S&P Global: a great business, priced above what I can justify

S&P Global now plans to buy back more than $7 billion of its own shares in 2026, about 5.8% of its market value, and roughly $2 billion of the money for the second half comes from new debt. Management guides to gross leverage of 2.7x to 2.8x EBITDA at the end of the year, above its own target range of 2.0x to 2.5x. That is a confident company, and the price it is paying is the question.

At $407.82 on October 9, 2026 the shares trade 19.1% above my triangulated Base value of $342.37 and 3.5% below my Bull value of $422.79. The Bear value of $196.46 is 51.8% below the price. The required margin of safety of 45% puts the maximum buy price at $188.30, and the price is 117% above it. For existing holders this is a hold, with no reason to add. For new money it is a watchlist name. The shares traded around $412 on October 12, up about 1.1% in a session that had not closed when I wrote this.

Route: SF-01 standard value. SF-07 does not apply: S&P Global is a profitable, cash-generating ratings and data company, not an ARR software story. Economic classification: Quality Growth, 6/12 on the sector scorecard, so the engine is GROWTH_ADJUSTED_MULTIPLE with an explicit multiple build. Primary method: owner free cash flow per share times a multiple. Secondary method: a six-year DCF at 10%. Values are present fair-value estimates, not 12-month price targets. Owner free cash flow here is adjusted free cash flow less the cash items management adjusts out and less stock pay.

What S&P Global is

S&P Global rates debt, runs the S&P 500 and other index families, sells financial data and analytics (Capital IQ, a terminal-style product for analysts), and publishes energy and commodity prices through Platts. In 2025, excluding Mobility, revenue was $13,589 million with Ratings $4,724 million, Market Intelligence $4,690 million, Energy $2,525 million and Indices $1,850 million, less $200 million of internal sales. Adjusted operating profit was $6,974 million, a margin of 51.3% (my calculation), and adjusted earnings per share were $15.85.

The businesses are not equal. Ratings and Indices are the high-quality franchises: in the second quarter they earned adjusted operating margins of 68% and 72%, grew revenue 17% and 20%, and together are nearly half of revenue (48% of the 2025 segment total, my arithmetic). Market Intelligence, the biggest data business, grew 6% at a 36% margin and is the part of the company that an AI chatbot could most plausibly compete with. Energy grew 2% to 3% at 48%. This split is the core of the analysis: the market is paying a premium for the franchises and a discount for the product that is most exposed to software substitution, and the discount is the debate.

On July 1, 2026 S&P Global spun off its Mobility business (the former Carfax and automotive data unit) as Mobility Global, listed in New York as MBGL, with one new share for each S&P Global share. Holders as of June 15 own both stocks now. The $407.82 price is for the company without Mobility, and the spin explains why 2026 guidance, recast figures and some data-provider histories do not line up. The 52-week range of $381.61 to $552.25 on the quote source includes months before the spin, so I do not read it as a like-for-like range.

The latest quarter

The quarter was strong at the top and in the margin. Ratings had billed issuance up 25%, helped by hyperscaler borrowing of about $169 billion in the first half, and management now expects $250 billion to $300 billion for the year against an initial assumption of about $200 billion. 2026 guidance, which now excludes Mobility, is for organic constant-currency revenue growth of 6% to 8%, adjusted margin expansion of 35 to 60 basis points (75 to 100 basis points excluding OSTTRA) and adjusted earnings per share of $17.50 to $17.75, which is 10.4% to 12.0% above the recast 2025 figure of $15.85 (my arithmetic). Ratings growth was raised to 5% to 8% and Indices to 12% to 14%. Market Intelligence (5.5% to 7%) and Energy (4.5% to 6%) were left unchanged. The earnings per share figure missed the consensus estimate on the day, according to a result summary I read, so the raised guidance and the buyback were the news.

The next report is due around October 27, 2026 (an estimated date), and it will be the first full quarter after the spin.

The hard metric: owner free cash flow

The company defines free cash flow as operating cash flow less capital spending and less distributions to noncontrolling interests (the cash that goes to the other owners of its index joint venture). It also reports an adjusted version that adds back severance, acquisition costs, tax on divestiture gains and legal payments. In the first half the gap was $113 million, or 4.8%, and stock pay was only $95 million. Management guides to adjusted free cash flow of $2.9 billion to $3.1 billion in the second half, excluding Mobility, but does not guide to a full-year figure.

