| Account | Shares | Price | Value | % of acct | Cost/sh | Gain $ | Gain % | Target |
|---|---|---|---|---|---|---|---|---|
| 401K | 14 | $453.60 | $6,350 | 0.26% | $392.16 | $860 | +15.7% | — |
| HSA | 19 | $431.71 | $8,202 | 7.60% | $397.94 | $642 | +8.5% | — |
| RLT | 1 | $453.60 | $454 | 0.03% | $466.28 | $-13 | -2.7% | — |
| ROTH | 2 | $453.60 | $907 | 0.35% | $482.06 | $-57 | -5.9% | — |
| Total | 36 | $15,914 | 0.35% | $1,432 | +9.9% | $1000 |
In short: Referenced only — held, no new view. Weight ~2.85%. It was named alongside Fairfax for an increase on 1 September, but it is still the third-smallest position and is not added to here. Sheet: EPS 17.83 → 23.03 + 0.9% = 9.80%/yr.
In short: Very resilient — "a lot of investors have this wrong." Its value is authoritative output, not reduced friction: ratings are regulated and written into contracts, and proprietary data "become[s] more useful when an agent can query them." AI research summaries raise demand for S&P's data: "Generic analysis may become cheaper. Trusted, licensable, auditable inputs do not."
Some investors fear AI will replace S&P Global because AI can write research summaries. Carlson thinks that gets it backwards: those summaries are built from S&P's data, so AI raises demand for it. Credit ratings are also required by regulators and written into loan contracts, and a bank lending billions needs a trusted, checkable source, not a cheap website. The value comes from authority, which an agent can't copy.
38:51S&P Global describes subscription revenue built from data, valuation services, analytics, research, and rating information alongside ratings, and index linked economics. So, a lot of investors have this wrong. They think that because AI can put together good research summaries that it will pressure SMB Global, but that's just the opposite.
In short: STRONG BUY (portfolio). Fair value $607.4 vs $411.76 (32.2% under); fwd PE 22.3 vs 29.5 (24.4% under); RDCF 10.4% vs 11.5% (+1.1pp) — the thinnest reverse-DCF margin of the seven Strong Buys that pass it. YTD −19.7%.
In short: The spin-off leg of the special-sits book, and a clean multiple argument. "S&P Global shares rallied last week on news that it was looking to spin off Capital IQ, which is the most AI-exposed part of the business. So that's obviously positive because Moody's trades at a higher multiple than SPGI." Shares "traded around 450" at the end of the week and "closed on the fourth around 443 after hours." The comp set: "FactSet is probably the best comp for the SPGI Cap IQ spin, which trades at around 15 times PE and 12 times EBITDA. So the Cap IQ business should trade at a 25% lower multiple. SPGI trades at 22 times forward EPS versus MSCI at 26 and Moody's at 27." The deck adds the strategic logic: coming two months after the Carfax/Mobility separation, a Cap IQ spin "would further weight the business towards the higher growth / margin portions of its business, namely Ratings & Indices."
S&P Global is several businesses in one wrapper: it rates the creditworthiness of bonds, it owns the S&P 500 and the index-licensing business behind it, and it sells financial data through a platform called Capital IQ. The ratings and index arms are extraordinary businesses — near-monopolies with very high margins. The data business is good, but it competes with FactSet, Bloomberg and others, and the market pays less for it.
The company is now looking at spinning Capital IQ out as a separate listed business. Two things make that attractive. First, whatever is left behind is a purer version of the high-margin ratings-and-indices company, which the market should value more highly — as evidence, Moody's, which is essentially just a ratings agency, trades at 27 times earnings and S&P at 22. Second, the piece being separated gets its own price rather than being buried in a conglomerate: FactSet, the closest comparable, trades at about 15 times earnings and 12 times cash profits.
This is the classic special-situations trade Singh runs repeatedly: a conglomerate discount closing because a company chooses to break itself up. It follows the same company separating its Carfax and Mobility businesses only two months ago — a management team that has shown it will actually do this rather than merely discuss it.
Full passage: premium transcript (PDF).
In short: $130k (+$23k), an underperformer (−17% YTD vs Moody's −5%) because of the Mobility spin, Market Intelligence restructuring and leadership changes. 11% growth × 27× = ~12–13%, and "the more attractive investment" of the two ratings agencies on valuation. Data embedded in Microsoft 365 Copilot. $750 (7.5%).
