Built by Grok 4.6 HighMoody's Corporation reported third-quarter 2026 results on Wednesday, October 21, 2026. This review was prepared Monday, October 26, 2026. Quarter ended September 30, 2026. Calendar fiscal year. Next print: Q4 2026 on February 17, 2027.
MCO | Earnings Review — Q3 2026
Verdict: decelerating. Revenue of $2,023 million grew only +0.8% YoY versus $2,007 million. That is a 1,432 bp step-down from Q2 2026's +15.1% and about 990 bp slower than Q3 2025's +10.7%. Adjusted diluted EPS of $4.14 grew +5.6% versus $3.92, down from +31.5% in Q2 and +22.1% a year ago. Adjusted operating margin of 52.6% compressed 30 bps versus 52.9% — the first YoY compression in this window, after +440 bps in Q2. Derived EBITDA of $1,030 million fell 1.0%. The dollar level is still high (third straight quarter above $2 billion). The slope is not.
Double miss, and the streak broke. FMP, pulled October 26, had $2,076 million of revenue and $4.25 of adjusted EPS. Both lines missed by about 2.6% (−$53 million and −$0.11). Adjusted EPS had beaten for 10 straight quarters after the Q4 2023 miss. L12Q is now 10/12 on EPS and 9/12 on revenue. Surprise magnitude was already narrowing into Q1; Q2's +$0.42 / +4.7% was the June pull-forward management refused to annualize; Q3 gave it back. GAAP diluted EPS of $4.24 rose 17.8% versus $3.60 because a divestiture gain sits in GAAP and is stripped from adjusted. Adjusted is the series the Street underwrites, and it slowed.
The miss is leveraged-finance mix inside MIS, not Moody's Analytics and not a cost blow-up. Rated issuance was $1,737 billion, −2% versus $1,764 billion. Transaction revenue fell faster, −5%, because the quarter was lower-yielding jumbo and frequent-issuer paper. Corporate finance revenue fell 8% to $530 million: investment grade $146 million was +18%, high yield $68 million was −40%, and bank loans $113 million were −27%. MIS external revenue of $1,087 million was −1.0%, below July's "low-single-digit" color. MA external revenue of $936 million was only +3.0% reported, but organic constant-currency revenue was $938 million versus a restated $867 million, +8.2%, and MA margin expanded 100 bps to 35.3%. Analytics did not cause the miss and was not large enough to offset it.
Guidance was reaffirmed, not raised, and the high-single-digit label only survives at the low end. Adjusted EPS stays $16.50–$17.00 (midpoint $16.75, which management called 12% growth versus FY2025 $14.94). Adjusted operating margin stays 52%–53%. Nine-month revenue is $6,287 million, +7.9%, and nine-month adjusted EPS is $13.15, +16.3%. The year is being carried by the first half. Q4 issuance is assumed down mid-teens, and MIS revenue only low-single-digit. The segment cadence implies a full year around +7%, not 8–9%. FMP's FY2026 EPS average of $16.77 sits on the midpoint. FMP revenue of $8,225 million is about +6.6%, a hair under that low end. The July "guide below the Street" EPS gap is gone. Do not carry it forward.
Tone, contradictions, and what is next. Fauber's frame is resilience inside an "uneven" market, and he reminded the Street, unprompted, that they chose not to raise in July. The bullish line he keeps repeating — AI infrastructure and power issuance up about fourfold, hyperscaler capex about $800 billion in 2026 and north of $1 trillion in 2027, a refinancing wall above $5 trillion — is not in this quarter's −2% issuance, and FMP's FY2027 slope (+6.8% revenue, +11% EPS) does not capitalize it. That is a 2027 duration argument, not a Q4 earnings argument: FMP's Q4 bar ($1,929 million, +2.1%; EPS $3.58 versus an implied midpoint of $3.60) is the company's own residual. Three document conflicts, none of them a silent operating-guide flip: a spoken 50% margin floor versus a stable 52–53% exhibit, a $1.8 billion versus ~$1.76 trillion issuance unit error, and a duplicated KYC 13% sentence. Model the exhibit. SharePoint, OneNote, Outlook, Excel, Visible Alpha, and Bloomberg were unavailable.
