

The content of the speech was not a surprise to anyone who had been paying attention. Warsh’s June dot-plot abstention, his July post-meeting language, and his congressional testimony all carried the same message: the Fed is not on a predetermined path, inflation is the primary variable, and any assumption of rate cuts in 2026 is a market construct rather than a policy signal. What Jackson Hole added was the formal institutional weight of the event itself. Every global central bank governor, every sovereign wealth fund CIO, and every major institutional fixed income manager receives the Jackson Hole speech as a definitive statement of Fed posture rather than an inter-meeting communication. The 30-year Treasury yield reaching 5.31%, its highest level since June 2007 and a 19-year record, is the bond market’s response to the cumulative weight of that communication over the past two months. The term premium, the additional yield investors demand for holding long-duration bonds rather than rolling short-term bills, has expanded to levels not seen since the pre-financial crisis era. That expansion reflects two simultaneous pressures: the Fed’s hawkish posture on the short end of the curve, and the structural supply of long-duration Treasury debt as the US fiscal deficit continues to require massive issuance at the long end.

The US Treasury acknowledged the term premium problem directly during the two-week period. On August 19, the Treasury Department announced it was doubling the cap on its long-term bond buyback operations to $4 billion, deploying a mechanism originally designed for liquidity management as a tool to suppress the long end of the yield curve. When a Treasury Department that is simultaneously the world’s largest bond issuer begins buying back its own long-term debt to prevent yields from rising further, the institutional signal is specific: the term premium expansion has reached a level that the fiscal authorities consider problematic for borrowing costs across the economy. The intervention is unlikely to be sufficient. The structural forces driving the 30-year toward 5.31%, specifically the Fed’s rate posture, the scale of Treasury issuance, and the AI capital expenditure cycle consuming long-duration capital, are not responsive to $4 billion in buyback operations.

The FOMC minutes released August 19 from the July 29 meeting revealed the nine-to-three voting split was accompanied by more contentious internal debate than the public statement communicated. Several officials who voted to hold expressed sympathy with the dissenters’ inflation concerns and stated their threshold for supporting a hike was close to current conditions. September’s decision depends on the August CPI print, scheduled for release before the September 16-17 meeting.

The Q2 earnings season finished with approximately 90% of S&P 500 companies having reported. The blended EPS growth rate came in at 50.4%, with 86% of companies beating estimates. The headline figure requires the same qualification it has required all season: a meaningful portion of the growth reflects non-cash mark-to-market gains on Anthropic investment stakes held by multiple technology companies. This quarter added Salesforce to that list. CRM reported Q2 FY2027 Agentforce ARR of $1.5 billion, up 240% year-over-year, alongside a $2.6 billion Anthropic investment gain that inflated reported net income. The Agentforce numbers are real and represent a genuine thesis confirmation for the portfolio position. The Anthropic accounting effect is now visible in four separate companies’ earnings this season. NVIDIA and Salesforce both reported on August 26, and the full analysis of both prints is in Earnings Decoder.

Share

The July retail sales data released during the period recorded the largest single-month decline in over a year, consistent with the consumer pressure that real disposable income declines have been building toward for several months. Applied Materials reported results that cleared every consensus estimate and saw its stock fall 5.1%, extending the pattern of strong operational results meeting elevated valuations across AI infrastructure names.

On August 24, Trump announced tariffs on Canadian automotive products and steel would rise to 50%. Canada is the United States’ largest trading partner, a significant source of steel and automotive inputs, and the legal domicile of Constellation Software. The tariff escalation introduces a trade dimension into an environment already managing energy price volatility, monetary policy uncertainty, and AI capital allocation at scale.

The S&P 500 Equal Weight index underperformed the cap-weighted index on the Thursday of Jackson Hole week, driven by technology sector strength from NVIDIA’s guidance revision. The seven consecutive weeks of equal-weight outperformance that characterized the June to August rotation did not resume. One company’s FY2028 guidance of 70% revenue growth is sufficient to temporarily redirect capital back toward cap-weighted concentration.

The September CPI print will determine whether Warsh’s Jackson Hole language translates into an actual 25 basis point hike or remains a policy threat. The portfolio enters September with cash above 40%, fully positioned across its core holdings, and two pending initiations scheduled for this period.

This Week’s Signal: Warsh said the Fed has work to do. Gold fell $134 the same afternoon. The 30-year is at a 19-year high. September CPI is the variable that resolves everything that two weeks of signals left ambiguous.

Module 2: Earnings Decoder

NVIDIA (NVDA) | Q2 FY2027 | August 26, After Close

NVIDIA reported adjusted EPS of $2.22 against a Wall Street consensus of $2.09, a beat of 6.29% that extended the chipmaker’s winning streak to five consecutive quarters of topping estimates. Revenue came in at $96.22 billion against a consensus estimate of $92.17 billion, a beat of approximately $4.05 billion, or 4.4%. Revenue grew 106% from $46.7 billion in the year-ago quarter.

Data Center contributed $89.02 billion in quarterly revenue, up 117% from a year ago, as the ramp of Blackwell Ultra architecture drove both hyperscale and enterprise custom deployments. Gross margin was 75.0%, up 2.6 percentage points year-over-year. Free cash flow was a staggering $21.3 billion generated during the quarter.

The forward guidance is where the report becomes analytically significant beyond the quarterly results. The forward guide was the release, and the quarter being reported was largely written three months ago by other people’s cash flow statements. A guide at or above consensus says the capex curve still compounds. This one is +4.0% against the $103.8 billion the street carried into the print. In a rare move, the chipmaker provided a revenue growth outlook for fiscal 2028 that forecast annual sales to increase 70% from the prior year. Analyst expectations had been approximately 44%. A company that guides 70% growth for a fiscal year that begins in February 2027 is describing committed order flows that exist today, not aspirational demand.

Jensen Huang said: “AI has reached its inflection point. It’s doing useful work. Its tokens are productive and profitable. Now, compute is revenue.”

Despite executing a seemingly flawless beat and raise quarter, Nvidia shares actually traded lower initially before reversing to gain more than 5% after hours as executives discussed the results on the conference call. The initial decline followed the AMAT and CSCO pattern from the prior period. The subsequent reversal reflects the FY2028 70% guidance landing above every sell-side model.

Thesis verdict: The FY2028 guide is the number that matters. Every Azure contract, every AWS workload, every Google Cloud deployment that Arista’s $9.7 billion in purchase commitments supports will ultimately run on compute that NVIDIA provides. The portfolio’s Blue Chip tier owns the monetization layer above this infrastructure. The infrastructure itself just confirmed it is not decelerating.

Salesforce (CRM) | Q2 FY2027 | August 26, After Close | Portfolio Holding

Salesforce reported fiscal 2027 second-quarter revenue of $11.345 billion, up 10.83% year over year and 0.25% above Street estimates. Adjusted earnings per share of $5.90 beat estimates by 80.38%, and net income of $4.844 billion climbed 73.31% year over year.

The EPS beat requires the same analytical surgery that has been applied to Amazon, Alphabet, and Microsoft this season. Salesforce attributed much of the net income lift to a $2.6 billion investment gain tied in part to its position in AI startup Anthropic. This is the fourth major technology company this earnings season to report inflated GAAP earnings from the same private company’s rising implied valuation. The operational result is more modest but still meaningful.

The bigger move came in bookings. Current remaining performance obligations grew 14% year over year in constant currency to $33.5 billion, a point ahead of guidance and the strongest quarter for net new annual order value in four years.

Agentforce ARR reached $1.5 billion, agentic workflow use hit 3.2 billion actions in the quarter, up 97% sequentially, and Slack bot grew to 1 million active users, up 150% quarter over quarter. Bookings from premium Agentforce and Slack bundles more than doubled sequentially.

Management used that momentum to raise the fiscal 2027 revenue outlook by $300 million in constant currency to a new range of $46.1 billion to $46.4 billion. The stock rose roughly 11% in extended trading.

Marc Benioff’s characterization on the call was direct: “This is not the SaaSpocalypse.” The Agentforce numbers support that claim. 3.2 billion agentic work units in a single quarter, growing 97% sequentially, is contracted workflow activity that does not reverse because a competitor releases a new model.

Thesis verdict: Agentforce ARR at $1.5 billion growing 240% year-over-year is the thesis confirmation the portfolio position has been waiting for since initiation. The Anthropic accounting distortion inflates the GAAP headline. The operational reality beneath it is the strongest CRM has delivered in four years. The Under Review status consideration for ACN is rendered more urgent by CRM’s success: one part of the enterprise software stack is clearly winning the AI transition. The question the ACN position must answer is whether IT consulting is in the same category.

HEICO Corporation (HEI) | Q3 FY2026 | August 25, After Close | Watchlist

HEICO reported an increase in net income of 33% to a record $235.4 million, or $1.67 per diluted share, in the third quarter of fiscal 2026, up from $177.3 million, or $1.26 per diluted share, in the third quarter of fiscal 2025. The company reported earnings of $1.67 per share, beating estimates of $1.52 by $0.15. The company also reported revenue of $1,413,050,000, beating estimates of $1,357,879,966 by $55,170,034.

Consolidated gross profit margin improved to 41.1% in the third quarter of fiscal 2026, up from 39.8% in the third quarter of fiscal 2025. The increase in consolidated operating income principally reflects a $47.0 million increase, a 24% increase, to a record $245.3 million in operating income of the FSG and a $44.6 million increase, a 55% increase, to a record $125.6 million in operating income of the ETG.

The company emphasizes cash flow as fact, generating $345.3 million in operating cash flow, representing nearly 150% of net income for the quarter. Strategic positioning in the defense sector led to upper-20s organic growth in FSG’s specialty products. Consolidated organic net sales growth reached 14%. Consolidated operating income as a percentage of net sales improved to 25.1% in the third quarter of fiscal 2026, up from 23.1% in the third quarter of fiscal 2025.

Every variable on the watchlist inclusion checklist resolved affirmatively in this single quarter. ETG operating income grew 55%, confirming the Q1 compression was transitory. Consolidated operating margin expanded 200 basis points to 25.1%. Operating cash flow was nearly 150% of net income, which is the management team’s own stated definition of earnings quality. The stock rose 1.54% after the print.

The only open question entering August 25 was whether the valuation, which had run to all-time highs following the Q2 beat in May, would compress enough to offer a disciplined entry point. The broader semiconductor and technology selloff during the period provided that compression. Price and thesis are now aligned simultaneously for the first time since the watchlist was established.

Thesis verdict: All inclusion criteria satisfied. HEI moves from Watchlist to Portfolio. Full initiation details in Portfolio Desk.

Walmart (WMT) | Q2 FY2027 | August 20, Before Open

Revenue hit $187.9 billion with 23% global ecommerce growth and 38% advertising surge. Walmart posted earnings per share of $0.81, which exceeded analysts’ expectations of $0.74 by 9.46%. Walmart U.S. comp sales grew 2.6%, led by a 1.5% increase in transactions.

Share

The operational results were solid. The consumer commentary was not. CFO Rainey said Walmart continues to see consumers stretched thin, especially with higher gas prices. Rainey said Walmart also expects to incur just over $2 billion of incremental cost headwinds related to higher fuel prices this year. The stock posted its largest single-day decline since 2022 despite the EPS beat, falling toward $100 per share.

The market’s reaction is analytically precise. A 2.6% comp driven by transaction count rather than ticket size means consumers are visiting Walmart more frequently and spending less per trip. That is the behavioral signature of a household managing a budget constraint rather than a household with discretionary spending capacity. The 23% ecommerce growth and 38% advertising growth are the high-margin businesses within Walmart that are growing fastest, but they are growing because consumers are optimizing for price discovery and value, not because spending power is expanding.

Walmart’s efforts to lower prices come as many shoppers have cut back on spending as they feel a strain from high fuel and food costs. The $2.9 billion in tariff refunds Walmart expects to receive and pass to consumers in Q3 will provide temporary relief. The structural pressure from real disposable income declines and elevated energy costs does not resolve with a tariff refund.

Thesis verdict: Walmart’s quarter is the most honest available read on where the US consumer actually is. A 9.46% EPS beat followed by the largest single-day stock decline since 2022 is the market pricing guidance rather than results. The consumer is transacting at Walmart because it is the value option of last resort, not because household finances are healthy. This validates the portfolio’s positioning in ROL, WM, and COST: non-discretionary demand and membership economics are more defensible than discretionary retail in this environment.

Module 3: Smart Money Monitor

1. Jackson Hole Repricing: What the Fed Funds Futures Market Said in Real Time

Warsh’s August 28 speech moved markets within minutes, and the most analytically precise measure of the institutional response was not the equity market reaction but the CME FedWatch tool. September hike probability moved from approximately 32% before the speech to above 45% by the time the Q&A session concluded. The two-year Treasury yield, the instrument most directly sensitive to Fed policy expectations, spiked on the short end as Warsh’s language was parsed by fixed income desks globally.

The institutional read on the speech was more nuanced than the initial rate move implied. Warsh said the Fed “would have work to do” if policymakers were not confident inflation was returning to target. That conditional phrasing is not a commitment to hike. It is a conditional statement that keeps optionality on both sides of the trade. The bond market initially sold off and then partially recovered as traders concluded that Warsh had deliberately preserved flexibility rather than committing to September action. The net weekly move in the 10-year of approximately 5 basis points reflects that ambiguity rather than a clean hawkish signal.

The September 9 CPI print now carries more institutional weight than any single FOMC communication. If August CPI comes in above 3.4%, the conditional in Warsh’s statement is satisfied and the market will rapidly price September as a live hike meeting. If it comes in below, the pause extends and the conditional is not triggered. Institutional fixed income desks are holding positions rather than committing directionally until that data arrives.

Signal: Warsh preserved optionality deliberately. The conditional framing was a negotiated communication, not an unambiguous signal. September CPI is the event that resolves the ambiguity the speech left open.

2. The Treasury Buyback Mechanism: What Happens When the World’s Largest Bond Issuer Becomes a Buyer

The US Treasury doubling its long-term bond buyback cap to $4 billion during the period is a structurally important policy development that received less analytical attention than it deserved. The Treasury is simultaneously the world’s largest issuer of long-duration bonds and, through this mechanism, an active suppressor of the yields on those same bonds. The tension in that dual role is real and has precedent. Japan’s Ministry of Finance and Bank of Japan operated similar mechanisms for years before the framework became unsustainable.

The immediate institutional beneficiaries of the buyback expansion are pension funds and insurance companies that hold long-duration bond portfolios and need to match them against long-duration liabilities. At 5.31% on the 30-year, these institutions are locking in yields not available since 2007. The buyback mechanism provides a liquidity backstop that allows them to transact in size without the market impact that would otherwise come from trading in thinly traded long-duration instruments. The Treasury is, in effect, subsidizing the duration management of institutions whose liability structures require long-dated bonds.

The structural problem the buyback cannot solve is the supply dynamic. The US fiscal deficit requires continued issuance of long-duration debt at a scale that the buyback program does not offset. Goldman Sachs and JPMorgan both published notes during the period flagging that term premium, the additional yield required by investors to hold long bonds rather than rolling short bills, is likely to remain elevated regardless of Fed policy direction. The buyback addresses a liquidity problem. It does not address the supply problem.

Signal: The Treasury buyback expansion is a liquidity tool being deployed against a supply problem. Institutional duration managers are using the liquidity backstop to lock in 5.31% on the long end. The structural pressure on term premium does not resolve with $4 billion in buybacks. The 30-year is likely to remain above 5% through year-end regardless of what Warsh does in September.

3. Sector Rotation: Jackson Hole Interrupted Eight Weeks of Equal-Weight Outperformance

The week of Jackson Hole produced the clearest single-week interruption to the sector rotation that has characterized markets since mid-June. On Thursday August 28, the S&P 500 Equal Weight index fell 0.29% while the cap-weighted index rose 0.7%, driven almost entirely by NVIDIA’s extended trading surge following the FY2028 70% guidance revision. Healthcare, utilities, and financials all underperformed on a session where a single technology company moved the headline index higher.

The ETF flow picture for the two-week period reflects the same dynamic. XLK received the strongest technology sector inflows of any two-week period since the AI rotation began in June, concentrated specifically in semiconductor and AI infrastructure names following the NVIDIA guidance. Healthcare and financial ETFs maintained their inflow trajectories but were overshadowed by the technology concentration on August 28. Energy ETFs continued receiving incremental inflows as Brent held above $89.

The more analytically important observation is what did not happen. Despite the NVIDIA-driven cap-weighted surge, the equal-weight index did not collapse. The defensive rotation that has been accumulating over eight weeks was not unwound in a single session. Institutional capital that had been systematically moving from mega-cap AI concentration into the broader earnings base did not reverse those positions because of one forward guidance revision from one company. The eight-week structural reallocation requires a multi-session sustained reversal to confirm it has ended. One Thursday is not sufficient.

Signal: Jackson Hole week produced a single-session interruption to eight weeks of equal-weight outperformance. The NVIDIA FY2028 guidance revision redirected cap-weighted flows for a day. The structural rotation from narrative to cash flow has not reversed. It has paused.

4. Pershing Square Q2 2026: Ackman Builds Visa, Mastercard, S&P Global, and Netflix While Adding to Microsoft

The Q2 2026 13F filings became publicly available during the period, with Pershing Square’s disclosure among the most analytically relevant for the portfolio. Bill Ackman’s fund grew to $19.47 billion across 14 positions, expanding from 10 positions in Q1, with four significant new initiations.

The new positions in Visa and Mastercard are the most directly portfolio-relevant disclosures. Ackman established meaningful stakes in both payment networks in a quarter when the portfolio trimmed V from 4.5% to 3% for capital reallocation purposes. Ackman’s entry into V at the same time reflects a thesis that is analytically identical to the one that keeps V in the portfolio: the payment network’s position as the settlement layer for global commerce is not threatened by AI, and the cross-border volume recovery from Hormuz normalization is a near-term catalyst for earnings acceleration. His simultaneous initiation of Mastercard suggests the thesis is specifically about payment network economics rather than any individual company.

The new position in S&P Global is analytically interesting for its proximity to Moody’s, which the portfolio holds at 2.25%. Both companies own the rating and data infrastructure that institutional capital markets require to function. Ackman’s entry into SPGI confirms that the financial data infrastructure moat, the same moat that underpins the MCO thesis, is being repriced from a discount to a premium by sophisticated institutional capital.

Ackman continued adding to Microsoft and fully exited Alphabet. The MSFT add is the third consecutive quarter in which Ackman has increased his Microsoft position, beginning with the initial $2 billion stake in Q1. The Alphabet exit is the more analytically interesting move: he initiated Alphabet in Q1, held it for one quarter, and exited in Q2. The most plausible interpretation is that the Google Cloud thesis was validated by the Q2 earnings print, the non-cash Anthropic gains inflated the EPS to a level that made the entry price less attractive on a forward basis, and the capex guidance revision reduced the near-term FCF profile he was expecting.

Separately, Akre Capital’s Q2 13F showed reductions in MCO, ROP, V, and MA. Akre’s trim of MCO and ROP from the same positions the portfolio holds is worth monitoring but does not constitute a thesis signal. Akre manages a concentrated long-term portfolio where trim decisions often reflect valuation discipline relative to position size rather than fundamental conviction changes.

Signal: Ackman’s simultaneous initiation of Visa and Mastercard validates the payment network thesis. His continued MSFT accumulation is now a three-quarter directional commitment. His SPGI initiation confirms the financial data infrastructure moat is attracting institutional capital from the most concentrated and thesis-driven fund managers in the market.

Module 4: Quality at a Discount

Fair Isaac (FICO) | The Software Miss That Buried the Scores Record

FICO fell 9.33% after hours on July 29 on a $18 million software segment revenue miss. The market priced the shortfall. It largely ignored everything else in the report.

The Scores segment delivered $458.9 million, up 41% year-over-year, with mortgage origination revenues growing 97% on low-single-digit volume increases. Platform ARR exceeded Non-Platform ARR for the first time. Non-GAAP operating margin reached 62%, up 479 basis points. Full-year guidance was raised. The $18 million miss is a one-quarter software timing issue. The Scores business, which earns revenue because US lending regulation requires it, did not miss anything.

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

📊 Market Lens | Two-Week Snapshot
Period: August 4 to August 15, 2026

• VIX: 15.02 (-1.82 pts over two weeks, intraday low 14.18)

• 10Y Treasury: 4.671% (+15.9 bps over two weeks)

• 30Y Treasury: 5.201% (+13 bps, highest since June 2007)

• WTI Crude: $82.41 (-2.1% over two weeks)

• Gold (XAU): $4,483.80 (+8.2% over two weeks)

• S&P 500 (SPX): 7,784.15 (+2.9% over two weeks, intraday ATH 7,798.99)

• USD Index (DXY): 101.47 (+1.6% over two weeks)

Share

The S&P 500 crossed 7,800 for the first time in history on August 7. It was the index’s 27th record high of the year. VIX fell to 14.18 intraday, its lowest reading of 2026. The semiconductor ETF SOXX gained 7% in the first week of the period alone, reversing a meaningful portion of the prior month’s losses. Three consecutive weeks of equity market gains. By every surface measure, this was a relief rally that absorbed the July dislocations and resumed the uptrend.

The same two weeks produced the following: the 30-year Treasury yield reached its highest level since June 2007. Preliminary August consumer sentiment from the University of Michigan fell as Americans grew more pessimistic about the economy. July retail sales recorded their largest single-month decline in over a year. The July non-farm payrolls report, released August 1 at the tail end of the prior period, showed a net decline of 23,000 jobs, with the prior two months revised down by an additional 103,000 positions. CME FedWatch futures assigned a 64% probability of at least one rate hike by year-end. Gold rose 8.2% over the two weeks, its strongest fortnight since the early months of the Hormuz crisis.

These two sets of data are not describing the same market. They are describing the same market from two different time horizons, and the divergence between them is the central analytical challenge of the current moment.

The equity market is looking at the next six to twelve months and pricing the AI infrastructure supercycle as a structural demand driver that will compound corporate earnings regardless of the macro environment. The bond market is looking at the same six to twelve months and pricing an inflation regime that is more entrenched than the equity market’s earnings models assume. The Atlanta Fed’s GDPNow estimate for Q3 fell sharply after the retail sales data, while lingering inflation and rising US fiscal deficits continued to push yields up despite soft economic readings. When growth forecasts fall and inflation forecasts hold, the word that describes the combination is stagflation, and stagflation is precisely the environment in which equity P/E multiples face structural compression from both the numerator and the denominator simultaneously.

Gold’s 8.2% gain over the two weeks is the most analytically honest signal in the period. Gold does not typically rise when real rates are elevated and the dollar is strengthening. Both of those conditions were present during this period. The 10-year real yield moved higher alongside the nominal yield. The dollar gained 1.6% against a basket of currencies. In that environment, gold should have been under pressure. Instead it reached its highest level since the prior period’s high, as the bond market began signaling that it sees trouble ahead for equities and capital rotated out of long-duration Treasuries and into real assets simultaneously. When gold rises despite high real rates and a strong dollar, the market is pricing something that interest rate differentials alone cannot explain. It is pricing the possibility that the current monetary policy framework is insufficient to address an inflation regime with structural components.

The FOMC minutes from the July 29 meeting, scheduled for release on Wednesday August 19, are the most important near-term data point for the portfolio’s rate sensitivity posture. The July meeting produced three dissenting votes in favor of an immediate hike and Warsh’s historic first-ever chair abstention from the dot plot. The minutes will reveal whether the hawkish language in the post-meeting statement reflected genuine committee conviction or a negotiated compromise between factions. There is now a 64% chance of a hike by year-end according to the CME FedWatch Tool, and the minutes will either confirm or complicate that pricing.

The two-week period’s most significant company-level development was a private company revenue disclosure. Anthropic posted $11.5 billion in revenue during the second quarter and is meeting with banks for a potential IPO, likely to be the biggest on record. This single data point materially changes the analytical context for the prior Super Week earnings. When Amazon, Microsoft, and Alphabet each recorded billions in non-cash gains from their Anthropic stakes, the analytical critique was that those gains reflected a private company’s implied valuation rather than demonstrated economic value. A company generating $11.5 billion in quarterly revenue has demonstrated economic value. The mark-to-market gains that appeared to distort the prior quarter’s earnings now have a more defensible foundation than they did when they were disclosed. The AI monetization that the portfolio’s Blue Chip holdings are positioned to capture is not speculative at this scale. It is contracted revenue from a market that has proven it will pay.

Applied Materials reported results that beat every consensus estimate and saw its stock fall 5.1%. Momentum stocks gave back significant gains during the two-week period, posing the threat of entering a stealth bear market as the market bifurcated between businesses with contracted demand and businesses priced for demand that has not yet been contracted. The AMAT reaction is the clearest single-company illustration of where the two-week period sits analytically: the AI infrastructure story is real, the earnings are real, and the stock declines anyway because the valuation had already priced in a supercycle that the actual results, while strong, cannot consistently exceed.

The semiconductor recovery, with SOXX gaining 7% in the first week of the period, deserves analytical scrutiny rather than celebration. The fund liquidations that drove the July selloff removed a leveraged buyer. The subsequent recovery did not restore that buyer. It was driven by fundamental-oriented institutional capital returning to semiconductor positions at improved valuations. That is structurally healthier than the prior positioning, which depended on a $45 billion fund running four times leverage in concentrated AI hardware names. The recovery was real. The prior enthusiasm was not the correct baseline.

Next week’s two most important events: FOMC minutes on Wednesday August 19, and AZO Q4 FY2026 on August 26, followed by HEI Q3 FY2026 on August 25. The minutes will set the rate policy tone. The HEI print will resolve the portfolio’s longest-standing watchlist decision. Jackson Hole begins August 26, and Warsh’s first appearance there as Fed Chair will be the most important central bank communication of Q3.

This Week’s Signal: S&P 500 at 7,800. 30Y at 5.20%. Gold at $4,484. Three assets moving in directions that are not consistent with each other. The one that is pricing the macro environment most accurately will determine what Q4 looks like.

Share

Module 2: Earnings Decoder

Arista Networks (ANET) | Q2 FY2026 | August 4, After Close

Arista Networks delivered its first $3 billion revenue quarter in Q2 2026, posting revenue of $3.036 billion, an increase of 37.7% year-over-year and 12.1% sequentially, well above its own guidance of $2.8 billion. Non-GAAP diluted EPS of $1.02 topped the $0.89 consensus estimate by 15.18%, while non-GAAP operating margin expanded to 49.9% from 48.8% a year ago.

The quarter’s most analytically important disclosures are not in the income statement. Arista raised its full-year 2026 revenue outlook to approximately $12.6 billion, implying 40% growth and marking the third upward revision this year, as supply-chain conditions improved. The company tripled multi-year purchase commitments to $9.7 billion, secured memory supply into 2027, and established a liquid-cooling supply chain. Cumulative Etherlink AI fabric customers exceeded 100, up from four to five at the start of 2024.

The $9.7 billion in purchase commitments is the number worth internalizing. Commitments of that scale, covering two to three years of supply, are made by customers who have locked in AI infrastructure expansion plans that are not subject to quarterly budget review. These are not discretionary purchases. They are contractual obligations. When Azure, Google Cloud, and the hyperscalers commit $9.7 billion in forward networking spend to Arista, they are telling you that the infrastructure they are building today requires networking at a scale they cannot yet fully deploy. The deferred revenue figure of $6.9 billion, up from $6.2 billion sequentially, confirms that the revenue recognition lag is real and growing.

CEO Jayshree Ullal mapped out the scale-across opportunity, pegging the current market at $3 to $4 billion and expanding to $15 to $20 billion by 2030, with Arista expecting to capture approximately $1.2 billion of that in 2026. Scale-across is the AI networking architecture that connects GPU clusters across physical locations rather than within a single rack or data center. It is the networking problem that becomes necessary when AI workloads exceed the capacity of any single site. The fact that Arista is already quantifying a $1.2 billion target for 2026 in a market that did not exist three years ago is the most forward-looking signal in the report.

Thesis verdict: ANET is the clearest single-company confirmation of what the AI infrastructure investment cycle looks like in a networking equipment business. Revenue growing 37.7%, margin expanding, $9.7 billion in purchase commitments, 100 AI fabric customers, and a third guidance raise in a single year. The MSFT position in the portfolio benefits from exactly the infrastructure Arista is deploying. Azure’s $678 billion contracted backlog runs on networks that companies like Arista build. The two data points are not independent.

Applied Materials (AMAT) and Cisco (CSCO) | The “Buy the Rumor, Sell the News” Pattern

Applied Materials reported Q3 FY2026 results on August 13 that beat every consensus estimate across revenue, EPS, and margin. The stock fell 5.1% on the day. Cisco reported Q2 FY2026 results on August 12 with a similar outcome: revenue and earnings above consensus, stock lower. Two companies, two strong beats, two negative stock reactions in the same week.

The pattern is analytically precise and has a specific structural explanation. Both AMAT and CSCO entered their earnings reports having substantially re-rated higher over the prior twelve months on the AI infrastructure thesis. AMAT was up more than 60% in June alone. CSCO’s stock had reflected expectations that Cisco’s networking and security stack would become embedded AI infrastructure at the enterprise level. When the actual results confirm the thesis but do not dramatically exceed the already-elevated expectations, the trade that was pricing the expectation unwinds. The business continues compounding. The stock gives back the expectation premium.

It is a signal that the valuation had already priced it in. The distinction matters because the same dynamic applies across the AI hardware complex. NVIDIA reports on August 26. The business is almost certainly compounding at the rate Arista and the hyperscaler capex disclosures imply. The question is not whether the results will be strong. The question is whether they will be strong enough to exceed a valuation that has priced in a supercycle at maximum optimism.

The AMAT and CSCO reactions this week are the market’s advance notice that “beat and fall” is now the base case for AI infrastructure names trading at peak multiples. Owning the monetization layer rather than the infrastructure layer is precisely the portfolio posture designed to avoid this dynamic. FICO does not trade at 60 times earnings because the market is pricing a AI networking supercycle into its multiple. It trades at a discount to its earnings growth rate because the market has not yet credited its regulatory moat with the same enthusiasm it gave to semiconductor equipment companies.

Thesis verdict: Two strong beats, two stock declines. The AI hardware trade is entering a phase where execution confirms the thesis but does not drive incremental multiple expansion. The monetization layer, where FICO, MSFT, and HLI operate, does not face this dynamic because its multiples were never elevated by AI capex enthusiasm in the first place.

Preview: NVIDIA Q2 FY2027 | August 26, After Close

NVIDIA reports the quarter ending July 31 on August 26, simultaneously with AZO Q4 FY2026 and HEI Q3 FY2026. The week of August 26 is the single most concentrated catalyst week remaining in the two-week forward window.

The analytical setup entering the NVIDIA print is precisely the one described above: the AMAT and CSCO reactions suggest the market has moved to a “beat and hold” or “beat and fall” regime for AI infrastructure names. NVIDIA’s Blackwell architecture order book has been confirmed by every major hyperscaler earnings call. The question is whether the Q2 revenue of approximately $40-43 billion that consensus expects will produce a reaction similar to AMAT’s, or whether the magnitude of the beat required to generate upside is achievable.

