GE Aerospace GE

Reference table

1. What the business is, and how the disclosure actually works

GE Aerospace traces to Thomas Edison's 1878 founding, is headquartered in Evendale, Ohio, employs 57,000 people, and is led by H. Lawrence Culp Jr. NYSE-listed; CUSIP 369604301.

The critical structural fact, and the one the knowledge base has not absorbed: this is no longer General Electric. Between January 2023 and April 2024 the conglomerate separated into three independent public companies — GE HealthCare, GE Vernova (power, wind and electrification) and GE Aerospace, which retained the GE ticker. Any claim about gas turbines, wind, medical imaging or small modular reactors is a claim about a different registrant. Section 7 shows how badly that has contaminated this name's research lane.

Two reportable segments

From the 10-Q: Commercial Engines & Services (CES) and Defense & Propulsion Technologies (DPT). The filing notes that DPT is itself an aggregation: "We have aggregated Defense & Systems and Propulsion & Additive Technology into one reportable segment, Defense & Propulsion Technologies, based on similarity in economic characteristics."

The economic engine is stated in one sentence in the 10-K:

> "installed base of approximately 50,000 commercial and 30,000 military engines, including parked aircraft in addition to fleet in service, supports our aftermarket services business."

And the revenue mix confirms the model. The vendor's FY2024 seg_prod row — the only one in the file that is both current and interpretable — splits revenue three ways and ties exactly to reported revenue:

Nearly two-thirds of revenue is service. That is the razor-and-blade model in a single line, and it is why analysts are willing to model this company's earnings five years out.

Note the third line. GE Aerospace still consolidates a run-off long-term-care and life insurance operation, which contributed $3.581 billion of revenue in FY2024 — 9.3% of the total. The 10-Q lists among its forward-looking risks "capital or liquidity needs associated with our run-off insurance operations or mortgage portfolio in Poland (Bank BPH), the amount and timing of any required future capital contributions and any strategic options that we may consider." This is why GE's balance sheet carries $38.788 billion of long-term investments and $111.269 billion of liabilities against $18.677 billion of equity — the insurance book, not aerospace leverage.

The segment and geographic tables are largely unusable and we say which parts

seg_prod FY2025 is uninformative rather than wrong. It contains two lines — "Operating Segments: $43.868B" and "Capital Segment: $1.987B" — which sum to $45.855B, matching reported revenue exactly, but which tell a reader nothing about the business. seg_prod FY2023 is genuinely broken: it sums to $67.954B against reported FY2023 revenue of $35.348B because it includes an "Aerospace Segment: $31.770B" and a "Power Segment: $17.731B" — i.e. it is on the pre-breakup consolidated basis while the income statement is on the continuing-operations basis. REJECTED.

seg_geo FY2025 double-counts $27.661 billion — the same defect pattern found on Bank of America in this batch. The six lines sum to $73.516B against reported revenue of $45.855B. The reason is arithmetic:

$3.664B + $10.819B + $8.603B + $4.575B = $27.661B — exactly the "Non-US" line. The vendor lists the subtotal AND its components. Removing the double-count gives $18.194B + $27.661B = $45.855B, which is reported revenue to the dollar.

Corrected geographic mix, FY2025: United States $18.194B (39.7%), non-US $27.661B (60.3%). GE Aerospace earns the majority of its revenue outside the United States, which is a genuine and under-appreciated fact about it.

Two further geographic defects. The FY2024 and FY2023 rows are byte-identical — the same five figures summing to $38.703B — against FY2023 reported revenue of $35.348B. The FY2023 geographic row is a duplicate of FY2024 and is REJECTED. And the FY2025 taxonomy changed: FY2024 discloses CHINA at $3.634B as a separate line while FY2025 has no China line, with "Asia, Excluding China Region" jumping from $3.602B to $10.819B. The year-on-year regional comparison is broken and no regional growth rate appears in this dive.

2. The quarter, and what the filing says is driving it

Five consecutive beats averaging +12.9%, alongside revenue beats of 6.3% to 17.7%. That is a company whose results the sell side has systematically underestimated for over a year.

The operating drivers, verbatim from the 10-Q:

> "In the first six months of 2026, there was a strong demand for engines and services, while commercial air travel departures were roughly flat."

> "These investments combined with improved material availability contributed to internal shop visit revenue growth of 30% in the first half of 2026, and total engine deliveries and LEAP engine deliveries increases of 37% and 41%, respectively."

> "Revenue increased due to internal shop visit volume and workscopes, increased spare parts volume, increased install engine deliveries and pricing."

> "Profit increased primarily due to internal shop visit volume and workscopes, increased spare parts volume, and price. These increases were partially offset by the impact of higher install engine deliveries, higher growth investment and inflation."

Read the last sentence carefully, because it is the razor-and-blade model appearing in the profit bridge. Higher original-equipment deliveries are a negative contributor to profit. GE makes more money when it overhauls engines than when it sells them, and the 10-Q's own attribution says so. A 41% increase in LEAP deliveries is a drag on today's margin and the foundation of tomorrow's revenue.