S&P Global (credit ratings, indexes and financial data) has lagged this year, down about 17%, mainly because of internal changes — spinning off its Mobility unit, restructuring a division and new leaders — rather than a weaker business. It is putting its data directly inside Microsoft's Copilot. He expects about 12–13% a year and prefers it to Moody's, which does similar things but costs more.
In short: TO BE UPWEIGHTED. The second-smallest position in the book at ~2.8%, bought only five weeks earlier at a $425 limit. 19x NTM P/E on a 15% EPS CAGR; −11% YTD and −15% over twelve months; 120 shares producing $3,363.60 of look-through free cash flow; modelled three-year return 9.80%. Also one of the seven STRONG BUYs a week earlier. Cited in the sector note as an example of how broad the "Financials" label is — "from a payment network provider like Visa to an insurance company like Kinsale Capital and a credit rating agency like S&P Global."
S&P Global rates corporate debt, runs the index business behind the S&P 500, and sells market data. Companies wanting to borrow effectively must be rated, and funds tracking its indices must pay to use them — so it charges a toll on activity it does not have to generate.
It is the second-smallest position in the portfolio at under 3%, and it is being increased. Again, the reason is recency rather than a change of view: it was only bought five weeks earlier, at a $425 limit price. On 19 times next year's earnings with growth expected around 15%, it sits in the middle of the book on valuation, and it was rated a Strong Buy on the firm's own scale a week before this letter.
In short: Trimmed to fund Meta, then bought back "to a large extent" after it fell; as a shareholder he "stamps approval" on Reddit's S&P 500 inclusion.
In short: Buy target $350 (fifth-largest, $127k / ~8.5%, +$20k). He flags it as one of the least likely to trigger — "S&P Global has not traded down below 350 this entire year… the last time was back in late 2023" — and treats that as the correct outcome of a strict rule: "We want to be disciplined with this new money… we're not just looking for modest dips." Even on 10.7% EPS growth and a 25 multiple ("modest… for a company as good as S&P Global"), $350 delivers almost 16% a year, "a very strong market beating return on very conservative assumptions."
11:58So, all of these buys so far at these target buy prices, assuming these very moderate assumptions, would have incredibly high returns. Next up, we get to S&P Global. It's my fifth largest position. We're at $127,000 total size, $20,000 in gains. It's about 8.5% of the portfolio. My buy-in target for S&P Global, if it's going to get this additional $10,000, is $350.
In short: STRONG BUY (portfolio). Fair value $611.1 vs $414.3 (32.2% under); fwd PE 22.3 against 29.5 (24.4% under); RDCF 10.4% required vs 11.5% expected — the thinnest margin of the seven Strong Buys, so the rating leans on the multiple rather than the DCF. Bought at a $425 limit five weeks earlier.
In short: One of Pershing Square's six new Q2 positions — the ratings/index duopoly leg of Ackman's financial-infrastructure basket (with ICE, V and MA).
In short: Bought by Ackman at 5.4% — "huge holding into this one. I think another excellent buy for the Ackman portfolio" — and added to by Hohn ("He continues to build conviction in Visa and S&P Global"). Carlson holds it and takes the opposite side of Kantesaria's trim: "he chose to reduce S&P Global and not Moody's. I would have done just the opposite… I would have kept S&P Global," because Moody's is the one holding up better and carries the higher valuation. Also on Dorsey's list — "I also like S&P Global, of course."
S&P Global rates debt and sells financial data — a business two firms effectively share with Moody's, which is why he calls them a duopoly. It shows up on three of the four managers' books this quarter: Ackman bought 5.4%, Hohn added to an already-large position, and Dorsey holds it. Carlson owns it too.
The interesting wrinkle is the one manager who sold it. Kantesaria trimmed S&P Global and left Moody's alone; Carlson would have done the reverse, and his reason is a trading rule rather than a quality judgement. When you're forced to raise cash, sell the one the market is still paying up for — here Moody's, which is holding its price better and carries the higher multiple — and keep the one that's already been marked down.
He's fair about the other view: Moody's is arguably the purer business because more of its revenue is credit ratings. But if the whole point of a forced sale is to keep the most upside, selling the cheaper of two near-identical assets is backwards.