The more important disclosure for the portfolio will be management’s commentary on Vera Rubin, the next-generation architecture, and any indication of whether AI inference workloads are growing as a proportion of total GPU demand. Inference is the phase where AI monetization becomes recurring and non-discretionary rather than front-loaded infrastructure spend. If NVIDIA’s customer mix is shifting toward inference, the AI monetization cycle the portfolio is positioned for is maturing on schedule.

Module 3: Smart Money Monitor

1. Gold at $4,484 With Rising Real Rates and a Strong Dollar: What Institutional Capital Is Actually Pricing

Gold rose 8.2% over the two-week period. The 10-year real yield moved higher. The dollar gained 1.6% against a basket of currencies. In the standard macro framework, both of those conditions are bearish for gold. Gold pays no income, so rising real yields increase its opportunity cost. A stronger dollar makes gold more expensive for foreign buyers, reducing demand. The standard framework predicted gold would fall. Gold rose 8.2%.

The explanation requires moving beyond the standard framework. Institutional capital that runs systematic macro models has been overriding the model signals on gold since late 2025, and the reason is structural rather than tactical. The 30-year Treasury yield hitting its highest level since June 2007 is not just a rate signal. It is a duration signal. When the long end of the curve moves to 5.20% despite falling growth expectations and soft retail sales data, the bond market is communicating that it has lost confidence in the policy framework’s ability to resolve the current inflation regime without economic damage. Lingering inflation and rising US fiscal deficits continued to push yields up despite soft economic data, which is the definition of a market that has stopped believing the central bank’s narrative about the inflation trajectory.

Institutional capital that has stopped trusting the policy framework does not move to long-duration Treasuries. It moves to gold. The 8.2% gain is a structural position on monetary policy credibility. When sovereign wealth funds, central bank reserve managers, and long-duration pension funds simultaneously reduce their confidence in a currency’s long-term purchasing power stability, gold is the asset that absorbs that reallocation regardless of what real rates are doing at any given moment.

Signal: Gold at $4,484 with real rates rising is the institutional equivalent of a confidence vote against the current policy framework. The portfolio’s defensive positioning, with above-40% cash rather than long-duration bonds, reflects the same analytical judgment without the explicit gold allocation.

2. The 30-Year at 5.20%: What Pension Funds and Insurance Companies Are Being Forced to Do

The 30-year Treasury yield hit its highest level since June 2007 during the two-week period, reaching 5.20%. The market commentary on this move focused on inflation and Fed policy. The more structurally important driver is institutional liability matching.

US pension funds and insurance companies hold approximately $15 to $20 trillion in long-duration liabilities that are typically hedged using long-dated Treasury bonds. When the 30-year yield rises rapidly, the present value of those liabilities falls, which creates a paradox: the pension fund’s balance sheet improves in accounting terms, but its existing long-duration bond portfolio generates mark-to-market losses simultaneously. The net effect depends on the hedge ratio and the speed of the move.

When the 30-year rises rapidly and unexpectedly, pension funds that are underhedged relative to their liabilities face pressure to buy more duration to close the gap. That buying pressure at 5.20% is structurally different from the buying pressure at 4.50%: at 5.20%, the liability discount rate is high enough that the funded status improvement is real and lasting, not just accounting. Some institutional buyers are therefore locking in 5.20% on the long end for the first time in nearly two decades, not because they believe inflation will stay at 4% but because the yield is finally high enough to fund their long-term obligations without relying on equity market returns.

Bank of America’s Michael Hartnett flagged that surging national debt and higher bond yields could derail the strong equity run, which is the institutional perspective that reconciles the equity market at 7,800 with the bond market at 5.20%: the equity market is pricing AI earnings growth, the bond market is pricing the cost of financing the government that backstops the entire system.

Signal: The 30-year at 5.20% is generating two simultaneous institutional flows: forced liability-matching buying from underfunded pensions at yields not seen since 2007, and forced selling from institutional equity holders who run valuation models that use the risk-free rate as a discount input. Both flows are real. Both are creating the volatility underneath the calm surface of a VIX at 15.

3. ETF Flows: The Rotation Is Slow But Persistent

Energy led performance over the prior week with approximately 6% gains, while healthcare and financials outperformed with more than 1% gains each. Defensive sectors including utilities, staples, and real estate gained and generally outperformed growth sectors over the last five sessions. Technology gained 4.4% over the past month but trailed energy and healthcare over that period.

Share

The ETF flow data tells the same story in capital terms. Healthcare and financial sector ETFs continued receiving net inflows during the period. Energy ETFs received inflows as oil prices partially recovered from the prior period’s lows. Technology ETFs received modest inflows on the back of the S&P 500 record high but the composition continued to shift toward diversified exposure rather than concentrated semiconductor positions.

The equal-weight S&P 500 has now outperformed the cap-weighted index for seven consecutive weeks. That duration is no longer a tactical signal. It is a structural signal. When the equal-weight index outperforms the cap-weighted index for seven consecutive weeks, it means the capital that drove the mega-cap AI premium is being systematically redistributed to the rest of the market at a rate that cannot be explained by short-term positioning changes. It requires institutional rebalancing decisions that operate on multi-quarter timescales.

The Russell 2000 small-cap index closed higher for four consecutive sessions during the period, which is the clearest signal that the rotation is broadening beyond the large-cap value names that led it in June and July. When small-cap outperforms over multiple sessions in an environment of rising rates and geopolitical uncertainty, the institutional buyers are not momentum traders. They are valuation-driven allocators who have concluded that the valuation gap between mega-cap AI names and the rest of the market has become too wide to ignore.

Signal: Seven consecutive weeks of equal-weight outperformance and four consecutive sessions of Russell 2000 gains are not noise. They are the institutional fingerprint of a multi-quarter reallocation away from mega-cap AI concentration and toward the broader earnings base that the portfolio has been building toward since June.

Module 4: Quality at a Discount

Accenture (ACN) | Forward PE 11.48x | Historical Average ~25x

The market is pricing AI as an existential threat to Accenture’s business model. At 11.48 times forward earnings, the lowest multiple in the company’s history, the stock offers a FCF yield of 10.97% on a business with 24.5% five-year average ROIC, $10 billion in cash, and a dividend approaching 4%.

The disruption thesis is partially correct: AI is compressing the time required for certain consulting tasks. Where it overreaches is in assuming that enterprise technology complexity decreases as AI adoption accelerates. It does not. Governance, security, and integration complexity grow alongside AI deployment, and Accenture’s $4.18 billion cybersecurity acquisition is a direct bet on that dynamic.

If Q4 bookings stabilize, the stock re-rates from 11.48 times toward 15 to 20 times, representing 30 to 75% upside without any earnings improvement. If bookings continue declining, 11.48 times may be fair value for a secularly challenged business. The Q4 print is the binary event that resolves the thesis in one direction.

Copart (CPRT) | Physical Moat, Temporary Pressure

Copart trades at approximately 28 to 30 times forward earnings, not cheap on any screen. What qualifies it here is the gap between what the market is pricing and what the business actually owns: 200 auction locations across 10 countries, 750,000 registered buyers in 170 countries, and physical land that no competitor can replicate in any timeframe that matters for a five-year thesis. The stock has declined approximately 45% from its 52-week high.

The market is pricing three concerns simultaneously: CEO transition, three consecutive quarters of earnings pressure, and used vehicle value normalization. The analytical response to each is specific. Jay Adair is not a new CEO, he ran the company for 14 years and knows the business at a level no external hire could replicate. Insurance unit volume softness is cyclical, not structural, as total loss frequency has a long-term upward trajectory driven by rising vehicle complexity. Q3 confirmed record insurance average selling prices even as volumes softened, which means pricing power is intact. The company holds $5.5 billion in cash with no debt.

The September 9 Q4 print is the first under Adair and the

Read more

On January 1, 2026, Greg Abel officially took over from Warren Buffett. Five months later, he handed in his first quarterly report.

Q1 2026 operating earnings came in at $11.35 billion, up roughly 18% from $9.64 billion in the same period last year. Net income attributable to shareholders reached approximately $10.1 billion, more than doubling from $4.6 billion a year ago. The headline numbers look solid. But several things beneath the surface deserve a closer read.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

The Cash Pile

By the end of March, Berkshire’s cash, cash equivalents, and short-term securities had swelled to a record $397.4 billion, up from $373 billion at year-end 2025. As a share of total assets, cash now represents 31.7% of Berkshire’s balance sheet, the highest proportion on record. That figure, adjusted for accounts payable on Treasury Bill purchases, comes down to approximately $380.2 billion in truly deployable capital. The difference is a rounding issue. The scale is not.

This is the fourteenth consecutive quarter in which Berkshire has been a net seller of equities. The accumulation of cash at this pace is not an accident or an oversight. It reflects a deliberate judgment, shared across both Buffett and Abel, that the current market offers few assets cheap enough to justify large-scale deployment. At Berkshire’s scale, “cheap enough” means a meaningful margin of safety on a transaction that can move the needle on a $900 billion enterprise. Those opportunities have been rare.

Then, shortly after the quarter closed, Abel moved. Berkshire agreed to purchase $10 billion in Alphabet shares directly from the company, bypassing the open market entirely. This single transaction is the most revealing signal of the Abel era so far. Buffett entered Apple because he understood it as a consumer products company with extraordinary brand loyalty. Abel’s decision to commit $10 billion to Alphabet suggests he is willing to pay for digital infrastructure at scale, a subtly different intellectual framework. Whether this represents a one-time opportunistic move or the beginning of a broader reorientation toward technology remains the central question heading into the next several quarters.

Insurance: Strong at the Top, Troubled at GEICO

The insurance segment was the primary driver of the quarter’s earnings improvement. Underwriting profit expanded 28% to $1.72 billion, compared with $1.34 billion in Q1 2025. The comparison was favorable partly because Q1 2025 was heavily impacted by the California wildfires, which generated substantial catastrophe losses. Absent a comparable event in Q1 2026, the insurance business recovered to a more normalized level of profitability.

Share

Insurance float, the net liabilities Berkshire assumes under its insurance contracts and effectively a source of interest-free capital, stood at $176.9 billion as of March 31, up roughly $500 million since year-end 2025. The float has compounded over decades into one of Berkshire’s most durable structural advantages. It costs nothing when the underwriting is disciplined, and it funds a portion of the investment portfolio that generates the vast majority of Berkshire’s investment income.

GEICO is a different conversation. The flagship auto insurance unit reported a 34% decline in underwriting earnings, driven by higher claims costs and increased marketing expenditure. Cathy Seifert, an analyst at CFRA Research, put it plainly: “Most of Geico’s peer group this quarter posted significantly improved underwriting results. They’re a big unit and that’s a big deterioration.”

GEICO has spent the past several years in a deliberate process of underwriting reconstruction: raising rates sharply, shedding high-risk policyholders, and rebuilding profitability at the cost of market share. That process produced a meaningful earnings recovery through 2024 and into 2025. The Q1 2026 deterioration suggests the recovery is incomplete. Higher claims frequency, rising repair costs, and competitive pressure from direct-to-consumer auto insurance competitors are making GEICO’s path to restored profitability longer than management initially projected.

Insurance investment income also declined, falling 7% to $2.68 billion, pressured by lower interest rates reducing income on the fixed income portfolio. This is a structural headwind that will persist until rates stabilize or Berkshire reallocates the portfolio toward higher-yielding assets.

BNSF and Energy: Reliable, Predictable, Unremarkable

BNSF Railway contributed $1.38 billion in earnings, a 13% increase from $1.21 billion a year ago. Higher revenues and operational efficiency improvements drove the gain. BNSF’s performance is rarely a source of surprise in either direction. It is a physical infrastructure asset with significant pricing power in certain freight corridors, and its earnings tend to reflect the general level of economic activity with a moderate lag.

Berkshire Hathaway Energy contributed $1.11 billion, roughly flat with the $1.10 billion it posted in Q1 2025. Natural gas pipeline operations and federal renewable energy tax credits provided the support. BHE has been a complicated story over the past two years, given the liability exposure from its role in certain wildfire events in the western United States. The near-flat earnings in Q1 suggest stabilization rather than acceleration.

Manufacturing, service, and retail operations added $3.2 billion, a 5% year-over-year gain. This segment is the broadest in Berkshire’s portfolio, encompassing businesses from industrial components to furniture retail to aircraft parts distribution. A 5% gain across such a diverse collection of businesses, in a macro environment that has been uneven, is a reasonable outcome.

What Abel Is and Is Not Doing

Abel resumed share repurchases in Q1, buying back $234 million of stock at an average price of $486.92 per share. The dollar amount is modest relative to the cash pile, but the decision to resume buybacks at all signals that Abel considers current valuations at least acceptable, if not particularly compelling.

He has not altered Berkshire’s posture of patient capital accumulation in any dramatic way. He has not made any transformative acquisitions. The $9.7 billion purchase of OxyChem from Occidental Petroleum, completed earlier this year, is the most significant transaction of his tenure so far. It is an industrial chemicals business with stable cash flows and a physical asset base, a Berkshire-style acquisition in every conventional sense.

What Abel has communicated clearly is that he intends to operate within the framework Buffett built rather than redefine it. He said at the annual meeting that Berkshire’s commitment to its owners remains unchanged, and that the institutional knowledge of the businesses has been fully preserved through the transition. These are the right things to say in a first quarter. The next several years will reveal whether the capital deployment decisions, particularly in moments of market dislocation, reflect the same quality of judgment.

The Assessment

The quarter was solid without being exceptional. Operating earnings grew 18%, which looks strong until you strip out the favorable comparison to a catastrophe-impacted Q1 2025 and adjust for foreign exchange tailwinds. On an organic, currency-neutral basis, operating earnings growth was closer to 7%. That is the real underlying run rate, and it is respectable for a business of Berkshire’s size, but it does not justify paying a significant premium to book value.

Share

GEICO remains the clearest operational concern. In a quarter when Berkshire’s other insurance units and the broader industry posted improved underwriting results, GEICO moved in the opposite direction. The business is structurally sound and its brand carries real value, but the gap between its underwriting performance and its peer group has widened in a way that demands explanation beyond one quarter’s claims variance.

The Alphabet investment is the most consequential development of the period, and it did not show up in the Q1 numbers at all. If Abel is willing to allocate $10 billion to a single technology holding, it suggests the cash pile may begin to shrink at a faster pace than the prior trend implied, assuming market conditions cooperate.

The Abel era is now several quarters old. The playbook looks familiar. The capital is accumulating. The first major deployment signal has been sent. The question the next several reports will answer is whether Abel’s version of patient capital eventually concentrates into a defining transaction the way Buffett’s Apple investment defined the final chapter of his tenure.

Key Metrics | Q1 2026

This analysis is part of the Thesis Rationale Due Diligence Desk series. It is provided for informational and educational purposes only and does not constitute investment advice. Investors should conduct their own due diligence before making any investment decisions.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

📊 Market Lens | Weekly Snapshot
Week Ending: August 1, 2026

• VIX: 16.84 (-1.86 pts WoW)

• 10Y Treasury: 4.512% (-19.4 bps WoW)

• 30Y Treasury: 5.009% (-6.1 bps WoW)

• WTI Crude: $84.20 (-7.5% WoW)

• Brent: $87.60 (-12.5% WoW)

• Gold (XAU): $4,142.30 (+2.9% WoW)

• S&P 500 (SPX): 7,567.80 (+2.1% WoW)

• Nasdaq: 26,154.22 (+4.1% WoW)

• USD Index (DXY): 99.87 (-1.4% WoW)

Share

This was the most important earnings week of 2026. Five of the most valuable companies on earth reported simultaneously, the Federal Reserve held rates for the fifth consecutive meeting with three dissents, and a $45 billion hedge fund was liquidated in a single afternoon. The S&P 500 gained 2.1% on the week. That number understates everything that happened inside it.

The week produced a pattern that requires immediate analytical attention. Three of the five technology giants reporting this week, Microsoft, Amazon, and Alphabet last week, each included massive non-cash gains from their investments in Anthropic in their reported earnings. Amazon’s Q2 net income of $62.6 billion included $53.4 billion in non-operating pre-tax income, primarily from marking its Anthropic stake to a higher value on paper. Microsoft’s quarter included a $3.2 billion gain from its investment in Anthropic. Alphabet reported a $99 billion mark-to-market gain the prior week from SpaceX and Anthropic combined. Strip these gains from all three companies and the earnings picture is still strong but considerably more modest. The more analytically significant observation is structural: when three of the five most valuable companies on earth are reporting inflated earnings from the same private AI company’s implied valuation, the AI investment ecosystem has become self-referential in a way that requires scrutiny rather than celebration.

The operational story beneath the accounting is, in several cases, genuinely strong. Microsoft reported Q4 FY2026 revenue of $90 billion, up 18%, with Azure revenue growing 43%, accelerating from 40% the prior quarter, and crossing $100 billion in annual revenue for the first time. Microsoft 365 Copilot reached 30 million paid seats, up from 20 million in April, while the contracted backlog jumped $51 billion in a single quarter to $678 billion. The stock surged 8.13% to $422 after hours. The backlog number is the most important disclosure in the report. Sequential growth in commercial remaining performance obligations was driven by commitments from clients other than AI model developers, meaning the enterprise AI adoption is broadening beyond the handful of frontier labs that initiated the original Azure hyperscaling. When 90% of Microsoft Cloud revenue comes from customers outside the AI labs, the monetization thesis has crossed from early adopter to mainstream deployment.

Amazon reported Q2 revenue of $200.6 billion, the first time the company has exceeded $200 billion in a single quarter, with AWS growing 36.7% year-over-year to $42.2 billion, its fastest pace in 18 quarters. The company raised its 2026 capital expenditure guidance to approximately $220 billion. The operating income of $27.5 billion, up 43% year-over-year, is the number that matters for long-term investors: it confirms that Amazon’s operating leverage is expanding even as capital expenditure escalates. The $53.4 billion Anthropic gain inflating the headline EPS from a real $9 billion to a reported $62 billion is an accounting event that will not repeat in Q3. The AWS acceleration is structural and will.

Meta reported Q2 revenue of $60.8 billion, up 28% year-over-year, but delivered the week’s most honest look at what aggressive AI capital expenditure actually costs. Net income fell 14% to $15.85 billion as single-quarter capital expenditure reached $31.08 billion. Operating margin compressed from 43% to 31%. Meta did not benefit from a large private equity stake gain to paper over the margin compression. It reported what AI spending looks like without an accounting cushion. The stock fell after hours before recovering partially as management guided Q3 revenue above consensus. Meta’s quarter is the most instructive of the week precisely because it shows the unvarnished cost of the AI infrastructure race.

Apple reported revenue of $109.4 billion, slightly above the $108.65 billion consensus, with EPS of $2.02. Apple’s quarter is notable primarily for what it lacks: a massive Anthropic stake gain, an AI infrastructure capex escalation, or a guidance compression from spending commitments. The business generated its revenue through hardware sales and services ecosystem monetization, compounding at a rate consistent with the prior several years. In a week where every other major technology company was navigating the complex interaction between AI capital expenditure and financial reporting, Apple’s relative simplicity was itself a data point. The stock has been the best performer among the Magnificent Seven year-to-date, which is not a coincidence.

The Federal Reserve’s July 29 decision arrived in the middle of this earnings avalanche and was almost immediately overshadowed by it. The FOMC voted 9 to 3 to hold the federal funds rate at 3.50% to 3.75%. Three regional Federal Reserve presidents voted for an immediate 25 basis point increase. Warsh described the current policy stance as not a pause and declined to provide any forward guidance on the trajectory of cuts. The 10-year Treasury yield fell on the week despite the hawkish hold, driven by oil price retreat as the Iran tactical pause reduced energy inflation expectations. This is the interest rate paradox of the current moment: a Fed signaling willingness to hike is being partially offset by a commodity market signaling disinflation from geopolitical de-escalation. The bond market is trying to price both simultaneously.

The week’s most dramatic non-earnings event was the complete collapse of Situational Awareness, the hedge fund managed by Leopold Aschenbrenner, a former OpenAI researcher who had grown the fund to nearly $45 billion in assets under management through the first half of 2026 on the strength of concentrated AI hardware positions. The fund employed approximately four times leverage on positions concentrated in SK Hynix, Micron, CoreWeave, Nebius, and other AI infrastructure names. When those names began their July correction, the leverage amplified the losses with mathematical precision. The fund lost approximately 67% in July alone. On July 24, Aschenbrenner sent investors a letter encouraging them to add to positions. Within days, JPMorgan, Bank of America, and Goldman Sachs issued margin calls. The fund’s equity positions were liquidated at a discount to Citadel and other large market participants in a single session. The Aschenbrenner episode is not primarily a story about a bad trade. It is a story about the structural danger of applying leverage to AI hardware thesis positions in a market where single-day moves of 8% to 16% in semiconductor names have become routine. The thesis may have been correct. The position sizing was the variable that determined the outcome.

Module 2: Earnings Decoder

FICO (Fair Isaac) | Q3 FY2026 | July 29, After Close | Portfolio Holding

FICO delivered non-GAAP EPS of $12.18, beating analyst estimates by 4.1%, and raised full-year guidance, though shares fell 9.33% in after-hours trading to $1,245 from the regular session close of $1,373.08.

Revenue reached $674.2 million, up 26% year-over-year, with Scores segment revenue of $458.9 million, up 41% year-over-year. Mortgage origination revenues surged 97% year-over-year, with volumes rising in the low-single digits. The revenue miss of approximately $18 million against a $679 to $692 million consensus was the primary driver of the after-hours decline.

For the first time, Platform ARR exceeded Non-Platform ARR, demonstrating that the company’s AI-native decisioning platform has reached critical mass. This is a structural milestone that the market largely ignored in the noise of the revenue miss.

The 9.33% after-hours decline requires analytical separation from the thesis. The revenue miss was real: $674 million against a $679 to $692 million range. But the miss was driven entirely by the Software segment underperforming. The Scores business, which is the thesis, delivered 41% growth with 97% mortgage revenue growth on low-single-digit volume increases. The pricing power is intact. The regulatory moat is intact. The Platform ARR milestone is structurally significant. The non-GAAP operating margin expanded to 62% from 57% a year earlier, an improvement of 479 basis points.

Full-year 2026 guidance was raised: revenue now expected at $2.53 billion, representing 20% to 27% growth, with non-GAAP net income at $979 million, up 33%.

Thesis verdict: The Scores thesis is intact and accelerating. The Software segment miss created a short-term pricing dislocation. A business with 62% non-GAAP operating margins, 97% mortgage revenue growth, and a Platform ARR milestone trading down 9% on an $18 million revenue miss is offering a re-entry point rather than a thesis breach. Monitoring variable: Software ARR acceleration in Q4 as Platform adoption compounds.

Qualcomm (QCOM) | Q3 FY2026 | July 29, After Close

Qualcomm reported $9.95 billion in revenue, beating the $9.67 billion estimate, while adjusted EPS of $2.21 fell slightly short of the $2.23 expectation. The headline miss on EPS is analytically less important than two other disclosures.

The first is the automotive segment. Automotive revenue grew 61% year-over-year in Q3, and Qualcomm raised its automotive annualized revenue run-rate outlook to approximately $7 billion exiting fiscal 2026, up from $6 billion. This is the most structurally durable revenue stream in Qualcomm’s portfolio: automotive compute content is contractually locked in years before a vehicle ships, making it immune to the smartphone inventory cycles that have dominated quarterly results.

The second is the strategic pivot CEO Cristiano Amon announced alongside the results. Qualcomm is raising prices across all products starting September 1, attributing the decision directly to the memory shortage driving up input costs. “Cost went up, prices are going to go up.” This is a notable test of Qualcomm’s pricing power with smartphone OEMs who already have few alternatives for premium Snapdragon chipsets. Non-handset revenues are now projected to grow to $40 billion by fiscal 2029, nearly double the target shared in November 2024, with data center revenue growth expected to accelerate from 24% in fiscal 2026 to greater than 60% in fiscal 2027.

The Q4 guidance of $9.7 to $10.5 billion with EPS of $2.05 to $2.25 was lighter than expected, and the stock fell after hours. Management noted that Apple revenue is expected to decline approximately 50% from September to December due to materially lower share in new iPhone launches, though this will be significantly offset by Android growth and the beginning of data center revenue ramp.

Thesis verdict: The handset business is structurally challenged in the near term. The automotive and data center pivots are real and accelerating. The price increase announcement is the most analytically interesting signal in the report: a company with genuine pricing power raises prices when costs rise. The market is not yet crediting Qualcomm for the diversification that is already in its revenue run rates.

Arm Holdings (ARM) | Q1 FY2027 | July 29, After Close

Arm reported first-quarter revenue of $1.29 billion, a 22% increase year-over-year, with adjusted EPS of $0.45 against a $0.40 consensus estimate. Royalties climbed 22% to $715 million, buoyed by data center royalties that more than doubled, and licensing fees advanced 23% to $574 million.

The data center royalties more than doubling is the single most important number in the report. ARM’s business model generates royalties on every chip shipped that uses its architecture. When data center royalties double, it means the number of Armv9-based chips entering hyperscaler deployments has reached a volume that produces meaningful recurring revenue, not just design wins. Both royalty and licensing revenues reached record first-quarter levels, driven primarily by surging demand in cloud AI infrastructure and continued adoption of Arm’s advanced Armv9 architecture and Compute Subsystem platforms.

The AGI CPU development is the longer-duration thesis. Arm projects AGI CPU revenue will reach approximately $15 billion by fiscal 2031, with cumulative customer demand already topping $2 billion, double the initial $1 billion opportunity identified. ByteDance and Oracle have begun utilizing the chip. The $25 billion total revenue target by fiscal 2031 implies the company believes it can grow total revenue approximately five times from current levels in five years.

The margin pressure is real and worth monitoring. Operating expenses rose 28% to $1.16 billion as R&D spending and headcount growth compressed operating margins to 7% from 11% a year ago. The stock dipped after hours despite the beat, reflecting the market’s concern that the investment required to pursue the $25 billion revenue vision is front-loaded while the revenue remains back-loaded.

Thesis verdict: ARM is building the foundational IP layer for the next phase of AI compute, and the data center royalty acceleration confirms the thesis is not speculative. The margin compression is a deliberate investment in the AGI CPU platform rather than a structural deterioration. The $25 billion revenue target by FY2031 is ambitious but grounded in a licensing model that is genuinely architecture-agnostic at the chip level. Not a portfolio holding, but the most structurally interesting semiconductor story of the earnings week alongside SK Hynix.

The Anthropic Accounting Theme

The week produced a pattern worth naming explicitly. Three companies, Microsoft, Amazon, and Alphabet the prior week, each reported earnings that were materially inflated by non-cash mark-to-market gains on their Anthropic investment stakes. Amazon’s $53.4 billion gain converted what would have been approximately $9 billion in net income into a reported $62.6 billion. Alphabet’s $99 billion gain the prior week produced a similar distortion. Microsoft’s $3.2 billion gain was smaller in absolute terms but followed the same structure.

The cumulative effect is that the S&P 500’s Q2 earnings season, which FactSet reported as producing a 15.7% net profit margin, a 15-year high, is substantially a reflection of a single private company’s rising implied valuation being recognized on three different public company balance sheets simultaneously. Anthropic is not yet public. Its implied value is derived from its most recent funding round. When that implied value rises, three of the most widely held stocks in the world report higher earnings without generating a single additional dollar of operating cash flow.

It means the current earnings reporting season requires more careful reading than the headlines suggest. The businesses that are genuinely compounding, AWS at 37% growth, Azure at 43% growth, Google Cloud at 82% growth, are doing so on real contracted revenue. The accounting gains sitting on top of those results are not.

Module 3: Smart Money Monitor

This module tracks the capital allocation decisions of institutional investors, sovereign wealth funds, and large fund complexes whose moves reveal structural positioning rather than short-term sentiment.

1. The Leopold Liquidation: What Three Prime Brokers Doing the Same Thing Simultaneously Means

The collapse of Situational Awareness is being covered as a story about a young fund manager who made a bad bet with too much leverage. The more analytically important read is institutional.

When JPMorgan, Bank of America, and Goldman Sachs simultaneously issue margin calls on the same fund in the same session, it means three separate prime brokerage risk desks independently concluded that the collateral supporting the fund’s positions was no longer sufficient to cover the exposure. Prime brokers do not coordinate margin calls. They make independent risk assessments using their own models. When all three arrive at the same conclusion on the same day, the underlying positions have moved so far and so fast that no risk model could justify continuing to extend credit against them.

The fund’s strategy was analytically coherent: long AI hardware infrastructure, short traditional software, four times leveraged. The thesis played out correctly for the first half of 2026. The problem was not the thesis. It was the leverage applied to positions in an asset class where 8% to 16% single-day moves have become routine. At four times leverage, a 16% single-day decline in a core position produces a 64% loss on that allocation before any recovery is possible. The July chip stock rout, driven by China AI manufacturing news, produced exactly that sequence across Situational Awareness’s core holdings.

The broader institutional signal is structural. The prime brokerage community has now observed multiple AI-themed leveraged funds experiencing forced liquidations in concentrated semiconductor positions. The next tier of similar funds will face more conservative margin terms going forward, which means the marginal buyer at the AI hardware thesis trade has become less leveraged and therefore less aggressive. That removes a source of incremental demand that had been supporting semiconductor valuations above what fundamental earnings models justify.

Signal: When three prime brokers liquidate the same fund simultaneously, the episode becomes a credit tightening event for the entire leveraged AI hardware trade. The leverage that amplified Situational Awareness’s gains on the way up was providing marginal price support for the entire category. That support has now been removed.

2. FOMC 9-3 and the Bond Market’s Contrarian Read

The July 29 FOMC decision produced one of the most analytically interesting market reactions of the year. Warsh held rates at 3.50% to 3.75% with three dissents in favor of an immediate hike, described the stance as not a pause, and refused to provide any forward guidance on cuts. By every surface measure, the communication was hawkish.

The bond market disagreed. The 10-year Treasury yield fell on the week from 4.706% to approximately 4.512%. Rate hike probabilities in CME FedWatch futures moved lower after the meeting, not higher. Institutional fixed income investors, who have the most direct financial stake in correctly pricing the Fed’s next move, interpreted Warsh’s communication as less hawkish than the headline suggested.

The analytical logic is specific. Warsh declined to provide forward guidance in either direction. A truly hawkish chair would have explicitly raised the probability of a September hike. Warsh did not. His statement that the policy stance is not a pause is technically accurate but strategically ambiguous: it neither commits to holding nor commits to hiking. Institutional investors read the ambiguity as a signal that Warsh is watching the incoming data, specifically oil prices and core PCE, before making any commitment. With WTI falling below $85 during the week on the Iran tactical pause, the inflation input that has most directly justified hawkishness is moving in the direction that reduces the probability of near-term action.

The equity market and the bond market produced opposite signals after the same FOMC meeting. Equities rallied on the mega-cap earnings beats. Bonds rallied on the dovish re-read of Warsh’s language. Both cannot be entirely correct, but the bond market’s interpretation, that the Fed is data-dependent rather than pre-committed to hiking, is structurally more consistent with how Warsh has conducted his first two meetings.

