Apollo Global Management APO
Financial Services · Asset Management · Synthos Deep Dive · 2026-08-04
The Overview
Apollo does two things that look unrelated and are not.
First, it manages money for other people — private equity, and especially private credit, meaning loans made outside the banking system. Second, since merging with a company called Athene, it sells retirement annuities. People hand Apollo money now in exchange for income later, and Apollo invests that money — largely in loans it makes itself.
That is the clever part of the design, and independent commentators in our research base rate it highly. An ordinary fund manager has to go and raise money every few years. Apollo's annuity business supplies money that stays for decades. So Apollo lends, holds the loans against the annuities, earns the difference, and charges a fee on top.
Right now the two halves are moving in opposite directions. In the first three months of the year, the fee business earned $728 million, up 30%. The annuity spread business earned $719 million, down 11% — mostly because what Apollo has to pay annuity holders rose 27%.
Apollo reported its April-to-June results this morning. Profit per share of $2.11 came in slightly below the $2.16 expected — the first miss in five quarters. The shares went up 2.85% anyway, to $133.11.
Two warnings you should hear.
The first is about the data. This is the messiest company file we have examined. Because Apollo consolidates an insurance company, the "revenue" figure in standard data feeds is meaningless — the same file contains three different revenue numbers for overlapping periods that differ by more than twenty times. The "free cash flow" figure is also wrong, because the capital-spending line has been recorded as exactly zero for three years running. We have thrown away every ratio built on either.
The second is about private credit. Several independent commentators — including the single most highly-weighted analyst we track — have been arguing since March that private credit funds are facing a wave of withdrawals and that the value they place on their loans is too optimistic. One names Apollo's own retail credit fund specifically, saying withdrawal requests rose from 11% to 16.8% of the fund. We could not check this. That fund reports separately and none of the documents we have covers it. We report the claim, we report that we cannot verify it, and we treat the uncertainty as a reason to wait.
Our fair value estimate is $144, about 8% above the price. Analysts on average say $150.67. Neither is far enough above $133 to justify buying into a name we cannot fully audit.
- Downside Risk 8/10. Leveraged credit, an insurance balance sheet, and an unverifiable redemption question.
- Growth Quality 7/10. Fees +30%. Spread −11%. They are now the same size.
- Exponential Potential 4/10. An excellent structural idea attached to a cyclical spread machine.
Putting a number on it: our fair-value estimate is $144 against a current price of $135.04 — real upside if our numbers are right.
Our summary metrics
"Rated 8 — a leveraged credit business inside a consolidated insurance balance sheet, in the middle of a private-credit redemption debate its own filings cannot settle, with the least usable payload of any name in this batch bar one. The supports: Fee Related Earnings of $728 million in the March quarter, up 30.2% year on year, which is the annuity-like part of the firm; a beta-adjusted franchise with 6,140 employees managing an institutional and retail asset base; $19,240 million of cash; total debt of only $13,364 million against $460,949 million of total assets; interest coverage of 28.6x; and a dividend just declared at $0.5625 per share for the June quarter. Against that, six things. ONE, the balance sheet is not the asset manager's: $460,949 million of total assets and $418,434 million of liabilities are overwhelmingly Athene's policyholder obligations, and $228,820 million of what the payload labels short-term investments is an insurance portfolio, not liquidity — `cashPerShareTTM` of $425.52 against a $133.11 share price is the arithmetic proof that the vendor's framework does not fit. TWO, Spread Related Earnings FELL 10.6% to $719 million in the March quarter, with alternative net investment income down 33% to $210 million, while the cost of funds rose 27% to $2,807 million. THREE, the knowledge base's most specific recent claims describe escalating redemptions in Apollo's own retail credit vehicle, and no filing in this archive tests them. FOUR, GAAP earnings are violently unstable: the March 2026 quarter reported a net LOSS of $1,906 million and diluted EPS of MINUS $3.24 on a GAAP basis, in the same quarter that adjusted EPS was $1.94. FIVE, beta is 1.508 and the maximum drawdown from peak over the trailing year was 25.5%. SIX, results were reported this morning and MISSED the adjusted consensus by 2.3%."
"Rated 7 — the growth is real, it is concentrated in the right half of the firm, and the other half is going backwards. The March 2026 quarter, filing-verified from the 10-Q's segment reconciliation: FEE RELATED EARNINGS $728 million against $559 million, up 30.2%, driven by fee-related performance fees of $64 million (from $54 million) against fee-related compensation of $333 million (from $259 million) and other operating expenses of $201 million (from $160 million). SPREAD RELATED EARNINGS $719 million against $804 million, DOWN 10.6%, with fixed income and other net investment income rising to $3,551 million from $2,914 million but alternative net investment income falling to $210 million from $315 million and the cost of funds rising to $2,807 million from $2,210 million. So the fee engine grew 30% while the spread engine shrank 11%, and for the first time the two are almost exactly equal in size. A separate 8-K filed 2026-07-01 gives a preliminary second-quarter alternative net investment income of approximately $350 million, an estimated 9% annualised return — a recovery from the March quarter's $210 million. Consensus wants adjusted EPS of $8.846 (FY2026, 11 analysts), $10.665 (FY2027, 11) and $12.495 (FY2028, only 3) — 20.6% then 17.2% growth. What holds this at 7: the FY2028 estimate rests on three analysts, the FY2029 row on ONE, and the consensus revenue series is internally incoherent, collapsing from $30,518 million in FY2028 to $8,425 million in FY2029, which is a basis change rather than a forecast."
"Rated 4 — a genuinely differentiated business model that is nonetheless a spread and fee business, not a compounding technology. The knowledge base describes the structural idea better than we can, at conviction 90: 'Merging asset manager with insurance balance sheet (Athene) creates principal-aligned merchant model that beats agent-only asset management.' The mechanism is real — permanent insurance capital removes the fundraising cycle from a large part of the asset base, and a 2025 claim puts it at '450B of 750B AUM' being perpetual — and it is why Apollo's Fee Related Earnings can grow 30% in a quarter its peers describe as a dead fundraising environment. A second claim (conviction 82) identifies the origination machine — 'Apollo's 4,000-person proprietary origination machine and cross-capital-structure culture is very hard to replicate' — which is the supply side of the same idea: an insurer needs investment-grade assets and Apollo manufactures them. Both are durable competitive advantages. Neither is exponential. What Apollo does is originate credit, hold it against annuity liabilities, and earn the spread plus a management fee; the growth rate is bounded by annuity sales, origination capacity and the credit cycle, and the last of those is the one the knowledge base is currently arguing about. A 4: an excellent structural idea attached to a cyclical, leveraged, spread-earning machine."
What does “fair value” mean?
Fair value is Synthos’s estimate of what one share is worth today, based on our model of the company’s future cash generation and the risks to it. It is not a price target or a prediction of where the stock trades next quarter — it’s the price where we think risk and reward are balanced. Above it, you’re paying for outcomes better than our base case; below it, the market is offering a margin of safety.
The model’s inputs and weightings are proprietary — they’re the product. What we publish is the output, plus an independent, public-math cross-check further down the page (“Check our number”) so you can judge it for yourself.
What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
The Road Ahead
What we expect to matter in each window, and the evidence that would prove us wrong.
Short term 0-6 months
No differentiated view- Driver
- "This dive is struck on the afternoon of an earnings day on which the company missed. Apollo reported second-quarter 2026 results this morning; the earnings calendar records adjusted EPS of $2.11 against a $2.16 estimate, a 2.3% shortfall, breaking a run of four consecutive beats (+4.3%, +12.6%, +21.1%, +2.6%). The stock nonetheless rose 2.85% to $133.11 from $129.42, opening at $127.95, trading as low as $126.4253 and closing within 0.5% of the day high of $133.73 on 3,722,429 shares — a session that opened down and closed up 5.3% from its low. The chief executive's quoted framing in the 8-K is 'record earnings across Asset Management and Retirement Services'. The technical position is constructive and stretched: 12.8% below the 52-week high of $152.70 and 32.7% above the low of $100.30, 6.1% above a 50-day average of $125.40 and 4.8% above a 200-day average of $126.97, with RSI at 67.96 — the highest in this batch and approaching overbought — and MACD at +0.92. Twelve-month return is −3.7% against SPY's +24.3%, a 28-point deficit, but the maximum drawdown from peak was 25.5%, so this is a stock recovering from a large decline rather than one in an uptrend."