16:44So, huge holding into this one. I think another excellent buy for the Ackman portfolio. The same thing for Visa and MasterCard. Combined, this is a 11% position, but we also have more buying with Bill Ackman. He bought more Meta, added 20% to his stake, raising it to 9%. I agree with him here. I think a great buy.
In short: Same call as Moody's — the agencies (with privately-held Fitch) supply "a language of credit," a shorthand humans need because "we're both very smart and very stupid at the same time." All three deliberately use the same letters, which is the point. "It's kind of one of the most stubborn oligopolies in the history of business probably." Caveat, aimed at the newer entrants rather than the big three: private-label ratings on private credit loans are "a lot iffier."
The same argument as Moody's, with the moat stated more explicitly. Wigglesworth quotes an industry phrase he came across while researching the book: credit needs "a language." Humans are "both very smart and very stupid at the same time" — we need shorthand, rules of thumb, simple models. A single B, a double A, a triple C is that shorthand, and it works only because everybody uses the same symbols.
This is why the three agencies, for all their marketing about methodological differences, still publish essentially identical letter scales. The standardisation is the product. As with all language it is sometimes wrong — he cites the statistician's line that all models are wrong but some are useful, adding that "the rating agency models are not as wrong as people think." And if the agencies vanished, "we'd have to reinvent them all over again." His summary: "one of the most stubborn oligopolies in the history of business probably."
One caveat he raises sits outside the big three: the "private label" ratings being attached to private credit loans so insurers can call them investment grade. Those he suspects are "a lot iffier" — which, if anything, is an argument for the incumbent brands rather than against them.
57:59both very smart and very stupid at the same time. And we like these shorthands. We like rules of thumb. We like simplistic models. And it's just nice to have something like, this is a single B, that's a double A, that's a triple C. And the reason why the rating agencies, despite having very different — they talk up all their difference, they still have basically the same letters as well.
In short: A $104,000 position (+18%, +$13,000) plus $18,000 in the Story Fund. February's 25% peak-to-trough fall on Anthropic's Claude Co-work launch took the multiple from 25× to 19×, "the lowest valuation in the previous 5 years" — and was mispriced: Capital IQ is under 7% of revenue and less of profit, over 40% of Market Intelligence revenue is proprietary data "deeply embedded in customer workflows," and growth is "almost exclusively determined by the company's benchmark business." Zeroing the entire disrupted line still leaves a 26 forward P/E.
S&P Global sells credit ratings, market indices (the S&P 500 itself) and financial data. In February the stock fell more than 25% because Anthropic launched Claude Co-work, a financial-analysis tool, and investors decided an AI could do what S&P Global does. The multiple dropped from 25 times earnings to 19 — the cheapest in five years for what is routinely called one of the highest-quality businesses in the world.
The fear is aimed at the wrong part of the company. Capital IQ, the desktop research product an AI plugin could plausibly compete with, is under 7% of revenue and an even smaller share of profit. Over 40% of the data division's revenue comes from proprietary data wired directly into customers' internal workflows — hard to rip out, and not something a chatbot supplies. And the real earnings engine is the benchmark business: ratings and indices, which an AI assistant doesn't replace at all.
Carlson's stress test makes the point bluntly: assume Claude destroys the entire contested business line, all the way to zero, and S&P Global would still trade at about 26 times forward earnings. That is the worst case, and it isn't a disaster price. He owns $104,000 of it (up 18%) plus another $18,000 in his Story Fund.
14:41It is incredible that S&P Global today, which is considered again one of the best companies in the world, is trading at 21 times multiple. And an interesting thing about this that I'll just highlight now is even if Claude was to hypothetically completely disrupt market intelligence, the thing that it actually competes with, if it made that business line of S&P Global go to zero, so it totally eradicated that entire business line, S&P Global's forward P/E ratio would be 26.
In short: "I do own this company and I like the situation of S&P Global better" than FICO — slower-growing but far less concentrated risk, and actually accelerating: 11% organic revenue growth ex-spin-offs, an "incredible" ratings business, the S&P Dow Jones indices, and Market Intelligence +6% (subscriptions +6%, volume +9%). Very efficient growth — "when their revenue grows by 11%, their earnings per share grow much faster." Still bullish; liked the report.