I need a full-year number excluding Mobility, so I estimate it. The second half midpoint of $3.0 billion plus about $2.05 billion for the first half (the reported $2.36 billion less my estimate of about $0.3 billion for Mobility, which is not disclosed) gives about $5.05 billion of adjusted free cash flow. Less 4.8% for the items adjusted out (about $0.24 billion) and about $0.19 billion of stock pay, owner free cash flow is about $4.62 billion, or $16.41 a share on the roughly 281.5 million shares I expect at the end of 2026 after buybacks. The hard metric is 91% of adjusted free cash flow. At the market value of $120.2 billion, the yield is 3.8% on the hard figure and 4.2% on the adjusted figure. I label the full-year figure ESTIMATED and test 8% either way.

Cash, debt and shares

Debt rose from $13,088 million at the end of 2025 to $15,170 million at June 30, after $1,986 million of senior notes were issued in the half, and cash rose to $4,141 million from $1,745 million. Short-term debt is $2,572 million, up from $718 million. On the call, management said the higher 2026 buyback of more than $7 billion (raised by nearly $3 billion) is funded by about $2 billion from a dividend paid by Mobility Global at the spin, about $2 billion of new debt in the second half, and free cash flow, with about $0.5 billion of the Mobility money retiring debt. Management expects to end 2026 at 2.7x to 2.8x gross debt to EBITDA and to return to its 2.0x to 2.5x range by the end of 2027.

I could not reconcile that leverage figure to my own estimate. About $16.7 billion of debt at year-end (June debt less $0.5 billion plus $2 billion) over about $7.7 billion of adjusted EBITDA (2026 adjusted operating profit of about $7.5 billion, my estimate, plus depreciation) is about 2.2x. The company's definition probably differs, so I use its 2.7x to 2.8x as the working number and treat the gap as a weak point. Interest expense is guided to $390 million to $410 million on an adjusted basis, so coverage by adjusted operating profit is about 19 times and the leverage is a risk to flexibility, not to solvency.

The share count is 294.8 million at June 30, down from 305.3 million at the end of 2025, and diluted weighted shares fell 3.5% year over year in the second quarter. If the remaining $5.5 billion of the 2026 programme is spent at about $412, the count falls by about 13 million more, to about 281.5 million, which is my estimate. In the model I assume a further 3% a year in 2027 and 2028 in the Base case. Stock pay is small at $95 million in the half. Redeemable noncontrolling interests of $5.0 billion are the other owner's share of the index business, and since my cash flow is already after its distributions, I do not deduct it again.

Primary valuation: owner free cash flow times a multiple

S&P Global is a Quality Growth company under the framework, so the multiple is built from a base and explicit premiums. The base is 12 times, the framework's very stable tier. I add premiums for growth per share, for the return on capital, for the recurring character of revenue and for the balance sheet, and subtract a risk discount for AI, leverage above target and a falling-share-count programme that depends on borrowing. The return on all capital is about 10% on my estimate (pro forma operating profit of $6.1 billion taxed at 22.5%, over about $47.7 billion of equity, net debt and the joint venture interest), which is low because $52 billion of the capital is goodwill and intangibles from acquisitions such as the IHS Markit merger. On tangible capital the return is not meaningful. I give a moderate premium for that.

The Base multiple of 18 times is 50% above the 12 times sector base, below the 75% line that triggers a mandatory DCF cross-check, though I run one as the secondary method. Today's price is 24.8 times my 2026 owner free cash flow per share, so even the Bull case is a multiple compression story. I did not rebuild a five to ten year history of the company's own multiple, so the framework's historical cross-check is UNAVAILABLE, and Confidence is lowered for it.

The primary Base value is $339.70. The cash flow path is my estimate: owner free cash flow of about $4.62 billion in 2026 grows 8% in 2027 and 7% in 2028 in the Base case ($5.00 billion and $5.35 billion), in line with revenue growth of 6% to 8% and margin gains, less higher interest on the new debt. The Bear holds owner cash flow flat at about $4.55 billion, the Bull reaches $5.70 billion. Buybacks are modelled as a falling share count at 2.5% to 3% a year, which is what the company's cash flow after dividends can pay for at today's price. I do not count the extra buyback above that.

Sensitivity: primary value per share (interim cash flow held at the Base figure)

Secondary valuation: six-year DCF

The secondary method discounts owner free cash flow per share (growth plus the buyback effect) for six years at 10%, and then a terminal value. Per share, owner free cash flow starts at $16.41 for 2026 and reaches $27.52 by 2032 in the Base case, which is growth of 7%, 7%, 6.5%, 6%, 5% and 4.5% in cash flow plus 3% a year from buybacks. The rate of 10% is the same as in the primary method and is my choice for a high-quality company with a rising debt load.