- What we’re watching
- "The detailed second-quarter figures, which are NOT in this archive: the 8-K carries only a summary release and directs readers to a presentation on the company's investor-relations website. Watch Fee Related Earnings against the March quarter's $728 million and +30.2%, and Spread Related Earnings against $719 million and −10.6%. Watch alternative net investment income against the preliminary $350 million the company flagged on 2026-07-01, which would be a recovery from $210 million in March. Watch the cost of funds, which rose 27% year on year to $2,807 million in the March quarter and is the single largest pressure on Spread Related Earnings. And watch for any disclosure of retail credit-vehicle flows: two knowledge-base claims from June 2026 assert redemptions in Apollo Debt Solutions escalating from 11% to 16.8% of shares with net flows negative, and nothing in this archive confirms or refutes it."
- Confidence
- Low
Medium term 6-24 months
No differentiated view- Driver
- "The medium term turns on whether the fee engine's 30% growth outruns the spread engine's decline, and on whether the private-credit redemption cycle is a fundraising problem or a credit problem. The constructive read is stated by an independent knowledge-base voice: 'Apollo down 40% is multiple compression from a dead fundraising environment, not a referendum on loan credit quality — if it were credit, it'd be down far more.' The filings partly support it: Fee Related Earnings grew 30.2% in the March quarter, which is not what a firm with a broken franchise produces, and a separate claim notes Apollo 'prudently reduced software lending early (early 2025), ahead of the market'. The bearish read is equally specific and comes from the store's highest-skill voice plus a second independent channel with a redemption percentage attached. Consensus splits the difference, wanting adjusted EPS of $8.846, $10.665 and $12.495 across the next three years — 20.6% then 17.2% growth — against a price of $133.11, which is 15.0x, 12.5x and 10.7x. If the fee engine is worth 15x and the spread engine is worth 8x, and they are now roughly equal in size, the blended multiple is around 11-12x, which is close to where the stock trades. The re-rating case requires Fee Related Earnings to keep compounding at 30% while Spread Related Earnings stabilises."
- What we’re watching
- "Whether Spread Related Earnings stops falling: $719 million against $804 million is the first decline of consequence in the filed record, and the cost of funds is the driver. Whether alternative net investment income normalises above the $350 million preliminary estimate for the June quarter. Whether the private-credit redemption claims are borne out in any disclosed flow data — this is the most important unresolved question on the name and the filings in this archive are silent on it. Whether the fee-related margin holds: fee-related compensation grew 28.6% and other operating expenses 25.6% in the March quarter, both close to the 30.2% Fee Related Earnings growth, so the operating leverage is thinner than the headline suggests. Whether the buyback expands from $773 million, 1.01% of market capitalisation. And whether the analyst coverage on the outer years thickens: FY2028 rests on three analysts and FY2029 on one."
- Confidence
- Low
Long term 2+ years
Tailwind- Driver
- "Long-run the structural argument is strong and the knowledge base makes it at high conviction. Apollo has combined an asset manager with a permanent insurance balance sheet, which converts a cyclical fundraising business into a partly perpetual one — one 2025 claim puts the perpetual portion at '450B of 750B AUM' and calls it 'Apollo's strategic breakthrough, driving the post-2022 stock doubling'. On the supply side sits a large proprietary origination operation that a second claim describes as 'very hard to replicate, creating durable diversified IG collateral'. Together they describe a genuinely differentiated model: Apollo does not compete for mandates so much as manufacture the assets its own insurer needs and sell the surplus. The secular tailwind behind it is documented in a third claim — '87% of US firms with over 100M revenue are now private... we live in a private-markets-heavy world' — and a fourth argues private markets will finance the next capital cycle in AI, robotics and energy 'via non-dilutive capital, crowding out public equities'. Against all of that: this remains a business that earns a spread on credit it originates and holds, and the long-run distribution of that business is set by the credit cycle rather than by the strategy. A knowledge-base claim describes the Berkshire float comparison and correctly rejects it as 'a genuinely different animal rather than a clean analog'."
- What we’re watching
- "Whether the perpetual-capital share of assets under management keeps rising, because that is the mechanism that de-cyclicalises the fee stream. Whether the origination machine scales without loosening credit standards — the whole model depends on manufacturing investment-grade assets faster than the insurer needs them. Whether the private-credit asset class survives its first genuine redemption cycle with its marks intact; the knowledge base has multiple independent voices arguing it will not. Whether the cost of funds on the annuity book normalises: a 27% year-on-year increase is the annuity market repricing, and Apollo does not set that price. Whether founder succession, which one claim says Apollo 'nearly fumbled', holds — Marc Rowan received 7,649,458 votes against at the 2026 annual meeting, the third-highest opposition of thirteen directors. And whether the GAAP and adjusted earnings series ever converge: a quarter reporting adjusted EPS of $1.94 and GAAP EPS of MINUS $3.24 is not a reporting framework a long-term holder can audit."
- Confidence
- Low
Exponential Potential
"Rated 4 — a genuinely differentiated business model that is nonetheless a spread and fee business, not a compounding technology. The knowledge base describes the structural idea better than we can, at conviction 90: 'Merging asset manager with insurance balance sheet (Athene) creates principal-aligned merchant model that beats agent-only asset management.' The mechanism is real — permanent insurance capital removes the fundraising cycle from a large part of the asset base, and a 2025 claim puts it at '450B of 750B AUM' being perpetual — and it is why Apollo's Fee Related Earnings can grow 30% in a quarter its peers describe as a dead fundraising environment. A second claim (conviction 82) identifies the origination machine — 'Apollo's 4,000-person proprietary origination machine and cross-capital-structure culture is very hard to replicate' — which is the supply side of the same idea: an insurer needs investment-grade assets and Apollo manufactures them. Both are durable competitive advantages. Neither is exponential. What Apollo does is originate credit, hold it against annuity liabilities, and earn the spread plus a management fee; the growth rate is bounded by annuity sales, origination capacity and the credit cycle, and the last of those is the one the knowledge base is currently arguing about. A 4: an excellent structural idea attached to a cyclical, leveraged, spread-earning machine."
What could take this further than the base case — and where the market may be underpricing it. Full forward-growth and acceleration math in Deeper analysis, §4 below.
Deeper analysis
Technicals, fundamentals, valuation, and the full expert-claim evidence panel — the detail behind the numbers above.
> # ⚑ EARNINGS TODAY — 2026-08-04, AND IT WAS A MISS
>
> Apollo reported second-quarter 2026 results this morning. The 8-K carries the summary release and the dividend declaration; the detailed financial presentation is NOT in this archive — the release directs readers to the company's investor-relations website.
>
> - Adjusted EPS $2.11 against a $2.16 consensus — a 2.3% MISS, breaking four consecutive beats.
> - The stock rose 2.85% to $133.11 anyway, opening at $127.95, trading down to $126.43 and closing near the high of $133.73.
> - Dividend declared: $0.5625 per share for the June quarter, payable 2026-08-31 — an annualised $2.25, a 1.69% yield, against the vendor's trailing dividendPerShareTTM of $2.0925.
> - Chief executive Marc Rowan's quoted framing: "Our strong second quarter results reflect record earnings across Asset Management and Retirement Services."
>
> Because the detailed presentation is absent, every segment figure in this dive comes from the 10-Q for the MARCH quarter, and is labelled as such.
The two predictions this dive was commissioned to test
Both are confirmed. A third defect, not predicted, is worse than either.