S&P Global rates the creditworthiness of bonds, owns the S&P 500 and Dow Jones indexes, and sells financial data by subscription. Carlson presents it as the answer to the FICO problem: same broad industry, but the revenue comes from several independent engines rather than one product, so no single regulatory or competitive shock can break it.
The quarter was quietly good and getting better: revenue up 11% once you strip out businesses it has spun off (Mobility Global), an "incredible" ratings business, growing indexes, and Market Intelligence — the piece investors watch most — up 6% with 9% more customer volume. Because the cost base barely moves, an 11% revenue increase turns into much faster earnings-per-share growth. He owns it and stayed bullish on the report.
13:14I do own this company and I like the situation of S&P Global better. Now, S&P Global is not growing as fast as FICO, but it also doesn't have the same level of risk. It's not as concentrated of a company. It's not as reliant on one thing going well. And S&P Global is not growing super fast, but it actually is accelerating its growth.
In short: Bought — $50,000 at a $425 limit, roughly 120 shares. The durability test opens it: "was the clear market leader 20 years ago; is still the market leader today; will (probably) still be the market leader in 20 years from now," with "+14.4% per year" since 1990 turning $10,000 into $1.4 million. The structural claim is the count: "It doesn't operate in just one oligopoly. It operates in five" — Ratings, Indices, Market Intelligence, Commodity Insights and Mobility, each facing "only 2–3 major competitors," each "protected by network effects, regulatory barriers, and proprietary data," and each reinforcing the rest. Quality: "very capital light: CAPEX equals just 1.1% of sales," gross margin 70.5%, operating margin 43.8%, net margin 32.8%; CEO Martina Cheung's compensation "is 92% performance-based and tied to long-term EPS growth." All three valuation methods agree — forward PE 22.1x vs a 27.0x ten-year average, an Earnings Growth Model return of 11.5%, and a reverse DCF needing "only 9.0% annual free cash flow growth to generate a 10% return… S&P Global should be able to grow more than that."
S&P Global sells the financial world its reference points. It rates the creditworthiness of borrowers, it owns and licenses the S&P 500 and hundreds of other indexes, it sells market data by subscription, it publishes the price benchmarks the commodity trade settles against, and it runs a vehicle-data business. It does not take positions and it does not lend; it charges for being the standard.
The argument for buying it is a counting argument. Most quality businesses have one protected position. This one has five, each with only two or three real competitors, and — the part that matters — each making the others harder to attack. A company that wants an index-linked ETF needs the index; the index business generates data worth selling; the data business makes the ratings relationships stickier. A competitor would have to break in at five points at once.
Two numbers show why that structure is so profitable. It spends barely 1% of its revenue on physical equipment, so almost all of the profit becomes cash available to shareholders rather than being consumed keeping the machine running. And it keeps about 33 cents of every dollar of sales as final profit. The chief executive's pay is 92% dependent on performance and tied to long-term earnings growth, which is the alignment substitute for a founder's shareholding.
The price is the reason it is bought now rather than admired. Three independent checks all say the same thing: the shares trade at 22 times next year's earnings against a ten-year norm of 27; a forward-return model gives about 11.5% a year; and working backwards from today's price, the company only needs to grow its cash flow 9% a year to hand shareholders 10% — less than it has historically managed. Compare that with the Eli Lilly analysis three days earlier, where the same reverse calculation demanded 18.9% growth and the stock was passed.
The trade is disclosed rather than described: $50,000, a limit price of $425 set just under the $426.4 quote, roughly 120 shares. It is the same name that topped the Best Buys list seven days earlier, which makes that monthly list a purchase queue rather than a reading list.
In short: Named as the other half of the ratings duopoly — some investors "have assumed that AI could eat into the Moody's and S&P duopoly." Eisman: "I do not believe that is possible."
S&P Global is the other credit-rating giant — the second half of the duopoly Eisman likes. It comes up only as the partner in that structure: some investors fear AI will erode the two firms' grip on ratings, and he says flatly he doesn't think that's possible. Referenced, not rated.
9:32I do not believe that is possible, so I continue to be an owner. The stock has been flat since the fall of 2024, so the multiple has come down. The 2026 PE multiple is now sub 30 times, which is the cheapest the stock has been in quite some time. Wednesday night, wow, Wednesday night was a big night. A bunch of very important companies reported.