The commercial wins, also from the 10-Q:

> "United Airlines and Delta Airlines selected our GENx engines to power their widebody Boeing 787 Dreamliner orders. American Airlines and Copa Airlines selected LEAP engines for their narrowbody orders. Ryanair signed a long-term material services agreement to support Ryanair's entire fleet of about 2,000 CFM56 and LEAP engines powering its Boeing 737 aircraft."

One margin item disclosed and worth flagging as a two-way adjustment: "In the first quarter of 2026, we recorded an unfavorable change in estimated profitability on long-term service agreements of less than $0.1 billion, including a $0.1 billion reversal of a majority of the tariff-related charge in the first quarter of 2025." Long-term service agreement profitability estimates are a genuine earnings-management surface in this industry, and GE flags changes in both directions. The amounts here are small — under $100 million — and we note the mechanism rather than the number.

And one risk the filing raises itself:

> "We are monitoring recent developments related to the conflict in the Middle East and the potential impact on the commercial aerospace industry, including lower utilization and increased prices, which can result in adverse effects on our airline customers... The conflict did not result in a material impact on our operations in the six months ended June 30, 2026."

3. Two revenue definitions, two EPS definitions — and the multiple that follows

GE reports on two bases and the vendor file contains both without labelling them, so this section establishes which number is which.

The revenue difference — $2.345 billion over twelve months — is approximately the insurance segment, which the FY2024 seg_prod row sizes at $3.581 billion annually. Consensus revenue estimates are on the adjusted basis: FY2025 revenueAvg of $41.706B against GAAP revenue of $45.855B, a gap of $4.1 billion consistent with insurance plus other adjustments.

The EPS relationship is the unusual part and deserves its own note. GAAP EPS EXCEEDS adjusted EPS here — $8.475 against $7.11 — which is the reverse of almost every other name in this batch. The reasons are visible in the income statement: FY2025 pre-tax income of $10.000B against operating income of $8.770B, i.e. $1.230B of non-operating income; and a trailing effective tax rate of 13.44%, unusually low. Adjusted EPS strips both. The conservative multiple is therefore the adjusted one — 53.1x trailing — and it is what we lead with. A reader who sees "44x GAAP" quoted elsewhere should understand that number is flattered by the insurance book and the tax rate.

4. Balance sheet, cash flow, and the conversion problem

As at 2025-12-31 (the vendor's newest annual balance sheet; the 10-Q gives selected 2026-06-30 figures in prose):

The FY2023 column describes a different company — $176.106B of assets against $130.169B — because GE Vernova had not yet been separated. No comparison across the FY2023/FY2024 boundary appears in this dive.

The standard correction of this batch — netting investments against debt — is deliberately NOT applied, and the reason matters. The $38.788 billion of long-term investments is the run-off insurance portfolio, held against long-term-care and life reserves that sit inside the $111.269 billion liability base. Netting it against debt would understate leverage by $38.8 billion and misrepresent GE Aerospace as a company with $30 billion of net cash. Net debt of $8.102 billion — 0.69x our computed trailing EBITDA of $11.765 billion — is the honest figure and it is comfortable.

The 10-Q gives the current liquidity position in prose, which is more recent than the vendor's balance sheet:

> "At June 30, 2026, our sources of liquidity of $10.3 billion included cash, cash equivalents and restricted cash of $9.3 billion and time deposits with maturities greater than three months classified as All other current assets of $1.0 billion. Of these amounts, $7.3 billion was held in the U.S. and $3.0 billion was held outside the U.S."

Cash has fallen from $12.392 billion at the 2025 year end to $9.3 billion at 2026-06-30 — a $3.1 billion decline in six months that the vendor's stale balance sheet does not show. Buybacks and dividends account for part of it; the cash-conversion problem below accounts for the rest.

The enterprise-value rebuild:

> Market cap $391.452B (1,037,563,000 shares × $377.28)

> + total debt $20.494B

> − cash $12.392B

> = enterprise value ≈ $399.55B

The vendor reports $401.264B, a 0.4% difference. Materially, the vendor is right here and we say so — this is one of the cleaner EV figures in the batch, because the insurance investments were correctly left out of the netting.

Trailing EBITDA is recomputed. ebitMarginTTM of 22.27% implies trailing EBIT of $11.284B, which is $749 million above the sum of the last four quarters' operating income ($10.535B). The gap reconstructs as trailing pre-tax income ($10.365B) plus trailing interest expense ($920M). The vendor's "EBIT" includes non-operating income. We use operating income of $10.535B plus estimated trailing D&A of ≈$1.23B — the June-quarter D&A field reads $0.0, another instance of the missing-D&A defect seen on Caterpillar, so we annualise the three available quarters and cross-check against FY2025 D&A of $1.220B — giving $11.765B. EV/EBITDA ≈ 34.0x against the vendor's 32.87x.

The cash-conversion problem, stated as an unresolved concern

The trailing figures are derived from km_ttm and ratios_ttm rather than stated: operatingCashFlowPerShareTTM of $4.805 × 1,037.6M shares gives $4.99B, and priceToFreeCashFlowRatioTTM of 109.01 against the share price gives $3.59B of trailing free cash flow. incomeQualityTTM of 0.555 corroborates.