Prediction 1 — a revenue gross-up the company does not publish: CONFIRMED, and there are THREE bases in one file
Apollo consolidates Athene Holding Ltd., a retirement-services company that issues and reinsures annuities. The 10-Q's own glossary defines it as "a leading financial services company specializing in retirement services that issues, reinsures and acquires retirement savings products... to which Apollo, through its consolidated subsidiary ISG, provides asset management and advisory services."
The consequence is that the vendor's revenue line is an insurance top line. It contains premiums, net investment income and investment-related gains and losses on a $460,949 million balance sheet, and it swings violently with markets rather than with the business.
| Basis | FY2023 | FY2024 | FY2025 |
|---|---|---|---|
Vendor inc_a revenue | $32,644M | $26,114M | $30,299M |
| year-on-year change | — | −20.0% | +16.0% |
Analyst consensus est.revenueAvg | $3,196M | $3,657M | $18,225M |
| ratio of vendor to consensus | 10.2x | 7.1x | 1.66x |
And a third basis appears in the earnings calendar. For the June 2026 quarter reported this morning, earn_cal records revenueActual of $1,343M against a revenueEstimated of $5,648,931K. For the March 2026 quarter it records $5,059M actual against $5,189M estimated. For the December 2025 quarter it records $9,864M actual against $4,771M estimated — a 107% "beat" that is a basis mismatch, not a result.
Three revenue definitions coexist in one payload and the largest differs from the smallest by a factor of twenty-three. Every revenue-derived ratio is therefore rejected: priceToSalesRatioTTM of 2.585x, evToSalesTTM of 2.304x, netProfitMarginTTM of 7.24%, operatingProfitMarginTTM of 31.05%, grossProfitMarginTTM of 89.33% and assetTurnoverTTM of 0.063. This is the confirmed gross-up defect class, and Apollo's instance is the most severe recorded in this programme — worse than the 101% and 70% cases previously found.
What this dive uses instead: the company's own segment measures, taken from the 10-Q.
Prediction 2 — enterprise value omitting noncontrolling interests: CONFIRMED at 28.0%
The balance sheet at 2025-12-31 reports total equity of $42,515 million, of which $23,341 million is attributable to Apollo and $19,174 million is NONCONTROLLING INTERESTS — 45.1% of total equity.
enterpriseValueTTM reads $68,390,246,155, which is market capitalisation of $76,740,246,155 less net cash of $8,350 million. It adds NOTHING for the $19,174 million of noncontrolling interests.
| Vendor | Corrected | |
|---|---|---|
| Market capitalisation | $76,740M | $76,740M |
| plus net debt | −$8,350M | −$8,350M |
| plus noncontrolling interests | $0 | +$19,174M |
| Enterprise value | $68,390M | $87,564M |
| understatement | — | 28.0% |
This is the defect this dive was told to look for, and it is present at 28.0% against the 35.6% previously confirmed on a peer. Every enterprise-value multiple in the payload is understated by roughly a quarter, and none is used.
Defect 3, not predicted, and worse — capital expenditure is EXACTLY ZERO for three consecutive years
capitalExpenditure reads exactly $0.00 in fiscal 2023, fiscal 2024 and fiscal 2025. In fiscal 2022 it read −$202,525,000. Capital expenditure did not become zero; the field stopped being populated.
And the diagnostic fails with it rather than independently. capexToDepreciationTTM reads exactly 0, capexToOperatingCashFlowTTM reads exactly 0 and capexToRevenueTTM reads exactly 0. The data contract states plainly that a capexToDepreciationTTM at or below zero is impossible. This dive was directed to check whether the datum and the diagnostic fail independently: on Apollo they fail TOGETHER, which is the simpler and more common failure and means the corruption is upstream of both.
The consequence: freeCashFlow is reported as identical to operating cash flow in all three years ($6,322M, $3,253M, $7,452M), freeCashFlowPerShareTTM equals operatingCashFlowPerShareTTM to the last decimal ($10.0908), and freeCashFlowYieldTTM of 7.82% and priceToFreeCashFlowRatioTTM of 12.78x are both REJECTED.
And a second, deeper reason to reject them. Even with correct capital expenditure, operating cash flow is not a meaningful lens for a consolidated insurer: it includes Athene's deposit and premium flows, which are policyholder money, not distributable earnings. The clearest demonstration is cashPerShareTTM of $425.52 against a share price of $133.11 — the vendor's own framework says every share is backed by three times its price in cash. The entire cash-flow block is discarded, exactly as it was on a deposit-funded bank elsewhere in this programme.
Reference table
| A share-count caveat | Market capitalisation divided by price implies 576,521,431 shares against inc_a's 607,913,669 weighted diluted. A 5.2% gap, consistent with market capitalisation being struck on basic rather than diluted shares. On the diluted count the market capitalisation is approximately $80,919M |
| Street consensus | $150.67 (+13.2%) · median $145 · high $165 (+24.0%) · low $142 (+6.7%, above spot) · 23 buy / 5 hold / 0 sell across 28 analysts |
| Valuation — on adjusted EPS only | 15.4x trailing adjusted EPS of ~$8.66 (four quarters from earn_cal) · 15.0x FY2026E $8.846 · 12.5x FY2027E $10.665 · 10.7x FY2028E $12.495 (3 analysts). All revenue, cash-flow and enterprise-value multiples are REJECTED |
| The filed shape of the firm (Q1 2026) | Fee Related Earnings $728M, +30.2% · Spread Related Earnings $719M, −10.6% · alternative net investment income $210M, −33.3% · cost of funds $2,807M, +27.0% |
| Capital return | Dividend just declared at $0.5625 per quarter — $2.25 annualised, a 1.69% yield. Buyback $773M in fiscal 2025, 1.01% of market capitalisation. Total shareholder yield ~2.70% |
| Data integrity | Revenue gross-up CONFIRMED (three bases, 23x apart) · NCI omitted from enterprise value CONFIRMED at 28.0% · capex EXACTLY ZERO for three years, with the diagnostic failing alongside it · seg_prod sums to 105.8% of revenue · seg_geo contains one line from 2020 |
| Conviction | Split by vintage — 25 raw KB hits, 25 surviving a case-sensitive re-run, 1 discarded as a quarantine_misattributed duplicate — the THIRD in this batch. 2025 claims: bullish on the model at conviction 90. 2026 claims: bearish on the cycle at conviction 88, with a redemption percentage attached |
| Technicals | −12.8% from the 52-week high of $152.70; +32.7% above the low of $100.30; +6.1% above the 50-DMA ($125.40) and +4.8% above the 200-DMA ($126.97); RSI 67.96 — the highest in this batch; MACD +0.92; 12-month −3.7% vs SPY +24.3%; max drawdown −25.5% |
What the experts actually said 8 traceable claims on APO · showing the highest-conviction voices
“The Athene merger and insurance-driven perpetual capital (450B of 750B AUM) was Apollo's strategic breakthrough, driving the post-2022 stock doubling.”
“Apollo Debt Solutions redemptions escalating from 11% to 16.8% of shares with net flows now negative signal investors doubt the marks and liquidity — a developing credit-cycle downturn, not a minor bump.”
Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.
Price & moving averages 12 months · 50 & 200-day averages · 52-week range
Solid line = price · dashed line = 50-day average · dotted line = 200-day average · the two thin horizontal lines mark the 52-week high and low. Price above both averages is an uptrend.
Data summary: last close $135.04, 7% above the 50-day average ($127), 6% above the 200-day average ($128) — an uptrend. 12% below the 52-week high of $153, 35% above the 52-week low of $100.
Bollinger Bands 20-day average ± 2 standard deviations
The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.
Data summary: price $135.04 is currently inside the band (band $126–$142).
RSI (14) momentum gauge · 0–100
Above 70 (overbought zone, shaded) = overbought, below 30 (oversold zone, shaded) = oversold. Currently 57.
MACD 12 / 26 / 9 · trend & momentum
The MACD line crossing above the signal line (bars flip to the up color) = momentum turning up; crossing below (bars flip to the down color) = turning down. Bar height = the size of that gap.
Data summary: MACD is currently below its signal line by 0.42, negative momentum.