Signal: Institutional fixed income investors are pricing a Fed that is watching, not moving. The 10-year falling despite a 9-3 vote with three hike dissents is the bond market’s judgment that Warsh is more flexible than his language implies. The portfolio’s positioning, with cash above 38% and no rate-sensitive long duration exposure, is correctly calibrated for either outcome.

3. ETF Flows After Super Week: The Rotation Is Not Over

The week’s ETF flow data reveals that the mega-cap earnings beats did not reverse the sector rotation that has been building since June. Financial sector ETFs continued absorbing inflows following the historic bank earnings from two weeks prior. Healthcare ETFs maintained their trajectory. Technology ETFs received net inflows on the back of the MSFT and Amazon beats, but the composition tells a more precise story: the inflows concentrated in broad market and large-cap technology ETFs rather than in semiconductor-specific vehicles.

The SOXX, the Philadelphia Semiconductor ETF, did not benefit meaningfully from the earnings week despite Arm’s beat and Qualcomm’s revenue outperformance. The fund liquidations in AI hardware-concentrated positions created selling pressure that offset the positive earnings reads. Institutional capital is not returning to the concentrated semiconductor positions that defined the first half of 2026. It is accepting technology exposure through diversified vehicles where no single semiconductor name can produce a 16% single-day drawdown on a thesis event.

The Dow Jones Industrial Average closed the week near its all-time high while the Nasdaq recovered but did not lead. The equal-weight S&P 500 continues to track at or above its cap-weighted counterpart, which is the most persistent signal that the rotation toward broad earnings quality is institutional rather than retail-driven.

Signal: The mega-cap earnings beats slowed but did not reverse the sector rotation. Institutional capital is accepting technology exposure through diversified vehicles while avoiding the concentrated semiconductor positions that generated the most volatility. The rotation from narrative to cash flow is now four weeks old and has survived the most concentrated earnings week of the year without reversing.

4. How Institutional Models Are Adjusting

The week produced a structural analytical problem for every institutional model that uses reported GAAP earnings as an input. Three of the five largest technology companies, Microsoft, Amazon, and Alphabet the prior week, reported earnings that were materially inflated by non-cash mark-to-market gains on their Anthropic investment stakes. Amazon’s $53.4 billion Anthropic gain converted approximately $9 billion in operating net income into a reported $62.6 billion. Alphabet’s $99 billion gain the prior week produced a similar distortion. Microsoft’s $3.2 billion was smaller but structurally identical.

Institutional investors running factor models based on earnings quality scores have already made the adjustment. The most widely used institutional earnings quality frameworks strip non-operating investment gains from reported income and weight them separately from recurring operating earnings. When those frameworks are applied to the current quarter, the S&P 500’s 15.7% net profit margin that FactSet reported retreats to approximately 14.4% excluding Alphabet’s gain, and further to roughly 13.8% excluding Amazon’s gain.

The more interesting institutional question is forward-looking. Anthropic’s implied valuation, which drives the mark-to-market gains at Amazon, Microsoft, and Alphabet, will fluctuate with Anthropic’s private funding rounds. If Anthropic raises its next round at a higher valuation, all three companies report another accounting gain. If Anthropic’s implied value declines, all three take a loss. The earnings of three of the world’s most widely held stocks are now partially dependent on the private market valuation of a company that has never been profitable and does not report public financial statements.

Signal: Institutional models have adjusted for the Anthropic distortion. The earnings quality scores that drive systematic factor allocations are using operating income, not GAAP net income, as their primary input. The companies in the portfolio that generate earnings from contracts, regulatory requirements, and non-discretionary demand do not carry this accounting dependency.

Module 4: Portfolio Desk

This Week’s Focus

No position changes were made this week. The deliberate choice to hold through the most concentrated earnings week of 2026 reflects a specific analytical posture: the information required to make the next set of portfolio decisions has now arrived, and the decisions themselves belong to the following week when the dust has settled. FICO delivered a mixed quarter that requires honest assessment. MSFT delivered the strongest single-quarter thesis confirmation since initiation. Jay Adair formally assumed the CPRT CEO role on July 31. The portfolio enters August with higher conviction on several positions and an unresolved question on one.

Defensive Anchors

Rollins (ROL) | Hold | ~4.0%
No material developments. The rotation toward non-discretionary service businesses that has been building for six weeks continues to benefit the thesis environment without requiring any company-specific catalyst. Monitoring variable: Q2 organic revenue growth.

Visa (V) | Hold | ~3.0%
No material developments. The Iran tactical pause and oil price retreat are constructive for the cross-border travel recovery thesis. Monitoring variable: cross-border transaction volumes in the next quarterly disclosure.

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

Share

The past two weeks produced the widest internal divergence in the equity market since the early months of this year. The headline S&P 500 declined 2.1%. The Dow Jones Industrial Average gained ground. The Nasdaq fell 4.4%. WTI crude rose 28%. The 10-year Treasury yield reached its highest level of the year. And JPMorgan Chase posted the highest quarterly profit of any bank in US history. These numbers are not part of the same story. They are competing descriptions of the same market, and understanding which one is real requires looking beneath each of them.

July 14 was the most concentrated single morning of financial information in 2026. All five major US banks reported simultaneously before the opening bell. June CPI dropped to 3.5%, the lowest reading since January, driven by energy price easing from the prior Hormuz de-escalation. Kevin Warsh delivered his inaugural testimony before the Senate Banking Committee as Federal Reserve Chairman. And IBM disclosed preliminary Q2 results that sent its stock down more than 23% before the market opened. Four events, four completely different signals, all arriving at the same moment.

The bank results were historic in a literal sense. JPMorgan’s profits jumped 41% year-over-year, and Goldman Sachs reported diluted earnings per share of $20.98, nearly double what it posted in the same period a year earlier, with net revenues of $20.34 billion, a 39% increase year-over-year. JPMorgan EPS of $6.14 versus the consensus of $5.85, and revenue of $58.02 billion against the $50.19 billion expected, represents one of the largest beats in recent JPMorgan quarterly history. The primary drivers were investment banking fees from the SpaceX IPO, elevated fixed income trading revenues from Hormuz-driven energy volatility, and net interest income that continues to compound in the higher-for-longer rate environment. Goldman Sachs CEO David Solomon noted that the firm’s deal backlog is at its highest level in five years, which is a forward-looking signal rather than a backward-looking one. The banks are not just reporting what happened in Q2. They are telling you what Q3 and Q4 look like.

IBM’s warning tells the opposite story about the same economy. Enterprise IT budgets are being cannibalized. Companies that committed to purchasing AI hardware, servers, and accelerators in Q1 and Q2 are discovering that those capital expenditures have consumed the portions of the budget that would otherwise have gone to software upgrades, system integrations, and consulting engagements. IBM’s results confirmed what the prior Accenture quarter had implied: the shift from AI planning to AI hardware deployment is creating a discrete trough in traditional enterprise software spending. The contradiction between Goldman Sachs at its best quarter ever and IBM at its worst earnings surprise in years is not a paradox. It is a description of where value is concentrating in the current cycle, in trading, deal-making, and infrastructure, and where it is being temporarily displaced.

Two days later, on July 16, a separate shock arrived from a completely different direction. Reports circulated that Chinese semiconductor manufacturers had achieved a meaningful breakthrough in domestically produced chip-making equipment, reducing reliance on foreign-supplied tools from ASML, Applied Materials, and Lam Research in specific process steps. The credibility of those reports remains contested, but the market reaction was immediate and severe. The Philadelphia Semiconductor Index fell more than 6% in a single session. Applied Materials, Micron, and AMD each declined approximately 8%. Korean equities fell so sharply that circuit breakers were triggered. SoftBank shares dropped 9.2% in Tokyo, and Tokyo Electron lost 9%. The scale of the reaction reflected not just the specific news but the accumulated tension in semiconductor valuations, which had been pricing in multi-year equipment export dominance as a structural certainty rather than a competitive assumption.

Meanwhile, the energy picture reversed everything the Iran peace deal had established. Following US strikes on Iranian positions in retaliation for attacks on commercial shipping in the Hormuz Strait, WTI crude surged from $71 to an intraday peak of approximately $102 per barrel, its highest level since the early weeks of the original conflict. Brent crossed back above $100. The moves reversed most of the oil price deflation that had occurred following the June ceasefire announcement. As of July 25, prices had partially pulled back to the $91 to $92 range as diplomatic contacts resumed and Iran signaled a tactical pause in offensive operations. But the structural damage to the peace deal narrative is real. Oil at $91 was supposed to be the floor under a normalization scenario. It turned out to be a ceiling under a re-escalation scenario. The distinction matters enormously for the Fed’s inflation path and for every position in the portfolio that was calibrated on energy normalization as a base case assumption.

The most analytically complex data point of the two-week period came on July 22, when Alphabet reported Q2 2026 earnings. Revenue rose 24% to $119.8 billion, while Google Cloud revenue accelerated to 82% growth. GAAP EPS came in at $9.11 against a consensus of $2.89. The headline beat of 217% is the single largest EPS surprise for a large-cap technology company in recent memory, and it is almost entirely illusory. The gain on equity securities of $99.0 billion, primarily from unrealized mark-to-market appreciation on Alphabet’s minority stakes in Anthropic and SpaceX, increased net income by $77.1 billion and diluted EPS by $6.26. Stripping this out, core operating EPS would have missed the $2.63 consensus estimate by approximately one cent. The S&P 500’s net profit margin reaching a 15-year high of 15.7% this quarter is therefore not a statement about the productive capacity of the US economy. It is a statement about the market value of private equity stakes held by one company. If Alphabet is removed from the calculation, the S&P 500 net profit margin reverts to approximately 14.4%.

The Google Cloud number, however, deserves to stand on its own. Google Cloud revenue increased 82% year-over-year to $24.8 billion, with a cloud backlog of $514 billion, growth that was ahead of Azure’s 40% and AWS’s most recent reported figures. Alphabet also raised full-year capex guidance to $180 to $190 billion, a further increase from the prior $175 to $185 billion range. The stock fell after hours despite the headline beat, which is the market’s correct interpretation: the one-time equity gains are not recurring, the core business delivered modestly, and the capex escalation continues to compress near-term free cash flow. Berkshire’s $10 billion private placement into Alphabet six weeks ago was validated by the cloud growth rate. It was not validated by the headline EPS.

The two-week period closes with the market in a state of genuine analytical uncertainty that the headline indices do not fully communicate. The 10-year yield at 4.706% is pricing a Fed that is not cutting. WTI at $91 is pricing a Hormuz that is not reopening. The Dow’s outperformance of the Nasdaq by more than 6 percentage points over the period is pricing a rotation from AI narrative to operational cash flow. The banks at record profits are pricing a credit environment that has not yet cracked. IBM’s collapse is pricing an IT services market that has been structurally disrupted. All of these markets are correct about something. The question the portfolio must answer is which of them is correct about what happens next.

Next week is the most concentrated earnings week of 2026. Meta reports Tuesday July 29 evening. Microsoft reports Wednesday July 30 after close. Amazon reports Thursday July 31. FICO reports Tuesday July 29 after close. The FOMC decision arrives Wednesday at 2 PM. Five market-moving events across four calendar days. The results of that week will determine whether the rotation of the past two weeks is a pause in the AI bull market or the beginning of a more durable repricing of AI capital allocation.

This Week’s Signal: Banks at record profits. IBM in freefall. Cloud at 82% growth with a $99 billion accounting distortion. Oil back above $90.

Module 2: Earnings Decoder

Goldman Sachs (GS) | Q2 2026 | July 14, Before Open

Goldman Sachs reported diluted earnings per share of $20.98, nearly double what it posted in the same period a year earlier, with net revenues of $20.34 billion, a 39% increase year-over-year, which is the best quarterly performance in Goldman’s history. Investment banking revenues surged on SpaceX IPO fees, which Goldman led as primary underwriter. The deal backlog is at its highest level in five years.

The forward signal matters more than the backward one. A five-year high backlog means Goldman is describing 2026 H2 and 2027 as visible revenue, not speculation. M&A advisory, ECM, and structured finance are all accelerating simultaneously, which typically happens when corporate confidence and credit availability are aligned. The current environment, with rates high but stable and credit quality intact, is precisely that alignment.

Thesis verdict: Goldman’s backlog is the most bullish forward indicator in the two-week period. Directly validates HLI’s restructuring pipeline thesis: when deal-making accelerates at the top of the market, restructuring mandates accelerate for the overleveraged names that cannot participate. Both sides of the capital markets cycle are active simultaneously.

Previews: The Super Week Ahead

FICO Q3 FY2026 | Tuesday July 29, After Close

Consensus EPS of $12.02 implies 40% year-over-year growth. Revenue consensus is $679 million, up 26.6%. FICO has beaten EPS estimates in each of the past four quarters with an average surprise of 8.78%. The structural setup entering the print is unusually constructive: mortgage origination revenues surged 127% year-over-year in Q2, FICO Score 10T adoption has expanded to 55 lenders representing $495 billion in annual serviceable mortgage originations, and Platform ARR grew 49% in Q2. The thesis monitoring variable is Scores segment pricing: if mortgage origination volumes continue recovering and Score 10T pricing commands a premium over standard FICO pricing, the upside to consensus is meaningful. This is the single most important earnings report for the portfolio this quarter.

Microsoft Q4 FY2027 | Wednesday July 30, After Close

The comparison that matters is Azure growth versus Google Cloud’s 82%. Azure has been running at approximately 40% growth. If that rate accelerates in Q4, the MSFT position at 4% allocation is structurally confirmed. If it decelerates toward 35%, the question is whether Copilot monetization is growing fast enough to compensate. Copilot enterprise seat count and any commentary on MAI model commercial adoption are the secondary variables.

Meta Q2 2026 | Tuesday July 29, After Close

Meta is not a portfolio holding, but its AI monetization result is the most direct test of whether AI capital expenditure is translating into advertising revenue acceleration. Meta’s AI recommendation and feed algorithms have been its primary revenue driver for three years. If AI spending is producing returns in the advertising layer, Meta shows it first. A strong Meta print validates the AI monetization thesis broadly. A miss raises the question of whether AI is producing revenue anywhere outside of cloud infrastructure.

Amazon Q2 2026 | Thursday July 31, After Close

AWS growth rate versus Azure and Google Cloud is the read. If AWS is decelerating while Google Cloud accelerates at 82%, the cloud market share dynamics are shifting in ways that benefit Alphabet and potentially MSFT. Amazon’s advertising and retail margin expansion are secondary reads on consumer health.

Module 3: Smart Money Monitor

This module tracks the capital allocation decisions of institutional investors, sovereign wealth funds, and large fund complexes whose moves reveal structural positioning rather than short-term sentiment.

1. Dimon vs Solomon: The Same Economy, Two Interpretations

The most analytically interesting signal from the July 14 bank earnings was not the numbers. It was the divergence between the two most important CEOs on Wall Street in their characterization of what those numbers mean.

Jamie Dimon, whose JPMorgan posted the highest quarterly profit in US banking history, spent his earnings call warning investors not to extrapolate. He cited geopolitical uncertainty, persistent inflation, and the risk that the current credit environment reflects a cyclical peak rather than a new structural baseline. His message, delivered at a moment of maximum financial strength, was one of caution.

David Solomon, whose Goldman Sachs delivered its best quarter ever, said the firm’s deal backlog is at its highest level in five years and described the environment for investment banking as structurally supportive rather than cyclically elevated. His message was that the pipeline ahead is more visible than at any point in recent years.

Both are correct about different time horizons. Dimon is describing the macro risks that could interrupt the cycle. Solomon is describing the revenue that is already contracted and waiting to close. For the portfolio, Solomon’s signal is more immediately relevant: a five-year high M&A backlog means the restructuring and advisory mandates that HLI feeds on are entering the pipeline at the top of the deal cycle, not the bottom.

Signal: When the most risk-averse banker in the world is cautious and the most deal-oriented banker in the world is optimistic, the correct portfolio posture is to own the businesses that profit regardless of which of them is right. HLI generates fees when deals close and when they fail.

2. The Chip Stock Rout Was Not a Rotation Out of AI. It Was a Rotation Within It.

The semiconductor selloff on July 16, triggered by reports of Chinese domestically produced chip-making equipment breakthroughs, wiped more than $500 billion in market capitalization from global chip stocks in a single session. The surface read was that institutional capital was exiting AI exposure. The granular data tells a more precise story.

The stocks that fell hardest were the most exposed to US equipment export dominance: Applied Materials, Lam Research, ASML, and KLA. These businesses derive their moats partly from the regulatory architecture that restricts Chinese access to advanced semiconductor manufacturing tools. A credible Chinese domestic alternative, even if years away from full commercial deployment, compresses the duration of that moat. Institutional investors who hold AMAT and KLAC on a 10-year thesis updated their probability estimates on a single piece of news.

Meanwhile, memory producers like Micron and SK Hynix sold off in sympathy rather than for fundamental reasons. The HBM supply constraint that drove SK Hynix’s 198% revenue growth and 72% operating margin in Q1 2026 does not reverse because of Chinese equipment news. Chinese fabs cannot produce HBM-capable chips regardless of their equipment progress. The selloff in memory stocks on July 16 was indiscriminate liquidation, not informed repositioning.

Signal: Institutional capital is not abandoning AI infrastructure. It is repricing the specific moats within it. Equipment export dominance is a regulatory moat with a finite duration. Memory supply constraints are a physical moat with a longer duration. The market treated them identically on July 16. It will not treat them identically over the next two years.

3. ETF Flow Data: The Great Sector Rotation in Numbers

The two-week period produced the clearest ETF flow divergence of the year, and the numbers tell a more precise story than the index moves.

Financial sector ETFs, led by XLF, absorbed significant inflows following the historic bank earnings. Healthcare ETFs including XLV continued receiving steady institutional allocations, extending the trend that began in June. Utilities ETFs, XLU, saw net inflows despite the 10-year yield rising to 4.706%, which is analytically significant: rate-sensitive utilities attracting capital even as rates rise is a signal that institutional buyers are pricing long-term contracted cash flows rather than near-term rate sensitivity. Consumer Staples ETFs including XLP also continued receiving inflows.

The outflows concentrated in two areas. XLK, the technology sector ETF, saw material redemptions following the chip stock rout, with the Nasdaq underperforming the Dow by more than 6 percentage points over the two-week period.

The Dow outperforming the Nasdaq by more than 6 percentage points over two weeks is institutional capital using the most efficient mechanism available to express a single view: earnings that are already recognized matter more than earnings that require AI monetization to materialize.

Signal: The rotation is out of multiples that require future monetization and into multiples that are supported by current cash flows. At 4.7% risk-free rates, the future is expensive. The present is cheap.

4. The Earnings Quality Divide: What Happens When You Remove the Accounting

The S&P 500 reporting a net profit margin of 15.7%, a 15-year record, in the same two weeks that IBM fell 23% and software budgets are being crowded out is not a contradiction. It is a description of how concentrated earnings power has become.

FactSet data shows that 86% of S&P 500 companies that have reported Q2 results beat EPS estimates. But remove Alphabet’s $99 billion mark-to-market gain on its Anthropic and SpaceX equity stakes, and the blended net profit margin reverts to approximately 14.4%. Remove the bank earnings spike driven by SpaceX IPO fees that will not repeat in Q3, and the underlying earnings trajectory is more modest still.

Institutional investors who run factor models are aware of this. The current quarter’s earnings quality score, which weights recurring operating earnings over one-time items, is materially lower than the headline margin implies. Sophisticated allocators are already adjusting their positions accordingly, which explains why the S&P 500 has declined 2.1% over the period even as the earnings headline looks strong.

Signal: The record profit margin is real on paper and misleading in practice. Institutional capital that runs earnings quality screens is already separating the signal from the noise. The businesses in the portfolio that generate their earnings from contracts, pricing power, and non-discretionary demand do not need accounting adjustments to defend their margins.

Module 4: Portfolio Desk

This Week’s Focus

No position changes were made during the two-week period. The decision to hold rather than act reflects a deliberate judgment: the most concentrated week of earnings in 2026 arrives immediately after this period closes, with FICO, Meta, Microsoft, Amazon, and the FOMC decision all landing between July 29 and July 31. Acting before that information arrives would be managing noise rather than thesis. The portfolio enters super week fully positioned and analytically prepared.

Core Compounders

Berkshire Hathaway (BRK.B) | Hold | ~6.5%
The Alphabet Q2 results validated Abel’s $10 billion private placement with unusual precision. Google Cloud growing 82% year-over-year with a $514 billion contracted backlog is exactly the infrastructure positioning Abel paid for. The validation arrived faster than most private placement theses resolve. The $397 billion cash reserve continues to function as deployment optionality in an environment where WTI hitting $102 intraday and the 10-year reaching 4.713% are both creating the conditions Berkshire was built for. Monitoring variable: any additional capital deployment announcement.

AutoZone (AZO) | Hold | ~4.5%
The ORLY-GPC acquisition bid remains pending regulatory review with no new developments in the two-week period. The structural read is unchanged: if ORLY acquires NAPA, the US automotive aftermarket consolidates into two dominant national platforms, raising barriers for any new entrant and deepening the route economics both ORLY and AZO have built over decades. AZO’s Q4 FY2026 results are expected August 26 and will be the first major data point following the competitive landscape reports. Monitoring variable: commercial same-store sales growth in Q4 and any management commentary on the ORLY-GPC developments.

Kinsale Capital (KNSL) | Hold | ~5.0%
No material developments. The E&S insurance market continues to benefit from standard carriers pulling back from complex risks. Monitoring variable: Q2 combined ratio and gross written premium trajectory.

Copart (CPRT) | Hold | ~4.0%

Read more

$1,000 invested in Ferrari ten years ago is worth $9,266 today. Invested five years ago, it is worth $1,971. Invested three years ago, it is worth just $1,153, and that last number is the one that should make you pause, because $1,000 in the S&P 500 over that same three year stretch grew to $1,681, more than four times Ferrari’s gain. Three data points, three different verdicts, and all three are true at once.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

For a closer look at how these principles translate into real positions, explore the 👉 Sample Portfolio — a demonstration of thesis-driven allocation and disciplined compounding.

Background

Ferrari has only existed as a standalone public company since October 2015, when it listed on the New York Stock Exchange at $52 a share as part of its spin off from Fiat Chrysler, engineered under the late Sergio Marchionne. At the time, the market genuinely did not know how to price it. Sell side analysts modeled it as a small, high fixed cost auto manufacturer with a few thousand units of annual volume, the kind of business that lives or dies on capacity utilization. Nobody was modeling it as a luxury house with return on equity approaching 45 percent and pricing power that resembles a watchmaker more than a carmaker.

That mispricing at the starting line, the market defaulting to the wrong comparison set, is a large part of why the ten year number looks the way it does. It took the market years to reprice Ferrari from car company with a nice logo to luxury goods company that happens to make cars, and a meaningful chunk of the stock’s return came from that re-rating rather than from unit growth, because unit growth was never really the plan. The company itself traces this identity back to its founder. Enzo Ferrari’s own operating philosophy, still cited by management today, was essentially that the company should always sell fewer cars than the market wants rather than one more than the market wants. That single instinct, encoded into the business seventy years before the IPO, is the root of everything that follows.

The Data

The figures below are pulled directly from total return charts, not approximated from annual price data, so these numbers are exact.

10-year performance

Ten years, entry around July 2016: Ferrari returned 826.64 percent cumulatively, turning $1,000 into $9,266.40, a CAGR of 24.94 percent. The S&P 500 over the identical window returned 250.29 percent, turning $1,000 into $3,502.90, a CAGR of 13.36 percent. Ferrari delivered more than three times the index’s cumulative return, and it was not a smooth three times. The chart shows Ferrari’s cumulative return climbing in fits and starts through 2017 and 2018, stalling through 2019 and the 2020 shock, then accelerating hard from 2023 onward into a peak above 1,250 percent in early 2025, before falling back to roughly 650 percent later that year and recovering to its current 826.64 percent. The S&P 500 line across the same decade looks almost flat by comparison, a slow, steady climb with none of the drama. That contrast is the entire risk profile of this stock in one image: the compounding is real, but so is the stomach it requires.

5-year performance

Five years, entry around October 2021: Ferrari returned 97.08 percent, turning $1,000 into $1,970.80, a CAGR of 14.53 percent. The S&P 500 returned 73.68 percent, turning $1,000 into $1,736.80, a CAGR of 11.67 percent. Ferrari led for nearly the entire five years. Its cumulative return spiked to almost 190 percent in early 2025, more than double where it sits today, before a sharp drawdown pulled it back toward 80 to 90 percent later that year. The S&P 500’s path was different in character: a real drawdown into negative territory through most of 2022, then a comparatively steady climb from 2023 onward with none of Ferrari’s amplitude. By July 2026 Ferrari is still ahead on this five year window, but only after giving back most of an enormous lead.

3-year performance

Three years, entry around October 2023: Ferrari returned just 15.27 percent, turning $1,000 into $1,152.70, a CAGR of 4.85 percent. The S&P 500 returned 68.07 percent, turning $1,000 into $1,680.70, a CAGR of 18.90 percent. This is the chart that should temper anyone tempted to buy Ferrari on a straight extrapolation of the ten year number. For the first year the two lines move together, then Ferrari pulls decisively ahead, spiking toward 70 percent by early 2025 while the S&P 500 is still sitting around 35 to 40 percent. Then the lines invert. Ferrari falls off a cliff in a sharp air pocket around April 2025, alongside a broader market selloff that briefly hit the S&P 500 too, and spends most of the following year oscillating between roughly negative five and twenty percent while the index, after its own dip in the same window, resumes a steady climb all the way to 68.07 percent. By July 2026, Ferrari has recovered to only 15.27 percent cumulative, a small fraction of where it stood at its own early 2025 peak, while the index has more than quadrupled Ferrari’s return over the identical three years.

These figures are price only, excluding Ferrari’s modest dividend and the index’s, and the entry prices are approximations built from annual and monthly public data rather than exact daily closes. The point is the shape of the story, not the second decimal place.

Share

The reversal is not random. Ferrari dramatically outperformed over ten years and modestly outperformed over five, then lagged over three, mostly because 2025 was a genuinely difficult year for the stock, down roughly 12 percent, coming right after an extraordinary 2023 to 2024 run that had already pulled forward a lot of the re-rating. Meanwhile the S&P 500 kept climbing through an AI driven bull market that took it from around 4,500 to past 7,500 in the same window. Most retrospectives quietly skip the window that does not cooperate. This one should not be skipped, because a three year stretch landing on the wrong side of a broad market rally is exactly the kind of result that tests whether an investor understands the business or was simply riding a chart.

The Luxury Pricing Power Effect

This is the part of the Ferrari story that a normal automotive analysis misses entirely, because Ferrari does not compete on the axis every other car company competes on. A conventional manufacturer is a volume business. It needs to sell more units to grow revenue, it is a price taker in a competitive category, and its margins get squeezed by input costs it cannot control. Ferrari inverted that model on purpose, and the inversion is the entire investment case.

Start with units. Ferrari delivered 13,640 units in 2025, a figure deliberately held substantially flat to ease a significant model changeover into 2026. For scale, a mass manufacturer builds that many vehicles in well under a day. Yet revenue for 2025 rose 7 percent to over 7.1 billion euros, with EBIT margin up 120 basis points to 29.5 percent. Revenue went up while units stayed flat. That single fact is the whole thesis in one sentence.

The mechanism is average selling price and personalization. Ferrari’s average selling price rose to around $500,000 by 2025, up from about $350,000 in 2022, and personalization revenue now exceeds 20 percent of core sales, up from 19 percent the prior year, and is a major driver of the positive mix and price effect. Personalization carries close to pure margin because it adds almost no incremental manufacturing complexity relative to what it adds in price. A buyer who might spend $280,000 on a base model routinely ends up spending meaningfully more once personalization is added, and that spending sits near the highest margin dollar in the entire business. This is a fundamentally different economic engine from selling more cars. It is selling the same number of cars for meaningfully more money, repeatedly, without triggering the price resistance that would show up instantly in almost any other consumer category.

None of this works without genuine scarcity, and Ferrari manufactures that scarcity as deliberately as it manufactures the cars. Management describes demand as very solid and disciplined across every market, reflecting an exclusivity model in which the order book already extends toward the end of 2027, meaning a buyer ordering today waits well over a year for delivery. Certain limited series cars are only offered to existing owners who already hold a qualifying model, turning ownership itself into a credential required to buy more. Ferrari has roughly 180,000 owners globally, a client base small enough that the company can allocate limited annual production toward its most profitable markets and its most loyal collectors rather than simply selling to whoever shows up first with a check. This is the definition of pricing power: the ability to raise price, richen mix, and grow revenue while deliberately refusing to grow volume, because volume is the thing that would destroy the very scarcity the pricing power depends on.

The Luxury Peer Group, Not the Auto Peer Group

The clearest way to see whether this strategy is real or just a story is to compare Ferrari against two different groups: hard luxury houses that run the same playbook, and a carmaker that tried to copy Ferrari’s playbook and failed to pull it off.

Against luxury peers, Ferrari looks credible but not yet best in class. Hermes, the closest true comparable in spirit, runs operating margins around 40 to 42 percent against Ferrari’s roughly 29.5 percent, and sustains that gap by refusing to expand production of its Birkin and Kelly bags even as demand has grown for decades. LVMH, as a sprawling multi brand conglomerate, runs meaningfully lower margins near 20 to 22 percent, closer to Ferrari’s own level, because a portfolio of 75 brands inevitably includes some that are aspirational rather than truly scarce. Ferrari sits between these two extremes, closer to Hermes in spirit and playbook, closer to LVMH in current margin, which is roughly what you would expect from a luxury business that is still only a decade into being priced as one.

Against its actual industry, the comparison is more dramatic and more useful. Porsche IPO’d in 2022 attempting a similar positioning, a storied performance brand asking to be valued above ordinary automakers. It has not worked. Porsche shares trade roughly 45 percent below their IPO price, weighed down by the fact that Porsche still ships close to 280,000 cars a year at an average price near $151,000, a real business but fundamentally a volume business exposed to the same EV transition costs, Chinese competition, and cyclical demand that pressure every other mass manufacturer. Ferrari, by contrast, has returned more than five times its IPO price over the same broader period, specifically because it never let volume creep up to capture more of that demand. The two companies sell some of the most desirable cars on earth. The market has treated them completely differently, and the difference is not brand admiration, which both companies have in abundance. It is whether the scarcity was actually defended when the temptation to grow volume was right there.

The Next Test

None of this is guaranteed to continue, and the honest version of this thesis has to say so. Ferrari’s next real test is the Luce, its first fully electric model, priced near 500,000 euros and launching into a customer base that has spent decades associating the brand with the sound and feel of a combustion engine. If buyers accept that the scarcity and craftsmanship travel intact into an electric car, the pricing power holds and the flywheel keeps turning. If they do not, and the Luce ends up discounted or under ordered relative to Ferrari’s combustion models, that would be the first real evidence that the moat is narrower than the last decade of numbers suggests. There is a second risk sitting underneath the first one. Ferrari’s valuation, at somewhere between 35 and 50 times trailing earnings depending on the measure, already assumes something close to flawless execution. A high multiple stock does not fall gently when it disappoints. A modest earnings miss on a richly priced compounder can produce a share price decline far larger than the miss itself, because the market re-rates the multiple at the same time it re-rates the earnings. That asymmetry is exactly what showed up in 2025, and it is worth remembering the next time the ten year chart looks too easy.