The secondary Base value is $346.38. Seventy percent of it is terminal value, so it depends on the 3.5% growth forever and the 10% rate. It uses the same cash flow path as the primary method, so the two are not independent, and their close agreement is weaker evidence than it looks.

Sensitivity: DCF Base value per share

Growth diagnostics

The growth valuation cross-check (SM-19) is a diagnostic only. Adjusted earnings per share are guided to grow about 11.2% at the midpoint ($15.85 to $17.625), and the price is 23.1 times that guidance, which gives a PEG (price to earnings divided by growth) of about 2.1. That is not cheap, and it is also flattered: about 3 points of that earnings growth is the falling share count, and the adjusted figure leaves out deal-related amortization of about $785 million (the GAAP guide is $16.35 to $16.60).

Free cash flow growth quality (SM-20, diagnostic only). Adjusted free cash flow in the first half rose 4.6% including Mobility ($2,362 million against $2,258 million), and operating cash flow was almost flat in the second quarter ($1,439 million against $1,445 million) (both including Mobility). My Base path implies owner free cash flow per share compounding at about 10.1% a year from 2026 to 2028, of which about 3 points are buybacks and 7 points business growth, supported by guided revenue growth of 6% to 8% and margin expansion of 35 to 60 basis points. I see no low-base distortion and no cyclical rebound in the numbers, but ratings issuance is cyclical and 2026 benefits from a record year of hyperscaler borrowing. Forecast reliability is MEDIUM: guidance is one year, the full-year free cash flow excluding Mobility is my estimate, and Mobility's cash flow is not disclosed. Stress test: owner free cash flow 8% lower is worth $314.98 in the triangulated Base and 8% higher $369.76.

Valuation triangulation

The two methods diverge by 1.9% of their midpoint, far below the framework's 20% warning line. They are not independent: both use the same cash flow path and the same share count, and the multiple of 18 and the 3.5% terminal growth were chosen with the other in mind, so I do not treat the agreement as proof. The primary gets 60% because the multiple is built from stated components, the DCF 40% because it relies on terminal value. Weighting them 50/50 gives $343.04.

What is the market already pricing in?

At $407.82 market value is $120.2 billion, and with $11.0 billion of net debt and $5.0 billion of the joint venture interest the enterprise value is about $136 billion, or 18 times my estimate of 2026 adjusted operating profit (about $7.5 billion). Holding my Base cash flow path, the price needs a multiple of 21.8 times 2028 owner free cash flow per share, against the 18 times I use. Said as growth in the DCF, it needs owner free cash flow to compound at about 9.5% a year for six years, on top of 3% a year from buybacks, with a 3.5% terminal rate. That is more than the 6% to 8% revenue guidance and margin gains deliver on their own, and it needs the buyback to keep working at $400 or more per share.

For the Base case to be right, an investor needs organic growth of 6% to 8% to hold, margin gains of 35 to 60 basis points a year to continue, Market Intelligence to hold its 5.5% to 7% growth while customers experiment with AI, and leverage to return to 2.0x to 2.5x by the end of 2027. Evidence so far: Ratings and Indices are ahead of plan, the margin is rising, and management reports more than 500 customers on its AI-ready data feeds with five times the API calls of the previous quarter. Market Intelligence has not yet shown the same acceleration, and management itself notes that some customers are scrutinizing AI token costs and building their own tools.

Bear, base and bull scenarios

Bear: Market Intelligence loses renewals to AI tools and in-house models, ratings issuance falls back after the 2026 hyperscaler peak, owner cash flow stays flat near $4.55 billion, leverage stays above target, and the market pays 11 times for a business it no longer sees as a toll road. Base: organic growth of 6% to 8%, margins up 35 to 60 basis points a year, 3% a year from buybacks, and 18 times owner cash flow. Bull: Ratings and Indices keep growing at double digits, AI turns into paid growth in Market Intelligence, owner cash flow reaches $5.70 billion in 2028, and the market pays 22.5 times. The Bear is further from the Base than the Bull is, as 52% below the price against 3.7% above it.

The probability-weighted value, at 25%, 55% and 20% (I give the Bear 25% because the AI risk is real and unresolved), is $321.98, which is 6.0% below the Base and 21.1% below the price.