This is the single most important unresolved item in the dive. FY2025 operating cash flow was $8.537 billion on net income of $8.704 billion — a 0.98x conversion, which for a business collecting cash on service contracts is exactly what one expects. Over the trailing twelve months to June 2026 that conversion has fallen to 0.555x, with operating cash flow at roughly $5.0 billion against net income of $8.971 billion — a shortfall of approximately $4 billion.

Candidate explanations, none of which we can verify: the inventory build visible on the annual balance sheet ($9.763B to $11.868B, +21.6%) as engine deliveries ramp 37%; the timing of long-term contract billings, which the 10-Q lists as a source of variability ("receipt of down payments on large equipment orders, timing of billings on long-term contracts, timing of customer allowances"); tax payments; or insurance-portfolio flows, since the 10-Q notes "higher net purchases of insurance investment securities of $0.4 billion" in investing activities.

We cannot narrow it further because the cash-flow statement is presented in numeric tables, which are stripped from this extraction. We state that as a limit rather than choosing an explanation. What we will say is the consequence: at $377.28 the stock trades at 109 times trailing free cash flow and yields 0.92% of free cash flow. That is not a valuation a reader should accept without an explanation for the conversion gap, and the 2026-10-20 print is where one should appear.

Capital returns, for completeness: buybacks of $7.551B in FY2025 (1.9% of market cap) after $5.827B in FY2024 and $1.233B in FY2023 — an accelerating programme that has taken diluted shares from 1,098M (FY2023) to 1,049M in the June 2026 quarter, a 4.5% reduction. Dividend: lastDividend and dividendPerShareTTM are both $1.66, a 0.44% yield, on an 18.2% payout ratio. The 10-Q states the policy — "We intend to return a portion of our free cash flow to shareholders through dividends and share repurchases" — but the current quarterly dividend rate is not stated in prose anywhere in this archive, so we use the trailing figure and label it as trailing. Total shareholder yield is approximately 2.3%.

5. Valuation — priced in or room?

At $377.28 (market cap $391.452B, EV ≈$399.55B, trailing adjusted EPS $7.11):

Estimate coverage is the deepest in this batch and it is itself informative. Twelve analysts still model FY2028 EPS and ten model FY2029 — three and a half years out. On most names in this batch the out-year rows collapse to two or four analysts and are rejected. Here they do not, and the reason is that an installed-base annuity is genuinely forecastable. We use FY2028 as the primary anchor and FY2029 as a cross-check, which we would not do elsewhere.

The consensus is on track and we verified it. First-half 2026 actual adjusted EPS is $1.86 + $2.02 = $3.88; the third-quarter consensus is $1.99; that leaves $2.05 needed in the fourth quarter to reach the FY2026 consensus of $7.918. Against $1.57 delivered in the December 2025 quarter that is a demanding step, but the trajectory (five consecutive beats) supports it.

The est EBITDA/EBIT block trips the corruption signature in one historical year and is NOT USED. FY2023 shows ebitdaAvg of $652.5 million against ebitAvg of $3.460 billion — EBIT exceeding EBITDA by 5.3x, which is arithmetically impossible. The forward rows do not trip it (FY2026: ebitdaAvg $9.911B, ebitAvg $8.171B, correctly ordered), but a file containing one impossible row does not earn trust in the others, and all forward valuation runs on epsAvg.

Peer context. The vendor peer set is genuinely relevant: RTX ($293.7B), Boeing ($187.4B), General Dynamics ($104.4B), Lockheed Martin ($136.0B), Northrop Grumman ($78.4B), L3Harris ($53.1B), TransDigm ($71.3B), Leonardo DRS ($12.1B), Woodward ($22.3B) — plus Caterpillar ($403.8B), which is in this same batch and is not an aerospace comparable. GE Aerospace at $391.5B is the largest pure aerospace name in the set and larger than Boeing and General Dynamics combined. TransDigm is the closest business-model analogue — a high-margin aftermarket parts franchise — and it is the one name a reader should benchmark GE's multiple against. No peer multiples are supplied in the file, so no peer-multiple comparison is drawn, and that is a real limitation on a name whose entire valuation debate is "what is an aerospace aftermarket annuity worth."

5a. What today's price assumes (the inversion)

At $377.28 — 41.4x FY2027 consensus and 36.4x FY2028 — the price embeds the following falsifiable claims:

• Adjusted EPS compounds from $7.918 in FY2026 to $10.372 by FY2028 and $11.681 by FY2029 — 14% a year for three years. (Consensus-derived; 13, 12 and 10 analysts.) The near test is 2026-10-20: consensus wants $1.99, which after $2.02 in the June quarter is essentially flat sequentially. Given five consecutive beats averaging 12.9%, a beat is the base case and the bar is low.
• Adjusted revenue reaches $55.852B in FY2027 (+11.1%) and $61.011B in FY2028 (+9.2%). (Consensus; 13 and 15 analysts.) The mechanism is documented — 30% shop-visit revenue growth, 41% LEAP delivery growth, a Ryanair agreement covering 2,000 engines — but note the filing's own qualifier: departures were "roughly flat." Sustained double-digit growth eventually needs traffic, not just workscope and price.
• Operating cash flow reverts toward net income. (Our assumption; the trailing data points the other way.) This is the most fragile assumption in the price and it is not a growth assumption. At 0.555x conversion the stock is on 109x free cash flow; at 0.95x conversion it is on roughly 60x. Nothing in consensus visible to us models the difference, because the estimate block contains no cash-flow line.
• The 13.44% effective tax rate persists, or adjusted earnings absorb its normalisation. (Our observation.) A tax rate this low is what makes GAAP earnings exceed adjusted earnings; a move toward a statutory rate removes roughly a fifth of GAAP net income.
• The market keeps paying roughly 36-41x forward for a low-double-digit compounder. (Our number.) At 30x FY2028E the stock is $311; at 36x it is $373 — essentially spot; at 45x it is $467. The multiple is the widest source of variance and the stock is at its 52-week high.

5b. The return bridge (why the multiple moves)

Expected return over the next twelve months decomposes as: EPS growth (+13.9%, FY2027E $9.107 to FY2028E $10.372) + multiple drift (COMPRESSION, from 41.4x to roughly 38x on the respective forward year, −8%) + shareholder yield (+2.3%) ≈ +7% to +8%, against a base fair value 4.7% above spot.

We assume mild compression, and the reason is the shape of consensus itself. Adjusted EPS growth runs +26.7%, +15.0%, +13.9%, +12.6%, +11.1% — a steady, orderly deceleration. A multiple appropriate to 27% growth is not appropriate to 11%, and the market normally begins narrowing that gap a year or two ahead.

To be explicit about the guardrail: 47.6x FY2026 becoming 41.4x FY2027 at a constant price is mechanical rolldown, not compression. What we assume is a genuine change in what is paid at each date — 41.4x forward today to roughly 38x forward in twelve months.

Nearly all of the expected return is earnings growth. If the multiple simply HELD at 41.4x on FY2028E of $10.372 the price would be $429 (+13.7%); if it compressed to 32x, $332 (−12.0%). The bull case at $500 requires 44x on the FY2028 consensus HIGH — that is an earnings beat AND a re-rating, which is why it is the least likely of the three.

5c. Variant perception (where we differ, what would surprise)

• We do not differ from the street on the business, and we differ from it by 5.8% on the price. Consensus is $419.10 with 24 buys, 11 holds and zero sells — a unanimously constructive sell side. Our base is $395. Note that the street's target LOW of $375 is BELOW spot, so at least one analyst has a price target the stock has already exceeded. We have no name-level informational edge here and we say so.
• We differ, materially and unusually, on CASH. Every screen shows GE Aerospace at 44.5x GAAP earnings with a 0.44% dividend yield and a clean balance sheet. The trailing cash statement shows operating cash flow of roughly $5.0 billion against $8.971 billion of net income — a 0.555x conversion against 0.98x for FY2025 — and a resulting free cash flow yield of 0.92% at 109x free cash flow. That gap is roughly $4 billion a year and it is not in any ratio a screen surfaces. Watchable number: operating cash flow in the 2026-10-20 print. A recovery toward 0.9x conversion would remove our main objection; a third quarter at 0.55x would move this to Watch.
• We differ on how to read the OE delivery ramp, and the filing is on our side. The consensus framing is that 37% engine delivery growth and 41% LEAP growth are unambiguously good. The 10-Q's own profit bridge lists "the impact of higher install engine deliveries" as a NEGATIVE contributor to segment profit. Deliveries are an investment: they depress today's margin and build tomorrow's annuity. That is a reason for patience, not for paying 53x trailing earnings today.
• We differ from our own knowledge base's apparent reading, and this is worth stating. A naive count of this lane would show three bearish claims from the highest-skill speaker available. All three are about GE Vernova, a company GE Aerospace spun off in April 2024. Correcting the attribution flips the lane from mixed to uniformly positive — and simultaneously reveals it is only six claims from two sources. Both facts matter and neither is visible without the correction.
• Positive surprise that would force a re-rate: operating cash flow converting at 0.9x or better in the September quarter; adjusted revenue materially above the $12.834B consensus, which implies almost no sequential growth; a further large long-term services agreement on the Ryanair model; or evidence that departures are turning up, since 30% shop-visit growth on flat traffic implies enormous operating leverage if traffic resumes.
• Negative surprise that would break the thesis: a third consecutive quarter of sub-0.6x cash conversion; an impairment of Boeing 777X contract fulfillment costs, which the 10-Q identifies as "the most significant program" in that category and which amortise only "once entered into service"; Middle East conflict effects becoming material to utilisation, as the filing warns they could; a required capital contribution to the run-off insurance operation; or the effective tax rate normalising toward statutory. Any of these hits a stock 0.37% from its 52-week high at 53.1x trailing adjusted earnings.