Share

The Moat - Pricing Power

Put together, the moat is not the engineering, although the engineering is real. It is the combination of an artificially constrained supply, a roughly two year order book, an ownership base treated as a qualification system rather than a customer list, and a personalization program that converts brand desire directly into margin. That combination produces numbers that look nothing like an automaker’s. Return on equity is approaching 45 percent. EBITDA margin has run close to 38 to 39 percent even through recent foreign exchange and cost headwinds. The stock trades at a multiple that only makes sense if the market is pricing Ferrari as a luxury compounder rather than a cyclical industrial name, which, given everything above, is exactly what it is, even if it has not yet earned the very top of the luxury tier where Hermes sits.

The Lesson

The lesson here is not simply that Ferrari won. It is that a business built on manufactured scarcity and luxury pricing power compounds unevenly, and judging it on the wrong window can make a genuinely exceptional business look mediocre, while judging it on the right one reveals something a normal industrial analysis would never find. Over ten years, scarcity value and pricing power did exactly what they are supposed to do and dramatically outran the index. Over three years, a strong but cyclical stock ran into a historic broad market rally and came up short. Both are true simultaneously, and neither invalidates the other.

The discipline is in knowing which horizon actually tests the thesis, and in respecting that a rich valuation cuts both ways. A luxury compounder whose entire model depends on refusing to sell more units is not meant to be judged quarter to quarter or even year to year, because short windows are dominated by whatever the broader market is doing. It is meant to be judged on whether the order book is still extending, whether average selling price and personalization are still rising, and whether the scarcity is still being defended rather than quietly relaxed for growth, the exact discipline that separated Ferrari’s stock from Porsche’s since each went public. When those things hold, a difficult three year window is noise. The moment any one of them breaks, and the Luce launch will be the clearest test yet, the entire model breaks with it, because the pricing power was never really about the car. It was always about what the company chose not to sell.

Pricing power is the most underrated quality in long term investing, more important in many cases than growth itself. A business that can raise prices without losing volume is a business that can pass inflation through, defend its margin in a downturn, and compound earnings even in years when unit sales barely move, which is precisely what Ferrari did in 2025. Growth without pricing power is fragile, because it depends on the world staying favorable. Pricing power without growth, as Ferrari has shown, can still compound at a very high rate, because the company controls its own economics rather than renting them from the market. This is why a name with real pricing power deserves a wider berth on valuation than a name that only has growth, and why losing it, whether through overexpansion, brand dilution, or a failed product transition, is a far more serious event than a single disappointing year.

If you want to see how we test a thesis like this across a full cycle rather than a single window, the Thesis Library breaks down the frameworks we use, and the Valuation Cases column walks through how we would underwrite a name like this across bear, base, and bull scenarios.

For informational and educational purposes only, not investment advice. Data is drawn from public sources and believed accurate as of publication but not guaranteed. Past performance does not indicate future results. Conduct your own due diligence or consult a licensed advisor before investing.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

Share

The S&P 500 closed the week 0.45% below its all-time high of 7,620.90, set on June 2. The Dow Jones crossed 53,000 on Monday July 6 for a new record close. The Magnificent Seven gained 4.9% for the week, leading the Nasdaq to a year-to-date gain of 29%. VIX sits at 15. By every surface measure, the market has absorbed the dislocations of the past two weeks and resumed its trajectory.

The 10-year Treasury yield tells a different story. It rose for eight of the last nine trading sessions, closing at 4.568%, the highest since May 22. The 30-year pushed back above 5%. These moves did not happen in a vacuum. They happened in the same week that the Federal Reserve released minutes from its June 16 to 17 meeting revealing a committee split directly down the middle, and in the same week that the Iran ceasefire formally unraveled.

The FOMC minutes released on Wednesday July 8 carried unusual analytical weight precisely because of what was absent from them. For the first time in modern Fed history, a sitting chair declined to submit a dot-plot projection. Kevin Warsh’s abstention means the June dot plot, which showed 9 of 18 participating officials projecting at least one rate hike before year-end 2026 against 8 projecting no change and 1 projecting a cut, does not include the chair’s own view. The committee’s inflation forecasts tell the story. Core PCE was revised to 3.3% for 2026, up from 2.7% in the March projection, the largest single-meeting upward revision in recent memory. Fed staff explicitly cited two structural drivers: the ongoing Middle East energy disruption and the inflationary effects of $750 billion in annual AI capital expenditure driving semiconductor and infrastructure costs higher across the economy. The latter is analytically important because it is not an energy shock. It is a demand shock. Energy shocks resolve when supply normalizes. Demand-driven inflation from AI infrastructure investment does not resolve when oil prices fall.

The oil market demonstrated this week that normalization of energy prices is not a straight line. Trump declared the ceasefire with Iran “over” on Wednesday July 8 while speaking at the NATO summit in Ankara, Turkey, following US strikes on Iranian positions in retaliation for attacks on three commercial ships in the Strait of Hormuz. His exact words: “For me, I think it’s over. It’s just a waste of time dealing with them.” Brent crude surged more than 5% on the day to nearly $78 per barrel. WTI climbed to $73 to $74 before settling back to $71.41 by Friday. The week’s net oil move of 3.1% upward partially reversed the 13% decline of the prior two weeks. The IEA issued its monthly oil market report this week projecting that global oil demand will decline by 1 million barrels per day year-over-year in 2026, the first annual demand decline since the 2020 pandemic, driven primarily by the Hormuz disruption reducing Persian Gulf consumption and shipping activity. That figure is structurally deflationary for global oil markets over the medium term. It does not help in the near term when the same strait is being attacked again.

The week’s most analytically significant market event was not macro. It was a listing. SK Hynix debuted on the Nasdaq on Friday July 10 under the temporary ticker SKHYV, priced at $149 per American depositary receipt, raising $26.5 billion in the largest US listing by a foreign company in history, surpassing Alibaba’s 2014 record. The stock closed at $168.01, up 13% from its offering price. The significance is not the first-day performance, which is a function of demand management in the book-build rather than fundamental assessment. It is what the SK Hynix listing confirms about the structural hierarchy of AI investment. SK Hynix controls approximately 60% of the global high-bandwidth memory market. HBM is the memory architecture that allows AI accelerators to move data fast enough to run the models that hyperscalers are deploying. There is currently no substitute, and there is not enough of it. The company’s Q1 2026 revenue surged 198% year-over-year to the equivalent of $35.5 billion, with an operating margin of 72%. Its chairman told CNBC on Friday that demand is “enormous, exponentially” growing and that AI agents and physical robots require far more memory than prior computing generations. The $26.5 billion raised goes directly into the Yongin cluster fab expansion and a $4 billion advanced packaging plant in Indiana. This is not a story about a company going public. It is a story about the physical infrastructure layer of AI receiving its largest single capital injection since the Hormuz war began and AI capex estimates started compounding.

The central contradiction of the week is this: the equity market sits 0.45% from its all-time high. The bond market has moved yields higher for eight consecutive sessions. The Fed is split 9 to 9 on whether to hike rates. Iran struck ships in the Strait of Hormuz. The IEA says global oil demand is falling for the first time in six years. SK Hynix raised $26.5 billion in one afternoon. Gold fell 1.4%. The VIX is at 15. These data points are not consistent with each other. Markets that are priced for continued AI-driven growth and markets priced for a hawkish Fed with a re-escalating energy war occupy different valuation regimes. The equity market is currently pricing the former. The bond market is pricing something closer to the latter. One of them will need to revise.

This Week’s Signal: The Fed is split 9 to 9. Iran is back. SK Hynix raised $26 billion. The VIX is at 15. The market is either right about all of them simultaneously, or it is right about none of them.

Module 2: Earnings Decoder

This module decodes the week’s most important financial disclosures.

SK Hynix (SKHY) | Nasdaq IPO Debut | July 10, 2026

The financial data embedded in SK Hynix’s IPO prospectus is the most important set of numbers published this week. Q1 2026 revenue reached the equivalent of $35.5 billion, up 198% year-over-year. Operating margin was 72%, a figure that puts virtually every software company’s margin profile to shame. The company controls approximately 60% of the global high-bandwidth memory market and raised $26.5 billion in the largest US listing by a foreign company in history, closing up 13% on its first day.

The market’s 13% first-day gain is analytically interesting not because of its magnitude but because of its composition. This was an institutional deal, not a retail story. The buyers who passed on SpaceX because of its $4 billion quarterly AI spending loss came to SK Hynix because the business model is the mirror image: SK Hynix is the beneficiary of AI spending, not its casualty. Every dollar a hyperscaler commits to Nvidia GPU clusters requires HBM that only three companies on earth can produce, and SK Hynix produces 60% of it.

Thesis verdict: The IPO confirms what AMAT and KLAC’s 60% gains in June already implied. The physical memory layer of AI is structurally scarce, structurally profitable, and now directly accessible to US investors.

Big Bank Q2 Preview | JPM, GS, BAC, WFC, C | Tuesday July 14, Before Open

Five of the largest US financial institutions report simultaneously next Tuesday morning, alongside the June CPI print. The collision of bank earnings and the most important inflation data of the summer in a single morning session is the most concentrated macro event of the quarter.

The consensus picture is constructive. S&P 500 Q2 earnings growth is expected at 23.9% on 11.7% higher revenues. For banks specifically, loan growth is expected to accelerate from an already strong Q1, investment banking revenues are broadly stable, and trading revenues should benefit from the elevated volatility in energy and fixed income markets that the Iran re-escalation and hawkish FOMC minutes generated this week. JPMorgan enters as the benchmark, having delivered Q1 net income of $16.5 billion and EPS of $5.94, beating estimates by 8.2%.

The number that matters most is not EPS. It is net interest margin and credit quality. With the Fed holding at 3.5 to 3.75% and rate hike probability at 40 to 55%, NIM expansion from the higher-for-longer environment remains the central bull case for bank profitability. Credit quality will tell whether the 42% year-over-year surge in Chapter 11 filings is beginning to transmit into consumer and commercial loan books. If it is, the restructuring thesis for HLI strengthens further. If it is not, it means stress is concentrated in private credit rather than bank balance sheets, which is a different and more contained risk.

Thesis verdict: Bank earnings next week are a macro read, not a sector call. Strong NIM with stable credit quality confirms the higher-for-longer environment is benefiting traditional lenders. Deteriorating credit quality would be the first hard evidence that rate pressure is transmitting into real economic defaults, which matters for every position in the portfolio.

Module 3: Smart Money Monitor

This module tracks the capital allocation decisions of institutional investors, sovereign wealth funds, and large fund complexes whose moves reveal structural positioning rather than short-term sentiment.

1. ETF Flow Data | ICI Week Ended July 1 | Capital Is Rotating, Not Retreating

The Investment Company Institute’s weekly flow report released July 8 shows a market in structural transition rather than panic. ETFs attracted $32.3 billion in net new issuance while long-term mutual funds bled $28.87 billion in outflows. The headline equity number looks bearish: equity funds saw net outflows of $10.22 billion, with domestic equity down $5.83 billion and world equity down $4.39 billion. But the composition of those flows tells a different story.

The biggest individual outflows were from iShares Gold Trust at $900 million, the SPDR 1-3 Month T-Bill ETF at $800 million, the WisdomTree Quality Dividend Growth Fund at $850 million, and the Pacer Global Cash Cows Dividend ETF at $789 million, which lost 19.4% of its entire asset base in a single week. Capital is not moving to safety. It is moving out of yield-adjacent and inflation-hedge strategies that worked in prior rate environments. The dominant inflows went to Vanguard’s low-cost broad market equity ETFs, which claimed seven of the ten highest inflow spots. Institutional capital is repositioning toward diversified equity beta and away from specific factor bets that the higher-for-longer rate environment has invalidated.

Signal: The rotation is not defensive. It is a repricing of which equity strategies remain rational at 4.5% risk-free rates. Gold and dividend income strategies are losing the argument. Broad market equity with quality earnings is winning it.

2. BlackRock H1 2026 ETF Report | Record $1 Trillion in Flows, But Gold Bleeding Out

BlackRock’s H1 2026 ETF market update, published this week, shows the first half of 2026 set an all-time record for global ETF inflows at $1 trillion, nearly double the previous record and an 86% jump year-over-year. Equity ETFs attracted $680 billion, more than double the prior year figure, as markets recovered from the March geopolitical selloff and corporate earnings consistently beat consensus estimates.

The more analytically important data point is commodity ETPs. Gold and commodity ETPs swung from $79.5 billion in inflows in all of 2025 to $12.4 billion in outflows in just the first half of 2026. In June alone, commodity ETPs bled $6.8 billion, the second-largest monthly redemption in at least two years. Gold ETPs accounted for most of the June withdrawals. BlackRock attributes this directly to higher interest rates and a stronger dollar compressing the appeal of non-yielding real assets.

The investor survey embedded in the report shows institutional clients planning to reduce cash by 11.6% and core bonds by 8.5%, while adding to alternatives at 13.4%, private markets at 6.0%, and US equities at 4.5%. This is not a flight to safety. It is a deliberate rotation into assets that generate contracted cash flows in an environment where the risk-free rate is doing real work for the first time in fifteen years.

Signal: The world’s largest asset manager is observing institutional clients systematically moving out of gold and bonds and into alternatives and private markets. The portfolio’s exits from EL and LIN in favor of FICO and HLI reflect the same analytical logic: contracted, non-discretionary cash flows beat inflation-hedge narratives when rates are structurally elevated.

3. Berkshire Hathaway | Weschler’s Quiet Winners and Abel’s Deployment Signal

While Greg Abel’s reconstitution of Berkshire’s public equity portfolio has received most of the analytical attention this year, a Motley Fool report published this week highlights a different story. Ted Weschler, the remaining stock picker at Berkshire, holds positions in DaVita and SiriusXM that are up 102% and 50% year-to-date respectively. Both are structurally non-discretionary businesses: DaVita provides kidney dialysis services where demand is biologically mandated, SiriusXM has a contracted subscription base that compounds through vehicle sales without requiring consumer discretionary spending to hold up.

The more significant development is Abel’s $10 billion Alphabet private placement from two weeks ago. That single transaction represents the largest capital deployment from Berkshire’s $397 billion cash reserve since Abel took over in January. It signals two things simultaneously. First, Abel is willing to deploy at scale when the risk-reward is compelling. Second, the structure of the investment, a private placement into Alphabet’s cloud infrastructure expansion rather than open-market buying, reflects a preference for negotiated entry with structural terms rather than passive market participation. The next Berkshire deployment, whenever it arrives, is likely to follow the same pattern.

Signal: Berkshire’s portfolio is bifurcating. Weschler is quietly compounding in non-discretionary service businesses. Abel is deploying selectively into AI infrastructure at private placement terms. Both strategies reflect the same underlying judgment: pay for structural certainty, not narrative potential.

Module 4: Portfolio Desk

This Week’s Focus

Three positions were added this week. CPRT moves from Under Review to an active add, MSFT reaches its highest allocation since initiation, and HLI reflects continued conviction in the restructuring cycle thesis. All three additions share the same underlying analytical logic: non-discretionary demand, pricing power that does not depend on macro tailwinds, and a five-year horizon that makes the current rate environment a pricing opportunity rather than a structural threat.

Defensive Anchors

Rollins (ROL) | Hold | ~4.0% No material developments. Non-discretionary pest control demand compresses through no macro environment. Monitoring variable: organic revenue growth and route density in Q2.

Visa (V) | Hold | ~3.0% No material developments. Iran re-escalation this week is net negative for cross-border travel volumes in the near term, which is the primary monitoring variable. Thesis intact at trimmed allocation. Monitoring variable: cross-border transaction volumes as Hormuz situation evolves.

WM, TMUS, COST | Hold No material developments across the three. All hold against stated theses with monitoring variables unchanged.

Core Compounders

Copart (CPRT) | Action: Add | Allocation: ~4.0% | Category: Core Compounder

CPRT is added from 3% to 4%, and the Under Review designation is removed. The decision requires direct explanation because the Q4 FY2026 report has not yet been released, with the next earnings date confirmed as September 9, 2026.

Read more

Every physical system trends toward disorder without continuous energy input. The US power grid, with an average transformer age above 40 years, is no exception. AI compute demand is accelerating the degradation of infrastructure that was already operating near its design limits. Hubbell sells the components that push back against that entropy. The thesis is not a capex cycle bet. It is a physics argument.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

The One-Line Thesis

Every megawatt of electricity that flows from a generation source to an AI data center, a manufacturing plant, or a residential neighborhood passes through connectors, switches, transformers, and distribution equipment. Hubbell makes a significant portion of those components. As the United States undertakes what amounts to a multi-decade reconstruction of its electrical grid, driven simultaneously by AI data center demand, industrial reshoring, electrification, and aging infrastructure replacement, the companies that supply the physical components of that reconstruction are structurally positioned for compounding demand that does not depend on any single end market, any single customer, or any single policy decision to materialize.

Hubbell is not a flashy business. It does not appear in AI headlines. It does not run on proprietary algorithms. It manufactures connectors, wiring devices, switchgear, and grid automation equipment, products that have existed for generations and will be required for generations more. That lack of narrative glamour is precisely what makes it worth understanding carefully.

Share

Business Overview

Founded in 1888 and headquartered in Shelton, Connecticut, Hubbell operates through two segments that together generated $5.8 billion in trailing twelve-month revenue as of mid-2026.

Utility Solutions ($3.7 billion in 2025 net sales) provides the physical infrastructure of the electrical grid: transmission connectors, substation components, distribution equipment, and grid automation technology. This segment serves the investor-owned utilities, electric cooperatives, and municipal utilities that operate the transmission and distribution network across the United States and internationally. Within Utility Solutions, Grid Infrastructure, which covers the physical transmission and distribution hardware, has been the consistent growth engine. Grid Automation, which includes smart meters and advanced metering infrastructure, has been a drag due to weak AMI project activity, a headwind that management expects to normalize over the medium term.

Electrical Solutions ($2.1 billion in 2025 net sales) provides the wiring devices, electrical boxes, fittings, and power management equipment used in commercial, industrial, and data center construction. This segment has been the more recent growth accelerant, driven by data center buildout and light industrial markets. The May 2026 announcement of the $3 billion acquisition of NSI Industries, a manufacturer of electrical fittings, connectors, and wire management products serving industrial, data center, and network infrastructure markets, will substantially expand this segment’s addressable market when the transaction closes in late 2026.

The Second Law of Thermodynamics

The Second Law of Thermodynamics states that entropy in any closed system increases over time. Without continuous energy input to maintain order, physical systems degrade. The United States electrical grid is not exempt from this law. The average American power transformer is over 40 years old, designed for a load profile that did not include data centers consuming hundreds of megawatts per campus. Every additional gigawatt of AI compute demand placed on a grid built for a different era accelerates the degradation rate of infrastructure that was already operating near the boundary of designed capacity.

Hubbell does not sell a discretionary product. It sells the physical components required to push back against entropy in a grid that is structurally losing the battle against its own obsolescence. Switchgear, transformers, distribution components, and utility connectors are not optional upgrades. They are the continuous energy input that the Second Law demands any ordered system must receive to avoid collapse. The investment thesis is not that Hubbell will benefit from a capex cycle. It is that entropy is non-negotiable, and the businesses positioned to supply the components that slow it are structurally non-discretionary regardless of interest rate environment, AI monetization timelines, or geopolitical noise.

The Structural Demand Case

Three forces are simultaneously driving demand for Hubbell’s products, and they are operating on different timelines with different drivers, which means the demand profile is not dependent on any one of them sustaining.

Force One: AI Data Centers

Every AI data center requires massive amounts of electrical infrastructure behind the meter, the wiring, connectors, switchgear, and power management equipment that routes electricity from the utility connection to the servers. In Q1 2026, organic net sales in the Electrical Solutions segment grew 10.6%, led by continued strength in the datacenter vertical. This is not a temporary project cycle. The infrastructure buildout required to support AI compute is a multi-year capital investment program with visible order pipelines and committed capital from hyperscalers.

The connection to Hubbell is direct and structural. Data centers do not build their own connectors. They buy them from companies like Hubbell that have the manufacturing scale, certification approvals, and distribution relationships to supply at the volume and lead time that hyperscale construction requires. The NSI acquisition, announced in May 2026, was explicitly framed around this opportunity. CEO Gerben Bakker stated: “As electrification megatrends drive attractive growth across the electrical industry over the next several years, NSI offers highly complementary products and industry-leading brands to our HES portfolio across strategic growth verticals including light industrial, datacenter and network infrastructure applications.”

Force Two: Grid Modernization and Transmission Investment

The United States electricity grid is old. Much of the transmission infrastructure was built in the 1950s and 1960s with a designed lifespan of 40 to 50 years. The combination of aging assets, increasing renewable generation requiring new transmission corridors, and the unprecedented load growth driven by AI data centers and industrial reshoring is forcing utilities to invest at a pace not seen in decades.

Core utility T&D markets remain strong, with load growth driving strength in transmission and substation markets, and aging infrastructure resiliency investment driving strength in distribution markets. Management has specifically called out a multi-year opportunity in high voltage 765 kV transmission, with early project wins and capacity expansion investments already underway. This is not a near-term cyclical uplift. Transmission projects require years of permitting, engineering, and construction, which means the order pipeline Hubbell is winning today converts to revenue over a three to five year horizon.

The DMC Power acquisition, completed in late 2025 for $825 million, directly expanded Hubbell’s exposure to this trend. DMC Power provides connectors and tooling for utility substation and transmission markets, with an “attractive growth and margin profile aligned to megatrends in load growth, datacenter interconnection and aging infrastructure.”

Force Three: Industrial Reshoring and Electrification

The broader industrial electrification trend, driven by manufacturing reshoring, electric vehicle charging infrastructure, and the general substitution of electricity for fossil fuels in industrial processes, is generating sustained demand for the electrical infrastructure products that Hubbell supplies. This demand is less headline-visible than AI data centers but more broadly distributed across Hubbell’s customer base, and it is less dependent on any single sector sustaining its investment pace.

Financial Performance

The financial record is consistent and improving.

Q1 2026 net sales reached $1.517 billion, up 11% year-over-year, with 8.2% organic growth. Adjusted operating profit was $301 million, up 18% year-over-year. Adjusted operating margin expanded 110 basis points to 19.8%. Adjusted EPS was $3.93, a 16% increase versus the prior year period.

The full year 2025 delivered 4% sales growth with 80 basis points of adjusted operating margin expansion and 10% growth in adjusted diluted earnings per share, compounding on a strong multi-year base. Management raised full-year 2026 guidance following Q1 results, with total sales growth now expected at 8% to 11%, organic sales growth at 6% to 9%, and adjusted EPS of $19.30 to $19.85 at the midpoint.

Return on equity stands at 24.54% and return on invested capital at 17.51%. These are not exceptional numbers by the standards of the highest-quality compounders in this portfolio, but they are solid and improving, with ROIC expanding as the business mix shifts toward the higher-margin grid infrastructure and data center verticals. Operating cash flow reached $1.0 billion on a trailing twelve-month basis, supporting both the acquisition program and the capital return commitment.

Hubbell has maintained dividend payments for 56 consecutive years and raised its dividend for 18 consecutive years. The quarterly dividend stands at $1.42 per share. The company repurchased $168 million in shares during Q1 2026 at an average price below $500 per share, a meaningful capital allocation signal given where management believed the stock to be trading relative to intrinsic value.

Competitive Position and Moat Assessment

Hubbell’s competitive advantage is not the kind that generates breathless analyst commentary. It is the kind that generates consistent earnings through market cycles without requiring brilliant strategic pivots or disruptive technology to defend it.

Certification and specification barriers. Electrical components used in utility and data center applications must meet rigorous safety, performance, and interoperability standards. Achieving and maintaining certifications with Underwriters Laboratories, major utilities, and data center operators requires years of qualification testing, manufacturing consistency, and technical support relationships. A new entrant with identical manufacturing capabilities cannot substitute for Hubbell’s products in a utility substation or a hyperscale data center on short notice, because the specification and qualification process takes years. This creates a structural stickiness that is not captured in traditional moat frameworks but is visible in Hubbell’s consistent pricing power and low customer churn.

Distribution network depth. Hubbell’s products reach customers through a network of electrical distributors, contractors, and direct utility relationships built over more than a century. This network is not easy to replicate. An electrical contractor who has specified Hubbell wiring devices on hundreds of projects is not going to switch to an unknown brand for a new data center build. The distribution relationship is itself a moat.

Scale advantages in manufacturing. Hubbell’s manufacturing footprint, developed over decades, provides cost advantages in raw material procurement, production efficiency, and the ability to respond to large orders with sufficient capacity and lead time. Smaller competitors cannot match both the product breadth and the supply reliability that large data center and utility customers require.

M&A capability as a compounding mechanism. The DMC Power and NSI acquisitions demonstrate a capital allocation discipline that is consistent with Hubbell’s long-term approach: acquiring businesses with complementary product lines, higher margins than the existing portfolio, and exposure to the same structural growth drivers. The company’s strategy is guided by four pillars: Serve the Customer, Grow the Enterprise, Operate with Discipline, and Develop People. That framework has produced a track record of acquisitions that have been accretive to both revenue and margin.

Key Risks and Open Questions

Grid Automation weakness. The Grid Automation segment, which includes Aclara smart meters, has been a consistent drag on results. Grid automation, including Aclara, experienced year-over-year organic sales declines, with management noting utilities have deprioritized AMI investments, resulting in “fewer projects come through.” T

his is a genuine headwind that has offset some of the strength in Grid Infrastructure. The open question is whether the AMI market recovers on a timeline consistent with management’s medium-term expectations or remains weak for longer than anticipated. The bull case requires Grid Automation to eventually contribute rather than subtract from overall growth.

Input cost and tariff pressure. Management described the net impact from Section 232 and related tariff changes as “about neutral” for 2026. This is a measured response, but it reflects ongoing management attention and pricing discipline required to offset material cost inflation. In a scenario where tariff pressure intensifies or raw material costs accelerate, margin expansion could be more modest than the current trajectory implies.

NSI integration execution. The $3 billion NSI acquisition is the largest in Hubbell’s recent history. Integrating a business of this scale while maintaining organic momentum in both existing segments requires management bandwidth and operational discipline. The acquisition was announced as accretive in 2026, but integration timelines have a way of extending.

Valuation

Hubbell trades at approximately 25 times forward earnings, against a 10-year historical average of 23.7 times. The premium to its own history is modest and structurally defensible given that the demand environment today, AI data center load growth layered on top of an already-undersupplied grid replacement cycle, did not exist during most of the period that produced that historical average. The more instructive comparison is to peers: Hubbell’s forward P/E of approximately 29 times compares favorably to its electrical industry peer average of 40.8 times, suggesting the market has not yet fully repriced the business for the secular tailwind it is positioned to capture.

Return on equity stands at 25.82% and return on invested capital at 16.80%, both consistent with a business that earns above its cost of capital across the cycle rather than in spite of it. Management guided 2026 adjusted EPS of $19.15 to $19.85, with free cash flow conversion of 90% or greater on adjusted net income, confirming that earnings quality is not distorted by working capital timing or accounting adjustments.

The primary valuation risk is the NSI Industries acquisition. The current debt-to-equity ratio is 0.72, elevated relative to Hubbell’s pre-acquisition balance sheet but manageable given the free cash flow generation capacity of the combined business. The acquisition was funded through bridge financing, and leverage normalization through organic free cash flow is the critical path variable for the next two to three quarters. If NSI integration proceeds without material disruption, the leverage concern resolves and the valuation case strengthens as the acquired revenue begins flowing through to earnings.

The next earnings release is scheduled for July 28, 2026, before market open, and will be the first print to include a meaningful NSI contribution. That report is the most important near-term data point for the valuation thesis: the combination of organic growth trajectory, NSI revenue contribution, and any leverage ratio update will determine whether the 25 times forward multiple is an entry point or a ceiling.

The Investment Thesis

Hubbell sits at the intersection of two of the most certain capital expenditure cycles of the next decade: the rebuilding of the American electrical grid and the physical infrastructure required to power AI compute. Neither of these cycles requires predicting a specific technology winner, a specific policy outcome, or a specific consumer behavior. They require only that electricity continues to be generated, transmitted, and consumed, and that the infrastructure carrying it continues to age and be replaced.

The business has compounded earnings at double digits through multiple market cycles, maintained pricing discipline through cost inflation, and allocated capital to acquisitions that expand its exposure to the highest-growth segments of an already growing market. ROIC of 17.5% and ROE of 24.5% are not exceptional, but they are consistent and directionally improving as the mix shifts.

The bear case requires believing that the utility capital expenditure cycle reverses materially, that data center buildout decelerates significantly, and that Hubbell loses pricing power to competitors who can replicate its certification infrastructure and distribution network. None of these conditions describe the environment that current data supports.

The bull case requires only that the structural demand drivers continue on the trajectory that management’s order book, utility capex guidance, and hyperscaler capital commitment data already indicate, and that the NSI and DMC acquisitions perform in line with the historical track record of Hubbell’s M&A program.

For investors willing to hold through the noise of quarterly earnings reactions and short-term macro uncertainty, Hubbell represents exposure to a secular infrastructure buildout with a business model that has been generating consistent returns for more than a century.

The grid does not care about AI sentiment. It needs to carry the electrons. Hubbell makes the components that carry them.

🔴 Bear Case — $356 | -27% downside

The bear case requires two things to go wrong simultaneously: NSI integration absorbs more capital than planned, and utility capital budgets moderate as sustained high rates slow state regulatory approvals. Organic growth reverts to 4 to 5% annually, NSI contributes minimal net free cash flow through 2027, and the overall FCF trajectory grows at approximately 4% per year, reaching $1.06 billion by 2030. Applying a terminal EV/FCF multiple of 20 times and subtracting $2.4 billion in net debt, the implied price target is $356.

🟡 Base Case — $541 | +10% upside

NSI integrates on management’s stated timeline with EPS accretion visible by Q4 2026. Utility Solutions demand continues at 6 to 9% organic growth, consistent with the 11% reported in Q1 2026. Free cash flow grows at approximately 9% annually, reaching $1.35 billion by 2030, in line with HUBB’s 2023 to 2025 compound annual FCF growth of 11%. A terminal EV/FCF multiple of 23 times implies a price target of $541. The July 28 Q2 report is the first data point that confirms or challenges this scenario.

🟢 Bull Case — $778 | +59% upside

Data center load growth has been consistently underestimated by utility planners for three consecutive years. If hyperscaler AI capex sustains at current levels and translates into physical grid connection requests at the rate the current pipeline implies, Hubbell’s addressable market expands beyond what historical utility capex models can capture. NSI synergies exceed initial estimates as the combined portfolio captures data center construction contracts that neither business could address independently. Free cash flow grows at 13% annually, reaching $1.61 billion by 2030. A terminal EV/FCF multiple of 27 times implies a price target of $778.