In short: Best Buy #1 — and bought a week later. "S&P Global operates two financial toll bridges: credit ratings and benchmark indexes. S&P Global gets paid every time: a company issues a bond; an investor buys an S&P 500 ETF (they own the S&P 500 index)." Both sides are oligopolies — "Credit Ratings: S&P Global, Fitch, Moody's; Indexes: S&P Global, MSCI, FTSE Russell" — and "oligopolies tend to be great businesses. They enjoy high barriers to entry, pricing power, and high margins." Also flagged: the recent spin-off of the mobility division, after which "the stock went up in the days after the spin-off."
S&P Global sits in two places where money has to pass through and takes a cut at each. When a company borrows by issuing a bond, it needs a credit rating, and there are only three firms that can give one — S&P, Moody's and Fitch. When an investor buys an S&P 500 tracker, the fund pays to use the index name and the list of companies in it, and S&P owns that index.
Slegers' point is that neither business is really about being clever. They are about being one of three permitted operators in a market everyone else must use. That is what he means by a toll bridge: the traffic decides the volume, but nobody gets to build a competing bridge. Oligopolies of that kind, he says, "enjoy high barriers to entry, pricing power, and high margins."
He ranks it the single most likely purchase of the month, and a week later he buys it — this is the shortlist the portfolio actually draws from. The company also recently separated its mobility division into its own listed business, which the market took well.
In short: Shares have rallied 12% since the Mobility (MGGL) spin to 17× '27 EBITDA — in line with the SOTP multiple for Moody's/MSCI/FDS; a cleaner pure-play data/ratings compounder.
Full passage: premium transcript (PDF).
In short: BUY. FV $646.6 vs $438.4 = 32.2% under; ER 13.6%; fwd PE 22.3 against 29.5 (24.4% under); RDCF 7.3% vs 11.5% expected. Ranked Best Buy #1 ten days later and bought on 26 July.
In short: Completed the separation of its mobility division (now Mobility Global, MGGL) Wednesday — "actually quite good for the S&P stock" as it becomes more of a pure-play data/ratings compounder.
S&P Global — the ratings-and-data giant — just finished splitting off its car-data division into a separate company (Mobility Global). Singh views that as a plus for the remaining S&P: it becomes a cleaner "pure play" on its highest-quality ratings and index businesses, which tends to earn a richer valuation.
Full passage: premium transcript (PDF).
In short: #6 — "Financial data and credit rating oligopoly." "Credit ratings, financial benchmarks (such as the S&P 500 index), and data analytics to the global capital markets… one of just a few big players that dominate the world market." The necessity argument: "Companies that borrow money need credit ratings. The global financial system can't work without them. As long as capital markets exist, S&P Global will take a cut of financial data and transactions."
S&P Global grades borrowers' creditworthiness, publishes the benchmarks the investment industry is measured against — including the S&P 500 itself — and sells the data underneath both.
Its durability comes from being wired into how the financial system works rather than from being preferred by customers. A company that wants to borrow in the bond market has to be rated, and only a handful of firms are recognised to do the rating. As long as anyone borrows or invests, S&P takes a small cut of the activity. It needs almost no capital to do so — the product is analysis, not a factory — which is why the cash conversion is so high and the position so hard to attack.
In short: Hohn's #5 at 10.3%; described, not re-rated here. "S&P Global is part of a oligopoly with Moody's and Fitch in credit ratings. When companies issue bonds, they must get a rating from at least 2 of these companies… This is a classic tollkeeper business with incredible pricing power. The ratings business requires minimal capital, generates massive free cash flow, and has high renewal rates." The section carries the post's most transferable formula, in Hohn's words: "If you can price 1% above inflation and you have a 20% profit margin, your profits will grow 5% faster than revenue." (SPGI is separately a March/April 2026 Best Buy and a Positive name on this hub in its own right.)
S&P Global does two things: it rates debt, alongside Moody's and Fitch, and it owns the benchmarks — including the S&P 500 itself — that trillions of dollars of index funds are obliged to track. Both are tolls. Neither requires much capital, both renew almost automatically, and both are protected by rules rather than by marketing.