| Total revenue | $2,023M (+0.8% YoY, −2.6% vs FMP $2,076M) | Adjusted diluted EPS | $4.14 (+5.6% YoY, −$0.11 / −2.6% vs $4.25) |
| GAAP diluted EPS | $4.24 (+17.8% YoY) — divestiture gain, not the trajectory | Adjusted operating margin | 52.6% (−30 bps YoY) · first compression in the window |
| MIS / MA external | $1,087M (−1.0%) / $936M (+3.0% reported, +8.2% organic CC) | Rated issuance | $1,737B (−2% YoY) · transaction revenue −5% |
| CFG mix | IG +18% · high yield −40% · bank loans −27% | MA ARR | $3,688M · reported stock +9.7% · management ~+8% on the post-divestiture base |
| FY2026 adj. EPS | $16.50–$17.00 reaffirmed · mid $16.75 vs FMP $16.77 | FY2026 revenue | High-single-digit held · segment math ~+7% · FMP +6.6% |
| Q4 implication | Issuance mid-teens lower · implied EPS mid $3.60 vs FMP $3.58 | L12 beat rate | Adj. EPS 10/12 · revenue 9/12 · 10-quarter EPS streak broken |
Columns are quarters. Rows are metrics. Growth is same-quarter YoY, never Q3 dollars versus Q2 dollars. Moody's does not report a gross margin; adjusted operating margin is the company KPI. EBITDA is Derived EBITDA (GAAP operating income plus depreciation and amortization). The gold rule marks Q3 2026. Italics are calculated, not a separate Daloopa series.
| Metric | Q4 24 | Q1 25 | Q2 25 | Q3 25 | Q4 25 | Q1 26 | Q2 26 | Q3 26 |
|---|---|---|---|---|---|---|---|---|
| Revenue ($M) | 1,672 | 1,924 | 1,898 | 2,007 | 1,889 | 2,079 | 2,185 | 2,023 |
| Revenue YoY | +13.0% | +7.7% | +4.5% | +10.7% | +13.0% | +8.1% | +15.1% | +0.8% |
| MIS external ($M) | 809 | 1,065 | 1,010 | 1,098 | 946 | 1,153 | 1,260 | 1,087 |
| MIS external YoY | +18.3% | +7.9% | −0.5% | +11.8% | +16.9% | +8.3% | +24.8% | −1.0% |
| MA external ($M) | 863 | 859 | 888 | 909 | 943 | 926 | 925 | 936 |
| MA external YoY | +8.4% | +7.5% | +10.7% | +9.4% | +9.3% | +7.8% | +4.2% | +3.0% |
| Adj. operating margin | 43.8% | 51.7% | 50.9% | 52.9% | 48.7% | 53.2% | 55.3% | 52.6% |
| Margin YoY (bps) | +120 | +100 | +130 | +510 | +490 | +150 | +440 | −30 |
| Derived EBITDA ($M) | 674 | 959 | 938 | 1,040 | 894 | 1,044 | 1,172 | 1,030 |
| EBITDA YoY | +13.3% | +6.4% | +6.0% | +22.9% | +32.6% | +8.9% | +24.9% | −1.0% |
| Adjusted diluted EPS | 2.62 | 3.83 | 3.56 | 3.92 | 3.64 | 4.33 | 4.68 | 4.14 |
| Adj. EPS YoY | +19.6% | +13.6% | +8.5% | +22.1% | +38.9% | +13.1% | +31.5% | +5.6% |
| Rated issuance YoY | +33% | +9% | −12% | +15% | +10% | +6% | +33% | −2% |
| MA ARR ($M) | 3,278 | 3,266 | 3,297 | 3,361 | 3,498 | 3,607 | 3,661 | 3,688 |
What moved the quarter — ratings mix, not the subscription book
PPIF is the only MIS line that accelerated. Corporate finance is the air pocket. High yield and bank loans together fell $88 million. Investment grade rose $22 million. That is not a broad ratings recession.