Probability-weighted target: $554, implying approximately 13% upside from the current price of $490. The market is currently pricing in execution slightly below the base case. The asymmetry is real, but requires the grid supercycle thesis to sustain and NSI to deliver above guidance. Entry at current levels makes sense for investors with conviction in the structural demand thesis. A pullback toward the $420 to $440 range would restore meaningful margin of safety in the base case.

Key Monitoring Variables

• Grid Infrastructure organic growth rate, specifically transmission and substation market strength

• Grid Automation recovery timeline, particularly AMI project activity

• NSI acquisition integration progress and accretion to adjusted EPS in 2026 and 2027

• Adjusted operating margin trajectory, targeting continued 100 to 150 basis point annual expansion

• Full-year 2026 adjusted EPS delivery against the $19.30 to $19.85 guidance range

• Data center vertical revenue growth within Electrical Solutions

Conclusion

Hubbell is not a bet on the AI cycle. It is a bet on the infrastructure that the AI cycle cannot function without. The US power grid is aging faster than it is being replaced, and the load being placed on it is growing faster than anyone projected two years ago. That combination does not produce a growth story. It produces a structural necessity.

At $490, the stock is priced at approximately base case execution. The probability-weighted target of $554 implies modest upside without requiring the bull case to materialize. The July 28 earnings report, the first to include a meaningful NSI contribution, is the near-term event that tightens the range between base and bull.

The thesis requires patience, not prediction. Entropy does not negotiate timelines.

Disclaimer: This report is provided for informational and educational purposes only. It does not constitute financial or investment advice. The views expressed reflect analysis at the time of writing and may change without notice. Investors should conduct their own due diligence or consult a licensed advisor before making any investment decisions.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

Share

The S&P 500 declined 0.6% over the past two weeks. That number is the single most misleading data point in this entire update.

Beneath a headline index that barely moved, the past fourteen days produced one of the most structurally significant sector rotations of the year. The equal-weight S&P 500, which assigns identical weight to Apple and to a mid-size industrial company, hit an all-time high on July 1. The market-cap-weighted index did not. That divergence tells a precise structural story: the megacap AI and technology positions that have dominated index performance since 2023 are distributing capital to the rest of the market at a rate that the headline index obscures entirely. This is not a market in decline. It is a market in reconstitution.

The energy picture over the past two weeks is the most analytically important development since the Iran peace deal was announced. WTI crude fell from $80.14 to $69.23, a decline of 13.6% in fourteen trading sessions. The mechanism was not a reversal of the peace agreement. It was the physical restoration of Persian Gulf export capacity. By the end of June, actual crude oil export volumes through the Strait of Hormuz had recovered to approximately 75% of pre-war levels, regardless of the residual geopolitical noise. A drone incident in late June, attributed by the White House to Iranian actors, briefly threatened to reverse the ceasefire framework and did prompt retaliatory US strikes on Iranian radar and drone installations. The market’s response to that escalation was a momentary spike followed by continued decline. The physical supply restoration was already in progress, and traders understood that the political drama could not undo the tanker traffic that was already moving. WTI at $69.23 represents a $36 collapse from its peak above $105. The inflation arithmetic changes dramatically at $69 oil. The Fed’s primary justification for the June dot plot, nine of eighteen officials supporting a hike, was written when oil was above $87. It was not written for $69 oil.

The June non-farm payrolls report, released the morning of July 3, confirmed what the oil price had already been implying. The US economy added 57,000 jobs in June, the fewest in four months and well below the 110,000 consensus forecast. Leisure and hospitality shed 61,000 positions despite World Cup tourism providing a seasonal tailwind. The unemployment rate fell to 4.2%, but for the wrong reason: workers leaving the labor force rather than finding employment. Wage growth eased to 3.5% year-over-year. CME FedWatch immediately repriced the September hike probability from 67% to 50%. Gold climbed to $4,170, its highest level since June 23, on its first weekly gain after four consecutive weeks of declines. The 10-year Treasury yield held at 4.484%, reflecting a market that is genuinely uncertain whether to price a hike or a pause. Warsh, speaking at the ECB’s Sintra Forum in Portugal this week in his first international appearance as Fed Chair, noted that inflation expectations are moderating while refusing to provide forward guidance. The phrase “moderating inflation expectations” was not in his vocabulary when he delivered his June FOMC press conference. It was not accidental that he used it now.

The sector rotation of the past two weeks is the most important analytical story the portfolio needs to process. Communication Services fell 6.2% over the two-week period. Technology fell 5.4%. These are not small moves for sectors that collectively represent over 35% of the S&P 500 by weight. The specific casualties tell the structural story. Super Micro Computer fell 36.4% over the month of June. Oracle fell 35.1%. Accenture, the portfolio’s only direct exposure to the consulting layer of AI adoption, crashed 18% in a single session on June 18, its largest single-day decline since going public, after reporting a 13% sequential decline in new bookings and cutting its full-year revenue guidance to 3 to 4% growth. Management cited three simultaneous headwinds: Middle East disruption causing approximately $100 million in lost revenue, major enterprise clients deferring managed-service contracts into fiscal 2027 due to the Fed’s higher-for-longer posture, and continued weakness in US federal consulting. Separately, both Apple and Microsoft announced price increases on consumer software and hardware products, with management attributing the cost inflation directly to AI data center demand driving DRAM and NAND flash memory prices to more than double their autumn 2025 levels, with further doubling expected by end of 2027. The AI infrastructure buildout is now visibly eating into the margin structures of the technology companies that are supposed to be its primary beneficiaries.

The rotation’s destination is as analytically important as its origin. Applied Materials gained 60.65% in June on the strength of its high-bandwidth memory processing and advanced packaging tool orders. KLA Corporation gained 57%. The Philadelphia Semiconductor Index recorded its best single-quarter performance since 1994. Healthcare gained 7.9% as a sector, led by Moderna’s 48% surge on a regulatory approval. Utilities gained 3.9%. Real Estate gained 4.0%. Consumer Staples gained 1.5%. The common characteristic of every winning sector in this rotation is the same: contracted, non-discretionary cash flow that does not require the AI buildout to monetize on schedule. Power utilities that sell electricity to data centers under long-term agreements do not care whether the inference workloads on those servers generate positive returns for the hyperscaler. Healthcare companies do not care about GPU availability. The market is not abandoning technology. It is applying a more rigorous filter to which technology businesses have monetization that is already contracted versus which businesses are spending on the assumption that monetization will follow.

Kevin Warsh spent this week at the ECB’s annual forum in Sintra, Portugal. He refused to provide forward guidance when pressed by moderators and fellow central bank governors. He acknowledged easing inflation expectations while reaffirming the Fed’s commitment to price stability. Contrast that language with his June 18 press conference, where the dot plot showed nine hike votes and no one was discussing easing. The oil price is doing the work that six months of rate policy could not. The question for the second half of 2026 is whether Warsh will take credit for the disinflation or acknowledge that the supply shock that drove it has resolved on its own terms.

The technology and communication services sectors lost 5.4% and 6.2% respectively over the two weeks. The casualties were not random. Super Micro Computer fell 36.4% in June. Oracle fell 35.1%. The common thread is not sector exposure. It is the distance between current price and demonstrable free cash flow. Apple and Microsoft raising consumer product prices due to AI-driven memory costs doubling since last autumn introduced a variable the market had not fully priced: the infrastructure buildout is now visibly compressing the margins of the companies that are supposed to benefit from it most.

The capital that left did not exit the market. Applied Materials gained 60.65% in June. KLA Corporation gained 57%. Healthcare gained 7.9%. Utilities gained 3.9%. Every winning sector shares one characteristic the AI software complex currently lacks: revenue that does not require a specific macro outcome to arrive on schedule.

The equal-weight S&P 500 hit an all-time high on July 1 while the cap-weighted index sat 0.6% lower. That divergence is the most honest summary of what is happening. The market is not in retreat. It is becoming more specific about what it is willing to pay for.

2. Portfolio Desk

This Week’s Focus

Four portfolio actions this week reflect a single coherent judgment: the sector rotation underway in global markets is not a temporary dislocation but a structural reassessment of where durable cash flows sit in the current macro environment. Visa was trimmed as cross-border travel volumes, while recovering, no longer justify the prior allocation weight relative to higher-conviction positions. FICO and HLI were both added, reflecting increased conviction that businesses with non-discretionary, structurally embedded revenue are being repriced upward as capital exits narrative-dependent software and AI application names. Estée Lauder was fully exited after the Iran peace deal timeline extended further and the Asia travel retail recovery thesis lost its near-term catalyst.

Defensive Anchors

Visa (V) | Action: Trim | Allocation: ~3.0% | Category: Defensive Anchor

V is trimmed from 4.5% to 3.0%, the first reduction since the position was established. The trim reflects a reallocation of capital toward positions with higher current conviction rather than any deterioration in the Visa thesis. The structural case remains intact: Visa’s network processes settlement for an increasing share of global commerce, the Agentic Ready expansion positions it as the payment layer for AI-initiated transactions, and cross-border volumes are recovering as Hormuz reopens and Gulf travel normalizes. The sizing reduction acknowledges that at 3%, Visa remains a meaningful Defensive Anchor position while freeing capital for FICO and HLI, where the near-term thesis development is more active. Monitoring variable: cross-border volume trajectory in Q2 as the Hormuz reopening completes and Gulf aviation traffic normalizes.

Waste Management (WM) | Hold | ~3.25% No material developments. Pricing power and route density compound quietly. Monitoring variable: landfill pricing in the Q2 earnings cycle.

Rollins (ROL) | Hold | ~4.0% No material developments. The macro environment of the past two weeks, with real consumer incomes under pressure and the equal-weight S&P rotating into non-discretionary services, validates the thesis structurally. Pest control demand does not respond to equity market volatility. Monitoring variable: organic revenue growth and route density improvement.

T-Mobile (TMUS) | Hold | ~2.25% No material developments. Postpaid account economics continue to compound. Monitoring variable: Q2 subscriber metrics and ARPA trajectory.

Linde (LIN) | Action: Exit | Final Allocation: 0%

LIN is fully exited. The thesis on industrial gas pricing power and long-term contract insulation remains structurally intact. The exit is a capital allocation decision: proceeds are redeployed into positions where the thesis is actively developing rather than compounding quietly at a premium multiple with limited room for further expansion.

Costco (COST) | Hold | ~1.0% No material developments. The membership renewal flywheel is intact. Monitoring variable: Q4 FY2026 results and Executive Member penetration trajectory.

Core Compounders

Berkshire Hathaway (BRK.B) | Hold | ~6.5% The $10 billion Alphabet private placement from two weeks ago remains the most significant capital deployment signal in the portfolio’s macro context. WTI at $69.23 and NFP at 57,000 both suggest the macro environment that justified holding $397 billion in cash is shifting. The deployment question is live. Monitoring variable: any additional capital commitment announcement in Q3.

AutoZone (AZO) | Hold | ~4.5% No material developments over the two-week period. Q4 FY2026 results are expected in late June or early July and will be the next material data point. The aging fleet thesis is structurally intact. Monitoring variable: commercial same-store sales growth in the upcoming quarterly print.

Kinsale Capital (KNSL) | Hold | ~5.0% No material developments. E&S insurance market conditions remain favorable. Monitoring variable: Q2 combined ratio and gross written premium trajectory.

Copart (CPRT) | Hold | ~3.0% | Under Review No new earnings over the two-week period. The Under Review status is maintained. Q4 FY2026 results are the final data point before a formal position decision. Monitoring variable: US insurance unit volume and total loss frequency in the upcoming print.

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

Share

The S&P 500 ended the week up 0.6%. VIX fell from 21.51 to 17.68. Oil dropped 6.3%. By every surface measure, the week resolved cleanly after last week’s chip-driven dislocation. Look at the inputs that produced that resolution and the picture is considerably more complicated.

The week’s most consequential event arrived on Thursday June 11 in a compressed window of several hours. The White House announced it was formally canceling planned strikes on Iranian oil infrastructure and confirmed that a sovereign framework agreement with Tehran, covering a ceasefire framework and the reopening of the Strait of Hormuz, had received approval from Iran’s highest leadership level. The agreement is in final signing stages. Within the same afternoon, SpaceX priced its IPO at $135 per share, raising $75 billion and implying a valuation of $1.75 trillion. Adobe reported Q2 results of $6.62 billion in revenue, a record, with AI-first ARR tripling year-over-year and crossing $500 million. WTI immediately dropped 2.6% on the Iran news. The S&P 500 surged 1.8% on the day. The Nasdaq jumped 2.5%. Three separate market-moving events landed simultaneously, and the net effect was the largest single-day rally in nearly two months.

But the underlying macro data released earlier in the week told a different story. May CPI, published Wednesday, came in at 4.2% year-over-year, the highest print since 2023 and materially above April’s 3.8%. Core CPI rose to 2.9%. The inflation that has been described as transitory, energy-driven, and war-related for three months is now running at the highest annual rate of this entire cycle. On the same day, the 10-year Treasury yield was, in the words of CNBC, “steady even after data showing highest inflation since 2023.” That steadiness is the tell. A bond market that does not sell off on a 4.2% CPI print is a bond market that is already pricing in a resolution to the energy shock, not dismissing the inflation. The bond market saw Thursday’s Iran deal coming before it was announced.

PPI released the following day confirmed what the supply chain has been building toward for months. May PPI rose 6.5% year-over-year, the largest increase since November 2022, with core PPI reaching 5.1%. These numbers represent the wholesale cost of production before retail inflation. They are a leading indicator. The pipeline from PPI to CPI runs with a three to six month lag. If PPI is at 6.5% and CPI is at 4.2%, the direction of travel for consumer prices is not ambiguous. It is upward, unless something disrupts the supply chain that is driving it. The Iran deal is that disruption. Goldman Sachs, reading the same data, pushed its first expected Fed rate cut to June 2027 and warned that a hike before year-end carries a higher probability than the market currently prices. Kevin Warsh delivers his inaugural address as Federal Reserve Chairman on Monday morning. The market will be listening for every word.

The chip sector spent the week absorbing the aftershock of Broadcom’s earnings. When Broadcom capped its full-year AI semiconductor revenue guidance at $56 billion, a figure far below the $100 billion the most aggressive sell-side models had been running, the repricing was immediate and brutal. AMD, Intel, and their peers shed a combined $1.3 trillion in market capitalization over the week. The divergence within technology became sharper. CoreWeave, confirmed this week as a new member of the Nasdaq 100, gained 5% while the chip index fell. The capital rotation is precise: money is leaving businesses that pitch AI story while moving toward businesses that have signed contracts for AI compute delivery. The market is applying an increasingly granular filter to what counts as AI exposure.

Adobe reported Q2 FY2026 results on Thursday evening that contained two separate narratives in a single press release. The operational story was unambiguous: record revenue of $6.62 billion, up 11% year-over-year, AI-first ARR tripling to above $500 million, non-GAAP EPS of $5.96 growing 18%, and a full-year guidance raise. The AI monetization that the thesis required is manifesting in contracted ARR, not pipeline. The corporate governance story was more complicated: CFO Dan Durn announced his departure effective June 15, four days after the earnings release, with no advance notice to investors. Adobe enters the second half of its fiscal year without a permanent CEO, Narayen having announced his intention to step down once a successor is named in March, and now without a permanent CFO. The stock fell 5.5% in after-hours trading. The business is performing. The institution around it is in an unusual state of transition that the market is discounting in the multiple.

SpaceX opened for trading on Friday June 12 and its market capitalization immediately crossed $2.1 trillion, surpassing the combined valuations of ExxonMobil, Bank of America, and Coca-Cola. The company carries a first-quarter net loss of $4.28 billion and allocates approximately $2.5 billion per quarter to AI infrastructure spending. The valuation arithmetic requires either a fundamental belief that Starlink’s global satellite broadband business will generate revenue at a scale that has never existed in the telecommunications industry, or a belief that the SpaceX brand, the xAI integration, and the Musk factor together justify a premium that financial modeling alone cannot produce. The IPO’s 30% retail allocation, unprecedented at this scale, suggests the company’s bankers understood that the latter is doing significant work in this equation.

This Week’s Signal: CPI hit 4.2%. Oil fell 6%. The market rallied. The Iran deal is doing more for inflation than the Federal Reserve has been able to do in six months of holding rates. Kevin Warsh speaks Monday.

Share

Module 2: Portfolio Desk

This Week’s Focus

Defensive Anchors

V, TMUS, WM, ROL, LIN, COST | Hold No material developments across the defensive tier. The Iran peace deal framework, if signed, is the most direct macro catalyst for V through cross-border volume normalization and for COST through supply chain cost reduction. ROL’s non-discretionary demand thesis was reinforced by the CPI 4.2% print: in a sustained high-inflation environment, the services that cannot be deferred are precisely the ones that compound. All six positions hold against stated theses with monitoring variables unchanged.

Core Compounders

Berkshire Hathaway (BRK.B) | Hold | ~6.5% The $10 billion Alphabet private placement disclosed last week and the Iran deal progress this week represent two separate signals pointing toward the same conclusion: the macro environment that has justified holding $397 billion in cash is beginning to shift. Energy-driven inflation may have peaked. If Hormuz reopens, the Fed’s primary justification for maintaining restrictive policy weakens. Berkshire’s deployment timing becomes a live analytical question rather than a theoretical one. Monitoring variable: any additional capital deployment announcement in the weeks following the Iran deal signing.

AutoZone (AZO) | Hold | ~4.5% No material developments. CPI at 4.2% and a 7%+ auto loan rate are structural tailwinds for the aftermarket parts thesis: new vehicle ownership remains prohibitively expensive for the median American household, extending the fleet age that drives AZO’s commercial segment. Monitoring variable: Q4 FY2026 results expected late June.

Kinsale Capital (KNSL) | Hold | ~5.0% No material developments. Thesis intact. Monitoring variable: Q2 combined ratio.

Copart (CPRT) | Hold | ~3.0% | Under Review No new earnings this week. Q4 FY2026 results are the final data point before a formal position decision. The May Q3 beat on pricing confirmed the moat is intact. Volume recovery is the remaining open variable. Monitoring variable: Q4 total loss frequency and unit volume.

Blue Chip

Microsoft (MSFT) | Action: Add | Allocation: ~3.5% | Category: Blue Chip

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

Share

Nine consecutive weeks of gains. Then a single jobs report.

Friday’s non-farm payrolls print showed 172,000 new jobs added in May, comfortably above consensus expectations. In most macroeconomic environments, a strong labor market is unambiguously good news.

In June 2026, with Core PCE running at 3.3% year-over-year, headline PCE at 3.8%, the Fed on hold with a 25% probability of a hike already priced by derivatives markets, and Kevin Warsh in his first weeks as Chair, a stronger-than-expected jobs number is the worst possible data point available. A labor market that refuses to cool gives the Fed no justification to ease and every justification to tighten. The 2-year Treasury yield spiked 12 basis points on Friday to a 15-month high. The 10-year held above 4.5%. Within hours, the Nasdaq had fallen 4.18%, its worst single-day decline since April 2025. Chip stocks lost over one trillion dollars in market capitalization in a single session. NVIDIA fell 6%. The semiconductor index fell 6.7%. VIX surged 40% to 21.51, erasing three weeks of calm in one afternoon. The nine-week winning streak that had carried the S&P 500 to all-time highs ended not because of a geopolitical shock or a policy surprise, but because the economy was too strong for the market’s own positioning.

The structural irony of this week is not the Friday selloff. It is the Dell Q1 earnings report that landed earlier in the week and immediately contextualized what the selloff was actually about. Dell reported total revenue of $43.8 billion, up 88% year-over-year. AI server revenue for the quarter was $1.61 billion, up 757%. That number is not a typo. The hyperscalers are converting their AI capital expenditure commitments into Dell’s income statement at a rate that has no historical analog. The full chain is visible and precise: NVIDIA ships Blackwell chips to Dell, Dell integrates them into server racks, the hyperscalers buy the racks with the capital they committed in their own earnings calls, and the revenue appears in Dell’s Q1 print as a 757% year-over-year increase. Every dollar of Dell’s AI server revenue represents a capital expenditure that must eventually generate a return for the hyperscaler that bought it. If the return does not materialize at the rate the equity market has priced in, the repricing does not happen gradually. It happens in an afternoon.

Against that backdrop, two macro signals arrived this week that the equity market’s nine-week rally had been ignoring. The World Economic Forum released its Chief Economists Outlook, in which 90% of surveyed chief economists projected that the global economy will weaken materially over the next twelve months, with Middle East energy disruption cited as the primary risk vector for European stagflation.

Simultaneously, Canada reported that its Q1 2026 real GDP contracted 0.1% on an annualized basis, following a revised Q4 2025 decline of 1%, the technical definition of a recession. This is Canada’s first recession since the 2020 pandemic, driven by a collapse in business investment and government spending under the weight of sustained high interest rates and US tariff pressure. For the portfolio, the Canada recession is directly relevant to Constellation Software, whose legal domicile is Canadian but whose revenue, reinvestment, and ROIC profile are globally dispersed. The analytical point is not that CSU is impaired by Canadian GDP. It is that CSU’s resilience in the face of a Canadian recession is itself a thesis confirmation: a business that compounds at 25% ROIC while its home country enters recession is demonstrating exactly the kind of structural independence the investment framework requires.

SpaceX launched its global IPO roadshow on Thursday, with a target price of $135 per share, a planned raise of $75 billion, and an implied valuation of $1.75 to $2.0 trillion. That valuation represents more than a doubling from the company’s December 2025 private tender offer at approximately $800 billion, achieved through the absorption of xAI and the accelerating commercialization of Starlink. In Q1 2026, SpaceX reported a net loss of $4.28 billion, with approximately $2.5 billion of that allocated to AI infrastructure spending per quarter. Elon Musk announced that 30% of the new shares would be directly allocated to retail investors, an unusual structure that makes the SpaceX IPO the most significant test of retail investor appetite for story stocks since the 2021 SPAC cycle. Pricing is scheduled for Thursday June 11, with the first day of trading on June 12. Until then, the SpaceX IPO functions as a liquidity vacuum, pulling institutional attention and capital allocation decisions toward a single event that will dominate next week’s market structure.

Gold’s 4.9% weekly decline is the most analytically significant data point in the snapshot. In every prior week of elevated market stress this year, gold held or rose as institutional capital sought real asset protection. This week it fell alongside equities, alongside bonds, and alongside crypto. That behavior suggests the Friday selloff was driven not by a flight to safety but by margin calls and forced deleveraging, conditions in which every liquid asset is sold to meet cash requirements regardless of its defensive profile. When gold falls in a risk-off environment, the selling is not selective. It is mechanical.

This Week’s Signal: Nine weeks of gains were built on AI earnings and geopolitical hope. One jobs print reminded the market that the Fed exists. The discount rate is not a narrative. It is a number.

Share

Module 2: Portfolio Desk

This Week’s Focus

One deliberate action was taken this week. FICO was added to 2.5%, the largest single allocation in the Blue Chip tier alongside CSU, reflecting a specific analytical judgment about what the Friday selloff reveals rather than what it obscures. The chip sector lost over a trillion dollars in market capitalization in a single session. The businesses that earn non-discretionary institutional fees did not change their revenue trajectory by a single dollar because the Nasdaq fell 4.2%. FICO is the clearest expression of that logic in the portfolio. The Friday dislocation may provide a more attractive entry point than was available last week.

Defensive Anchors

Rollins (ROL) | Hold | ~4.0% No material developments. The thesis on ROL, structural non-discretionary demand insulated from consumer confidence cycles, was validated by the macro context this week. Real disposable incomes under pressure and a labor market that is simultaneously strong and inflationary do not reduce pest control demand. They make it more relevant as a portfolio anchor. Monitoring variable: organic revenue growth and M&A integration cadence in Q2.

Visa (V) | Hold | ~4.5% No material developments. Oil’s bounce from $87 to $90.54 on the week reflects renewed Hormuz uncertainty as the Iran negotiation timeline extended. Any cross-border volume normalization from a peace deal remains the primary upside catalyst for V. Thesis intact. Monitoring variable: cross-border volume in the Q2 disclosure.

WM, TMUS, LIN, COST | Hold No material developments across any of the four. All hold against stated theses with monitoring variables unchanged.

Core Compounders

Berkshire Hathaway (BRK.B) | Hold | ~6.5% The $397 billion cash reserve is no longer purely theoretical optionality. On June 1, Berkshire committed $10 billion to Alphabet through a private placement, structured as $5 billion in Class A shares at $351.81 per share and $5 billion in Class C shares at $348.20 per share. The transaction is part of Alphabet’s $80 billion equity capital raise, the largest in US corporate history, announced to fund AI compute infrastructure at a scale that Google Cloud’s own CEO described as supply-constrained: “Our Cloud revenue would have been higher if we were able to meet the demand.”

AutoZone (AZO) | Hold | ~4.5% No material developments. The NFP print of 172,000 new jobs is net neutral for AZO. A stronger labor market means employed consumers, which supports DIY spending. Elevated rates and high vehicle prices mean those consumers continue to repair rather than replace. The structural demand thesis remains intact. Monitoring variable: Q4 FY2026 results expected late June.

Kinsale Capital (KNSL) | Hold | ~5.0% No material developments. The E&S insurance market continues to benefit from standard carriers pulling back from complex risks. Monitoring variable: Q2 combined ratio and gross written premium trajectory.

Copart (CPRT) | Hold | ~3.0% | Under Review No new earnings this week. The

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

Opening

Consider a simple arithmetic problem.

From 2015 through 2024, the S&P 500 delivered a CAGR of approximately 13%. If you had spent 1% of your portfolio value each year on SPY Put Options as continuous downside protection, your net CAGR would have been reduced to roughly 12%. That sounds like a small price for peace of mind. But compounded over ten years, that 1% annual drag reduces a $100 portfolio’s final value from $340 to $310. You paid $30 for insurance against losses that, in eight of those ten years, never arrived.

In 2020 and 2022, those puts would have delivered real value. In 2015, 2016, 2017, 2018, 2019, 2021, 2023, and 2024, they were pure cost. The math of continuous hedging is almost always negative. Protection purchased when it is not needed is not prudent risk management. It is a recurring tax on compounding.

Share

This does not mean hedging is wrong. It means continuous hedging is almost always wrong. The discipline is knowing precisely when the cost of protection is justified by the probability and magnitude of the risk being hedged, and executing in the one window when protection is actually cheap: before anyone else thinks they need it.

The Two-Layer Monitoring Framework

Before the specific triggers, the tools. Every hedging decision should be preceded by reading two distinct layers of market information, each answering a different question.

Layer One answers: Is protection cheap or expensive right now?

This layer determines the timing of any hedging action. The cost of protection is not constant. It varies enormously based on current market conditions, and buying protection at the wrong moment in the volatility cycle can make even a correct hedging thesis financially unworkable.

Layer Two answers: Has the macro environment changed structurally?

This layer determines whether a hedging action is warranted at all. Short-term price volatility and genuine structural shifts in the economic environment require completely different responses. Most situations that feel like they require hedging are the former. Only a small subset qualify as the latter.

Layer One: Hedging Cost Indicators

VIX and the VIX/VXV Ratio

The VIX measures the implied volatility of S&P 500 options over the next 30 days. In normal market conditions, it oscillates between 15 and 20. Above 30 signals significant market stress. The historical peaks tell the full story: during the 2008 financial crisis, the VIX reached 89.53. During the COVID crash of March 2020, it peaked at 82.69. During the April 2025 tariff shock, it spiked to its highest level since 2020 before subsiding.

The implication for hedging is counterintuitive. When the VIX is at 80, Put Options are extraordinarily expensive. An at-the-money SPY put that costs $5 in a VIX-15 environment might cost $25 in a VIX-80 environment. The investor who waited until the crisis was visible to buy protection is paying five times the price for protection that is already too late to prevent most of the damage.

Share

The optimal hedging window is VIX below 15, when protection is cheapest and the probability of a volatility event feels lowest. This requires the psychological discipline to buy insurance when nothing appears to be wrong, which is precisely why most investors never do it.

The VIX/VXV ratio is a more precise signal than the VIX alone. VXV measures three-month implied volatility, so the ratio captures the relationship between near-term and medium-term fear. When the ratio falls below 0.9, the market is more complacent about near-term risk than medium-term risk, a classic signal of self-satisfaction. When the ratio rises above 1.0, near-term fear exceeds medium-term concern, typically a sign of acute but transient stress rather than structural deterioration.

Put/Call Ratio

The Put/Call Ratio measures the volume of put options traded relative to call options on any given day. A high ratio means investors are buying protection aggressively. A low ratio means they are not. Because this is a contrarian indicator, extreme readings are the meaningful ones.

The historical reference point: the PCR hit its all-time low of 0.39 in March 2000, at the exact peak of the dot-com bubble. The market had never been more complacent about downside risk, and it had never been more wrong. When the 10-day moving average of the PCR rises above 1.1, markets have historically been near a trough. When it falls below 0.8, markets have historically been near a peak.

As of early 2026, the PCR weighted ratio had declined to its lowest level since the top of the prior bull market in late 2021, with persistently low readings reflecting call-heavy options activity. Such readings have historically been associated with corrections ranging from modest, around 9% to 10%, to significantly larger moves. Low PCR is not a precise timing signal. It is a condition that indicates the market is poorly positioned for downside surprise.

The practical application: a low PCR in combination with a low VIX is the most favorable environment for purchasing protective options. You are buying when supply is high and demand is low, which is the only rational time to purchase insurance.

Fear and Greed Index

The CNN Fear and Greed Index synthesizes seven market indicators into a single 0 to 100 reading, covering stock price momentum, new highs versus new lows, breadth, put/call ratios, junk bond demand, safe haven demand, and market volatility. At its peak in 2021, the index reached 97, deep into Extreme Greed territory, before the S&P 500 experienced its 18% drawdown in 2022. During the COVID crash of March 2020, it fell to single digits before the fastest market recovery in history.

Three rules govern its use in a hedging framework.

First, only extreme readings carry signal value. Readings between 20 and 80 provide limited actionable information. The distribution of forward returns from neutral sentiment readings is barely distinguishable from random.

Second, the index is a confirming tool, not a standalone trigger. A Fear and Greed reading below 20 combined with VIX below 15 and PCR below 0.7 is a powerful convergent signal. Any one of these indicators alone is insufficient to drive a hedging decision.

Third, extreme greed does not predict the timing of a correction. Markets can remain in Extreme Greed for months or even years before a significant drawdown occurs. The index reading of 80 in June 2023 preceded continued market gains for another eighteen months. It is a measure of current vulnerability, not an imminent event prediction.

Layer Two: Structural Risk Indicators

Investment Grade Credit Spreads

Credit spreads are the single most important leading indicator available to equity investors, and the most underused. The mechanism is straightforward: institutional bond investors, who manage significantly more capital than equity investors and operate under strict risk mandates, react to deteriorating credit conditions before equity markets do. When they begin demanding higher yields to hold corporate debt, they are pricing in default risk that equity markets have not yet incorporated.

In 2007, investment grade credit spreads began widening in June, four months before the S&P 500 peaked in October. In late 2018, credit spreads widened through Q4 alongside the equity selloff, providing a concurrent confirmation rather than a leading signal. In 2022, the credit market was relatively slow to react because the primary driver was the rate shock, not credit deterioration, which is exactly when credit spreads are less predictive.