Relative performance vs S&P 500 & its sector (XLF (sector)), set to 100 a year ago
Solid = APO · dashed = S&P 500 · dotted = XLF (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.
Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate
Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.
Key stats an RIA wants
1. The business, and the only segment numbers worth using
Apollo Global Management is an alternative investment manager operating through two reportable segments — Asset Management and Retirement Services — with the latter comprising Athene. Chairman and chief executive Marc Jeffrey Rowan; approximately 6,140 employees; NYSE-listed since 2011-03-30.
The seg_prod block is over-counted and is not used. For fiscal 2025 it reports Retirement Services $27,049M + Asset Management $5,000M = $32,049M against inc_a revenue of $30,299M — 105.8%, with no eliminations line. seg_geo contains a single entry, dated 2020, for a "Private Equity Segment" of $631,203,000 — a segment Apollo no longer reports — and is discarded entirely. So the payload contains no usable segment data for a two-segment company.
The 10-Q's own segment reconciliation is the source for everything below (three months ended 31 March, $M):
| Asset Management | Q1 2026 | Q1 2025 | change |
|---|---|---|---|
| Fee-related performance fee | 64 | 54 | +18.5% |
| Fee-related compensation | (333) | (259) | +28.6% |
| Other operating expenses | (201) | (160) | +25.6% |
| FEE RELATED EARNINGS | 728 | 559 | +30.2% |
| Retirement Services | Q1 2026 | Q1 2025 | change |
|---|---|---|---|
| Fixed income and other net investment income | 3,551 | 2,914 | +21.9% |
| Alternative net investment income | 210 | 315 | −33.3% |
| Cost of funds | (2,807) | (2,210) | +27.0% |
| Other operating expenses | (118) | (114) | +3.5% |
| Interest and other financing costs | (153) | (130) | +17.7% |
| SPREAD RELATED EARNINGS | 719 | 804 | −10.6% |
Three things follow from that table and they are the whole analysis.
First, the two engines are now the same size. $728 million against $719 million. A year earlier the split was $559 million against $804 million — the insurer was 44% larger than the asset manager. It is now smaller.
Second, the spread decline is a funding-cost story, not an asset-quality story. Fixed income net investment income actually rose 21.9%; the cost of funds rose 27.0%, which is what happens when annuity crediting rates reprice upward faster than the asset book turns over. Apollo does not set the price of annuities; the market does.
Third, the alternative investment line is volatile and is the swing factor. $210 million against $315 million — a 33.3% fall. The 8-K filed 2026-07-01 gives a preliminary June-quarter figure of approximately $350 million, an estimated 9% annualised return, with the pooled vehicle holding "the large majority" of the alternatives portfolio returning an estimated 10% and other alternatives including retirement-services platforms returning an estimated 6%. That is a recovery, and it is the single most useful forward datum in this archive.
Fourth, the fee-related operating leverage is thinner than 30.2% suggests. Fee-related compensation grew 28.6% and other operating expenses 25.6% — both close to the earnings growth rate. This is a business growing its revenue and its cost base at similar rates, not one expanding its margin.
2. Why the GAAP numbers are unusable, and what replaces them
Apollo's GAAP earnings swing violently because a consolidated insurance balance sheet marks its investments through income. The evidence is in the quarterly series:
| Quarter | GAAP net income | GAAP diluted EPS | Adjusted EPS (earn_cal) | vs estimate |
|---|---|---|---|---|
| Q1 2025 | $425M | $0.68 | — | — |
| Q2 2025 | $630M | $1.02 | $1.92 | vs $1.84 (+4.3%) |
| Q3 2025 | $1,736M | $2.78 | $2.14 | vs $1.90 (+12.6%) |
| Q4 2025 | $1,690M | $2.78 | $2.47 | vs $2.04 (+21.1%) |
| Q1 2026 | −$1,906M | −$3.24 | $1.94 | vs $1.89 (+2.6%) |
| Q2 2026 | (not in payload) | — | $2.11 | vs $2.16 (−2.3%) |
The March 2026 quarter reported a GAAP net LOSS of $1,906 million and diluted EPS of MINUS $3.24, in a quarter in which adjusted EPS was $1.94. A $5.18 per-share gap between the two measures in a single quarter is not an adjustment; it is a different accounting universe.
Consequently: priceToEarningsRatioTTM of 39.27x is computed on netIncomePerShareTTM of $3.613, a trailing GAAP figure containing that loss quarter, and is rejected. Every multiple in this dive is struck on the adjusted basis the analyst estimates use, and the trailing adjusted figure — the sum of the four quarters in earn_cal — is $8.66, giving 15.4x.
returnOnAssetsTTM of 0.46% is also rejected, because the denominator is $460,949 million of Athene's investment portfolio; returnOnEquityTTM of 10.03% is computed against equity that is 45.1% noncontrolling interests and is likewise not the shareholder's return.
3. Valuation — on adjusted earnings, because nothing else survives
At $133.11:
| FY2026E | FY2027E | FY2028E | FY2029E | |
|---|---|---|---|---|
| Adjusted EPS | $8.846 (11) | $10.665 (11) | $12.495 (3) | $13.442 (1) |
| EPS growth | — | +20.6% | +17.2% | +7.6% |
| P/E | 15.0x | 12.5x | 10.7x | 9.9x |
| EPS range | $8.727–$8.996 | $10.198–$11.192 | $10.224–$14.705 | — |
| Analysts on revenue | 5 | 5 | 1 | 2 |
Estimate coverage is thin and the revenue array is incoherent. Eleven analysts cover FY2026 and FY2027 adjusted EPS — usable. THREE cover FY2028 and ONE covers FY2029, and both years are excluded from every conclusion. And the consensus revenue series collapses from $30,518 million in FY2028 to $8,425 million in FY2029 — a 72.4% fall in a growing business — which is a basis change inside the array rather than a forecast, and is the strongest single reason to distrust it.
est.ebitdaAvg and est.ebitAvg are rejected for the fixed-ratio fabrication signature. ebitdaAvg is exactly 30.87% of revenueAvg and ebitAvg exactly 27.82% in FY2025, FY2026, FY2027, FY2028 and FY2029 alike. Separately, the FY2023 row reports ebitAvg of MINUS $3,245,761,209 against ebitdaAvg of PLUS $276,159,736 — a $3.5 billion depreciation charge on a $3.2 billion revenue estimate, which is impossible. All valuation uses epsAvg.
3a. What today's price assumes (the inversion)
At $133.11 — 15.0x FY2026 consensus adjusted EPS and 12.5x FY2027 — the price embeds:
- Adjusted EPS reaches $8.846 in 2026 and $10.665 in 2027. (Consensus; 11 analysts each.) The four quarters in
earn_calsum to $8.66 trailing, so FY2026 requires only 2.2% growth on the run rate — the near year is close to already earned. FY2027's 20.6% is the leg that must be won. - Spread Related Earnings stop declining. (Our derivation from the 10-Q.) They fell 10.6% in the March quarter on a 27.0% increase in the cost of funds. This is the most fragile assumption in the price: Spread Related Earnings are roughly half of segment earnings, and a business earning a spread cannot grow while its funding cost grows faster than its asset yield.
- Fee Related Earnings keep compounding near 30%. (Our derivation.) $728 million against $559 million. Against a knowledge-base claim that the sector faces "a dead fundraising environment", a 30% fee-earnings growth rate is the strongest fact in Apollo's favour and it is filing-verified.
- The retail credit vehicles do not experience a disorderly redemption cycle. (Knowledge-base sourced; NOT filing-verified.) Two independent claims from June 2026 assert redemptions escalating from 11% to 16.8% of shares with net flows negative. The price contains no visible discount for it and the filings contain no way to check it.
- The market keeps paying 12-15x forward adjusted earnings. (Our number.) At 10x FY2027E the stock is $107; at 16x it is $171. The observed twelve-month range on this estimate — from the 52-week low of $100.30 to the high of $152.70 — is 9.4x to 14.3x, so the price sits near the top of the band the market has actually paid this year.