| Q3 line | Q3 2025 | Q3 2026 | YoY |
|---|---|---|---|
| CFG investment grade ($M) | 124 | 146 | +17.7% |
| CFG high yield ($M) | 114 | 68 | −40.4% |
| CFG bank loans ($M) | 155 | 113 | −27.1% |
| Corporate finance ($M) | 576 | 530 | −8.0% |
| Structured finance ($M) | 146 | 153 | +4.8% |
| Financial institutions ($M) | 208 | 213 | +2.4% |
| PPIF ($M) | 161 | 179 | +11.2% |
| IG / HY / loan issuance YoY | IG −6% | IG +16% | HY −21% · loans −35% |
Issuance growth rolled over: +15% in Q3 2025, +10% in Q4 2025, +6% in Q1 2026, +33% in Q2 2026, −2% in Q3 2026. AI-related issuance was a little over half of year-to-date issuance growth and closer to 40% of ratings revenue growth, because jumbo hyperscaler paper sits on frequent-issuer contracts (Fauber). Volume outran yield. That is why a −2% issuance quarter became a −1% MIS revenue quarter.
Five completed years
FY2026 is incomplete. Annual revenue already slowed from +19.8% in FY2024 to +8.9% in FY2025, with margins still expanding. Q3 2026 is the first quarter in the window where that margin expansion flipped negative at the same time revenue growth fell under 1%.
Green is a beat versus FMP. Red is a miss. There is no in-line band. The gold rule is this quarter. Street EPS is adjusted diluted EPS; it matches the Daloopa series on every quarter in the window, including $4.14. FMP EBITDA and EBIT averages are a different construct and are not scored.
| Metric | Q4 24 | Q1 25 | Q2 25 | Q3 25 | Q4 25 | Q1 26 | Q2 26 | Q3 26 |
|---|---|---|---|---|---|---|---|---|
| Revenue surprise | −1.8% | +2.6% | +2.6% | +2.6% | +1.5% | +0.5% | +4.7% | −2.6% |
| Adj. EPS $ surprise | +$0.35 | +$0.29 | +$0.17 | +$0.22 | +$0.21 | +$0.11 | +$0.42 | −$0.11 |
| Revenue YoY | +13.0% | +7.7% | +4.5% | +10.7% | +13.0% | +8.1% | +15.1% | +0.8% |
| Adj. EPS YoY | +19.6% | +13.6% | +8.5% | +22.1% | +38.9% | +13.1% | +31.5% | +5.6% |
Pattern: a consistent beater whose streak just ended, with deteriorating magnitude. L12Q adjusted-EPS beat rate is 10/12 (83%). L4Q is 3/4. Revenue L12Q is 9/12; the misses are Q4 2023 (−0.3%), Q4 2024 (−1.8%), and this quarter (−2.6%), and they are getting larger. Strip the Q2 spike management labeled a pull-forward and the EPS dollar surprise still narrows into the miss, but it is not a straight line: +$0.35, +$0.29, +$0.17, then a step up to +$0.22, then +$0.21, +$0.11, −$0.11. L4Q average EPS surprise is +$0.16, down from +$0.26 in the prior four quarters. Read Q2 and Q3 together and the two-quarter EPS surprise is +$0.31, not +$0.42.
This quarter's actuals: revenue $2,023 million versus FMP $2,076 million; adjusted EPS $4.14 versus FMP $4.25. The 14-analyst EPS mean ($4.25) and the 13-analyst revenue mean ($2,076 million) give the same call. July had described Q3 MIS revenue as low-single-digit. Actual MIS external revenue was −1.0%. The miss is versus the Street and versus Moody's own qualitative cadence. It is not a cut of the full-year range.
Every P&L line on the October 21 outlook table is marked unchanged versus July 22. The range did not move. The mix did. High yield and leveraged loans were tempered. Structured finance, financial institutions, and PPIF were raised. Q4 issuance is now a mid-teens decline. The speculative-grade default assumption rose from 3.0% to 3.8%. U.S. high-yield spread assumption tightened from 330 bps to 300 bps. Credit got worse. The technical got tighter. GDP, inflation, unemployment, and FX were unchanged. The earnings release, not the call transcript, still assumes one more Fed hike and one more ECB hike by year-end. That sentence is not in the call file and has no Daloopa id.
Adjusted EPS waterfall — the only move this year was July's floor
The midpoint is +12.1% versus FY2025 $14.94, a deceleration from FY2025's +19.8%. The low end is +10.4%. The high end is +13.8%. Nine-month adjusted EPS is already $13.15 of a $16.50 floor, so the range survived a weak quarter. Implied Q4 adjusted EPS, a residual and not a guide, is $3.35 / $3.60 / $3.85. Versus Q4 2025 $3.64 that is −8.0% / −1.1% / +5.8%. The midpoint no longer needs Q4 EPS growth. H2 at the midpoint is only about +2.4% versus H2 2025.