Synthos fair values

All three anchors are multiples of the FY2028 consensus adjusted EPS distribution (mean $10.372, low $9.017, high $11.402, 12 analysts), cross-checked against FY2027E of $9.107 (11 analysts) and FY2029E of $11.681 (10 analysts).

• Bear ~$270 — 29.9x the FY2028 consensus LOW of $9.017. Cross-check: 29.6x FY2027E; roughly 14.5% below the 200-day moving average of $315.82 and 1.5% above the 52-week low of $265.93. The scenario: cash conversion does not recover and the market re-prices the earnings quality; departures weaken on Middle East disruption; a 777X impairment lands; and a low-double-digit compounder settles at a market-plus multiple rather than double it. −28.4%.
• Base ~$395 — 38.1x the FY2028 consensus MEAN of $10.372. Cross-check 1: 33.8x FY2029E of $11.681. Cross-check 2: at today's 41.4x forward multiple applied to FY2028E the price is $429, so our base embeds an 8% de-rating. Sensitivity, stated openly: 36x gives $373 — essentially spot — and 41x gives $425, essentially the street. The entire answer lives in a 36-41x band. +4.7%.
• Bull ~$500 — 43.9x the FY2028 consensus HIGH of $11.402. Cross-check: 42.8x FY2029E; 9.9% above the street's target high of $455. The scenario: cash conversion normalises, revealing the free cash flow multiple to have been an artefact; traffic turns up on top of 30% shop-visit growth; the LEAP installed base compounds faster than modelled; and the market pays a TransDigm-style multiple for an aftermarket annuity. +32.5%.

Base is 4.7% above spot; asymmetry roughly 1.14:1 to the upside (28.4% down, 32.5% up). That is a marginally favourable ratio and it is the reason this is a Hold rather than a Watch. What keeps it from being a Buy is that the base sits below the street, the stock is at its 52-week high after a 34.5% quarter, and the cash conversion is unexplained.

6. Knowledge base — 23 hits, nine about companies GE no longer owns

Raw hits: 23. Entity matches: 20. Text matches: 3. Used as name-level conviction: 6. Discarded: 17. Search terms: GE, General Electric, GE Aerospace, Larry Culp, word-boundary matched across entity lists and claim text over all 51,928 distilled claims.

Discard class 1 — the spin-off attribution failure (9 claims). This is the most consequential knowledge-base finding in this dive and it is a structural problem, not an accident. GE separated into three public companies between January 2023 (GE HealthCare) and April 2024 (GE Vernova). The knowledge base has continued tagging claims about all three to the GE ticker. Discarded:

• 2025-07-27, jordi_visser (skill 2.0), entities: GEV — "Gas turbines were 'dead' in 2022-23 under ESG; now makers can't meet data-center surge, backlogs beyond 2029 — Crusoe ordered 19 GE Vernova units."
• 2025-10-04, jordi_visser (skill 2.0), entities: ORCL, GEV, BEARISH — "Oracle and GE Vernova have insatiable demand but are sold out ~5 years with high PEs; the buildout is already priced in."
• 2026-01-11, jordi_visser (skill 2.0), entities: GE Vernova, BEARISH — "Data-center infrastructure names at 30-50x already price in three years of demand; bottlenecks make upside surprises unlikely over next two years."
• 2025-08-15, dwarkesh, BEARISH — "GE has looming structural problems; ~$35/MWh just for the Brayton-cycle spinning components makes turbines inherently expensive to scale." A Brayton-cycle gas turbine argument, tagged to GE, about GE Vernova.
• 2026-02-12, anthony_pompliano_show — "Vernova/GE Power/Siemens orders up 260% with 36-month equipment queues."
• 2026-03-05, macrovoices — "Marketed SMRs like the Westinghouse AP300 and GE-Hitachi BWRX-300 are neither truly small nor modular."
• 2026-03-27, jensen_huang, entities: GE Vernova, Caterpillar — power generation as an AI beneficiary.
• 2026-07-16, Rome Nam on anthony_pompliano_show — "The big turbines made by companies like GE are sold out 5 to seven years."
• 2023-05-25, business_breakdowns, entities: GE HealthCare — "Healthcare is a lower-mid-single-digit grower, best FCF converter (~100%)."

The most important consequence: ALL THREE claims from Jordi Visser — the only skill-2.0 speaker in this lane — are about GE Vernova, and all three are BEARISH. Had they been counted, the highest-weighted voice available would have driven GE Aerospace's net conviction negative on the strength of a gas-turbine valuation argument about a company GE Aerospace does not own. This is precisely the class of error the attribution policy exists to prevent, and it is the reason this section leads with the discards.

Discard class 2 — the pre-breakup conglomerate (5 claims). Claims about a company that no longer exists:

• 2020-11-23, mike_green — GE's Halloween-2018 junk downgrade freezing the high-yield market.
• 2021-01-04, lyn_alden — credit mispricing, with GE as an example.
• 2021-07-26, Jack Farley via darius_dale — "Old-guard tech like IBM, Intel and GE bought back stock to goose EPS instead of investing in innovation."
• 2024-03-29, arthur_hayes — "Debt-laden old-economy corporates are squeezed in a high-rate environment."
• 2023-05-25, business_breakdowns — "GE's historic returns were propped up by GE Capital (up to ~60% of earnings) and earnings smoothing, not managerial brilliance." A historically important observation about the old GE; not a view on the current registrant.