The practical rule: when IG credit spreads are widening while equity markets are still near highs, this is a structural warning that warrants a careful review of existing hedging levels. The divergence between calm equity markets and stressed credit markets has historically been a more reliable early warning signal than any volatility measure.

The inverse signal matters equally. When credit spreads compress to historically tight levels alongside equity market highs, this indicates that the fixed income market sees limited credit risk, which reduces the structural justification for broad equity hedges. Tight spreads alongside high equity valuations are not safe. They are a signal that both markets are simultaneously vulnerable to the same shock. But they reduce the asymmetry that makes hedging financially rational.

Shiller CAPE

The cyclically adjusted price-to-earnings ratio currently sits between 36 and 38, roughly double its historical average of 17 and approaching the levels seen at the peak of the dot-com bubble in 2000. This single fact does not tell investors to sell equities or buy puts. The CAPE has been “elevated” by historical standards for most of the past decade, and investors who exited equities on that basis missed extraordinary returns.

What CAPE actually measures is the forward margin of safety. At current levels, the historical distribution of 10-year forward returns suggests annualized returns in the 3% to 6% range rather than the 10% historical average. This has two specific implications for the hedging framework.

First, the Spare Tire tier of the portfolio should be at full allocation. In a low-return environment, the optionality value of deployable capital during a dislocation is proportionally higher. The opportunity cost of holding defensive assets is lower when overall equity return expectations are compressed.

Second, the tolerance for concentration risk should be lower. A portfolio that generates 20% annual returns through a concentrated, high-conviction position can absorb a 30% drawdown in that position without catastrophic consequences to the overall compounding trajectory. A portfolio in a 4% expected-return environment cannot absorb the same loss without permanent impairment.

CAPE does not trigger hedges. It calibrates how aggressively the rest of the framework should be applied.

Gold and the Gold/SPY Ratio

Gold’s behavior during market stress is not uniform, and understanding its different modes is essential to using it correctly as a risk signal.

When gold falls alongside equities, this is a liquidity crisis. Investors are selling everything to raise cash, and asset correlations collapse to 1. This is the most dangerous market environment and also the worst time to implement new hedging strategies. In March 2020, gold initially fell with equities before recovering as the Federal Reserve began its emergency interventions.

When gold rises while equities fall, this is traditional risk-off behavior. The bond market is functioning, credit spreads are manageable, and the decline is driven by risk aversion rather than forced selling. This environment is more structured and more amenable to systematic hedging responses.

The Gold/SPY ratio trend over a 12-month period is a useful structural signal. A rising trend suggests that real assets are being preferred over financial assets, typically associated with rising inflation expectations, currency debasement fears, or geopolitical risk premiums. A sustained ratio trend change in either direction is a prompt to review whether the portfolio’s Spare Tire allocation remains appropriate.

WTI Oil and PCE

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

Opening

Praveen Neppalli Naga, Uber’s Chief Technology Officer, thought he had budgeted correctly for AI in 2026. He had not. By April, the company had consumed its entire annual allocation for AI coding tools. “I went back to the drawing board,” he told The Information, “because the budget I thought I would need has already evaporated.”

Uber was not alone. Microsoft gave thousands of engineers, product managers, and designers access to Claude Code in late 2025, encouraging experimentation. Within months, the company began canceling licenses across its Experiences and Devices group, the team behind Windows, Microsoft 365, and Surface, with a June 30 cutoff. Bryan Catanzaro, Nvidia’s Vice President of Applied Deep Learning, stated publicly what his own company’s infrastructure team had concluded: “For my team, the cost of compute is far beyond the costs of the employees.”

Share

This is the irony at the center of the current AI moment. The companies most aggressively deploying AI are openly acknowledging that it has become, in many contexts, more expensive than the human labor it was supposed to replace. The narrative of inevitable cost reduction has collided with the reality of how token-based pricing actually works at scale.

None of this means AI is overrated as a technology. It means most organizations are making the same set of calculable errors when they evaluate the economics. Understanding those errors, and understanding precisely where AI genuinely does reduce cost and where it does not, is the difference between deploying AI as a competitive advantage and deploying it as an expensive liability.

Part One: What Is Inside a Token

Before evaluating the economics of AI versus human labor, it is worth tracing what actually happens when an enterprise deploys a large language model at scale. The cost of a token is not a price. It is the visible surface of a multi-layer cost structure that most organizations never fully account for.

Share

Layer One: The Chip

Every token generated by a frontier model requires GPU or specialized AI accelerator compute. The H100, Nvidia’s flagship inference chip, costs between $30,000 and $40,000 per unit to purchase outright. Cloud providers charge between $1.49 and $6.98 per hour depending on the provider and commitment level, a range that has fallen significantly: H100 cloud prices dropped from $7 to $8 per hour at peak to under $2 per hour for some configurations by mid-2026.

Token prices reflect this underlying chip economics. GPT-4-equivalent performance now costs approximately $0.40 per million tokens, down from $20 per million in late 2022, a 98% reduction. This number is cited frequently as evidence that AI is becoming dramatically cheaper. It is accurate as far as it goes. It stops too early.

Layer Two: Energy

The IEA’s 2025 Energy and AI report projected that global data center electricity consumption could double by 2030, with AI workloads accounting for the majority of incremental demand. Major markets including Northern Virginia, Silicon Valley, and Northern Europe have seen power approval timelines for new facilities stretch to 24 to 36 months, regardless of capital availability. Electricity is no longer an unconstrained resource for AI infrastructure.

The energy cost per token does not appear on any API invoice, but it is real. Current AI inference hardware consumes approximately 0.0001 to 0.002 watt-hours per output token. An H100 GPU draws 700 watts at peak. Data center Power Usage Effectiveness ratios of 1.2 to 1.5 mean that for every watt delivered to the GPU, an additional 0.2 to 0.5 watts is consumed by cooling, power distribution, and infrastructure. None of this appears on the token price.

Layer Three: The Inference Tax

Training receives the headlines. The $100 million model. The breakthrough architecture. Inference, the continuous cost of serving that model in production, is where the budget dies quietly. GPT-4’s training cost approximately $150 million. Within two years, its inference costs exceeded $2.3 billion. That is a 15x ratio.

Inference now represents approximately 55% of total AI infrastructure spending in early 2026. As models shift from simple question-and-answer interactions to agentic workflows that plan, use tools, execute code, and self-correct, token consumption per task scales exponentially. Models like GPT-4 Pro now process approximately 22,000 tokens in a single planning-and-execution exchange, compared to a few hundred for a simple query. Gartner projects that by 2030, even a 90% reduction in token unit prices will be outpaced by this consumption growth, leaving enterprise AI total costs higher than today.

Layer Four: Verification and Oversight

This is the layer that breaks most enterprise AI business cases. Frontier models maintain an average hallucination rate of approximately 9.2% on general knowledge tasks. Well-architected production systems using retrieval-augmented generation can reduce this to below 3%, but not to zero. Forrester Research estimates that each enterprise employee spends resources worth approximately $14,200 per year on AI hallucination mitigation: checking outputs, correcting errors, handling escalations, and rebuilding trust with downstream stakeholders who received incorrect information.

Microsoft’s own 2025 data found that knowledge workers spend 4.3 hours per week, more than 10% of their working time, verifying AI outputs. A 20-agent deployment requires dedicated human oversight running at approximately $23,000 per agent per year. This cost does not appear in the model API bill. It appears in the headcount budget of the teams that have to catch what the model gets wrong.

The complete cost chain of a deployed AI token, from chip depreciation through energy through inference infrastructure through human verification, is typically two to four times the visible API cost for enterprise deployments. Organizations that model the economics using API price alone are systematically underestimating their total cost of ownership.

Part Two: Where AI Is Genuinely Cheaper Than Humans

With the full cost structure in view, a more precise picture emerges of where the economics actually favor AI deployment.

High-volume, standardized, low-stakes tasks

The clearest case for AI cost advantage is in tasks that are highly repetitive, structurally predictable, and where the cost of an error is low or easily caught before causing damage.

But the honest accounting does not stop there. Add enterprise infrastructure, integration costs, quality assurance oversight, and the human escalation layer required for complex or emotionally sensitive interactions, and the economics become viable only when the team size exceeds approximately five to eight agents. Below that threshold, human representatives are often cheaper on a total cost basis. The BLS projects customer service representative employment will decline 5% from 2024 to 2034, confirming the economic case is real, while also confirming the transition is slower than the simple token math implies.

Additional task categories where AI delivers genuine cost advantage:

• Large-scale document classification and data extraction, where accuracy requirements are moderate and human review is the final gate

• Code review for standard patterns and security vulnerabilities, where the output is verified by a human engineer before acting

• Initial contract drafting for standard transaction types, where a lawyer reviews the AI output rather than writing from scratch

• Translation and localization at volume, where the cost differential is large enough that human review of AI output is still cheaper than human translation from scratch

The common thread is structure. AI outperforms human labor economically when the task has a predictable form, the inputs are well-defined, and the output is reviewed before high-stakes use.

Companies capturing this advantage:

Salesforce’s Agentforce deploys AI agents for standardized customer service workflows, charging on a per-conversation basis rather than a per-seat model. For customers with high-volume, structured inquiry patterns, the unit economics work because the deployment is confined to the structured, lower-stakes tier. ServiceNow applies the same logic to IT service management, automating common ticket resolution while routing complex issues to human technicians.

Tasks Where AI Augments Rather Than Replaces

MIT’s study of automation economics found that AI is only economically justifiable for 23% of the wages tied to the tasks examined. The remaining 77% does not mean AI has no role. It means the economic case is for augmentation rather than replacement: a human with an AI tool outperforms either alone, at a combined cost lower than two humans operating independently.

The empirical evidence on what augmentation actually produces is now specific enough to be actionable. A study published in the Quarterly Journal of Economics in May 2025 examined 5,172 customer support agents given access to a generative AI assistant. Productivity, measured by issues resolved per hour, increased 15% on average. Critically, the gains were largest for lower-skilled workers, who benefited most from access to institutional knowledge embedded in the AI tool, while the highest-performing agents saw smaller gains. The AI did not replace the agents. It compressed the performance gap between the best and the rest.

In financial analysis, a natural experiment using FactSet’s 2023 AI platform launch found that AI-assisted analysts produced reports with 40% more distinct information sources, 34% broader topical coverage, and 25% greater use of advanced analytical methods, while also improving timeliness. However, forecast errors rose 59%, as AI-assisted reports conveyed a more complex mix of information that proved harder to synthesize, particularly for analysts under heavy cognitive load. That last finding matters. AI made analysts faster and broader. It did not make them more accurate. The judgment layer remained indispensable, and in some cases, the additional information volume created by AI increased the cognitive burden on the human responsible for the final call.

A February 2026 review of the empirical evidence across multiple studies found that roughly 80% of U.S. workers have at least 10% of their tasks exposed to LLM assistance, and about 19% have 50% or more of their tasks exposed. The research characterizes current AI systems as primarily augmenting human labor rather than automating it outright, with AI adoption increasing complexity in augmentation-prone roles while reducing skill requirements in automation-prone roles.

Three domains illustrate the augmentation dynamic with particular clarity.

In legal research, AI tools reduce the time required to survey case law, identify relevant precedents, and draft initial briefs. The productivity gain is real. The lawyer’s judgment in applying that research to the specific facts of a client’s situation, and the professional liability that attaches to the advice given, remains entirely human. The total cost of legal work delivered falls because the AI absorbs the low-judgment research hours. It does not fall because the lawyer is eliminated.

In medical diagnosis support, AI systems have demonstrated detection rates for specific conditions that match or exceed experienced clinicians in controlled settings. In practice, no major jurisdiction permits AI to make clinical decisions without physician validation under current regulatory frameworks. The physician who reviews AI-flagged imaging studies is not freed from the task. They are redirected to higher-judgment interpretation and patient communication, with AI handling the first-pass pattern recognition. The cost per diagnosis falls. The physician does not disappear from the workflow.

In software engineering, generative AI tools allow developers to generate, test, and document code more rapidly, improving data quality and accelerating system integration work. BLS analysis suggests these productivity enhancements are likely to be outweighed by strong business demand for software infrastructure, meaning employment in these roles is not projected to decline despite significant AI-driven productivity improvement. More output per engineer. Not fewer engineers.

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

1. Market Lens

Share

The S&P 500 closed at a new all-time high on Friday. The Dow crossed 51,000 for the first time. WTI crude has fallen from $105 six weeks ago to $87.36 today. Yields are declining. Every surface indicator is pointing in the same direction.

The question worth asking is: what changed?

The answer is oil. WTI fell 9.6% this week as Iran peace deal optimism moved from statement to framework. Trump’s declaration that the agreement was “largely negotiated” was followed by four days of quiet diplomatic confirmation from Gulf intermediaries. By Friday, the market was pricing a meaningful probability of a formal signing within weeks. The transmission mechanism was immediate: lower oil means lower energy-driven inflation, lower energy inflation means the Fed’s hands are less tied, and less tied hands mean every long-duration asset in the equity market re-rates higher. Ten-year yields fell nearly 12 basis points on the week, not because the PCE data was dovish, but because the energy price path that has been the primary driver of inflation for three months is beginning to reverse.

The PCE data released Thursday told a layered story. Headline PCE rose 3.8% year-over-year in April, in line with consensus, driven almost entirely by energy. Core PCE came in at 3.3% year-over-year on a 0.2% monthly gain, a cooling from March’s 0.4%. The Dallas Fed trimmed mean PCE showed a twelve-month rate of 2.3%, a number that reveals where the inflation is actually concentrated: it is war-driven and energy-driven, not structurally embedded across the consumer price basket. If Hormuz reopens and WTI stabilizes in the high $80s, the Fed’s primary justification for hawkishness begins to dissolve without any policy action required. The consumer picture beneath the headline is less comfortable. Real per capita disposable income fell year-over-year for the second consecutive month. The personal saving rate dropped to 2.6%, a multi-year low. Americans are spending more in nominal terms because energy costs more, not because they have more to spend.

That context makes this week’s earnings slate analytically important. The businesses reporting this week are not benefiting from consumer confidence. They are benefiting from structural demand that does not require it.

AutoZone reported Q3 FY2026 sales of $4.8 billion, up 8.4% year-over-year, the largest quarterly revenue increase in over three years. EPS of $38.07 beat the $36.13 consensus by 5.4%, and adjusted EPS growth, excluding a $20 million LIFO accounting headwind, was 12.5%. The number that defines the thesis is commercial same-store sales growth of 10.4%, against DIY of 2.2%. People are not repairing their cars because consumer sentiment is strong. They are repairing their cars because new vehicles cost $48,000, auto loan rates are above 7%, and the average age of the US vehicle fleet is at a record high. That is structural demand. It compounds regardless of what PCE prints.

Salesforce reported Q1 FY2027 revenue of $11.13 billion, up 13% year-over-year, with non-GAAP EPS of $3.88, a 24% beat against consensus. Non-GAAP operating margin reached a record 34.8%, expanding 250 basis points year-over-year. Agentforce premium SKU bookings grew approximately 60% year-over-year. Slack active work units grew 350% sequentially. Three million custom apps were built on the platform in a single quarter, of which 250,000 were third-party AI agents, more than doubling sequentially. The company raised full-year guidance to $45.9 to $46.2 billion. The stock closed essentially flat. The market’s inability to reward a 24% EPS beat and a record operating margin is not a signal about Salesforce’s business quality. It is a signal about the price already embedded in the multiple, and about concern over the $25 billion debt-funded accelerated buyback that returned $27.5 billion to shareholders in the quarter. The thesis is confirmed. The valuation is demanding.

Costco reported Q3 FY2026 net sales of $69.15 billion, up 11.6%, with EPS of $4.93 slightly below the $4.97 consensus. Comparable sales grew 9.8%, or 6.6% excluding gasoline and currency effects. Paid executive memberships reached 41.2 million, up 9.6%. The US and Canada renewal rate held at 92.2%. Digitally-enabled comparable sales grew 21.5%. The stock dipped modestly after the print. At approximately 50 times forward earnings, Costco’s price does not accommodate even minor misses. The business is compounding exactly as designed. The membership flywheel is intact. The slight revenue comp miss is a function of gasoline price deflation flowing through the top line, not a deterioration in consumer engagement. The thesis is intact at the current 1% allocation.

The week’s most analytically significant earnings result came from a watchlist name, not a portfolio holding. HEICO reported Q2 FY2026 EPS of $1.66 against a consensus of $1.33, a 24.6% beat that also cleared management’s own guidance of $1.60. Revenue reached $1.38 billion, up 25.3% year-over-year, beating the $1.24 billion estimate by 10.7%. Organic net sales growth was 18%. Flight Support Group operating margin expanded to 26.2% from 24.1%, with EBITA margin of 28.6%, up 160 basis points. Electronic Technologies Group net sales grew 34%. Operating income rose 41% to $350.4 million, lifting consolidated operating margin to 25.5% from 22.6%. Net income attributable to HEICO was a record $233.8 million, up 49% year-over-year. Every variable on the watchlist inclusion checklist resolved affirmatively in a single quarter. HEI moves from watchlist to portfolio next week.

The week’s signal is not the all-time high. It is the convergence: oil is falling, inflation may be peaking, the Fed has room to pause, and the businesses in the portfolio are printing results that confirm structural demand rather than cyclical optimism. That combination, if the Iran deal is signed, resets the macro framework for the second half of 2026 in a way that benefits the portfolio’s positioning more directly than any single earnings report could.

This Week’s Signal: The inflation that was war-driven is starting to unwind. The earnings confirmed what the portfolio was built for. The market noticed. The portfolio is watching the signing ceremony.

Share

2.Portfolio Desk

This Week’s Focus

The macro environment, with real consumer incomes under pressure and the saving rate at a multi-year low, favors businesses whose demand does not require consumer confidence to exist. ROL is the clearest expression of that logic in the portfolio. Separately, the AI monetization thesis continues to develop concrete evidence through CRM’s Q1 print and NVIDIA’s confirmed order book, which increases conviction in the software layer that sits above the infrastructure. MSFT and CSU are the two most structurally defensible positions in that layer. HEI reported a 24.6% EPS beat this week, resolving every inclusion criterion on the watchlist. The position has not yet been initiated pending entry price discipline in the week ahead.

Defensive Anchors

Rollins (ROL) | Action: Add | Allocation: ~4.0% | Category: Defensive Anchor

ROL is added from 3% to 4%, the largest single-position increase among the defensive tier this week. The timing is deliberate. April PCE confirmed that real per capita disposable income has now fallen year-over-year for two consecutive months, and the personal saving rate has dropped to 2.6%. In that environment, the distinction between discretionary and non-discretionary spending becomes the primary analytical variable for any consumer-facing business. ROL sits at the extreme non-discretionary end of the spectrum. Pest control is not postponed when household budgets tighten. It is legally required in commercial properties, contractually embedded in multi-family residential agreements, and driven by health and safety imperatives that do not respond to consumer confidence surveys.

ROL’s business model compounds through route density rather than pricing power alone. Each additional customer on an existing route reduces the marginal cost of service delivery. As the company continues acquiring regional operators in fragmented local markets, the density flywheel accelerates. This is a business that generates above-average returns on incremental capital without requiring macro tailwinds. In a period where energy-driven inflation is compressing real incomes and the Fed has no room to cut, that combination is precisely what the Defensive Anchor tier is designed to provide. The add brings ROL to 4%, making it the largest individual position in the Defensive Anchors tier. Monitoring variable: organic revenue growth rate and route density improvement as the integration of recent acquisitions completes.

Visa (V) | Hold | ~4.5% No material developments this week beyond the macro context. The Iran deal progress and WTI decline benefit Visa indirectly through cross-border travel normalization as Hormuz reopens. Monitoring variable: cross-border volume in the next quarterly disclosure.

Costco (COST) | Hold | ~1.0% Q3 FY2026 confirmed the membership flywheel is intact. Net sales of $69.15 billion, up 11.6%, EPS of $4.93, and a 92.2% US renewal rate. The slight miss against the $4.97 consensus reflects gasoline price deflation flowing through the top line rather than any deterioration in member engagement. Executive membership reached 41.2 million, up 9.6%. The stock’s muted reaction is a valuation story at 50 times forward earnings, not a business story. Thesis intact at 1% allocation. Monitoring variable: Executive Member penetration and tariff cost pass-through in Q4.

TMUS, WM, LIN | Hold No material developments. All three hold against stated theses with monitoring variables unchanged.

Core Compounders

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

For a closer look at how these principles translate into real positions, explore the 👉 Sample Portfolio — a demonstration of thesis-driven allocation and disciplined compounding.

The Defining Lens

Chuck Akre runs his investment firm from a small town in rural Virginia, far removed from Wall Street. He does not appear on financial television. He does not manage quarterly earnings expectations or offer macro commentary. What he does, with remarkable consistency across five decades, is find businesses capable of compounding capital at rates well above average, and then hold them long enough for that compounding to become the dominant force in the portfolio.

The numbers confirm the discipline. From 1997 through August 2009, while managing the FBR Focus Fund, Akre delivered an annualized return of 12.3 percent. The S&P 500 returned 4.4 percent over the same period, and the average mid-cap growth fund returned 4.9 percent. His fund ranked in the top one percent of its category. After breaking from FBR, he launched the Akre Focus Fund in August 2009 at age 67. From inception through September 2024, the Fund compounded at 15.70 percent annually, against 14.34 percent for the S&P 500 total return. A dollar invested at inception in the Fund grew to over nine times its original value, compared to roughly 7.5 times for the index over the same fifteen years.

These are the numbers of a practitioner who has solved a specific problem, solved it clearly, and refused to deviate from the solution regardless of what the market was doing around him.

The Making of the Framework: An English Major Figures It Out

Akre did not arrive at his investment philosophy through a conventional path. He studied English literature at American University in Washington, D.C., having started as a pre-med student before switching majors. He graduated in 1968 with no formal training in finance, economics, or accounting, and entered the securities industry the same year.

His first job was at Johnston, Lemon and Co., an NYSE member firm based in Washington, D.C. He stayed for twenty-one years, moving from stockbroker through various roles before becoming CEO of the asset management division and director of research. Those two decades inside a traditional securities firm gave him something more useful than a credential: the time and proximity to observe what actually separated investment outcomes at the extremes.

Two books shaped his thinking in ways he has since credited publicly. John Train’s The Money Masters introduced him to Warren Buffett’s approach to business analysis, specifically the idea that buying shares means buying pieces of a business, and that the quality of the business determines the quality of the investment. Thomas Phelps’s 100 to 1 in the Stock Market opened his understanding of what compounding produces over long enough time horizons when the reinvestment rate is high enough. Neither book was a technical manual. Both changed the way he defined the investment problem.

In 1989, Akre left Johnston Lemon to found Akre Capital Management. He operated the firm under the umbrella of Friedman, Billings, Ramsey and Co. from 1993 until 2000, when he took the firm independent again. He moved Akre Capital to Middleburg, Virginia in early 2002, where the firm remains today. When FBR management sought to cut his fees by 20 percent, he declined and left, taking his track record and his framework with him. The Akre Focus Fund launched in August 2009 with a clear mandate and no institutional parentage.

The Three-Legged Stool

Every serious analysis of Akre eventually arrives at the same place: a three-legged milking stool that sits in the conference room of Akre Capital Management and serves as the firm’s organizing metaphor. Akre’s own description of it is precise: “The essence of our investment approach is perfectly captured by the visual of a three-legged stool. This metaphoric three-legged stool describes what we look for in an investment: extraordinary business, talented management, and great reinvestment opportunities and histories.”

The first leg is the quality of the business itself. Akre looks for companies generating enduring, high returns on equity and free cash flow, with identifiable competitive advantages that make those returns sustainable. The source of the advantage can vary: brand, network effects, patents, switching costs, or economies of scale. What matters is that the advantage is real and durable. He does not screen by industry, by geography, or by valuation multiple as a primary filter. The starting question is whether the business model itself earns above-average returns on the capital it deploys, and whether it will continue to do so.

The second leg is management quality, assessed specifically through the lens of capital allocation. Akre has made this point explicitly: “Not only do we want to have great business managers, but we want to see they treat public shareholders as partners even as though they don’t know them.” He is looking for managers who behave like owners, who make capital decisions as if their own wealth depended on the outcome, and whose historical record shows a pattern of disciplined reinvestment rather than value-destroying acquisitions or excessive compensation.

The third leg is the reinvestment opportunity, and Akre has said plainly that as he has gotten older, this has become the most important of the three. A business can have superior economics and excellent management, and still produce mediocre long-term compounding if it has no productive place to reinvest its free cash flow. The compounding engine requires fuel. What Akre is looking for is not just a high-return business, but a high-return business with a runway long enough that it can keep deploying capital at those rates for years, even decades, into the future. The growth of economic value per share over time is the only number that matters for long-term investors, and it requires this third leg to sustain itself.

The stool’s physical design is not incidental to the metaphor. A three-legged stool is inherently stable on uneven terrain in a way a four-legged stool is not. Akre is making a structural argument: a portfolio built on all three legs can survive the uneven terrain of markets in a way that portfolios built on fewer criteria cannot.

Share

Portfolio Architecture: What the Holdings Reveal

The portfolio at Akre Capital Management as of late 2024 contained eighteen stocks, with the top five positions accounting for the majority of total exposure. Mastercard held the largest weight at approximately 16.87 percent, followed by KKR at 13.16 percent, Moody’s at 12.61 percent, Brookfield Corporation at 9.43 percent, and O’Reilly Automotive at 9.18 percent.

Reading these holdings as a group reveals the three-legged stool applied at scale.

Mastercard is the clearest example of all three legs working simultaneously. The payment network model generates returns on invested capital that are among the highest in the S&P 500, driven by a business with minimal physical assets and near-zero marginal cost on incremental transaction volume. Management has deployed free cash flow through a combination of buybacks and network investment that has sustained earnings growth well above nominal GDP for over a decade. The reinvestment opportunity remains substantial: global cash transactions continue converting to electronic payment, and cross-border volume growth adds a second long-duration growth driver above the core domestic business. Akre has held Mastercard for approximately fifteen years. His original investment is reported to have produced gains exceeding 390 percent on the position.

Moody’s represents a different expression of the same framework. The credit rating agency operates in a structural oligopoly alongside S&P Global. The two firms together dominate global bond rating, and that dominance is protected by regulatory recognition, issuer relationships, and the reputational capital that took decades to accumulate. New entrants cannot replicate those barriers quickly. Moody’s Analytics, the data and research division, provides a parallel reinvestment channel at attractive returns. The position as of late 2024 represented Akre’s second-largest holding by weight.

O’Reilly Automotive is perhaps the most instructive example of Akre’s patience. He began building the position in 2011 at prices in the mid-fifty-dollar range. The stock has since appreciated substantially, and Akre’s reported cost basis on the position turned a $172 million investment into over $1 billion in value at peak. O’Reilly operates in the automotive aftermarket parts category, a business with structural non-discretionary demand: cars require maintenance regardless of economic conditions. The company’s distribution model, store-level unit economics, and pricing discipline against independent competitors have produced a ROIC profile that has consistently justified Akre’s long holding period.

The portfolio’s concentration is not an artifact of size constraints. It is an intentional expression of conviction. Akre’s stated timetable is five to ten years ahead. A quarterly earnings miss is assessed as a potential buying opportunity, not as a sell trigger. The portfolio’s low turnover reflects the underlying logic of the framework: if you have correctly identified all three legs and the three legs remain intact, the rational action is to stay invested and let compounding operate without interruption.

Track Record in Context

Akre’s performance record is best understood in the context of what he faced during the periods it covers, rather than against a backdrop of favorable conditions.

The FBR Focus Fund period from 1997 through 2009 encompassed the collapse of the technology bubble, the September 2001 shock, and the 2008 financial crisis, three of the most structurally damaging environments for equity portfolios in the modern era. Across those twelve years, Akre compounded at 12.3 percent annually. The index returned 4.4 percent. That is not a cyclical advantage. That is a structural one.

The Akre Focus Fund period beginning in 2009 has been more favorable for equities broadly, yet the comparison still holds. From inception through the third quarter of 2024, the Fund compounded at 15.70 percent against the S&P 500’s 14.34 percent. The annualized gap appears narrow, but compounding erases that impression: a dollar in the Fund grew to over nine times its value, while the same dollar in the index grew to 7.5 times. The Fund has also navigated the COVID drawdown, the 2022 rate shock, and a period of extreme market concentration in mega-cap technology names that structurally disadvantaged any fund without heavy exposure to the five or six largest companies in the index.

The one year where the framework visibly struggled was 2023, when the Fund returned 14.77 percent against the S&P 500’s 24 percent return. That gap was a direct consequence of the narrow AI-driven market leadership in that year, where a handful of large-cap technology companies accounted for the majority of index returns. Akre’s portfolio, oriented toward financial infrastructure and durable compounders rather than AI-adjacent semiconductor and software names, did not participate in that leadership. Whether that underperformance reflects a framework limitation or a timing artifact of an unusually concentrated market is a question that requires more than one year of data to answer.

Where He Was Wrong

Akre’s CarMax position illustrates the structural risk inside any concentrated compounder portfolio. CarMax appeared to satisfy the three-legged stool criteria: the company was the largest used-vehicle retailer in the United States with a clearly differentiated customer experience, a management team with a strong multi-decade record, and a long runway for store expansion and digital integration. Akre built a meaningful position over time.

The investment has underperformed relative to expectations. Rising interest rates drove a significant increase in consumer financing costs, compressing used-vehicle demand and squeezing CarMax’s finance subsidiary margins simultaneously. The stock declined from approximately $265 to below $60 at its trough during the 2022 to 2023 period. Akre trimmed the position materially through 2024 and into 2025.

The lesson embedded in this case is specific to the third leg of the stool. The reinvestment opportunity for CarMax was always dependent on consumer access to affordable vehicle financing. When rate conditions shifted structurally, the reinvestment runway contracted faster than the business model’s quality metrics suggested it should. For investors applying the three-legged stool framework, this argues for stress-testing the reinvestment assumption across interest rate scenarios, not just under the conditions prevailing at the time of purchase.

The American Tower position tells a different version of the same story. The cell tower business was a textbook Akre holding for years: essential infrastructure, long-duration lease contracts, pricing power, and a reinvestment opportunity tied to global wireless network densification. Rising interest rates compressed real estate and infrastructure valuations across the board, and American Tower’s capital structure became a concern as refinancing costs increased. Akre trimmed the position significantly across 2024 and 2025, reducing it by approximately 65 percent through the period. The exit reflected not a change in the quality of the business but a reassessment of whether the third leg, the reinvestment opportunity, remained as attractive under a structurally higher cost of capital.

Both positions demonstrate something important: the three-legged stool framework is not a static checklist applied once at purchase. It is a continuous evaluation. When any leg weakens in a material and durable way, the rational response is to reduce or exit, regardless of how long the position has been held.

The Succession Question

In 2020, Akre transitioned day-to-day investment management responsibility to John Neff and Chris Cerrone, the two portfolio managers who had worked alongside him for years. He retained his role as Chairman of Akre Capital Management but stepped back from active portfolio decisions. The Fund continues under Neff and Cerrone’s management.