3b. The return bridge (why the multiple moves)
Expected return over the next twenty-four months decomposes as: adjusted EPS growth (+41.3% cumulative, from FY2026E $8.846 to FY2028E $12.495) + multiple drift (roughly HELD at 12-13x on the then-forward year) + shareholder yield (+2.70% per year) ≈ +15% to +20% a year if the base lands.
Our base of $144 applies 13.5x FY2027E, which is a modest expansion from today's 12.5x and is the fragile leg of the bridge — we name it as such. The justification is narrow: the market paid up to 14.3x on this estimate within the last twelve months, so 13.5x is inside the observed band rather than beyond it. But the honest statement is that at an unchanged 12.5x the fair value is $133 — precisely the current price — and the entire base case is a one-turn re-rating plus a year of earnings.
That is a weaker structure than it looks, and it is the main reason the verdict is Watch.
3c. Variant perception (where we differ, what would surprise)
- We are BELOW the street and the gap is small. Our $144 is 4.4% under the consensus $150.67, and both are inside the analyst target range of $142 to $165. We have no meaningful variant perception on the level and say so.
- We do differ on the composition, and it is checkable. The market appears to be pricing Apollo as a growing asset manager. The filings say it is a 30%-growth asset manager bolted to a shrinking spread business of identical size. Watchable number: Spread Related Earnings in the third-quarter print against $719 million. A second consecutive decline makes the blended multiple harder to defend; a return to growth resolves the argument in the bulls' favour.
- We think the redemption question is the whole risk and it is unpriced because it is unobservable. Apollo Debt Solutions is a non-traded vehicle that files separately from the parent. Its flows appear in no document in this archive. A research process that concluded "no evidence of redemptions in the filings" would be describing the filings, not the world, and this dive declines to do that.
- Positive surprise that would force a re-rate: Spread Related Earnings returning to growth on a stabilising cost of funds; alternative net investment income sustaining the preliminary $350 million level; or disclosed retail credit-vehicle flows turning positive.
- Negative surprise that would break the thesis: a second consecutive Spread Related Earnings decline; a disclosed redemption gate or fee waiver on any Apollo vehicle; a mark-down in the private credit book; or a second consecutive adjusted EPS miss after this morning's 2.3% shortfall.
Synthos fair values
All three anchors are multiples of the FY2027 consensus adjusted EPS distribution (mean $10.6651, low $10.19751, high $11.19239, 11 analysts), cross-checked against the observed twelve-month multiple range of 9.4x–14.3x.
- Bear ~$102 — 10.0x the FY2027 consensus LOW of $10.198, and roughly the 52-week low of $100.30. Cross-check: 11.5x FY2026E; a 2.21% dividend yield on the declared $2.25. The scenario: the private-credit redemption cycle proves to be a credit cycle, marks come down, Spread Related Earnings keep falling, and an alternative manager trades at the bottom of its observed band. −23.4%.
- Base ~$144 — 13.5x the FY2027 consensus MEAN of $10.665. Cross-check: 16.3x FY2026E; 11.5x FY2028E; 4.4% below the street's $150.67; 5.7% below the 52-week high. The scenario: Fee Related Earnings keep compounding near 30%, Spread Related Earnings stabilise as alternative income recovers toward the preliminary $350 million, redemptions prove to be a fundraising rather than a credit event, and the multiple gains a turn. +8.2%.
- Bull ~$196 — 17.5x the FY2027 consensus HIGH of $11.192. Cross-check: 28.4% above the 52-week high; 15.7x FY2028E on a three-analyst estimate. The scenario: the private-credit scare passes entirely, the perpetual-capital thesis re-rates the fee stream toward a traditional asset-manager multiple, and both engines grow together. +47.2%.
Base is 8.2% above spot; asymmetry roughly 2.02:1 (23.4% down, 47.2% up), plus a shareholder yield of 2.70%. The asymmetry technically clears the 2:1 bar. The verdict does not follow, for the reasons in Section 10.
4. Knowledge base — twenty-five claims, split by vintage, and a third quarantine duplicate
Raw hits: 25. After a case-sensitive entity re-run: 25. Text-only: 0. Discarded as a quarantine duplicate: 1. Claims used: 24.
The sweep ran APO, Apollo, Apollo Global Management and Athene across all 52,021 distilled claims.
The homograph test PASSED and it is worth recording why that is not trivial. "Apollo" is a common proper noun. Zero case-insensitive-only matches were returned and every hit resolves to this company on content inspection — no mythological, aerospace or medical collisions. Nothing was discarded on homograph grounds.
THE DEFECT FOUND, and it is the third instance in this batch. The claim dated 2026-03-08 under the channel slug quarantine_misattributed (bearish, conviction 85, skill 2.0, entities: private credit, Blue Owl, Blackstone, BlackRock, Apollo) — "A virtual run on private credit is underway; funds hid tech/software exposure, marks are stale — we're at the start, not bottom, of the crisis" — is a near-verbatim restatement of the jordi_visser claim dated the SAME DAY, at the SAME conviction of 85 and the SAME skill weight of 2.0, with the same entity set. It is EXCLUDED and the view counted once. This programme has now found the quarantine slug leaking duplicates into entity sweeps on three separate names in one batch, which is a store-level retrieval problem rather than three coincidences.
The 2025 lane — a high-conviction structural bull case
> 2025-06-03 · bullish · conviction 90 · thesis · invest_like_the_best
> "Merging asset manager with insurance balance sheet (Athene) creates principal-aligned merchant model that beats agent-only asset management."
> 2025-06-03 · bullish · conviction 82 · thesis · invest_like_the_best
> "Apollo's 4,000-person proprietary origination machine and cross-capital-structure culture is very hard to replicate, creating durable diversified IG collateral."
> 2025-03-26 · bullish · conviction 78 · thesis · business_breakdowns
> "The Athene merger and insurance-driven perpetual capital (450B of 750B AUM) was Apollo's strategic breakthrough, driving the post-2022 stock doubling."
> 2025-03-26 · bullish · conviction 75 · thesis · business_breakdowns
> "Apollo's edge is a balance-sheet/credit focus that finds complexity and illiquidity without compromising credit quality, a durable cultural moat."
> 2025-03-26 · neutral · conviction 55 · thesis · business_breakdowns
> "Founder-succession execution differentiates alt managers; Apollo nearly fumbled it but de-risked into Mark Rowan, who is reshaping the firm as an evangelist-CEO."
Note the concentration: five of the twenty-four surviving claims come from a single business_breakdowns episode dated 2025-03-26, and two more from a single invest_like_the_best episode dated 2025-06-03. Seven of twenty-four — 29% — come from two episodes. Remove them and the bull lane is materially thinner.
One claim is directly testable and it holds. The perpetual-capital claim describes the mechanism by which Fee Related Earnings can grow 30.2% while an independent claim describes the sector's fundraising environment as "dead." The 10-Q confirms the outcome even though it does not confirm the mechanism.
The 2026 lane — a bear case with numbers
> 2026-06-23 · BEARISH · conviction 82 · thesis · eurodollar_university · skill 0.7
> "Apollo Debt Solutions redemptions escalating from 11% to 16.8% of shares with net flows now negative signal investors doubt the marks and liquidity — a developing credit-cycle downturn, not a minor bump."
> 2026-06-27 · BEARISH · conviction 72 · fact · eurodollar_university · speaker Jeff · skill 0.7
> "Apollo's fund is now cash-flow negative — more redemption requests than inflows even after blocking redemptions, forcing draws on cash reserves that further hurt performance."
> 2026-03-07 · BEARISH · conviction 88 · thesis · jordi_visser · SKILL 2.0
> "Private credit is mispriced and unwinding via retail redemption runs; we're at the start of the crisis, not the bottom—stay away."
> 2026-02-03 · BEARISH · conviction 70 · thesis · compound_and_friends
> "Alt asset managers got crushed because they acquired or lent heavily to SaaS names now being disrupted; their marked-up software portfolios are at risk."