What was reaffirmed, and what actually changed
| Item | New guide | vs July | vs FMP |
|---|---|---|---|
| MCO revenue growth | High-single-digit | Unchanged | Street +6.6% ($8,225M) |
| Adj. operating margin | 52%–53% | Unchanged | No FMP margin |
| Adj. diluted EPS | $16.50–$17.00 | Unchanged | $16.77, on the mid |
| GAAP diluted EPS | $16.00–$16.50 | Unchanged | FMP tracks adjusted |
| Free cash flow | $2.7B–$2.9B | Unchanged | No FMP FCF |
| Share repurchases | Up to $3.0B | Unchanged | YTD $2.6B |
| Rated issuance | Mid-single-digit | Total held; mix worse | No FMP issuance |
| Leveraged loans / PPIF | −10% / +20% | Loans tempered; PPIF raised | — |
| Investment-grade issuance | +10% | Unchanged | Q3 just did +16% |
| MA ARR / MA margin | HSD · 34%–35% | Unchanged | No FMP ARR |
| First-time mandates | 900–950 | Raised from 750–850 | Call only, not Daloopa |
| Spec-grade default, year-end | 3.8% | +80 bps vs 3.0% | — |
Why "high-single-digit" is a low-end statement. Nine-month revenue of $6,287 million is already +7.9%. MIS nine-month growth of about +10% plus a low-single-digit Q4 on the $946 million base still leaves full-year MIS around +8%. MA nine months are already inside the mid-single-digit reported guide. Putting those together, Q4 consolidated revenue lands around $1.95–$1.97 billion, about +3% to +4% versus $1,889 million, and the full year around +6.7% to +7.0% ($6,287 million plus a $1.95–$1.97 billion Q4, on the $7,718 million FY2025 base). A true 8% year would need Q4 revenue near $2,048 million, hotter than the cadence just described. FMP Q4 revenue of about $1,929 million (+2.1%) sits inside that single-digit bridge, a bit below it. FMP Q4 EPS of $3.58 is $0.02 under the $3.60 residual. The earnings-file next-print figure is $3.62. Neither is a heroic bar.
Revenue-growth bridge, the label versus the math. FY2025 actual growth was +8.9%. Nine months of 2026 are +7.9%. Q3 was +0.8%. The segment-implied full year is about +7%. The spoken guide is still "high-single-digit." FMP is +6.6%. Issuance went from low-single-digit (April) to mid-single-digit (July, raised) to mid-single-digit (October, held). The October edit is inside the stack.
Management test. April held the guide through an issuance air pocket. July banked a blowout as timing, raised issuance and the EPS floor only, and pre-announced a softer second half. October delivered a quarter a bit worse than that low-single-digit MIS comment and reaffirmed rather than cut. That is a team that guides the year tightly. It is not a team still ahead of the second-half path it described in July. The next tell is Q4 issuance, not another adjective on the funding supercycle.
Three inflections, not a smooth trend. Acceleration is the change in the YoY rate versus the prior quarter's YoY rate, in basis points. It is not a comparison of sequential dollars.
| Metric | Q4 24 | Q1 25 | Q2 25 | Q3 25 | Q4 25 | Q1 26 | Q2 26 | Q3 26 |
|---|---|---|---|---|---|---|---|---|
| Revenue YoY | +13.0% | +7.7% | +4.5% | +10.7% | +13.0% | +8.1% | +15.1% | +0.8% |
| Revenue accel (bps) | −1,019 | −525 | −327 | +624 | +228 | −492 | +707 | −1,432 |
| Adj. EPS YoY | +19.6% | +13.6% | +8.5% | +22.1% | +38.9% | +13.1% | +31.5% | +5.6% |
| EPS accel (bps) | −1,246 | −598 | −511 | +1,358 | +1,681 | −2,588 | +1,841 | −2,585 |
- Q4 2024 through Q2 2025 — the deceleration leg. Both rates fell for three straight quarters. Revenue troughed at +4.5%. Adjusted EPS troughed at +8.5%.