How the buy price is set

Confidence C starts the margin of safety at 40%, the middle of the framework's 35% to 50% range. I add 5 points for structural disruption (AI and in-house models could erode demand for third-party data in Market Intelligence) and 5 for terminal value dependence (70% of the Base DCF), and subtract 5 for the ratings and index franchises, with margins above 50%, repeat revenue and low model risk. I add nothing for leverage: it is above management's target but investment-grade and covered about 19 times by operating profit, and the Bear already carries it. That leaves 45%.

A 45% discount to the $342.37 Base gives a maximum buy price of $188.30. At the bottom of the Confidence C range (35%) the price would be $222.54, and with no add-ons (40%) $205.42. Even the lightest of those is 45% below the price. The shares would need to fall by about half, or the earnings need to prove stronger growth, before the framework says buy.

Dividend and capital return

The quarterly dividend is $0.97, or $3.88 a year, a yield of 0.95% at $407.82. Dividends paid were $575 million in the first half, about 25% of my owner free cash flow, so the dividend is small and safe. The buyback is the real capital return: $1.5 billion in the first half and more than $7 billion planned for 2026, about 176% of my owner free cash flow when added to dividends. It is funded with the Mobility dividend and new debt, and I do not count the part above what free cash flow can fund when valuing the shares. At 24.8 times free cash flow, each dollar of buyback retires shares at a cash yield of about 4%, below the 10% return I require.

Reasons to own SPGI

• Franchises with pricing power. Ratings grew 17% in the second quarter at a 68% margin, Indices 20% at 72%, and guidance for both was raised.
• Margins above 50% and rising. Adjusted operating margin was 54.3% in the quarter, up 200 basis points, with 2026 guidance of a further 35 to 60 basis points.
• Strong cash conversion. Adjusted free cash flow of $2.4 billion in the first half and $2.9 billion to $3.1 billion guided for the second, with capital spending of only $65 million in the half.
• A shrinking share count. Diluted shares fell 3.5% year over year in the second quarter, and the 2026 programme is more than 5% of market value.
• A simpler company. The Mobility spin simplifies the group and puts the supply chain business inside Energy, and management says it has ruled out transformational acquisitions.

What could go wrong?

• AI and in-house tools in Market Intelligence. This unit is about a third of revenue, grew 6% and earns a 36% margin. The shares fell earlier in 2026 on a forecast below expectations and AI concerns, and management says some customers are building their own tools.
• The issuance cycle. Ratings issuance was up 25% on hyperscaler borrowing, with $250 billion to $300 billion expected for the year. A fall in debt issuance would hit the highest-margin business first.
• Leverage and a debt-funded buyback. Gross debt to EBITDA of 2.7x to 2.8x at the end of the year is above the target range, and about $2 billion of the 2026 buyback is new borrowing. I could not reconcile the leverage figure to my own estimate.
• Price paid for the buyback. The company is retiring shares at about 25 times free cash flow. If the shares are worth my Base value of $342.37, each dollar spent above that value transfers wealth to sellers.
• Recast and adjusted numbers. The spin changed segments and cost allocations, and the adjusted measures leave out deal-related amortization of about $785 million, so historical comparisons need care.

Management execution

Management has raised guidance for Ratings and Indices, grew adjusted margin by 200 basis points, completed the Mobility spin on schedule, and bought back shares each quarter. Against that, the 2026 forecast given in February disappointed the market, the new Market Intelligence structure was just reorganized under new leadership, the buyback was raised by nearly $3 billion in the same period that leverage rose above target, and the company cited free cash flow and margin measures that exclude real costs. The chief executive, Martina Cheung, named market leadership, high-growth adjacencies and AI-enabled enterprise capabilities as priorities. Management credibility scores 14/20: strong delivery on margins and cash, a more aggressive balance sheet, and an AI strategy that has not yet changed the Market Intelligence growth rate.

Stock Analyza scorecard

The moat score is high because the ratings brand and index licences are very hard to replicate, and it loses points for Market Intelligence. Economic return is middling: the return on all capital is about 10%, held down by $52 billion of goodwill and intangibles, though margins and cash conversion are exceptional. The balance sheet gets 8 for an investment-grade profile and loses points for leverage above target and a buyback funded partly with debt.

The valuation score is 28/100: discount to Base 3/40, protection against the Bear 2/20, agreement between methods 12/15, historical valuation 4/10, market implied expectations 4/10, data and model penalty 3/5. The score is low because the price is above the Base and the Bear is 52% below it. It does not say the business is poor; the verdict follows from the price being 16% above the Base.

Confidence score

What I would watch from here

The next review point is the third-quarter 2026 report, expected around October 27, 2026, the first full quarter without Mobility. One dashboard line is red today (leverage above the target range), which is part of the reason for the 45% margin of safety.