These are the most specific and most recent claims in the file, and NONE of them is testable against this archive. Apollo Debt Solutions is a separately-reporting non-traded vehicle; its flows appear in no document here. The 10-Q, the 10-K and the earnings 8-K are silent on retail credit-vehicle redemptions. We report the claims, we report that we cannot verify them, and we treat the gap as a reason to wait rather than as an absence of risk.
Note the skill weights, because they cut both ways. The most alarming claims come from eurodollar_university at skill 0.7 — below baseline — while the broader private-credit warning comes from jordi_visser at skill 2.0, the highest in the store. The specific claim is weakly sourced and the general claim is strongly sourced, which is the least helpful possible configuration.
The two constructive 2026 claims, which are the honest counterweight
> 2026-03-24 · neutral · conviction 65 · thesis · compound_and_friends
> "Apollo down 40% is multiple compression from a dead fundraising environment, not a referendum on loan credit quality — if it were credit, it'd be down far more."
> 2026-03-02 · bullish · conviction 50 · fact · odd_lots
> "Apollo prudently reduced software lending early (early 2025), ahead of the market — well-positioned on AI disruption risk."
The first is the most useful claim in the lane because it is a falsifiable distinction rather than a direction. If the decline were credit, the argument runs, the stock would be down more than 40%. It has since recovered to 12.8% below its 52-week high, which is weak evidence for that reading and is reported as such. The second directly addresses the software-exposure charge in the February bear claim, from an independent source, and narrows it.
Attribution. Twenty-three of twenty-five carry speaker: null. Two name "Jeff" at eurodollar_university. No claim in this lane is management voice, so the half-weight rule does not engage. speaker_role is absent throughout.
Conclusion. Breadth 24, claim count 24, net conviction split by vintage — bullish on the model, bearish on the cycle. The knowledge base thinks Apollo built the right structure and picked the wrong moment. Both halves may be true and the filings settle neither.
5. Data integrity — the three predicted defects and five more
Eight findings. Apollo's payload is the second-least usable in this batch.
DEFECT 1 — the revenue gross-up, three incompatible bases, up to 23x apart. Detailed above. Rejected: priceToSalesRatioTTM 2.585x, evToSalesTTM 2.304x, grossProfitMarginTTM 89.33%, operatingProfitMarginTTM 31.05%, netProfitMarginTTM 7.24%, ebitdaMarginTTM 33.91%, assetTurnoverTTM 0.063, revenuePerShareTTM $49.887, daysOfSalesOutstandingTTM 137.3 and daysOfPayablesOutstandingTTM 444.4.
DEFECT 2 — enterpriseValueTTM omits $19,174M of noncontrolling interests, a 28.0% understatement. Detailed above. Rejected: evToSalesTTM, evToEBITDATTM 6.79x, evToOperatingCashFlowTTM 11.39x, evToFreeCashFlowTTM 11.39x and netDebtToEBITDATTM −0.83x.
DEFECT 3 — capitalExpenditure is EXACTLY ZERO in three consecutive years, with the diagnostic failing alongside it. $0.00 in FY2023, FY2024 and FY2025 after −$202,525,000 in FY2022. capexToDepreciationTTM, capexToOperatingCashFlowTTM and capexToRevenueTTM all read exactly 0, a value the data contract states is impossible. freeCashFlow therefore equals operating cash flow to the dollar in all three years and freeCashFlowPerShareTTM equals operatingCashFlowPerShareTTM to four decimals. freeCashFlowYieldTTM of 7.82% and priceToFreeCashFlowRatioTTM of 12.78x are REJECTED — and would be rejected even if capex were correct, because operating cash flow for a consolidated insurer includes policyholder flows.
DEFECT 4 — seg_prod sums to 105.8% of revenue and seg_geo is a single 2020 line. The FY2025 seg_prod block totals $32,049M against $30,299M with no eliminations. seg_geo contains one entry — fiscal 2020, "Private Equity Segment", $631,203,000 — for a segment Apollo no longer reports. There is effectively NO segment data in this payload for a two-segment company.
DEFECT 5 — GAAP quarterly earnings are unusable and the trailing multiple is rejected. The March 2026 quarter reports GAAP diluted EPS of −$3.24 against adjusted EPS of $1.94. priceToEarningsRatioTTM of 39.27x is struck on a trailing GAAP figure containing that quarter. Rejected in favour of the adjusted basis.
DEFECT 6 — the est revenue array is internally incoherent. FY2028 revenue of $30,518M collapses to FY2029's $8,425M — a 72.4% fall in a growing business — and FY2024's estimate of $3,657M sits against a delivered $26,114M. The revenue array switches basis at least twice and no revenue estimate is used anywhere in this dive.
DEFECT 7 — est.ebitdaAvg and est.ebitAvg carry the fixed-ratio signature, and one year is arithmetically impossible. ebitdaAvg is exactly 30.87% of revenueAvg and ebitAvg exactly 27.82% in every year from FY2025 to FY2029. The FY2023 row reports ebitAvg of −$3,245,761,209 against ebitdaAvg of +$276,159,736, implying $3.5 billion of depreciation on a $3.2 billion revenue estimate. Rejected.
DEFECT 8 — one insider record is entirely null and the share count is inconsistent. The 2026-06-09 entry for Kelly Martin (officer) carries a blank transaction type, zero shares, zero price and zero holdings. Separately, market capitalisation divided by price implies 576,521,431 shares against inc_a's 607,913,669 weighted diluted — a 5.2% gap consistent with market capitalisation being struck on basic shares. On the diluted count the market capitalisation is approximately $80,919M and every per-share multiple in this dive would be 5.4% higher. We use the vendor's basis for comparability and flag the gap.
Also noted: cashPerShareTTM of $425.52 against a $133.11 share price — the single clearest demonstration that the vendor's framework does not fit a consolidated insurer, since the "cash" includes $228,820M of Athene's investment portfolio classified as short-term investments. workingCapitalTTM of −$74,657M and investedCapitalTTM of −$68,745M are similarly artefacts of that classification and are used nowhere. km_ttm and ratios_ttm are present and complete, which is the one structural thing this payload gets right.
Vendor composite rating — noted and disregarded. C+ / 2 overall, with 1 out of 5 on price-to-earnings and price-to-book. Both are computed on the rejected GAAP earnings and on equity that is 45.1% noncontrolling interests. The block should not be used on this name.
Non-equity tripwire — checked and passed. APO is common stock, NYSE-listed. Price of $133.11 is not par-like; beta is 1.508, the highest in this batch; the dividend is variable and was just declared at $0.5625; volume was 3.7M shares (roughly $496M of turnover); the 52-week band of $100.30 to $152.70 is a 52% range. This is common equity.
6. Technicals
- Price $133.11, up 2.85% on an earnings day that carried a miss, on 3,722,429 shares. Opened $127.95 — below the prior close — traded down to $126.4253 and closed at $133.11, 5.3% above the day low. The session reversed intraday, which is more informative than the close.
- −12.8% from the 52-week high of $152.70; +32.7% above the 52-week low of $100.30. Position within the annual range: 63rd percentile.
- Above both moving averages: +6.1% above a 50-day average of $125.40 and +4.8% above a 200-day average of $126.97, with the 50-day just below the 200-day — a downtrend in the process of turning.
- RSI 67.96 — the highest reading in this batch and approaching the conventional overbought threshold. MACD +0.92, positive.
- Maximum drawdown from peak over the trailing year: −25.5%, against a 12.8% distance from the high — so the stock has already recovered roughly half of a large decline.
- Relative performance: 3-month +3.0% against SPY +7.6%; 6-month −0.1% against SPY +11.1%; 12-month −3.7% against SPY +24.3% and QQQ +30.8% — a 28-point deficit.
- Sentiment: 23 buy, 5 hold, 0 sell across 28 analysts; consensus target $150.67 (+13.2%), median $145, high $165 (+24.0%), low $142 — 6.7% ABOVE spot. Not one of 28 analysts has a target below the price, which after a 25.5% drawdown and an earnings miss is a consensus that has not marked to market. We treat it as a caution.