- Q3 2025 through Q2 2026 — a choppy reacceleration. Revenue's Q2 2026 +15.1% (+707 bps) was a one-quarter MIS issuance spike, not a new plateau. Q1 2026 had already given back 492 bps. EPS peaked earlier, at +38.9% in Q4 2025, air-pocketed in Q1 (−2,588 bps), then echoed in Q2.
- Q3 2026 — both series break at once. Revenue +0.8% is below the prior trough. EPS +5.6% is the slowest EPS rate in the eight-quarter window. The −1,432 bp revenue move is the largest in the window. The Q2 peak did not hold for a single quarter.
$2,023 million is not a broken franchise. It is $16 million above last year's record third quarter, and $4.14 is the third-highest adjusted EPS in the window. Nine-month growth of +7.9% hides the break. It should not be used as the run-rate. GAAP EPS growing faster (+17.8%) does not improve the operating read: adjusted is below GAAP this quarter because gains are being taken out.
The forward read is the reaffirmed guide, not a new raise. Q4 issuance down mid-teens and MIS revenue only low-single-digit is an explicit statement that the next ratings quarter stays slow. FMP's FY2026 averages ($8,225 million of revenue, $16.77 of EPS) live inside that deceleration, absorbed by a strong first half.
The catalyst the Street underweighted in Q3 was leveraged finance, not AI. Management did not cut the year. Five days after the print, FMP's FY2026 EPS average sits on the midpoint, and the Q4 EPS average is the company's own implied fourth quarter. The open debate is 2027 duration: further hyperscaler issuance in 2026 is upside to a guide that assumes that financing is already done.
| Catalyst | Timing | What is already in the bar | What would change the view |
|---|---|---|---|
| Q4 issuance, mid-teens lower | Now through year-end; print 2027-02-17 | FMP Q4 revenue about $1,929M (+2.1% vs $1,889M). FMP EPS $3.58 vs implied mid $3.60 and vs $3.64. | A hyperscaler or M&A print the guide excluded is upside. Leveraged finance staying shut slides the year toward $16.50. |
| 2027 AI capex and hyperscaler debt | First hard read is the February guide | FMP FY2027 is $8,787M revenue and $18.65 EPS, about +6.8% and +11%. A normal year, not a capitalized supercycle. | Fauber: about $800B of 2026 capex and north of $1T in 2027, with issuance growth continuing. Yield dilution (about 40% of revenue growth vs half of issuance growth) means the translation is smaller than the volume. |
| MA Q4 renewal and ARR | Largest sales quarter is Q4 | High-single-digit ARR is the guide. Q3 stock $3,688M. Retention 95% is a call figure. FMP does not require ARR to accelerate. | A price step-up on AI workflows would have to show up in renewals. Management said those motions are still proof-of-value. |
| MCP, agents, 2027 pricing | Usage now; Christina early 2027 | Not in any FMP line. Users more than quadrupled and usage rose more than tenfold in three months, off a small base. Paying MCP customers more than doubled. | Not a Q4 P&L catalyst. Do not add it on top of ARR until there is a price. |
| Private credit vs BSL rotation | Q3 was the rotation quarter | Rated private-deal count +40% (call). PitchBook, via Fauber: about $12B of direct lending refinanced into BSLs, the highest quarter since 2022 tracking began. Loan issuance −35%. | Moody's rates both sides, so the migration is not lost volume. The public-market slowdown dominated the print. Mandates raised to 900–950 are the leading indicator; MIS recurring revenue $373M was only +8%. |
| S&P Global Q3 | 2026-10-27 | Same end markets. MCO has already shown the IG-versus-leveraged split. | If S&P underwrites more 2026 or 2027 hyperscaler issuance than Moody's did, the "upside if they issue" framing gets tested within a week. A matching leveraged-finance cut means the mix is an industry fact, not share loss. |
| Rates, defaults, midterms | FOMC Oct 27–28; midterms Nov 3; December hike is the assumption | Default assumption already 3.8%. An October hold does not change the guide. The release's year-end hike assumption sits beside the mid-teens issuance decline. It is not a quoted call sentence. | Issuance pulled into November ahead of a new Congress would help Q4 and steal from Q1 2027. A skipped December hike would be the dovish surprise. |
Year-to-date free cash flow was $2,200 million, +22% versus $1,798 million. Management said the quarter itself was only about +2%. The 22% is an H1 story. Q4 needs roughly $0.5–$0.7 billion of free cash flow to land in the $2.7 billion–$2.9 billion range. That is not a new catalyst. About $0.4 billion of repurchase capacity is left under the $3 billion cap. FMP grades are 18 buy / 13 hold / 1 sell — buy-leaning, not one-sided, and not a substitute for the issuance forecast. No price target is used here.