The section is where the post's most portable idea appears, in Hohn's own words: if a company can raise prices 1% faster than inflation and runs a 20% profit margin, its profits grow about 5% faster than its revenue. The reason is arithmetic — a price rise costs nothing to deliver, so it falls straight to profit, and the thinner the margin the bigger the multiplier. It is a one-line way to see why a modest, durable pricing edge is worth so much more than it sounds.
Marked Neutral only because this write-up reports Hohn's rationale rather than restating a Compounding Quality view; S&P Global is separately a Best Buy on this hub in its own right.
In short: Preferred (with Moody's) over FactSet as a wide-moat data/ratings compounder; also a special situation — its Mobility Global division (MGGL) began when-issued trading Friday and goes regular-way this week.
Singh prefers S&P Global and Moody's — the dominant, wide-moat ratings-and-data duopoly — over FactSet, which he doesn't like. There's also a catalyst: S&P is spinning off its automotive-data arm (Mobility Global / MGGL, anchored by CARFAX), which just began trading on a "when-issued" basis.
Full passage: premium transcript (PDF).
In short: Slower top line (~7.2% ex-acquisitions) but ~13% EPS on very high-margin data/ratings. Rejects the "Claude makes you not need S&P/Moody's" fear — banks lending billions "can't make these loans based on whatever Claude spat out"; they pay for the real analysis. His favored buy of the ratings pair on valuation.
S&P Global grows its top line slowly (~7% once you strip out acquisitions) but converts it into ~13% earnings growth thanks to very high margins. Carlson rejects the worry that AI chatbots replace it: a bank lending billions of dollars "can't make these loans based on whatever Claude spat out" — when that much capital is on the line, institutions pay for the real, authoritative data and analysis. It's his preferred buy of the ratings pair on valuation.
In short: BUY. FV $626.1 vs $424.4 = 32.2% under; ER 13.6%; fwd PE 22.3 against 29.5 (24.4% under); RDCF 9.8% vs 11.5% expected. YTD −17.2%; ranked Best Buy #1 in July and bought the week after.
In short: Grouped with Moody's as a buy — a wide-moat ratings/data duopoly franchise he'd own through the rotation into quality value.
Full passage: premium transcript (PDF).
In short: Down 26% from highs, the cheapest in 5 years on both P/E and P/FCF; a ~21 forward P/E for this quality is "almost unheard of," and unlike software it isn't being AI-disrupted. Buy.
S&P Global runs the credit-rating agency and the data/index business behind the famous S&P 500. The stock has fallen 26% from its high and is the cheapest it's been in five years — whether you measure it against profits (price-to-earnings) or against the cash it generates (price-to-free-cash-flow).
Carlson loves that for a business this dependable. Paying about 21× next year's expected earnings for this quality is, in his words, "almost unheard of." And unlike software companies, it isn't really threatened by AI — people will still need trusted credit ratings and market data. So he rates it a buy.
4:15And I do find right within my portfolio a number of companies that I believe are good value. We first start off with S&P Global. We have that one right here. This one's down 26% from its highs. The stock is at the lower range of its 52- week range. So, it's trading down. You're not buying it at top dollar. And then when we look over the past 5 years and look at the historical valuation range in which it's traded at, we have it right now.
In short: The index maker — critiqued for arbitrary-committee reclassification "games" (Apple/Amazon as consumer-discretionary, Meta as telco) that hide a "borderline preposterous" concentration; credited for rejecting SpaceX's fast-track inclusion.
21:22Um we're watching the the semi the direct semi market from the sidelines with um a little bit of shock, but uh it's it's certainly interesting to see the movements here. — Well, it's interesting you talk about tech. Even this week the S&P index makers announced they're kicking out Campbell Soup.
In short: BEST BUY #3. "S&P Global is a financial toll bridge. You literally cannot issue corporate debt without a rating from S&P Global, Fitch or Moody's. It is a legal oligopoly." Two tolls, not one: "S&P Global gets paid every time a company issues a bond" and every time "an investor buys an S&P 500 ETF (they own the S&P 500 index)" — which "rides the wave of more and more people switching to passive investing". The drawdown is attributed to an AI fear and rebutted: "The market sold off S&P Global out of fear that AI will disrupt their business. I think that's unlikely. Their proprietary data, like Platts commodity pricing and their market data goes back over a century. Large Language Models actually need S&P's private, licensed data to be accurate." (The conclusion calls it a duopoly having named three raters in the body.)