The three most important thesis breakers

1. Market Intelligence organic growth falls below 4% as customers move to AI tools or in-house models.
2. Ratings issuance reverses sharply after the 2026 hyperscaler peak, with Ratings guidance cut.
3. Leverage stays above 2.5x into 2027 while the buyback continues at current share prices.

Adversarial review

I rebuilt the key figures from S&P Global's second-quarter 2026 earnings release and the July 6, 2026 recast release: revenue, adjusted operating profit, earnings per share, cash flow lines, cash, debt, redeemable noncontrolling interests, buybacks, dividends, guidance and the 2025 pro forma figures ($13,589 million of revenue, $6,974 million of adjusted operating profit, $15.85 of adjusted earnings per share). Net debt of about $11.0 billion is June 30 debt of $15,170 million less cash of $4,141 million, my calculation. The funding of the buyback, the leverage target, the issuance expectations and the AI data points come from call summaries, not from the transcript itself. The price is the October 9 close from the history page of the quote source.

Weak points, in order. First, full-year free cash flow excluding Mobility is my estimate: Mobility's share of the first-half cash flow is not disclosed, and I assumed about $0.3 billion. Second, I could not reconcile the 2.7x to 2.8x leverage guide to my own estimate of about 2.2x, so the company's definition probably differs. Third, the multiples and the 10% discount rate are my judgment on the SF-01 tiers, with no peer table and no rebuilt history of the company's own multiple. Fourth, both methods share the same cash flow path, so their 1.9% divergence is weaker evidence than it looks. Fifth, the share count after the buyback assumes the remaining $5.5 billion is spent at about $412, and the pace and price could differ. Sixth, the return on capital of about 10% is my estimate, using pro forma operating profit and a book capital base that includes the joint venture interest.

The strongest counter-argument is that S&P Global deserves a much higher multiple than 18 times. Ratings and index licences are close to toll roads, margins are above 50%, and a buyback of this size lifts earnings per share by several points a year. If the market keeps paying 23 times earnings and the AI fears fade, the Bull value of $422.79 is already close to the price and the shares could keep compounding. The reply is that the price already needs about 9.5% annual cash flow growth plus buybacks, or 21.8 times 2028 owner cash flow per share, that the buyback is partly financed with debt at a leverage above the company's own target, and that a third of revenue sits in the product most exposed to AI substitution.

Single changes keep the Base between $314.98 and $370.09, and the maximum buy price between $173.24 and $203.55, so the price is above every single-change Base value. Stacking all cautious inputs gives $254.88, 37% below the price, and stacking all generous inputs gives $461.78, 13% above it. The errors are two-sided: the generous stack is 35% above the Base and the cautious stack 26% below it. The price sits between the Base and the generous stack, which is why the verdict is a hold and not a sell. The publication gate status is PASS_WITH_WARNING.

Final verdict: hold at $407.82, do not add

S&P Global is one of the best business franchises in public markets, with margins above 50%, a ratings and index core growing at double digits and a management that is shrinking the share count. At $407.82 the price already sits close to my Bull value, and a company buying its own shares with new debt at 25 times cash flow is paying a full price for the same story.

Verdict: HOLD at $407.82, do not add. Triangulated Base value $342.37, downside 16.0%, Confidence C (16/25). Required margin of safety 45%, maximum buy price $188.30.

For an existing holder the franchises, the margin trend and the cash conversion justify holding. New money waits for $188.30 or below, or for proof that Market Intelligence can grow above 7% while leverage returns to 2.5x. The main thesis breaker is Market Intelligence growth falling below 4%. Management is a buyer at $407.82. My model is a buyer at $188.30.

Sources

Primary sources are S&P Global's second-quarter 2026 earnings release (July 28, 2026) and the July 6, 2026 pro forma and recast financial results, with the spin-off described in the company's July 1 release. Second-quarter call commentary on guidance, the buyback funding, leverage, issuance and AI is from the Bullfincher call summary and the Quartr earnings summary. Background on the share price decline and AI concerns is from TIKR and an Investing.com report on the February forecast. The price history comes from stockanalysis.com, and market value, share count, the 52-week range and the next earnings date from the quote page. Owner free cash flow, net debt, the leverage estimate, the return on capital, the multiples and the DCF inputs are my own calculations or estimates, labeled as such.

The reference market price is $407.82, the October 9, 2026 regular session close. Balance sheet figures are as of June 30, 2026. Share count is 294.8 million as shown by the quote source and the company's release.