Banking reported revenue looks broken and is not. The earnings release put organic constant-currency banking revenue at +19% against a reported decline, because Learning Solutions and regulatory reporting sat in that segment and were sold. Do not model another year of reported banking decline as the franchise. Insurance ARR of $736 million is the clean MA growth line; management said about +10%, with casualty still small. KYC ARR of $489 million grew on the release's comparative; the call said +12%. October AML fines confirm a demand story already printing. They are not a 2026 revenue bridge, and the Orbis-to-Maxsight migration is late.
Fauber answered every question. Heuland gave the prepared financials and never took one. The operator allowed one question each. Nine analysts got through. Tally: 5 well answered, 4 deflected. The deflections are all quantification. The well-answered set is mechanism, mix, and stance.
| Analyst | Topic | Verdict |
|---|---|---|
| George Tong, Goldman Sachs | Is the 4x MCP-user / 10x usage jump incremental ARR, or retention? | Deflected |
| Andrew Steinerman, J.P. Morgan | Q4 issuance puts and takes, AI and M&A financing, a shorter calendar | Well answered |
| Jason Haas, Wells Fargo | Do recent KYC / AML fines become a demand driver? | Well answered |
| Scott Wurtzel, Wolfe Research | Mature banking is slowing — how do you accelerate it? | Well answered |
| Craig Huber, Huber Research | Hyperscaler rated-debt dollars and share of global issuance, 2026 and 2027 | Deflected |
| Alex Kramm, UBS | Private-credit appetite and smaller agencies' rating quality | Well answered |
| David Motemaden, Evercore | Is third-party casualty capital already in the results? | Well answered |
| Owen Lau, Clear Street | MCP distribution versus Moody's own channel, without cannibalizing the platform | Deflected |
| Peter Christiansen, Citi | Pricing into the largest MA renewal quarter | Deflected |
What was usable. Steinerman got the Q4 algorithm: mid-teens issuance decline, hyperscalers assumed done for 2026 so any sizable deal is upside, a healthy November pipeline, opportunistic leveraged finance selective as Treasuries widened, and headline risk from two wars and the U.S. midterms. He did not get a dollar on M&A financing. Haas got a straight read that the fines are a market-wide reminder, plus an admission that the Maxsight migration is behind. Kramm got the private-credit stance: same analysts on public and private, rate the BSL rotation too, and partnerships with MSCI and Allvue to push an independent probability of default. Motemaden was told casualty is still small. Wurtzel got the banking cleanup — learning and regulatory reporting were in that segment; the focus is lending (ARR +12% on the call) and a decision-intelligence layer. He did not get a bridge from "banking ARR +8%" to the reported stock of $412 million versus a pre-divestiture $466 million, −11.6%.
What was dodged, and why it matters. Tong asked for incremental ARR versus retention. Fauber named the dodge ("I know people want that") and gave usage, a doubled-but-small paying cohort, and an unquantified pipeline. No share of the $3,688 million ARR stock, and no test for when AI lifts growth above the high-single-digit guide that was just reaffirmed. Huber asked for hyperscaler rated-debt dollars and the percent of global issuance. He got capex of about $800 billion this year and north of $1 trillion next year, and the 40% revenue-versus-half-of-issuance-growth split, and not the debt number. Lau asked how MCP entitlements do not cannibalize Moody's own platform. The answer was "we will do all three channels," with the pricing model deferred to early 2027. Christiansen asked what customers are paying into the largest renewal quarter. The answer was that budgets exist, proof-of-value comes first, and the ROI is clear — no renewal price, no uplift.
That is the contrarian gap on the call, and it is narrow. Management has been repeating that AI is already in the product and that the financing supercycle extends into 2027. The Street cannot underwrite either one in a number that is not already in the high-single-digit ARR guide or the mid-single-digit issuance guide. The CFO, who owns the bridge from usage to revenue, did not get a question.