S&P Global collects a fee twice over from the financial system, which is why it is described as a toll bridge.
The first toll is credit ratings. A company issuing bonds effectively must have a rating from S&P, Moody's or Fitch — investors and regulations require one. Three firms, and no realistic way to bypass them: a legal oligopoly.
The second toll is the indices. S&P owns the S&P 500 itself, and rents it and its other benchmarks to the exchange-traded funds that track them. Every time someone buys an S&P 500 tracker, S&P Global gets paid — so it earns money from the shift of the whole world into passive investing.
The shares sold off because investors feared artificial intelligence would make its data business obsolete. The argument here is the reverse: its data is private and licensed, some of it — like Platts commodity prices — stretching back over a century, and language models need exactly that kind of proprietary data to be accurate. AI becomes a customer rather than a competitor.
In short: A phase-4 "predictable winner" — distribution, longevity, customer lock-in, unique data sets that won't commoditize. Under pressure but he stays "very bullish"; owns over $100k.
S&P Global runs the credit-ratings agency and the data/index business behind the S&P 500. In Carlson's "phase 4" software split, it's firmly on the winners' side: it has distribution, longevity, customer lock-in and unique data sets that AI can't easily copy — so AI makes it more useful, not obsolete.
The stock has been under pressure and investors are nervous, but he stays "very bullish" and owns over $100,000 of it. To him it's a predictable, hard-to-disrupt compounder being unfairly lumped in with vulnerable software.
18:01That's when phase one and phase 2 come to a conclusion. We also have companies that I believe still have the distribution, the longevity, the customer lock-in, the unique data sets, all the stuff that will make these companies not commoditize. That's S&P Global, Moody's, FactSet, MSCI. We also have companies that are at some risk in some aspects of AI, but overall they have a lot of non-AI aspects, and I'd put Uber and DoorDash in that category.
In short: BUY. EPS growth 11.5%, dividend 0.9%, FWD PE 22.3 against a fair exit 25.0, expected return 13.6%, fair value 629.5 against 426.8 = 32.2% undervalued. Ranked Best Buy #1 in July and bought a week after that.
In short: UPGRADED Hold → Buy, with Moody's, as the "duopoly in global credit ratings." 19.6x forward against a 29.5x five-year average (23.4% under), expected return 10.5%, +1.5pp reverse-DCF margin. Down 12.2% year to date.
In short: Best Buy #3. Owns the S&P 500 index and is the world's leading credit-ratings agency — "You cannot issue corporate debt without a rating from S&P or Moody's. It is a legal duopoly." Paid every time a bond is issued and every time someone buys an S&P 500 ETF, with regulation requiring rated debt and index network effects compounding ("a self-reinforcing flywheel"). Down more than 20% on AI fear: "S&P Global owns proprietary market data going back over a century… Platts commodity prices and CARFAX records are private data… This is short-term noise if you ask me. They own the data that LLMs need to be accurate."
S&P Global sits at two of the financial system's toll booths. It is one of two credit-rating agencies (with Moody's) whose rating a company effectively must have to issue bonds — regulations often require pension funds to hold only debt rated by one of them, which makes it a legal duopoly. And it owns the S&P 500 itself, so every dollar in an S&P 500 ETF pays it a small fee. Both grow on their own: more debt issued, more money indexed.
The stock is down more than 20% because investors fear AI will disrupt its data and analytics business. Slegers thinks that inverts the truth. S&P owns proprietary data nobody else has — a century of market history, Platts commodity prices, CARFAX vehicle records — and an AI model is only as good as the data it is fed. It is deploying its own AI (Kensho) to make that data more valuable, not less: "They own the data that LLMs need to be accurate."
In short: Named as a SaaSpocalypse casualty: −25% in a month. "The companies above are great companies. It has been a while we've seen them fall so much so quickly." No stance or valuation is offered here — the full toll-bridge write-up and the Best Buy #3 rating arrive two weeks later in March.
Nothing matches this filter.
Verbatim excerpts from the public transcripts (auto-pulled at each mention's timestamp, lightly cleaned). Timestamps link into the video; "source page" opens that commentator's full analysis at this row. Click a mention's header line to expand it (one open at a time). For personal study — not investment advice.