Discard class 3 — GE Vernova sub-business claims and management principles (3 claims). The 2023-05-25 claims on Renewables (bearish) and gas Power (bullish) are GE Vernova; the 2023-05-25 Berkshire decentralisation claim is a management principle with GE as an illustration.

The six claims that survive — and their breadth problem

> 2023-05-25 · bullish · conviction 85 · entities: General Electric, GE Aviation, CFM/Safran, Boeing, Airbus · thinker: business_breakdowns · skill 1.0

> "Aviation is a wide-moat razor-and-blade business; sole/dual-sourced LEAP engine sold at a loss drives decades of high-margin parts and service revenue."

> The highest-conviction claim in the lane and the correct structural description of the business. Corroborated by the 10-Q's profit bridge, which lists higher install engine deliveries as a NEGATIVE profit contributor.

> 2023-05-25 · bullish · conviction 70 · entities: GE Aviation · thinker: business_breakdowns

> "Expect double-digit commercial aero revenue growth next couple years plus margin recovery from operating leverage on pent-up demand."

> Made three years ago and delivered: quarterly revenue growth has been 17.6% to 24.7% in each of the last five quarters.

> 2023-05-25 · bullish · conviction 62 · entities: General Electric, GE · thinker: business_breakdowns

> "GE is attractively undervalued and one of the best industrial operators under Larry Culp; breakup re-rates the stock closer to peers."

> The breakup completed in April 2024 and the re-rating occurred. This claim has been fully paid.

> 2026-02-01 · bullish · conviction 62 · entities: GE · speaker: Romesh Naranaswami · thinker: business_breakdowns

> "GE Aerospace has ~$175B backlog (~6-7yrs commercial services), 70% predictable high-margin aftermarket revenue and dominant engine share, supporting steady growth."

> Test: "fairly steady growth of revenue and earnings over the next 5 to 10 years, absent pandemics/recessions"

> We attempted to verify the $175 billion figure against the 10-Q's remaining-performance-obligation disclosure and COULD NOT: RPO is presented in numeric tables, which are stripped from this extraction. The claim is retained as unverified and labelled so. What we CAN corroborate is the 70% aftermarket share — the FY2024 revenue split shows Service at 64.2% of total revenue including insurance, which on an aerospace-only basis is consistent with 70%.

> 2026-02-01 · bullish · conviction 60 · entities: GE · speaker: Romesh Naranaswami · thinker: business_breakdowns

> "Engine makers sell OE engines at a loss/break-even (GE revenue ~$6M vs ~$20-22M list) but earn ~60% gross margins on aftermarket; aftermarket = 3-5x OE value over engine life."

> The most specific economic claim in the lane and the single most useful sentence in this section for understanding why 41% LEAP delivery growth is a good thing that hurts current margin.

> 2026-04-29 · bullish · conviction 30 · entities: GE · speaker: Colin · thinker: invest_like_the_best

> "Added GE to his watch list."

> Retained for completeness at conviction 30. It carries no information beyond attention.

The breadth problem, stated plainly

Five of the six surviving claims come from a single channel, business_breakdowns — three from a May 2023 episode and two from a February 2026 episode, the latter with a named speaker (Romesh Naranaswami). The sixth is a one-line watchlist mention. Effective breadth is two, and one of the two is trivial.

That is a thin lane presented as a thin lane. It is positive, it is name-level, the structural claims are corroborated by the filings, and one of them — the 2023 breakup-re-rates thesis — has been fully realised. But five claims from one podcast is a single analyst's view, not a consensus of independent thinkers, and treating it as the latter would be exactly the failure the attribution policy names.

Net conviction: positive-low. Breadth: 2. Claim count used: 6. The knowledge base likes this business, from one source, and has nothing to say about whether 53x trailing adjusted earnings is the right price for it.

7. Data integrity — what we rejected from the vendor file and why

Eleven findings.

1. seg_geo FY2025 DOUBLE-COUNTS $27.661B — corrected. The block lists Americas ex-US $3.664B + Asia ex-China $10.819B + Europe $8.603B + Middle East and Africa $4.575B = $27.661B and then lists "Non-US: $27.661B" as a sixth line. The stated sum of $73.516B overstates reported revenue of $45.855B by exactly the subtotal. Corrected: United States $18.194B (39.7%), non-US $27.661B (60.3%), total $45.855B — matching reported revenue to the dollar.

2. The FY2023 seg_geo row is a byte-identical DUPLICATE of FY2024 — REJECTED. Both show the same five figures summing to $38.703B, against FY2023 reported revenue of $35.348B. Not used.

3. The seg_geo regional taxonomy changed at FY2025 — year-on-year regional comparison REJECTED. FY2024 discloses CHINA at $3.634B as a separate line; FY2025 has no China line and "Asia, Excluding China Region" jumps from $3.602B to $10.819B. No regional growth rate appears in this dive.