The succession matters analytically because Akre’s framework, unlike many investment approaches, is explicit enough to be transmitted. The three-legged stool has a defined set of criteria. The investment philosophy published on Akre Capital’s website uses Akre’s own language and has not changed. The quarterly letters from Neff and Cerrone continue to apply the same evaluative structure to both existing positions and new opportunities. In the third quarter 2024 letter, the team explicitly cited compounding as the organizing objective of the portfolio and described the Fund’s fifteen-year track record in terms Akre himself would have used.

The practical test of succession is performance continuity over a full market cycle, and that data is still accumulating. What the current record shows is philosophical continuity and disciplined portfolio management in a period that has been structurally difficult for the style. The 2024 Q3 one-year return of 38.60 percent, against the S&P 500’s 36.35 percent, suggests the framework is operational in capable hands.

The Extractable Principle

Three principles emerge from Akre’s record that are directly applicable to how a portfolio manager should evaluate businesses and construct a long-duration portfolio.

The first is that the reinvestment opportunity is the most critical and most frequently underestimated variable in a long-term investment thesis. A business with high current returns on capital and no productive reinvestment channel will compound at its dividend yield plus organic growth, and no more. A business with high current returns and a decade-long reinvestment runway will compound at a rate that makes it appear expensive every single year until, in retrospect, it was obviously cheap the entire time. Akre’s portfolio construction begins with this variable, and investors who evaluate businesses without systematically asking where the next decade of free cash flow will go are missing the most important question.

The second is that selling great businesses early is a cost that never appears on a performance attribution report but compounds as severely as any explicit loss. Akre has stated that one of the biggest mistakes investors make is selling exceptional businesses prematurely. The tax event, the re-deployment risk, and the foregone compounding all work against the investor who exits a high-quality compounder to rotate into something that looks cheaper on a near-term multiple. This does not mean never selling. It means the bar for selling a confirmed compounder should be set at fundamental impairment of one of the three legs, and nothing less.

The third is that portfolio concentration, applied with genuine conviction rather than false precision, produces better long-term outcomes than diversification applied as a risk management substitute for research quality. Akre’s eighteen-stock portfolio is concentrated because each position reflects a high-conviction assessment of all three legs, held simultaneously. Diversifying into a fiftieth position because it is available is a different activity from owning eighteen businesses you understand deeply enough to hold through a 30 percent drawdown without reassessing the thesis.

The Closed-Market Test

If you could not buy or sell any of Akre Capital’s disclosed holdings for the next five years, would you be comfortable with the current portfolio?

The portfolio as of late 2024 is anchored in financial infrastructure businesses with dominant market positions, recurring revenue models, and reinvestment runways tied to global payment network expansion, alternative asset management growth, and insurance reserve compounding. These are businesses whose competitive positions do not change materially over five-year periods. The primary risk is valuation: businesses of this quality rarely trade cheaply, and a significant multiple compression would affect near-term returns even when underlying business quality remains intact.

The succession question adds a second variable. Neff and Cerrone have demonstrated the philosophical continuity of the framework, but they have not yet managed through a full market cycle as the primary decision-makers. That is a genuine uncertainty.

Chuck Akre spent five decades arriving at a framework so simple that it can be written in one sentence and so difficult to execute that almost no one does it consistently. He found extraordinary businesses, run by managers who treated capital with discipline, with long enough reinvestment runways to sustain compounding across full economic cycles. Then he sat still. For investors willing to apply the same patience to evaluating their own portfolios, the Akre record is among the clearest available demonstrations of what that patience actually produces over time.

This article is part of the Thesis Rationale Legend Series. All analysis is hypothetical and for informational purposes only. This does not constitute investment advice.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

Market Lens

The Dow Jones Industrial Average closed at a new all-time high of 50,579 on Friday. The S&P 500 is within 0.4% of its own record. VIX collapsed from 21.4 to 16.70 in five trading sessions. By every surface measure, this was a risk-on week driven by relief and optimism.

1. NVIDIA Q1 FY2027: $81.6B and the AI Factory Declaration NVIDIA reported revenue of $81.6 billion, up 85% year-over-year, with Data Center at $75.2 billion, up 92%. Non-GAAP gross margin held at 75%, and Q2 guidance came in at $91 billion, ahead of every analyst estimate. The company announced an $80 billion share buyback and doubled its quarterly dividend to $0.25. Jensen Huang’s framing was precise: the world is building AI factories at a scale that has no historical precedent, and Nvidia is the primary infrastructure tax collector for that buildout.

2. Kevin Warsh Sworn In, FOMC Minutes Release Hike Risk Kevin Warsh was formally sworn in as Federal Reserve Chairman on Friday May 22. The FOMC minutes released Wednesday showed a committee deeply concerned about persistent inflation, with PPI at 6% year-over-year and CPI at 3.8%. Fed Governor Waller stated explicitly that rates at 3.5 to 3.75% may be insufficient, and derivatives markets now assign a 25% probability to at least one hike before year-end. The transition from Powell to Warsh introduces a new communication style into a Fed that is already divided by the most dissents since 1992.

3. Trump: Iran Deal “Largely Negotiated,” Hormuz to Reopen In a development that broke Saturday morning, President Trump stated that a peace framework with Iran has been “largely negotiated,” covering an official end to the war, reopening of the Strait of Hormuz without tolls, and a two-month negotiation window on Iran’s nuclear program. Secretary of State Rubio told reporters he expected “good news within hours.” Eight regional governments including Saudi Arabia, Qatar, UAE, Pakistan, Turkey, Egypt, Jordan and Bahrain participated in brokering the framework. If the deal is signed, the single largest input to the current inflation regime, energy prices, begins to normalize structurally rather than cyclically.

4. WTI Falls 8.4% to $96.60 on Gulf Diplomacy WTI crude declined from $105.42 to $96.60 on the week as Gulf state diplomatic intervention signaled de-escalation momentum. The physical supply disruption from the Hormuz closure has not resolved, and the IEA warned the market could remain undersupplied through October even if fighting ends soon. The oil price decline contributed directly to the VIX compression and equity rally, but the structural supply damage from ten weeks of closure continues to transmit through energy-driven inflation with a lag.

5. 10Y Breaks 4.6%, Private Credit Defaults Hit Record High The 10-year Treasury yield touched 4.60% this week, a one-year high, before settling at 4.572% on Friday. The 30-year briefly exceeded 5.1%. CNBC reported that private credit defaults have reached a record high as sustained rates above 4.5% compress refinancing capacity for leveraged borrowers. Chapter 11 bankruptcy filings rose 42% year-over-year in April. The transmission mechanism is now visible: elevated rates are not abstract macro risk. They are producing real defaults in real businesses, concentrated in BBB-rated and below industrial, retail, and real estate credit.

6. Workday Q1 FY2027: Enterprise Software Inflection Confirmed Workday reported Q1 FY2027 EPS of $2.66 against a consensus of $2.25, an 18% beat, with revenue of $2.54 billion up 13.5% year-over-year. Management described it as the strongest Q1 for net new Annual Contract Value in five years, with EMEA ACV growing 50% year-over-year. Non-GAAP operating margin expanded significantly. The result is a direct read-through for CRM: enterprise software demand is not collapsing under rate pressure, and AI-driven workflow adoption is accelerating bookings rather than compressing them.

7. SpaceX Files Historic IPO, Starship Launch Delayed SpaceX formally filed its IPO prospectus this week in what is expected to be the largest technology listing in years. The planned Starship launch was scrubbed at the last moment due to technical compliance review. The IPO is expected to redraw the valuation architecture across aerospace, satellite communications, and commercial defense, and to attract significant institutional capital reallocation from existing technology positions. Index funds have begun building liquidity models for the expected inclusion.

8. Salesforce Reports Tuesday May 27: The Agentforce Test CRM reports its Q1 FY2027 results on Tuesday May 27 after market close. This is the most consequential near-term catalyst for the portfolio’s Blue Chip tier. The single variable that matters is Agentforce paid seat count and its contribution to net new ARR. Workday’s Q1 beat establishes that enterprise AI workflows are generating real bookings. The question is whether Salesforce has converted its Agentforce pipeline announcements into contracted, recurring revenue at a scale that changes the thesis from speculative to structural.

9. HLI Restructuring Thesis Accelerates The 42% year-over-year surge in Chapter 11 filings and record private credit defaults reported this week are direct tailwinds for Houlihan Lokey’s restructuring advisory business. HLI’s Q4 FY2026 missed on revenue but reported a record backlog entering FY2027. The macro conditions generating that backlog, sustained high rates, leveraged buyout vintage stress from 2021 to 2022, and now visible default acceleration, are not easing. They are intensifying. The thesis for HLI’s restructuring mandate pipeline has strengthened materially this week without any new company-specific news.

2. Portfolio Desk

Defensive Anchors

Visa (V) | Hold | ~4.5% The Berkshire 13F disclosed this week that Greg Abel liquidated the V position as part of his broader clearing of Todd Combs legacy allocations. This is procedural rather than analytical. Abel’s exit reflects portfolio concentration discipline at Berkshire’s scale, not a judgment on Visa’s structural business quality. The thesis on V in this portfolio rests on network effects, cross-border volume compounding, and the Agentic Ready expansion that positions Visa as the settlement layer for AI-initiated payments. Neither Abel’s exit nor any development this week changes that assessment. Monitoring variable: cross-border volume trajectory as the Hormuz deal progresses and travel normalization accelerates.

Costco (COST) | Hold | ~1% Similarly, Berkshire’s 13F showed a full liquidation of COST, again attributed to the Combs allocation clearing. The thesis for this portfolio’s 1% position is not Berkshire-dependent. COST reports Thursday May 28, with consensus EPS of $4.97. Executive Member penetration and any tariff cost pass-through commentary are the primary monitoring variables. Thesis intact at current sizing.

WM, ROL, LIN, TMUS | Hold No material developments this week. All four positions hold against their stated theses with monitoring variables unchanged.

Core Compounders

Copart (CPRT) | Hold | ~3% |

CPRT reported Q3 FY2026 results on Thursday that represent the first meaningful positive datapoint in three quarters. EPS came in at $0.43 against a consensus of $0.41, a 4.9% beat. Revenue reached $1.24 billion against an estimate of $1.20 billion, a 3.3% beat. The stock rose 4.09% on the day.

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

The Consensus That the Data Keeps Contradicting

The narrative has been consistent and loud. AI tools will commoditize software functions. ChatGPT can write code, draft emails, and summarize documents. Cursor replaces the IDE. Claude designs interfaces. Why would enterprise customers keep paying for specialized SaaS subscriptions when a general-purpose AI can do the same job for a fraction of the cost?

The market bought this argument with conviction. Adobe’s stock fell 40% from its 2025 highs. Atlassian traded sideways through stretches where the broader market was making new records. Figma’s stock collapsed from $143 to $16 after its IPO, a decline of nearly 90%, as investors concluded that AI-generated design would make the platform obsolete. Monday.com was consistently framed as a commoditized project management tool with no defensible edge against an AI that could organize tasks from a chat prompt.

Then the earnings reports started arriving.

What follows is not a theoretical argument for why SaaS should survive AI. It is a forensic reading of what the financial data from four specific companies is showing about what is actually happening when AI meets deeply embedded enterprise software.

The Core Misunderstanding

Before the cases, the framing.

The disruption thesis rests on a category error: it confuses feature replacement with workflow replacement. An AI tool can replicate an individual function. It cannot replicate the organizational memory that a SaaS platform accumulates over years of active use.

When a company has used Jira for five years, Jira contains something that no AI tool can generate from scratch: the decision history of the engineering organization. Every ticket, every sprint, every escalation, every resolution. The Teamwork Graph that Atlassian has built now contains more than 100 billion objects and relationships across first and third-party tools. That is not a feature. It is institutional infrastructure.

The same logic applies across the category. Figma contains the design system of every team that has ever built on it. Monday.com contains the workflow architecture of every operation that has been configured inside it. These platforms are not selling access to functions. They are the system of record for how organizations work. Replacing them is not a product decision. It is an operational reconstruction.

AI does not disrupt this layer. It amplifies it. The evidence is in the numbers.

Share

Case One: Figma | The Most Direct Refutation

No company in software had a more dramatic narrative reversal than Figma in the twelve months following its July 2025 IPO.

The stock debuted at $33 per share and immediately surged to $142, valuing the company at more than $60 billion. Then Anthropic launched Claude Design. Google released Stitch. Adobe accelerated its own AI design tools. The market concluded that generative AI would commoditize interface design and that Figma’s seat-based subscription model would be the first casualty. By early 2026, Figma had fallen more than 80% from its post-IPO peak, trading near its 52-week low of $16.60, battered by the conviction that AI would make the design tools it sells redundant.

Then Q1 2026 results arrived.

Revenue grew 46% year over year to $333.4 million, accelerating for the second consecutive quarter. Non-GAAP EPS came in at $0.10, against consensus of $0.06. Figma raised its full-year revenue guidance by $55 million to between $1.422 billion and $1.428 billion. Net Dollar Retention reached 139%, up three percentage points from the prior quarter and the highest in more than two years.

The AI monetization data inside that report is the more instructive story.

About 30% of customers spending over $100,000 in annualized revenue are using Figma Make, the AI-powered design generation feature, on a weekly basis. New Pro team conversions grew more than 150% year over year, with the company attributing that growth directly to AI feature adoption. Paid customers with more than $100,000 in ARR grew 48% year over year to 1,525, well ahead of overall customer growth. Figma began enforcing AI credit limits for higher-tier seats in Q1, and the data on what happened when it did is striking: 75% of users who exceeded the free credit limit converted to paid AI credits. The 5% who left represents normal churn for any software product.

The CEO Dylan Field’s framing of the competitive dynamic is worth quoting directly. “When code is a commodity, design is the competitive edge.” The argument is structural: as AI coding tools make it trivially easy to generate functional software, the craft of designing what that software looks and behaves like becomes the scarce input, and Figma is the platform where professional design work happens at scale.

The market priced Figma as a victim of AI. The earnings data suggests it is a beneficiary.

Case Two: Atlassian | AI as an Upgrade Engine

Atlassian’s story is not about a dramatic narrative reversal. It is about a company methodically converting AI adoption into a revenue expansion mechanism that was not available to it before.

In Q2 FY2026, Atlassian delivered its first-ever $1 billion Cloud revenue quarter, growing 26% year over year. Total revenue was $1.586 billion, up 23% year over year. Rovo, Atlassian’s AI platform, surpassed 5 million monthly active users.

The mechanics of how AI is translating into revenue deserve attention.

Customers using Rovo are growing their annual recurring revenue at roughly two times the rate of customers not using Rovo. This is not a product marketing claim. It is a disclosed financial metric from an earnings call. It means AI adoption is the single strongest predictor of customer expansion that Atlassian has ever had. Net revenue retention remains above 120% and has ticked up for several consecutive quarters.

The packaging model explains the mechanism. Rovo’s enhanced AI credits are bundled into Atlassian’s premium Teamwork Collection tier, not available in the standard subscription. In under nine months, more than 1,000 customers upgraded to Teamwork Collection and purchased over one million seats, with double-digit seat expansion rates compared to standalone buyers. Teamwork Collection customers use more than three times as many Rovo credits per user and have more than twice as many active agents.

AI has given Atlassian a new pricing lever that did not exist in 2022. Customers who want AI capabilities need to upgrade. When they upgrade, they use the platform more intensively, which increases switching costs and reduces churn. The Service Collection crossed $1 billion in ARR, with more than 65,000 customers, representing 50% of the Fortune 500. More than two-thirds of Service Collection customers now use it for non-IT functions including Human Resources, Finance, and workplace management.

The CEO’s framing of AI’s structural impact is direct: “Our customers don’t use models, they use applications. We don’t sell chips, we sell apps.” The Teamwork Graph that underpins Rovo, containing more than 100 billion objects and relationships, is the reason Atlassian’s AI produces better outputs than a generic model. The context is proprietary. The switching cost is not just contractual but operational.

Case Three: Monday.com | AI Expands the Addressable Market

Monday.com’s response to AI disruption has taken a different form than either Figma or Atlassian. Rather than defending an existing market, Monday has used AI to expand into new ones.

Q1 2025 revenue was $351 million, growing 24% year over year, while simultaneously delivering record GAAP operating income and the highest-ever adjusted free cash flow for a single quarter. Q2 2025 revenue grew 27% year over year to $299 million, adding a record number of net new customers with over $100,000 in ARR. Monday CRM reached $100 million in ARR just three years after launch.

The most important development for Monday’s long-term thesis is not the revenue growth rate. It is what the company has done with its addressable market.

Monday launched as a work management and project tracking tool. It has since expanded to CRM, service desk, marketing campaigns, and resource planning. New products now account for more than 10% of total ARR. AI is the mechanism enabling this expansion: Monday’s AI features allow teams to build custom workflows, automate repetitive processes, and connect data across its expanding product suite without requiring technical implementation.

The platform has crossed 1 billion dollars in ARR. Net dollar retention for all customers stood at 111% in H1 2025, with 90% gross margins and $174 million in adjusted free cash flow, representing 30% of revenue.

Monday’s model demonstrates a specific SaaS-AI dynamic that the disruption thesis did not anticipate: AI does not just protect existing revenue streams in enterprise software. It creates new ones by enabling companies to serve adjacent use cases with lower implementation friction than was previously possible. A company that already has Monday for project management can adopt Monday CRM without a full enterprise sales cycle. AI lowers the barrier to expansion within an existing platform relationship, which is structurally different from a new vendor trying to displace an incumbent.

Case Four: Constellation Software | The Meta Proof

Constellation Software does not build AI products. It acquires vertical market software companies and manages them as a permanent holding company. Its 2025 results are included here not as an AI story but as a structural argument.

CSU’s entire investment thesis depends on one proposition: that vertical SaaS businesses serving specific industries have switching costs high enough to generate durable, recurring free cash flow that grows predictably over time. If AI were genuinely commoditizing vertical SaaS, CSU’s acquisition pipeline would be drying up and its existing portfolio companies would be losing customers. Neither is happening.

CSU’s organic revenue growth has remained stable through the period of maximum AI disruption anxiety. The company continues to acquire vertical market software businesses at historical rates, deploying capital at returns that its management has described as consistent with prior years. The companies it acquires are still generating the kind of durable free cash flow that justifies CSU’s valuation model.

This is the meta evidence. CSU has 40 years of experience evaluating which software businesses have real switching costs and which ones are vulnerable to competitive displacement. The fact that it continues to find and acquire vertical SaaS companies at scale, and that those companies continue to perform, is a direct market signal that the AI disruption narrative has not materialized in the way the broader market feared for the segment CSU knows best.

The Three Patterns

Across these four companies, three patterns emerge that the disruption thesis did not account for.

Pattern one: AI is a monetization layer.

In each of these businesses, AI is not being added to retain customers who might otherwise leave. It is being packaged as a premium capability that existing customers pay more to access. Atlassian’s Rovo is in the Teamwork Collection upgrade. Figma’s AI credits are tiered. Monday’s AI features drive upgrades to higher-tier plans. The revenue implication of this packaging model is that AI creates a new upsell motion that did not exist before, without requiring the company to acquire a single new customer.

Pattern two: Enterprise customers with high switching costs are accelerating adoption.

Across all four companies, the growth rate of large enterprise customers (greater than $100,000 in ARR) is outpacing overall customer growth. Figma’s large customer base grew 48% year over year. Monday’s high-ARR cohort grew 39%. Atlassian’s enterprise sales generated a record number of deals above $1 million in ACV, nearly doubling year over year.

This is the opposite of what the disruption thesis predicted. If AI were making these platforms replaceable, enterprise customers would be the first to act: they have the procurement resources, the technical teams, and the financial incentive to switch to cheaper alternatives.

They are not switching. They are expanding.

Pattern three: Net Dollar Retention is stable to rising.

Figma’s NDR reached 139%. Monday’s NDR was 111%. Atlassian’s net revenue retention was above 120% and trending upward. An NDR above 100% means that even if a company stopped acquiring new customers entirely, revenue would continue to grow because existing customers are spending more. These numbers are moving in the wrong direction for the disruption thesis. If AI tools were substituting for these platforms, NDR would be falling as customers downgraded or churned.

Selecting SaaS Companies That Survive AI: A Framework

The evidence across these cases points to five characteristics that distinguish SaaS businesses that are structurally strengthened by AI from those that are genuinely vulnerable to displacement.

First: Workflow memory depth. The most durable SaaS businesses do not just store data. They store the accumulated logic of how an organization works: decision histories, escalation patterns, integration architectures, and team structures built over years. This organizational memory is proprietary to the platform and cannot be migrated without significant operational cost. A useful test: if a company cancelled its subscription today, would the loss of data be painful to recover, or catastrophic to replicate? The more catastrophic, the deeper the moat.

Second: Net Dollar Retention above 110%. NDR is the most direct financial signal of whether AI is helping or hurting a SaaS business. Above 110% means existing customers are expanding faster than any churn is occurring. Below 100% means the platform is losing ground with its existing base. In an AI environment where substitutes are proliferating, an NDR above 110% is evidence that customers are voting with their budgets to stay and spend more.

Third: AI as a revenue driver, not a defensive feature. The critical distinction is whether AI capabilities are creating new billing opportunities or being bundled into existing plans to prevent churn. Companies using AI to create premium tiers, consumption-based add-ons, or upgrade motions have a structural revenue tailwind. Companies adding AI features to defend their base without monetizing them are absorbing cost without generating return.

Fourth: Vertical depth or horizontal breadth, with no middle ground. Vertical SaaS businesses serving specific industries with deep regulatory, compliance, or workflow complexity have AI moats because general-purpose AI cannot replicate industry-specific institutional knowledge without years of proprietary data accumulation. Horizontal platforms that have become operating system-level infrastructure, where switching would require reconstructing the entire workflow architecture of an organization, have a different but equally powerful moat. SaaS businesses that are neither vertically deep nor horizontally broad are the most vulnerable: they provide point solutions that a general-purpose AI can increasingly substitute.

Fifth: Enterprise customer concentration trending higher. Enterprise customers with complex, multi-product deployments have the highest switching costs and the lowest tolerance for the operational disruption that platform migration causes. If a SaaS company’s large-customer cohort is growing faster than its overall customer base, that is evidence that the most sophisticated buyers in the market are making deliberate long-term commitments, not entertaining alternatives.

What This Means for Portfolio Positioning

The SaaS disruption thesis was based on a plausible analogy that turned out to be wrong in practice. The analogy was: AI will do to enterprise software what the internet did to physical retail. The reality is closer to the opposite. The internet commoditized physical distribution because the fundamental value of retail was in inventory access, and digital channels provided superior inventory access at lower cost. Enterprise SaaS’s fundamental value is not in feature access. It is in organizational infrastructure, workflow memory, and cross-team coordination. AI requires that infrastructure to function effectively, which means deeply embedded SaaS platforms are more valuable to AI adoption, not less.

The investment implication is that the selloff in enterprise SaaS driven by AI disruption fears has, in several cases, created entry points into businesses whose competitive positions have structurally improved. The screening criteria matter: workflow memory depth, NDR trajectory, AI monetization mechanics, and enterprise customer expansion rates are the variables that separate the genuinely durable businesses from those with real structural exposure.

Figma at $16 was priced for obsolescence. Its Q1 2026 results showed acceleration. That gap between narrative and data is where durable investment returns are found.

Disclaimer: This article is provided for informational and educational purposes only. It does not constitute financial or investment advice. The views expressed reflect analysis at the time of writing and may change without notice. Investors should conduct their own due diligence or consult a licensed advisor before making any investment decisions.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

The Number That Hides Inside the Index

The S&P 500 is a market-capitalization-weighted index. When Apple, Microsoft, and Nvidia go up, the index goes up, regardless of what the other 496 stocks are doing. This mechanical fact produces a specific and recurring distortion: index gains can look broad and healthy while the majority of the underlying companies are treading water or declining.

In 2023, the S&P 500 returned 26%. The equal-weighted version of the same index, where every company counts equally, returned approximately 12%. The 14-point gap was not noise. It was the entire story of that market in a single comparison. Seven companies, the group now known as the Magnificent Seven, were responsible for the difference. In June 2023, only 1% of S&P 500 stocks, representing just five companies, outperformed the median Magnificent Seven return on a rolling six-month basis.

In the fourth quarter of 2023, the Magnificent Seven accounted for 225% of the overall earnings growth of the S&P 500. That figure requires a moment to absorb. The seven largest companies in the index were growing earnings so fast that they offset net earnings contraction across the remaining 493 companies and still produced positive index-level growth. The index was healthy. The market was not.

Market breadth is the set of tools that measures this gap.

Share

What Market Breadth Actually Measures

Market breadth describes the degree of participation in a market move. When an index rises and most of its components are rising with it, breadth is wide. When an index rises but only a handful of companies are carrying it, breadth is narrow. The distinction matters because narrow breadth markets are structurally more fragile than broad breadth markets.

Three indicators do most of the analytical work.

Advance-Decline Line (A/D Line)

The A/D line is the cumulative daily count of advancing stocks minus declining stocks. Each trading day, you add the net number of advancers to the running total. A rising A/D line alongside a rising index confirms the move. A falling A/D line alongside a rising index is a divergence, and historically one of the most reliable warning signals in market analysis.

A practical example: in the months before the dot-com peak in March 2000, the S&P 500 continued to make new highs while the NYSE A/D line had already peaked and begun declining. The index was being carried by a shrinking number of technology names while the broader market had already turned. The divergence was visible in real time for anyone looking at breadth rather than index levels alone.

Percentage of Stocks Above Their 200-Day Moving Average

The 200-day moving average is the standard institutional dividing line between a stock in a long-term uptrend and a stock in a long-term downtrend. The percentage of S&P 500 components trading above their own 200-day moving average is a direct measure of how many companies are in structural uptrends.

The thresholds that matter:

• Above 70%: broad market participation, generally constructive

• 40% to 70%: mixed conditions, increasing selectivity required

• Below 40%: structural deterioration, the index level is likely misleading

• Below 20%: capitulation territory, historically associated with durable market lows

After the White House announced proposed tariffs on April 2, 2025, the percentage of stocks trading above their 200-day moving average dropped to 16%, before climbing back to 64% by the end of the following month. That swing from 16% to 64% in under 60 days is one of the fastest breadth recoveries on record, and it preceded the S&P 500’s 36% rally from its April 8 low.

New 52-Week Highs vs. New 52-Week Lows

This indicator tracks the daily count of stocks hitting new annual highs versus new annual lows. A healthy bull market produces an expanding population of new highs. When indexes are rising but new highs are declining, or when new lows are expanding alongside a flat or rising index, the market’s leadership is deteriorating even if the headline number looks fine.

As of September 2025, both the S&P 500 and Nasdaq were trading at all-time highs, but the share of stocks within those indexes hitting new 52-week highs was 4.6% for the S&P 500 and 3.5% for the Nasdaq. Compare that to late 2024 when indexes were rising alongside an expanding population of new highs. The September 2025 reading was a textbook narrow-breadth signal: record index levels sustained by a shrinking group of leaders.

The 2023-2025 Case Study: What Narrow Breadth Looks Like in Practice

The period from mid-2023 through late 2024 is the cleanest recent example of how narrow breadth distorts the investment environment, and what eventually happens when it resolves.

Phase 1: Extreme Concentration (Mid-2023)

By June 2023, the Magnificent Seven dominated S&P 500 performance to a degree not seen since the late 1990s technology concentration. The Magnificent Seven account for 34.8% of the S&P 500 as of May 12, 2026, a significant increase from 2016 when they made up 12.5% of the index.

As of April 2026, the S&P 500’s Herfindahl-Hirschman Index, the standard measure of market concentration, stood at 185 against a five-year average of 142, and the index’s effective number of independent positions was just 54 stocks despite having 503 holdings. In practical terms, the S&P 500 was behaving like a portfolio of 54 equally-weighted companies, not 503.

For passive index investors, this was essentially irrelevant. The index went up, the ETF went up, the statement looked good. For active investors and for anyone trying to understand the structural health of the market, it was a significant distortion.

Phase 2: The Consequences of Concentration

Narrow breadth creates a specific tail risk: when the few companies carrying the index face any fundamental challenge, the index drawdown is disproportionate to what the underlying fundamental deterioration would suggest, because the companies involved carry more index weight than their fundamental contribution would justify at normal valuations.

The week of April 21, 2025, Morgan Stanley’s equity-finance desk reported that 60% of every gross dollar sold by hedge funds worldwide over those five days was in Magnificent Seven stocks, despite the group representing nowhere near 60% of total gross exposure. When concentrated positions unwind, they unwind fast, because everyone who loaded up for the same reasons unloads for the same reasons at the same time.

Phase 3: Breadth Recovery and What It Signals

On a rolling six-month basis through June 30, 2025, the percentage of stocks outperforming the Magnificent Seven median reached 51%, representing 251 securities, up from just 1% in June 2023. This recovery in breadth, from historic concentration to something approaching normal participation, is a constructive structural signal. It means the earnings cycle is broadening, that companies outside the mega-cap technology group are generating enough fundamental improvement to attract capital, and that the index’s future performance is less dependent on whether seven specific companies keep growing at extraordinary rates.

In 2025, the S&P 500 gained 14%, but excluding the Magnificent Seven, the remaining 493 stocks returned just 8%. The median stock in the S&P 500 remained 19% below its all-time high while the index was at an all-time high. The breadth recovery had begun, but it was not yet complete.

What Breadth Does Not Tell You

Market breadth is frequently misused as a market timing tool, and this misuse produces poor results. Three clarifications on what breadth indicators do and do not do.

Narrow breadth does not predict a crash. The A/D line can diverge from index highs for months or even years before a correction. In 1999, the NYSE A/D line peaked in April, a full eleven months before the S&P 500 peaked in March 2000. An investor who exited equities on the A/D divergence signal missed 25% of gains before the correction arrived. Breadth divergences are risk signals, not timing signals.

Wide breadth does not guarantee continued gains. A broad, healthy market can still correct on valuation, macro shock, or earnings disappointment. Breadth describes participation, not valuation. October 1987 had relatively healthy breadth before the crash.

The relevant question breadth answers: Is this market’s strength structural and broad, or is it concentrated and therefore fragile? Structural breadth environments reward long-term equity exposure. Concentrated breadth environments reward selectivity and demand a more careful look at what is actually in the portfolio versus what the index level suggests.

How to Use Breadth in a Long-Term Portfolio Framework

For long-term investors, breadth is most useful as a risk calibration tool rather than a trading signal. The practical applications are straightforward.

When breadth is wide and recovering:

• Broad equity exposure becomes more reliable; the rising tide is actually lifting most boats

• The Engine and Chassis tiers of a structured portfolio can carry more weight with higher confidence

• Individual stock selection matters less, because the macro environment is supportive across a wider range of companies

When breadth is narrow and deteriorating:

• Index level performance is increasingly unreliable as a proxy for portfolio performance

• Selectivity becomes the primary driver of returns; owning the right companies matters far more than owning equities generally

• The Tire and Spare Tire tiers of a structured portfolio provide increasing structural value, not because a correction is imminent, but because the risk distribution across the market has become more asymmetric

When breadth is at extremes:

At the lows, below 20% of stocks above their 200-day moving average, the market is pricing capitulation. Historically, these readings have preceded some of the strongest forward returns available in equity markets. They are not comfortable moments to deploy capital, which is precisely why they tend to be rewarding ones.