Today's move and what it does to the entry
APO closed 2026-08-04 at $133.11, up $3.69, on a day it reported an adjusted earnings MISS. The session opened 1.1% below the prior close, fell a further 1.2%, and then rallied 5.3% off the low to close near the high.
The honest read: this is not an entry discount. The base case is +8.2%, RSI is 67.96, and the stock has recovered from a 25.5% drawdown to sit 12.8% below its high. Whatever the market liked in this morning's release is not in this archive — the detailed presentation is on the company's website, not in the 8-K — and issuing a Buy on a document we do not hold would be exactly the failure this programme exists to avoid.
7. Insiders — seven identical director awards and one null record
| Date | Person | Role | Type | Shares | Price | Held after |
|---|---|---|---|---|---|---|
| 2026-07-01 | Marc A. Beilinson | director | A-Award | 1,589 | $0 | 115,296 |
| 2026-07-01 | Jessica M. Bibliowicz | director | A-Award | 1,589 | $0 | 17,635 |
| 2026-07-01 | Kerry Murphy Healey | director | A-Award | 1,589 | $0 | 22,294 |
| 2026-07-01 | Pamela J. Joyner | director | A-Award | 1,589 | $0 | 16,524 |
| 2026-07-01 | Mitra O'Neill | director | A-Award | 1,589 | $0 | 33,704 |
| 2026-07-01 | Lynn C. Swann | director | A-Award | 1,589 | $0 | 20,051 |
| 2026-07-01 | Patrick Toomey | director | A-Award | 1,589 | $0 | 11,380 |
| 2026-06-09 | Kelly Martin | officer | (null record) | 0 | $0 | 0 |
Seven identical awards of 1,589 shares to seven non-employee directors on a single date, at a price of zero — the annual board equity grant. They carry no information about anything. The eighth record is malformed and is reported as a data defect.
What the file does not contain: a single open-market purchase or sale by any officer or director, including Marc Rowan. The insider block for Apollo is uninformative and is scored as such.
Governance, from the 8-K filed 2026-06-09, which is not in the payload. At the 2026 annual meeting held 2026-06-08, thirteen directors were elected. Marc Beilinson received 14,232,088 votes against — the highest opposition — followed by Gary Cohn at 13,820,258 and Marc Rowan at 7,649,458. The remaining ten directors each drew under 8 million votes against. The say-on-pay proposal drew 131 million-plus votes against out of roughly 453 million cast — the block is truncated in the extract, but even on the visible figure that is an unusually high level of opposition to executive compensation.
8. Verdict, kill-criteria and flip conditions
Watch.
The arithmetic technically clears the bar and the verdict still does not follow, so we will explain rather than assert. At $133.11 Apollo trades at 15.0x the FY2026 consensus of $8.846 and 12.5x FY2027's $10.665. Our base of $144 is 8.2% above spot with 2.02:1 asymmetry and a 2.70% shareholder yield. On a mechanical reading that is a marginal Buy.
Four reasons it is not.
First, the base case is a re-rating, not an earnings case. At an unchanged 12.5x multiple the fair value is $133 — the current price. The entire 8.2% comes from assuming the market pays one more turn, and Section 3b says so explicitly. A base case that consists of a multiple assumption is the weakest structure in this batch.
Second, half the firm is going backwards. Spread Related Earnings fell 10.6% to $719 million while Fee Related Earnings rose 30.2% to $728 million, and the two are now the same size. The decline is a 27.0% increase in the cost of funds, which Apollo does not control.
Third, the single most actionable claim on this name cannot be tested from any document we hold. Two independent knowledge-base sources assert escalating redemptions in Apollo's own retail credit vehicle, with a specific escalation from 11% to 16.8% of shares. Apollo Debt Solutions files separately. Its flows appear in no filing in this archive. We will not issue a Buy on a name whose largest identified risk is invisible to our evidence base, and we will not dismiss the risk because it is invisible.
Fourth, the data quality is among the worst this programme has seen. A revenue line with three incompatible bases up to 23x apart; an enterprise value understated 28.0% by omitting $19,174 million of noncontrolling interests; a capital-expenditure line reading exactly zero for three consecutive years with its own diagnostic failing alongside it; a segment block over-counting to 105.8% and a geography block containing one line from 2020; and a GAAP quarter reporting minus $3.24 against an adjusted $1.94. Every one of those was found and corrected here, but a company this hard to audit deserves a wider margin of safety than 8.2%.
And the timing: results were released this morning and MISSED by 2.3%, at an RSI of 67.96, with the detailed presentation absent from this archive.
Pre-registered KILL criteria — what would take this to Avoid:
- A second consecutive Spread Related Earnings decline in the third-quarter print, against $719 million.
- Any disclosed redemption gate, fee waiver or suspension on an Apollo retail credit vehicle.
- A material mark-down in the private credit portfolio, or a disclosure that non-accruals are rising.
- Alternative net investment income below $200 million in any quarter, against a preliminary June figure of approximately $350 million.
- A second consecutive adjusted EPS miss after this morning's 2.3% shortfall.
- A price break below the 200-day average of $126.97 on company-specific news.
Pre-registered UPGRADE conditions — what would take this to Buy — Tactical:
- A price below roughly $118, at which the base case is +22% and the asymmetry moves toward 3:1 with no change to any number here. That is 11.4% below spot and just under the 50-day average.
- Spread Related Earnings returning to year-on-year growth, which would resolve the largest visible negative in the filings.
- Disclosed retail credit-vehicle flow data, in either direction — the value is in the disclosure, not the number, because it would convert the biggest unknown in this dive into something gradeable.
- Two more quarters of Fee Related Earnings growth above 25% with fee-related compensation growing slower than earnings, which would demonstrate the operating leverage the current 28.6% compensation growth does not.
Where APO sits in the Synthos Framework Portfolio. No position. On the watch list at the $118 trigger, for a first tranche of 1.0% in the financials sleeve. Logged as a tracked Synthos call (Watch) as of 2026-08-04 at $133.11, with the fair-value anchors, kill criteria and upgrade conditions all gradeable. The specific item to grade this dive on is the data-integrity section: three defects were predicted before the file was opened and all three were confirmed, which is a result about the vendor rather than about Apollo, and is worth recording as such.
Single biggest risk: that the private-credit redemption cycle is a credit event rather than a fundraising event — and that no document in this archive can distinguish between them. Apollo earns roughly half its segment income as a spread on credit it originates and holds against annuity liabilities. If those credits are marked optimistically, both halves of the firm are impaired at once: the spread compresses and the fee base shrinks as investors withdraw. The knowledge base's highest-skill voice says we are "at the start of the crisis, not the bottom"; a second independent voice says Apollo is "down 40% from multiple compression... not a referendum on loan credit quality." The 10-Q shows Fee Related Earnings up 30.2%, which favours the second reading, and Spread Related Earnings down 10.6%, which does not settle it. At 12.5x forward adjusted earnings with a 1.69% dividend there is a modest valuation cushion and no informational one.