4. The FY2023 seg_prod row is on the PRE-BREAKUP basis — REJECTED. It sums to $67.954B against reported FY2023 revenue of $35.348B because it includes a "Power Segment: $17.731B" that is now GE Vernova. The FY2025 row (two lines: "Operating Segments" and "Capital Segment") ties to revenue but conveys nothing. Only the FY2024 product/service/insurance split is used.

5. inc_a before FY2024 describes a different company — no comparison crosses that boundary. FY2021 shows revenue of $56.469B and a net LOSS of $6.337B; FY2020 shows $75.834B; FY2023 shows net income of $9.482B on $35.348B of revenue (26.8% net margin), inflated by separation accounting. The FY2023 balance sheet shows $176.106B of total assets against $130.169B in FY2025. Only FY2024 onward is used for any trend in this dive.

6. GAAP and adjusted bases are BOTH present and unlabelled — disentangled here. inc_q revenue is GAAP ($13.349B for Q2 2026); earn_cal revenue is adjusted ($12.634B for the same quarter); est.revenueAvg is adjusted (FY2025 $41.706B against GAAP $45.855B). The difference is approximately the insurance segment, which the FY2024 seg_prod row sizes at $3.581B. Trailing EPS is $8.475 GAAP and $7.11 adjusted, and GAAP EXCEEDS adjusted — an inversion caused by $1.230B of FY2025 non-operating income and a 13.44% effective tax rate. We lead with the adjusted multiple (53.1x) as the conservative one.

7. inc_q.depreciationAndAmortization is $0.0 for the June 2026 quarter — a plain data error. The three preceding quarters report $312M, $307M and $303M. We annualise to ≈$1.23B for the EBITDA calculation, cross-checked against FY2025 D&A of $1.220B, and label the figure as derived. (The same defect appears on Caterpillar in this batch, suggesting a systematic problem with the most recent quarter's D&A field.)

8. Trailing EBITDA is not an operating figure — RECOMPUTED. ebitMarginTTM of 22.27% implies EBIT of $11.284B, which is $749M above the sum of the last four quarters' operating income ($10.535B); the gap reconstructs as pre-tax income plus interest expense. We use $10.535B + $1.23B = $11.765B, giving EV/EBITDA of ≈34.0x rather than 32.87x.

9. est.ebitdaAvg/ebitAvg trip the corruption signature in FY2023 — the block is NOT USED. FY2023 reports ebitdaAvg of $652.5M against ebitAvg of $3.460B — EBIT exceeding EBITDA by 5.3x, arithmetically impossible. The forward rows are correctly ordered, but one impossible row disqualifies the block. All forward valuation runs on epsAvg.

10. bal_a is two quarters stale and misses a $3.1B cash decline. The newest balance sheet is 2025-12-31 showing cash of $12.392B; the 10-Q states that at 2026-06-30 "our sources of liquidity of $10.3 billion included cash, cash equivalents and restricted cash of $9.3 billion." We use the filing figure for the current position and label the vendor's as at 2025-12-31.

11. quote.yearHigh/yearLow disagree with tech.hi52/lo52 — we use tech. The quote block gives $382.97 and $263.80; the technical block gives $378.68 and $265.93. We use tech throughout. Note that the resulting pct_from_hi of −0.37% and max_dd_from_peak of −0.37% are not defects — they correctly report that today's close is within a third of a percent of the highest close in 252 sessions.

Long-term investments are deliberately NOT netted against debt — the standard correction is inverted here and the reason is stated. Short-term investments are zero; the $38.788B of long-term investments is the run-off insurance portfolio held against long-term-care and life reserves inside the $111.269B liability base. Netting would understate leverage by $38.8B and portray GE as having $30B of net cash. Net debt of $8.102B (0.69x EBITDA) is the honest figure. For the same reason, financialLeverageRatioTTM of 7.24x, debtToEquityRatioTTM of 1.09x, priceToBookRatioTTM of 22.23x and returnOnEquityTTM of 49.0% are all distorted by an insurance balance sheet sitting behind an aerospace business and are rejected as lenses. returnOnInvestedCapitalTTM of 10.4% is the honest return measure. The vendor's composite rating of B− / 2, scoring 1 out of 5 on price-to-book, debt-to-equity and price-to-earnings, is generated by exactly these distortions and is rejected.

The cash-conversion figures are DERIVED, not stated, and are labelled as such. Trailing operating cash flow of ≈$4.99B comes from operatingCashFlowPerShareTTM × shares; trailing free cash flow of ≈$3.59B from priceToFreeCashFlowRatioTTM against the share price; both are corroborated by incomeQualityTTM of 0.555. The cash-flow statement itself is presented in numeric tables and is stripped from this extraction, so we cannot decompose the gap and we do not try.

Not defects, correctly reported and independently confirmed: enterpriseValueTTM of $401.264B against our rebuilt $399.55B — a 0.4% difference and one of the cleanest EV figures in the batch, because the insurance investments were correctly excluded from the netting; the seg_prod FY2024 and FY2025 rows both tying exactly to reported revenue; earn_cal actuals reconciling to the quarterly income series on the adjusted basis; the employee count, chief executive and segment names matching the filings; and interestCoverageRatioTTM of 11.45x consistent with the debt and income figures.