At the highs, when fewer than 5% of stocks are making new 52-week highs while the index is at records, the narrowing of leadership is a warning that the index’s performance has become increasingly dependent on a small number of names sustaining extraordinary valuations. Historically, extreme concentration has occurred only 16% of the time since 1990. In the 12 months following such periods, the equal-weight S&P 500 outperformed the market-cap-weighted index 43% of the time, with average excess returns of 3.1%.

The Current Breadth Environment

As of mid-2026, the breadth picture is mixed and improving. The April 2025 tariff shock produced one of the sharpest breadth collapses in recent history, but the subsequent recovery has been equally sharp. The percentage of stocks above their 200-day moving average recovered from 16% to 64% within roughly 60 days of the April low.

The Magnificent Seven’s weight in the S&P 500 has lessened somewhat from its peak, and the broadening of earnings growth beyond the mega-cap technology group is continuing. The equal-weight index has begun to close some of its performance gap against the cap-weighted index. These are constructive structural signals.

The residual risk is concentration. Seven companies still represent approximately 30% of the S&P 500, and the index still behaves like a portfolio of roughly 54 equally-weighted stocks despite having more than 500 constituents. The breadth recovery is real but the concentration legacy has not fully resolved. For investors running actively managed portfolios, the most important implication of the current environment is that returns are becoming more broadly available, but the structural fragility that comes from extreme concentration has not fully unwound.

In practical terms: the environment rewards selectivity, favors quality businesses with durable earnings over speculative positions, and continues to penalize passive exposure to the cap-weighted index in proportion to whatever the top seven companies decide to do next quarter.

Summary: Three Things to Remember

First: Index performance and market performance are frequently different things. Check the equal-weight version of any index you are evaluating before drawing conclusions about the health of the underlying market.

Second: The three breadth indicators that matter most are the Advance-Decline line (direction and divergence), the percentage of stocks above their 200-day moving average (structural trend participation), and the new highs versus new lows ratio (leadership quality). Together they describe the market’s internal structure in ways the index level cannot.

Third: Breadth is a risk calibration tool. Wide breadth lowers the structural risk of broad equity exposure. Narrow breadth raises it, and shifts the premium toward selectivity and quality. The investors who understand this distinction build portfolios that perform through cycles rather than in spite of them.

This article is part of the Thesis Library series. All Thesis Library content is free to access. For portfolio-specific analysis of how the current breadth environment affects individual positions, see Thesis Weekly, available to Thesis Insider subscribers.

Disclaimer: This article is provided for informational and educational purposes only. It does not constitute financial or investment advice. Investors should conduct their own due diligence or consult a licensed advisor before making any investment decisions.

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

Market Lens

Share

Markets ended the week almost exactly where they started. That flatness is the most misleading data point of the week.

What happened between Monday’s open and Friday’s close was a collision between two forces that cannot both be right at the same time: a diplomatic breakthrough that suggested the worst of geopolitical risk was fading, and an inflation print that confirmed the worst of geopolitical risk is still fully embedded in the price of every gallon of gasoline, every airline ticket, and every food item that requires a supply chain running through the Middle East. The S&P 500 opened the week with optimism and closed it with a 1.24% selloff on Friday. The net weekly change was near zero. The internal damage was not.

President Trump’s three-day state visit to China, from May 13 to May 15, produced agreements that markets have been waiting months to see: a joint statement committing both nations to keeping the Strait of Hormuz open, the formation of two new bilateral institutions covering trade and investment, a Chinese commitment to purchase 200 Boeing aircraft with options that could extend to 750, a minimum of $17 billion in annual agricultural purchases, and an opening on rare earth and critical mineral supply chain coordination. For a market that has been pricing existential risk into the US-China relationship since the beginning of the year, the summit functioned as a geopolitical pressure valve. Risk assets rallied in the first half of the week on the read that the trade war’s tail risk had been structurally capped.

Then Tuesday morning arrived.

April CPI came in at 3.8% year-over-year, the highest reading since May 2023, against a consensus of 3.7%. The monthly gain was 0.6%, the same pace as March. Energy drove over 40% of the headline increase, with gasoline up 28.4% year-over-year and fuel oil up 54.3%. But the number that matters most for Fed policy is the one that excludes energy: core CPI accelerated to 2.8% year-over-year, with a monthly gain of 0.4%, the highest monthly core reading since January 2025. Real average hourly wages fell 0.3% annually for the first time since April 2023. For the first time in three years, Americans’ paychecks are no longer keeping pace with prices. The data landed on Tuesday morning, and Treasury yields immediately repriced. By Thursday, following a similarly hot PPI print, the 10-year had climbed to 4.56%, a new year-to-date high. The 30-year briefly touched 5.1%. CME futures traders raised the probability of a Fed rate hike by year-end to approximately 30%.

The macro logic of the week is a single sentence: the Trump-Xi summit reduced the political risk premium on US-China trade, but it cannot reduce the physical energy shock of a war that has closed the Strait of Hormuz for over ten weeks and removed more than 10 million barrels per day from global supply. WTI crude closed at $105.42 on Friday, up 11% for the week, with the day’s gain alone exceeding 4%. The IEA warned this week that even if fighting ended next month, the global oil market could remain materially undersupplied through October. The summit communique said both leaders “want the straits open.” Iran was not at the table.

This creates the macro framework that governs the portfolio for the foreseeable future. Geopolitical progress at the diplomatic level does not translate immediately into energy price relief at the pump. WTI at $105 means CPI at or above 3.8% through the summer. CPI above 3.8% means there could be no rate cuts in 2026 and a genuine probability of a hike in early 2027. A genuine probability of a hike in 2027 means the long-duration valuation multiple that has sustained the top of the equity market since 2023 is under structural pressure, not cyclical pressure.

Against this backdrop, the most consequential unreported story of the week is that NVIDIA CEO Jensen Huang was added to the US presidential delegation to Beijing at the last moment. He was not on the original manifest. His presence signals that the negotiation over AI chip supply chains, export restrictions, and the technology decoupling between the US and China is being conducted at a level well above what public statements reflect. NVIDIA reports its own quarterly results next Wednesday, May 20. The Blackwell architecture order book and the Vera Rubin roadmap disclosure will determine whether the AI infrastructure capex cycle continues to accelerate or whether the market begins to discount a demand plateau. Every AI-adjacent holding in the portfolio, from MSFT to AMD to FICO’s compute-intensive analytics infrastructure, has some exposure to that outcome.

The week’s final signal is gold. It closed at approximately $4,788, up 1.2% even as the dollar strengthened nearly 1.4% against a basket of currencies. Gold and the dollar do not typically rise together. When they do, it is because the demand for real assets is outpacing the normal currency correlations, which is itself a signal that institutional capital is hedging against something more fundamental than a short-term rate move. In 2026, that something is an inflation regime that is being driven by war, not by central bank error, and that therefore cannot be resolved by central bank tools alone.

This Week’s Signal: The summit bought goodwill. The oil market bought nothing. Diplomacy and inflation are on different timelines, and the bond market just voted on which one it believes.

Super Investor 13F | Q1 2026 Positioning Update

This quarter’s 13F filings produced one of the most analytically revealing datasets of the past decade. Five of the most sophisticated institutional investors in the world looked at the same market, the same inflation environment, the same AI infrastructure cycle, and arrived at contradictory conclusions on three of the most widely held stocks in the world: Alphabet, Amazon, and Microsoft.

Berkshire Hathaway | Greg Abel’s First Reconstitution

Abel used his first quarter as portfolio steward to execute a comprehensive restructuring, reducing the 13F equity portfolio from 40 positions to 26. The explanation, per the Wall Street Journal, is structural: the majority of the liquidated positions were managed by Todd Combs, who departed Berkshire for JPMorgan in December. Abel cleared the slate entirely.

New Positions

• Delta Air Lines (DAL): New buy, betting on the free cash flow inflection as the airline industry’s heavy capex cycle rolls off

• Macy’s (M): Contrarian value buy, with the underlying real estate footprint likely the primary thesis rather than the retail operating business

• Alphabet Class C (GOOG): Initiated alongside the GOOGL addition

Added

• Alphabet (GOOGL): +204%, adding 36.4 million shares, moving it into the top seven holdings

• New York Times (NYT): +199%, ownership rising to 9.4%, treated as a high-retention digital subscription business

• Lennar (LEN): +43%, adding to the homebuilder position

Fully Liquidated

• Amazon (AMZN)

• Visa (V)

• Mastercard (MA)

• Costco (COST)

• HEICO (HEI)

• Domino’s Pizza (DPZ)

• UnitedHealth (UNH)

• Charter Communications (CHTR)

• Diageo (DEO)

• Pool Corp (POOL)

• Lamar Advertising (LAMR)

• Allegion (ALLE)

• AON (AON)

• Formula One Group (FWONK)

• Atlanta Braves Holdings (BATRK)

• Liberty Latin America (LILAK)

Reduced

• Chevron (CVX): -35.17%, with proceeds rotating toward Alphabet

The Alphabet addition at +204% is Abel’s clearest statement of conviction. At approximately 20 times forward earnings, with the largest digital advertising network in the world and the third largest cloud platform, Abel is making a straightforward valuation argument: the market has priced AI disruption risk into Alphabet at a level that significantly exceeds the actual probability of that disruption materializing within a five-year horizon.

The HEICO exit is directly relevant to the watchlist. Berkshire’s position was a Combs allocation rather than a core Abel conviction, which means its removal is procedural rather than analytical. It does not change the fundamental thesis on HEICO’s PMA parts moat or its ETG defense business. But it removes an institutional anchor from the stock’s valuation, which is a factor in sizing any entry position after the May 27 earnings report.

Pershing Square | Bill Ackman

Ackman’s quarter reflects a single thesis: identify the most defensible digital toll roads in enterprise technology and concentrate into them while eliminating businesses where the competitive moat is structurally ambiguous.

New Positions

• Microsoft (MSFT): Approximately $2 billion initiation, making it the AI infrastructure centerpiece of the fund

Added

• Amazon (AMZN): +19%, buying 1.84 million additional shares

Fully Liquidated

• Alphabet (GOOG and GOOGL): Complete exit, abandoning the digital advertising thesis in favor of the enterprise infrastructure thesis

Ackman’s simultaneous MSFT initiation and GOOGL exit is an explicit choice between two types of AI monetization. He is betting that enterprise workflow lock-in, the kind generated by Azure contracts, Office 365 seat counts, and Copilot integrations, produces more durable revenue than advertising-based discovery monetization, which is structurally exposed to AI-generated search disruption. This directly validates the portfolio’s MSFT add over the past two weeks.

Duquesne Family Office | Stanley Druckenmiller

Druckenmiller concentrated the portfolio into what he views as the most physically irreplaceable layer of the AI infrastructure stack, custom silicon, while exiting businesses he believes are exposed to disruption from that same stack.

New Positions

• Broadcom (AVGO): Nearly 200,000 shares, targeting the custom ASIC business that designs dedicated AI accelerators for Google and Meta

• Caris Life Sciences: 1.89 million shares in an unlisted AI-driven precision oncology business

Fully Liquidated

• Alphabet (GOOGL): Complete exit

• Amazon (AMZN): Complete exit

Reduced

• Nvidia (NVDA): Continued trimming at elevated prices, locking in gains on the general-purpose GPU position

The Broadcom initiation is the most analytically precise move of the quarter. Custom silicon sits above general-purpose compute in the efficiency hierarchy: it delivers more performance per watt for a specific workload at a lower total cost of ownership, which is the most relevant metric when electricity is the binding constraint on AI scaling. Druckenmiller is rotating from the infrastructure layer that succeeded in phase one of the AI cycle toward the infrastructure layer he believes will dominate phase two.

Appaloosa Management | David Tepper

Tepper’s thesis is operating leverage: find businesses that have already absorbed their fixed cost bases and are positioned to convert incremental revenue directly into free cash flow.

Added

• Amazon (AMZN): +98.2%, buying 4.32 million shares, making it 15.16% of the fund at nearly $900 million

• Uber (UBER): +242%, treating the platform as a physical mobility toll road with a largely sunk cost structure

• Micron (MU): +11%, targeting memory chip cycle asymmetry

• Intel (INTC): New buy of 8 million shares, as a low-cost option on the foundry business finding customers as TSMC capacity constraints tighten

Reduced

• Nvidia (NVDA): Staged profit-taking at elevated levels

• Alphabet (GOOGL): Moderate reduction

• Alibaba (BABA): Partial exit from China exposure

The Amazon doubling is Tepper’s most concentrated statement. He is not buying the AI story or the advertising story. He is buying the margin expansion story: a cost structure built over the past four years that has now stabilized, positioned to convert the next several years of revenue growth into disproportionate free cash flow as capital expenditure intensity normalizes.

Third Point | Daniel Loeb

Loeb took the most explicitly contrarian position of the quarter, buying the stock that three other managers were exiting simultaneously.

New Positions

• Alphabet (GOOGL): 175,000 shares, directly counter to Ackman and Druckenmiller

Reduced

• Super Micro Computer (SMCI): Significant reduction, trimming overheated hardware exposure

Loeb’s Alphabet thesis is valuation-driven rather than competitive. At approximately 20 times forward earnings with a clean balance sheet and compounding free cash flow, he is arguing that the market has already priced in a worst-case AI disruption scenario that is unlikely to materialize at the speed or magnitude being assumed. He is not making a claim about who wins the AI search war. He is making a claim that the price of losing is already embedded in the stock, and that the price of that pessimism is too high.

Thesis Portfolio Desk

This Week’s Focus

Read more

If you haven’t explored our previous research, you may revisit some of our earlier due diligence reports and thematic notes below. Each piece reflects the same thesis-driven framework we apply across every investment case.

For a closer look at how these principles translate into real positions, explore the 👉 Sample Portfolio — a demonstration of thesis-driven allocation and disciplined compounding.

Opening

On May 13, 2026, Air Force One touched down at Beijing Capital International Airport. Chinese Vice President Han Zheng, dispatched personally by Xi Jinping, stood at the base of the steps to receive Donald Trump. The highway from the airport was lined with American and Chinese flags. City skyscrapers were lit with characters reading “Beijing Welcome.”

The pageantry was deliberate. So was the guest list.

Trump did not arrive in Beijing alone, or with diplomats. He arrived with Elon Musk, Jensen Huang, Tim Cook, Larry Fink, Stephen Schwarzman, David Solomon, Jane Fraser, Kelly Ortberg, and more than a dozen other chief executives representing companies with a combined market capitalization in the trillions. The composition of that delegation is not incidental. It is the agenda, expressed in human form. This visit is, at its structural core, a negotiation about who controls the economic infrastructure of the next decade: the semiconductor supply chains, the AI compute layer, the aircraft manufacturing relationships, the agricultural commodity flows, and the rare earth minerals that everything else depends on.

For long-term investors, the Beijing summit is not a geopolitical event to observe from a distance. It is a capital allocation signal of the first order.

Share

The Delegation as a Stock Screen: Nine Years of Performance

The 17 companies Trump brought to Beijing this week are not a random sample of American business. They represent the businesses whose competitive positions are most structurally significant to the U.S.-China bilateral relationship. And if you had simply bought all of them in equal weight on January 20, 2017, the first day of Trump’s first presidency, the data tells a striking story.

The performance was not evenly distributed across categories. The Hardware Engine group, which includes Nvidia, Tesla, Apple, Micron, and Qualcomm, produced the most extreme outcomes in both directions. Nvidia’s return is so large that it distorts any equal-weight average. Even excluding Nvidia, the Hardware group averaged well over 500% across five names. The Financial Chassis group, which includes Visa, Mastercard, Blackstone, BlackRock, Goldman Sachs, and Citi, produced more consistent but still compelling results. Every name in this group generated positive returns, ranging from Citi’s +9% to Blackstone’s +348%. These are businesses with structural moats, not cyclical stories, and the nine-year return distribution reflects that. The Frontier Turbo group was the most bifurcated. Meta returned +289% while Illumina fell -25%, a reminder that structural positioning within a category matters as much as the category itself.

The most important finding is not the absolute returns. It is the selection logic that produced them. The companies that a sitting U.S. president brings into the room during high-stakes trade negotiations tend to share a specific set of characteristics: genuine bilateral economic dependency, structural competitive advantages that neither side can easily replicate, and business models whose success or failure has consequences for both economies simultaneously. Those characteristics are not a bad definition of a long-term investment screen. The Beijing delegation did not choose these companies because they had already performed well. They chose them because they are structurally important. That the performance followed is not a coincidence.

The Context: From Escalation to Managed Tension

Understanding what this visit means requires understanding what preceded it.

The relationship between Washington and Beijing deteriorated sharply through 2025 in a cycle of escalating tariffs and retaliatory export controls. The United States raised tariffs on Chinese goods to 145%, then paused them. China responded with counter-tariffs and, more consequentially, with restrictions on rare earth exports, a move that struck at the earliest stages of global manufacturing rather than at finished goods. The leverage was asymmetric and deliberate. China controls the extraction, processing, and export of more than 70% of the world’s most critical industrial minerals. When Beijing restricted rare earth exports, it was not making a trade argument. It was demonstrating a structural dependency that no tariff schedule could immediately resolve.

The Busan Summit in October 2025 produced a temporary truce: both nations suspended the most punitive tariffs and export controls for one year. The Beijing summit this week is the follow-on. Both sides are arriving with a shared interest in converting that temporary truce into something more durable, without conceding the structural leverage each has worked to accumulate.

The Iran war complicates this calculus. The conflict, which began in February 2026 when the United States and Israel launched coordinated strikes, has dragged on far longer than the Trump administration anticipated. Iran is China’s largest oil trade partner, and Beijing’s continued purchase of Iranian crude gives it a degree of geopolitical insulation from American pressure. At the same time, China has provided assurances that it will not transfer weapons to Iran, a commitment that Defense Secretary Pete Hegseth attributed to the “strong and direct relationship” between the two presidents. The war has not derailed the summit. It has reshaped the bargaining dynamic in ways that favor Beijing.

The Council on Foreign Relations, in an analysis published before the summit, described the meeting as an effort to stabilize bilateral relations rather than resolve long-standing disputes. That framing is correct and important. The agenda is not resolution. It is management, and managed tension between the world’s two largest economies is a profoundly different investment environment than the escalating confrontation of 2025.

The Delegation: Who Went to Beijing and Why It Matters

The CEO delegation is the most instructive signal of what the Trump administration considers negotiable and what it considers worth fighting for. Each company represented carries a specific economic relationship with China, and each presence at the table reflects a specific set of interests.

Elon Musk | Tesla and SpaceX

Musk traveled on Air Force One itself, alongside Jensen Huang, a distinction that communicates proximity to the principal rather than merely participation in the delegation. Tesla’s relationship with China is foundational to its economics. The Shanghai Gigafactory is one of Tesla’s most productive manufacturing facilities globally, and China represents one of its largest consumer markets. Tesla also depends on Chinese suppliers for battery components and rare earth materials used in electric motors. Musk’s presence signals that the administration views Tesla’s China operations as a legitimate interest to protect, not a vulnerability to exploit as leverage.

The strategic question for Tesla is whether the summit produces any movement on the tariff and component supply chain issues that have elevated production costs on both sides of the Pacific. A meaningful tariff reduction or an extension of the rare earth export exemption would be directly beneficial to Tesla’s margin structure.

Jensen Huang | Nvidia

Huang’s presence is the most consequential single development of the delegation, and he almost did not come. Initially absent from the invite list, he joined at the last minute during a refueling stop in Alaska. The symbolism is impossible to miss: Nvidia’s absence would have suggested that the semiconductor issue was off the table. His presence suggests it is not.

Nvidia has identified China as a $50 billion market opportunity for its AI chips, and that opportunity has been largely foreclosed by U.S. export controls. The Trump administration has allowed the sale of some less advanced chips, taking a 25% fee on those transactions, but the Chinese government has itself restricted the purchase of even these approved versions, creating a regulatory standoff on both sides of the Pacific. Huang’s participation in the summit does not guarantee a change in policy, but it puts the chip access question directly into the room where the decisions are being made.

The investment implication here is significant. Any meaningful easing of chip export controls to China, even partial, would expand Nvidia’s addressable market materially. Conversely, any tightening would accelerate Chinese domestic chip development, which is itself a long-term structural concern for the semiconductor supply chain.

Tim Cook | Apple

Cook is participating in what is expected to be his final major diplomatic effort before handing leadership to John Ternus on September 1. That timing gives this trip a particular weight. Apple’s manufacturing relationship with China is one of the most complex bilateral dependencies in global business. Approximately 90% of iPhones are assembled in China, primarily by Foxconn. Apple recently committed to a $600 billion investment in the United States over four years, a move that secured critical tariff exemptions for its products. That commitment buys diplomatic goodwill but does not change the fundamental manufacturing geography in the near term.

For Cook, the Beijing visit is about protecting the supply chain relationships and market access that underpin Apple’s current economics while managing the longer-term pressure to diversify production toward India, Vietnam, and other geographies. The most relevant near-term outcome for Apple investors is any clarity on tariff treatment of consumer electronics and components, and any signal about whether the Chinese consumer market will remain accessible to Apple’s premium product lineup amid rising domestic competition from Huawei and others.

Kelly Ortberg | Boeing

The commercial aviation relationship between the United States and China is one of the most straightforward bilateral trade stories of the current moment. China is one of the largest commercial aircraft markets in the world, and Beijing has been ordering almost exclusively from Airbus since imposing 125% import taxes on American goods in April 2025. Ortberg’s presence is specifically tied to the widely anticipated announcement of a major Boeing aircraft order, which congressional delegation member Steve Daines described with characteristic plainness: “Boeing, beef and beans.”

A large Chinese aircraft order would be a direct positive for Boeing’s order book and cash flow generation, both of which have been under pressure from regulatory challenges and the prior tariff environment. It would also be a tangible, photogenic deliverable that both governments can present as evidence that the summit produced concrete economic results.

Larry Fink | BlackRock

BlackRock manages approximately $10 trillion in assets globally and has been one of the most persistent advocates for sustained Western engagement with Chinese capital markets. Fink’s presence reflects the financial sector’s specific interest in the summit: access to China’s domestic financial markets, the treatment of U.S. financial institutions operating in China, and the broader question of whether dollar-denominated assets retain their role in Chinese reserve management. Any opening of Chinese capital markets to foreign asset managers, or any relaxation of restrictions on cross-border financial flows, would be directly relevant to BlackRock’s China ambitions.

Stephen Schwarzman | Blackstone

Schwarzman has decades of personal relationships in Beijing that predate almost anyone else in the current American business establishment. Blackstone has substantial real estate and private equity exposure in China, and Schwarzman’s presence is as much about the maintenance of those long-standing relationships as it is about any specific policy outcome. His inclusion signals that the administration is treating the summit as an opportunity for broad economic engagement rather than a narrow negotiation over specific trade categories.

David Solomon | Goldman Sachs and Jane Fraser | Citi

The presence of the two largest U.S. investment banks reflects the financial sector’s direct interest in cross-border capital flows, M&A activity, and market access. Goldman and Citi both have significant operations in China that have navigated the regulatory environment of the past several years under substantial uncertainty. Any stabilization of the bilateral relationship reduces the tail risk that those operations face from regulatory retaliation or forced restructuring.

The Broader Delegation

The remaining executives from Mastercard, Visa, Qualcomm, Micron, GE Aerospace, Cargill, Coherent, Illumina, and Meta represent a comprehensive cross-section of American economic interests in China. Mastercard and Visa have long sought access to China’s domestic payment processing market, which is dominated by Alipay and WeChat Pay. Qualcomm and Micron have deep semiconductor supply chain dependencies on Chinese manufacturing and customers. GE Aerospace is the American engine manufacturer most exposed to commercial aviation in China. Cargill represents the agricultural commodity trade that is consistently the most politically visible and easily quantifiable element of U.S.-China trade negotiations.

The Macro Environment: What the Numbers Are Saying

The visit is occurring against a specific macroeconomic backdrop that shapes both the urgency of the negotiations and the investment implications of their outcomes.

U.S. Inflation: The Persistent Constraint

Core U.S. inflation remains above the Federal Reserve’s target, with core PCE running at approximately 3.1% as of early 2026 according to IMF data. The Iran war has introduced an energy price shock that has complicated the Fed’s policy path. WTI crude oil has been trading near or above $94 per barrel, and the IMF’s adverse scenario for 2026 models oil prices rising 80% from January levels if the conflict escalates further. Tariffs on Chinese imports have contributed to supply shocks that sustain consumer price pressure across electronics, household goods, and manufactured components.

A meaningful reduction in U.S.-China tariffs, or an extension of the current truce beyond its one-year expiration, would be directly disinflationary in categories that have been persistently elevated. The Federal Reserve’s ability to cut rates, which the market has been pricing in as a near-term catalyst for equity valuations, is constrained by any sustained inflationary pressure. A tariff-driven reduction in goods price inflation would materially improve the Fed’s ability to ease policy without risking a second wave of above-target inflation.

China’s Trade Position and the Rare Earth Leverage

China’s export controls on rare earth minerals have evolved from a reactive countermeasure to a strategic asset. China controls the refining of 19 of the 20 most important industrial minerals globally, and that dominance was demonstrated in 2024 when Beijing drove lithium prices down 80% through predatory pricing, forcing Western mining projects to close before U.S. domestic supply chains could be established. The rare earths deal, confirmed as still in force by U.S. officials just days before the summit, is one of the most sensitive files on the table.

An extension or formalization of the rare earths arrangement would reduce the structural vulnerability of U.S. manufacturing in semiconductors, defense systems, electric vehicles, and medical equipment. The Busan agreement suspended China’s export restrictions for one year. A permanent or multi-year arrangement would remove one of the most significant tail risks facing advanced manufacturing supply chains globally.

Agricultural Commodities: The Most Visible Deliverable

The congressional delegation member who visited Beijing the week before Trump characterized the expected deliverables as “Boeing, beef and beans.” This framing captures a consistent pattern in U.S.-China trade negotiations: China uses agricultural purchases as a visible, quantifiable signal of goodwill that generates domestic political support in American farm states without requiring structural changes to its own industrial policy.

A large-scale Chinese commitment to purchase U.S. soybeans, corn, wheat, and beef would support agricultural commodity prices in the near term and would be significant for companies across the food production and agricultural logistics value chains.

Investment Implications: What Investors Should Watch

An Observation: The Delegation as a Stock Screen

There is one more finding worth noting, and it did not come from the geopolitical analysis. It came from the data.

Share

The caveat is important: by 2026, 79% of that portfolio’s value was Nvidia alone. The “diversified” delegation basket effectively became a single-stock bet on AI semiconductors. Without rebalancing, the other 16 companies were nearly irrelevant to the final outcome.

But the underlying observation is worth taking seriously. The companies Trump chooses to bring into rooms where global economic architecture is being negotiated are not chosen randomly. They represent the businesses whose competitive positions are most structurally significant to the bilateral relationship, which means they are also the businesses most likely to benefit from any stabilization or improvement in that relationship. Nvidia, Apple, Tesla, Visa, Boeing, and the others in Beijing this week were not invited because they are important to Trump personally. They were invited because they are important to the economy he is trying to manage.

Historically, the companies that a sitting U.S. president brings into high-stakes trade negotiations tend to be the ones with the strongest structural moats, the deepest bilateral dependencies, and the most consequential policy exposure. Those characteristics are not a bad starting point for a long-term investment screen.

The Beijing delegation is not a portfolio recommendation. But it is a useful signal about which businesses the most powerful economic actors in the world consider too important to leave out of the room.

The summit is ongoing as this analysis is published, and specific outcomes are not yet known. The investment framework for evaluating results should focus on four categories of potential outcome.

Scenario One: A Broad Trade Stabilization Agreement

The most likely outcome is an extension and formalization of the Busan truce, potentially with additional commitments on agricultural purchases, Boeing aircraft orders, and some easing of the most restrictive tariff layers. This scenario is constructive for: companies with significant China revenue exposure across technology, finance, and manufacturing; agricultural commodity prices and the companies that produce and trade them; equity markets broadly, through the reduction in trade-war tail risk that has been a persistent headwind to multiple expansion; and the Federal Reserve’s policy flexibility, through the disinflationary effect of reduced goods tariffs.

Scenario Two: Technology and Semiconductor Progress

Any signal of movement on chip export controls, even incremental, would be significant for the semiconductor sector. Nvidia is the most obvious direct beneficiary, but easing of controls would also affect Qualcomm, Micron, and the broader semiconductor supply chain. Conversely, any tightening of controls would accelerate Chinese domestic chip development and increase the strategic importance of non-Chinese semiconductor capacity.

Scenario Three: Rare Earth Supply Chain Clarity

An extension or multi-year formalization of the rare earth arrangement would reduce the structural vulnerability of industries that depend on Chinese mineral processing. This affects defense contractors, electric vehicle manufacturers, semiconductor companies, and renewable energy infrastructure builders simultaneously.

Scenario Four: Financial Market Access

Any opening of Chinese capital markets to U.S. financial institutions would be directly relevant to BlackRock, Goldman Sachs, Citi, and other financial firms with existing or aspirational China operations. This is the least likely near-term outcome, as Chinese financial market liberalization has been a long-standing aspiration of U.S. negotiators with limited historical progress.

The Structural Investment Judgment

For long-term investors, the Beijing summit represents not a trading event but a signal about the direction of a structural relationship that will determine the competitive landscape of global business for the next decade.

The most important observation is this: both governments arrived in Beijing wanting stability. China’s goal, in the words of Trivium China senior analyst Joe Mazur, is to put the Busan truce “on a stabler, longer-lasting footing.” The United States is simultaneously managing an active military conflict in Iran, navigating domestic inflation pressure, and approaching an economic cycle that requires trade clarity to allow capital investment to proceed.

These aligned incentives do not guarantee a transformative outcome. They do make a managed, stable relationship more likely than either continued escalation or dramatic resolution. For investors with a multi-year horizon, managed stability between the world’s two largest economies is a positive environment for the businesses that operate across both. Companies with durable competitive advantages, strong pricing power, and exposure to the structural themes on the summit agenda, from semiconductor infrastructure to agricultural trade to financial services, are positioned to benefit from whatever stabilization this summit produces.

The most expensive scenario for investors is not the one where the summit produces limited results. It is the one where the summit fails and the relationship deteriorates toward the confrontational dynamic of 2025. The composition of the delegation, the diplomatic groundwork laid in the months before arrival, and the aligned incentives of both governments suggest that outcome is the least likely of the available alternatives.

Trump’s own assessment, posted on Truth Social before departure: “Great things will happen for both Countries.”

For once, the market’s job is simply to assess whether that is correct.

Disclaimer: This article is provided for informational and educational purposes only. It does not constitute financial or investment advice. The views expressed reflect analysis at the time of writing and may change without notice. Investors should conduct their own due diligence or consult a licensed advisor before making any investment decisions.