No Q3 operating-guide flip. Management's message is internally consistent: hold the full-year P&L guide, issuance still mid-single-digit with a worse mix, Q4 issuance down mid-teens, MA ARR still high-single-digit, buybacks still up to $3 billion. Three document conflicts will corrupt a model if they are taken literally. One is high. Two are medium. They are source errors, not a walk-back.
On the February 18, 2026 call Heuland said FY2026 adjusted operating margin was "expanding by 150 bps to the 50% to 53% range." On this call she said the company is "maintaining" adjusted operating margin of 52%–53%. The exhibit never printed 50%. It is low 52% and high 53%, and it was 52%–53% at the original guide, at Q1, and at Q2.
FY2025 margin was 51.1%. Adding 150 bps lands at 52.6%, inside 52–53 and on this quarter's 52.6%. A 50% floor would be compression, which fights the same sentence. Do not model 50%, and do not treat Q3 as a 200 bp raise of the floor. The coherent reading is a speech-to-text error. The authoritative range is the exhibit.
The year-ago transcript says issuance "totaled nearly $1.8 billion," the record third quarter. This call says about $1.7 trillion, down 2%, the second-highest third quarter. Those cannot both be the history. The filing is $1,764 billion in Q3 2025 and $1,737 billion in Q3 2026, a change of −2%.
Use the filing. The 2025 transcript's unit is wrong by three orders of magnitude. The just-reported call and the exhibit agree. This is not a Q3 2026 management contradiction.
Heuland's Q1 and Q2 prepared remarks use the same freeform clause — "KYC grew 13%, driven by deeper penetration within existing banking customers and expansion beyond financial services" — while the surrounding paragraph was rewritten. Q3 breaks the pattern: "KYC ARR grew 12%," with a different reason. Two independent year-over-year prints three months apart are unlikely to share that sentence.
Do not cite "KYC grew 13%" for both Q1 and Q2 until a press release or the 10-Q confirms both. Q3's 12% is a separate claim and is not part of the copy. The reported KYC ARR stock is $489 million versus a pre-divestiture $407 million. That stock growth is not the same series as the spoken rate.
Checked and not contradictions: the first-time-mandate raise (labeled, and a different metric from the P&L guide that was reaffirmed); the issuance mix shift inside an unchanged mid-single-digit total; the July EPS-floor raise that October held; the earlier free-cash-flow trim; AI issuance at half of volume growth and about 40% of revenue growth (reconciled in the same answer as frequent-issuer pricing); "users quadrupled" versus "paying customers doubled" (different populations); and hyperscalers "mostly done for 2026" versus growth in 2027 (different years). Q1's promise that KYC would pick back up into the mid-teens, versus Q3's 12%, is a miss against their own language. It is not two descriptions of the same past fact.
October's macro message is narrower than July's. Inflation and the consumer were not discussed. The rate story got worse for opportunistic credit. The industrial story got more specific, not softer. Private credit is being questioned and partly refinanced into public loans, while Moody's rated private-deal count still grew.
Rates and credit. Treasuries widened on deficits and a heavy issuance calendar. Spreads widened about 10% late in September and stayed near historical tights (Heuland). High-yield revenue −40% and bank-loan revenue −27%. Issuers with real funding needs are still coming. Opportunistic leveraged-finance issuers are waiting. The default assumption moved the wrong way even as the spread assumption tightened. That is worse for sub-investment-grade issuers and for anyone whose Q4 model still has a high-yield window. It is not a shut investment-grade market: IG issuance was +16%.
The supercycle that did not show up in the consolidated rate. AI-infrastructure and power issuance was up about fourfold versus last year's third quarter (Fauber). Rating-analyst capex is about $800 billion in 2026 and north of $1 trillion in 2027, and he expects hyperscaler issuance to grow next year. The four-year U.S. and EMEA refinancing wall still tops $5 trillion. Investment-grade maturities grew about 5%, the U.S. about 11%. On a five-year lens, U.S. refinancing needs grew about 4% to about $3.5 trillion, and the peak maturity year moved from 2028 to 2031. Better for data-center and power suppliers feeding that spend, and for 2027 duration. Not a 2026 bond-supply surge: management thinks this year's hyperscaler financing is largely done. Do not map the dollars onto a named hyperscaler. None was named. Slightly worse, tactically, for other investment-grade issuers competing for the remaining windows.