Provenance & disclosures
- Traceability: 24 name-level knowledge-base claims name Apollo out of 52,021 distilled claims. Raw hits 25; surviving a case-sensitive entity re-run 25; text-only 0; discarded as a
quarantine_misattributedDUPLICATE 1; breadth 24; net conviction split by vintage. The homograph test PASSED with zero case-insensitive-only matches, which is not trivial for a common proper noun. THE DEFECT FOUND — the THIRD instance in this batch: the 2026-03-08 claim underquarantine_misattributed(bearish, conviction 85, skill 2.0) is a near-verbatim restatement of thejordi_visserclaim of the SAME DATE at the SAME conviction and skill, with the same entity set. Excluded. Three names in one batch have now returned quarantine duplicates to an entity sweep, which is a store-level retrieval problem. CONCENTRATION NOTE: seven of the 24 surviving claims — 29% — come from just two episodes (five frombusiness_breakdowns2025-03-26 and two frominvest_like_the_best2025-06-03), and both are bullish. The bull lane:invest_like_the_best2025-06-03 (conviction 90 on the Athene merchant model, 82 on the origination machine),business_breakdowns2025-03-26 (78 on perpetual capital, 75 on the balance-sheet moat, 68, 65 and 55). The bear lane:jordi_visser2026-03-07 (conviction 88, SKILL 2.0) and 2026-03-08 (85),eurodollar_university2026-06-23 (82) and 2026-06-27 (72, speaker "Jeff", skill 0.7),compound_and_friends2026-02-03 (70). The counterweights:compound_and_friends2026-03-24 (neutral 65) andodd_lots2026-03-02 (bullish 50). THE MOST ACTIONABLE CLAIMS ARE UNVERIFIABLE FROM THIS ARCHIVE: Apollo Debt Solutions is a separately-reporting non-traded vehicle and its flows appear in no 10-K, 10-Q or 8-K here. Twenty-three of 25 carryspeaker: null; no claim is management voice;speaker_roleis absent throughout. All quotes are verbatim from the stored claim text. - Data as-of: fundamentals — segment reconciliation, Fee Related Earnings and Spread Related Earnings through 2026-03-31 from the 10-Q filed 2026-05-07; annual figures for FY2023-FY2025 from the 10-K filed 2026-02-25 · June-quarter 2026 adjusted EPS from
earn_calONLY, because the 8-K filed this morning carries a summary release without financial tables and directs readers to a presentation on the company's website · estimates 2026-08-04 · prices 2026-08-04, quote timestamp 1785873599 = 2026-08-04T19:59:59Z ($133.11, +2.85%; 50-DMA $125.40; 200-DMA $126.97; RSI 67.96; MACD +0.92) · knowledge-base claims 2026-08-04. Apollo's fiscal year is the calendar year. All figures come from the Synthos vendor data file for APO or from the SEC filings in the APO archive. - Filing archive contents: 10-K filed 2026-02-25 (fiscal 2025); 10-Q filed 2025-11-10 (September 2025 quarter); 10-Q filed 2026-05-07 (March 2026 quarter — the source of the Fee Related Earnings and Spread Related Earnings reconciliation and the Athene and Athora definitions); 8-K filed 2026-06-09 (Item 5.07 — the certified voting results of the 2026 annual meeting held 2026-06-08); 8-K filed 2026-07-01 (Items 2.02 and 7.01 — preliminary second-quarter alternative net investment income of approximately $350 million, an estimated 9% annualised return); 8-K filed 2026-08-04 (Item 2.02 — second-quarter results and the $0.5625 dividend declaration; the summary release is present but the detailed financial presentation is NOT — it is hosted on the company's website). The 2026-08-04 8-K carries no
[TABLE]markers, which is why no June-quarter segment figure appears in this dive. - Where the filings contradicted or corrected the vendor (detailed in Section 5) — and note that THREE of these were predicted before the file was opened: the revenue line, which for fiscal 2025 reads $30,299M in
inc_aagainst an analyst consensus of $18,225M and a June-quarterearn_calactual of $1,343M — three incompatible bases in one file, up to 23x apart — because Apollo consolidates Athene's premiums, net investment income and investment gains, invalidating every revenue-derived ratio;enterpriseValueTTMof $68,390M, which omits $19,174M of noncontrolling interests — 45.1% of total equity — understating enterprise value by 28.0% against a corrected $87,564M, exactly the defect confirmed on a peer at 35.6%;capitalExpenditure, reading EXACTLY ZERO in FY2023, FY2024 and FY2025 after $202,525,000 in FY2022, withcapexToDepreciationTTM,capexToOperatingCashFlowTTMandcapexToRevenueTTMall reading exactly zero — a value the data contract states is impossible — so the datum and the diagnostic fail TOGETHER, andfreeCashFlowYieldTTMof 7.82% is rejected both for that reason and because operating cash flow for a consolidated insurer includes policyholder flows;seg_prod, summing to $32,049M against $30,299M of revenue — 105.8% — with no eliminations line;seg_geo, containing a single 2020 entry for a segment Apollo no longer reports; the GAAP quarterly series, which reports −$3.24 diluted EPS for the March 2026 quarter against an adjusted $1.94, invalidatingpriceToEarningsRatioTTMof 39.27x; theestrevenue array, which collapses 72.4% from FY2028's $30,518M to FY2029's $8,425M;est.ebitdaAvg/ebitAvg, carrying a fixed-ratio signature of exactly 30.87% and 27.82% of revenue in every year, with the FY2023 row reportingebitAvgof −$3,245,761,209 againstebitdaAvgof +$276,159,736; and the implied share count of 576,521,431 againstinc_a's 607,913,669 weighted diluted, a 5.2% gap. Where vendor and filing AGREED:km_ttmandratios_ttmare present and complete, not truncated;dividendPerShareTTMof $2.0925 is a correct trailing figure against the newly declared $0.5625 quarterly rate; and the 8-K's dividend and the annual-meeting results reconcile to the payload's director list. - Basis note: all valuation in this dive is struck on ADJUSTED earnings per share, which is the basis the analyst estimates use and the basis on which
earn_calrecords actuals. The GAAP series is rejected outright because a consolidated insurance balance sheet marks investments through income, producing a March 2026 quarter with a GAAP loss of $1,906 million and diluted EPS of minus $3.24 alongside adjusted EPS of $1.94. The trailing adjusted figure of approximately $8.66 is the sum of the four quarterly actuals inearn_caland is used for the trailing multiple of 15.4x. No revenue, cash-flow or enterprise-value multiple appears anywhere in this dive. - Estimate coverage: 11 analysts on both FY2026 and FY2027 adjusted EPS — the anchor for all three fair values, with an FY2027 range of $10.198 to $11.192 (9.3%); 3 on FY2028 and ONE on FY2029, both excluded from every conclusion; and only 5 analysts on revenue in FY2026 and FY2027, falling to 1 in FY2028, in an array that is internally incoherent and is used nowhere.
- Peer note: the vendor peer set — Affiliated Managers, Ares Capital, Ares Management, Brookfield, BlackRock, Blackstone, Carlyle, Carlyle Secured Lending, CION, KKR, Blue Owl, Patria and TPG — is well constructed and is the best peer set in this batch, containing every relevant alternative manager. But no earnings or estimate data for any peer is present in this file, only price and market capitalisation, so no peer-multiple comparison is drawn. Note that five of these names appear together inside the bearish private-credit claims in Section 4, which is the closest thing to a peer read available here.
- Fair-value caveat: the $102 / $144 / $196 anchors are multiples of the FY2027 consensus adjusted EPS distribution — 10.0x the low of $10.19751, 13.5x the mean of $10.6651, and 17.5x the high of $11.19239 — cross-checked against the observed twelve-month multiple range of 9.4x to 14.3x derived from the 52-week low of $100.30 and high of $152.70. Stated arithmetic, not a discounted cash flow. The base assumes MODEST MULTIPLE EXPANSION from 12.5x to 13.5x and this is named as the fragile leg: at an unchanged 12.5x the fair value is $133, the current price, and the entire base case disappears.
- Timing: second-quarter 2026 results were released THIS MORNING, 2026-08-04, and MISSED the adjusted consensus by 2.3% ($2.11 against $2.16), breaking a run of four beats (+4.3%, +12.6%, +21.1%, +2.6%). The detailed financial presentation is not in this archive. The board declared a dividend of $0.5625 per share payable 2026-08-31 to holders of record 2026-08-19. The next print is 2026-11-03, 91 days away, against a consensus of $2.30. All insider records are dated 2026-07-01 (seven identical director awards) and 2026-06-09 (one malformed null record).
- Accessibility note: no information in this dive is conveyed by colour. All emphasis is carried by bold text, table structure and explicit labelling.
- Not investment advice. Independent research, educational and informational only, never personalised. No recommendation to buy, sell or hold any security is made to any person.
- Version: 2026-08-04-full.