| Name | Relationship | Read-through |
|---|---|---|
| Hyperscalers, unnamed | Rated issuers; funding end of the AI build-out | Better for 2027 capex and for the suppliers feeding it. 2026 financing is assumed mostly done, so more size this year is upside to Moody's, not the base. No issuer was named. |
| Apollo (AMAP VI) | Issuer; private-credit structure Moody's rated | Better for Apollo's structuring franchise. Evergreen multi-strategy vehicles are still getting rated in size. Not a credit-quality criticism. Sits next to the refinancing rotation into public loans. |
| Broadly syndicated loans | Destination of the private-credit refi | PitchBook, via Fauber: about $12 billion of direct-lending deals refinanced into BSLs, the highest quarter since 2022. Better for loan arrangers. Worse, on the margin, for direct lenders who lost the refi. Moody's rates both. |
| MSCI and Allvue | Partners on private-credit risk language | Better for both as distribution of an independent probability of default. The tell is that investors are asking harder questions after the headlines. Mildly worse for the "trust the yield" private-credit pitch. Not framed as MSCI taking share from Moody's. |
| Smaller rating agencies, unnamed | Competitors in private credit | Worse for whichever agencies the market sees as easier: Fauber said recent deals priced closer to Moody's when the rating diverged or was absent. Do not attach this to a named competitor. None was named. |
| SAP Ariba | Channel; supplier intelligence embedded in procurement | Better for SAP Ariba and for Moody's distribution. Worse for the displaced supplier-data incumbent, who was not named. Supplier-risk data is being pulled into the workflow. |
| Lloyd's of London | Partner on casualty and financial lines | Confirms casualty is a strategic underwriting problem at a core market. Still a small part of Moody's insurance results. Intent, not a current earnings driver. |
| European Commission (Arachne) | Customer; public-sector fraud control | Better for EU funds-protection spend. The relationship was renewed, not rolled off. Pairs with an unnamed German federal agency on investigative workflows. |
| PhilRatings | Affiliate; minority stake being acquired | Sixth domestic rating affiliate in Asia Pacific. Management cited more than $100 billion of Philippine infrastructure investment planned by 2030. Onshore bonds dwarf cross-border holdings. The $100 billion is their statement, not a booked Moody's number. |
| Meta, as Facebook | Mass-tort example, not a customer | Casualty models flagged addictive-software tort risk around 2018; Fauber cited this year's settlement as confirmation. A proof point for casualty analytics. Not a 2026 operating update on Meta. |
Unnamed and not inferred: a top-three U.S. auto and property insurer more than doubled ARR on the Intelligent Risk Platform; a top-tier reinsurance broker signed five years at nearly +14% ARR; a top-five U.S. bank displaced an incumbent on screening; one of the largest U.S. banks has 8,000 users on an AI surface and wants Moody's content in it; a new Mexican bank logo completed the top five on the lending suite; a quantitative firm above $1 trillion of AUM converted a OneView trial; one of the world's largest asset managers embedded Orbis in a private-markets platform. Each is one logo. None is an industry conversion. The pattern is embed-don't-portal, and the monetization is still conditional on proof of value.
S&P Global reports Q3 on October 27. Moody's has already shown the mix. The useful question is whether S&P describes the same investment-grade-versus-leveraged split and the same "hyperscalers mostly done for 2026" assumption. M&A was described as pipeline support and as an upside case, not as a realized surge. No positive read-through that sponsor M&A has reaccelerated.
Moody's is still the ratings franchise, and the year they guided in July is still the year. The quarter is not that year. Revenue growth fell to +0.8%, adjusted EPS growth to +5.6%, and the margin stopped expanding, because leveraged finance gave back a hard compare and the investment-grade and AI paper that replaced it paid a lower fee. The 10-quarter EPS beat streak broke by 2.6% on both lines. Management's choice not to raise, twice, was the right description of the second half and a miss of their own low-single-digit MIS color. FMP has already moved onto the midpoint. The number that is not in the model is 2027 hyperscaler issuance, and they would not put a debt dollar on it. Until February, treat further 2026 hyperscaler paper as upside and a mid-teens Q4 issuance decline as the base. Hold.