Non-equity tripwire — checked and passed. GE is common stock, NYSE, CUSIP 369604301, ISIN US3696043013. Price $377.28 is not par-like; beta 1.348; the dividend is variable; the 52-week band of $265.93 to $378.68 is a 42% span; volume 3.59M shares. This is common equity. A related tripwire worth naming: GE Vernova (GEV) and GE HealthCare (GEHC) are separate registrants with separate tickers, and this dive covers only GE Aerospace. Section 6 shows that the knowledge base has not made that distinction.

Extraction limits, stated rather than estimated around. Numeric tables are stripped, so remaining performance obligation (which would verify the knowledge base's $175 billion backlog claim), the cash-flow statement (which would explain the conversion gap), the segment revenue and profit tables, and the insurance reserve balances are all unavailable. Earnings-call transcripts are not available on this data plan. The three 8-Ks in the archive (2026-06-11, 2026-06-25, 2026-07-16) are cover documents.

8. Technicals

• Price $377.28 — 0.37% from the 52-week high of $378.68. Maximum drawdown from peak over the trailing year: −0.37%, i.e. essentially none. +41.9% above the 52-week low of $265.93. Position within the annual range: 99th percentile.
• Moving averages, both supportive and both well below: +8.4% above a 50-day average of $348.05; +19.5% above a 200-day average of $315.82. Both rising.
• RSI 59.58 — firm, not overbought, which for a stock at its high after a 34.5% quarter is notable and suggests an orderly advance. MACD +4.47, positive and meaningful.
• Relative performance, with the recent acceleration as the headline: 3-month +34.5% vs SPY +7.6% and QQQ +7.7% — a 27-point outperformance in one quarter, the largest three-month relative move in this batch; 6-month +22.2% vs SPY +11.1%; 12-month +40.1% vs SPY +24.3% and QQQ +30.8%. Note the shape: more than four-fifths of the twelve-month outperformance was generated in the last three months.
• Sentiment: 24 buy, 11 hold, zero sell across 35 analysts — the only name in this batch with no sell ratings. Consensus target $419.10 (+11.1%), median $423, high $455 (+20.6%), low $375 — 0.6% BELOW spot.

Today's move and what it does to the entry

GE closed 2026-08-04 at $377.28, up 2.26% or $8.35 from a previous close of $368.93, on 3.59M shares against a 5.09M average — 71% of normal volume. It opened at $374.52, traded $368.60 to $378.85, and closed 0.4% below the day's high. No company-specific news for GE is in this file for 2026-08-04; the last company event was the 2026-07-16 earnings release and 10-Q, nineteen days earlier.

The honest read. A quiet 2.3% advance to a new 52-week closing high, on below-average volume, after a 34.5% three-month run, with RSI at 59.6. That is a healthy chart and an expensive entry. There is no gap to decline and no drawdown to exploit; the stock has simply gone up in a straight line for a quarter. Our base fair value of $395 is 4.7% above the close and 5.8% below the street's target. The valuation, not the price action, is the reason not to add.

9. Insiders — eight transactions, one officer, and a genuine open-market sale

Every transaction belongs to one officer on one day, eight days after the second-quarter print, and the structure is a textbook exercise-and-sell.

The mechanics. Options struck at $69.55 and $90.01 — grants from several years ago, when the pre-breakup GE traded at those levels — were exercised for 8,096 shares, and 8,096 shares were sold in four tranches at $353.64 to $353.75. The exercise-to-sale ratio is exactly 1.00: every share acquired was sold. Gross proceeds were approximately $2.86 million; the intrinsic gain over the strike prices is roughly $2.20 million.

The reading, stated carefully. This is a genuine open-market disposal, not a tax-withholding event, and it is worth distinguishing from the F-InKind and phantom-stock tables found elsewhere in this batch. But an officer selling 100% of an option exercise, at strikes representing a five-fold gain, eight days after an earnings release, is the most routine transaction pattern in corporate America. It is what a scheduled exercise of expiring or vested options looks like.

What is mildly notable is the residual: after the sale, holdings fall from 25,727 to 20,666 shares — about $7.8 million at today's price — so the officer retained more than he sold. And the file contains zero open-market purchases by anyone, and no transactions at all by the chief executive or chief financial officer.

Net: a neutral-to-very-mildly-negative signal, and we report it as such. One senior vice president monetising an option grant is not information about the business.

10. Verdict, kill-criteria and flip conditions

Hold.

This is the best business model in the batch and one of the two most expensive stocks in it.

What is genuinely excellent, and all of it is filing-verified: an installed base of "approximately 50,000 commercial and 30,000 military engines"; internal shop visit revenue up 30% in the first half and LEAP engine deliveries up 41%, achieved while "commercial air travel departures were roughly flat"; second-quarter revenue up 21.1% to $13.349B; five consecutive earnings beats averaging 12.9%; new GEnx wins at United and Delta, LEAP wins at American and Copa, and a Ryanair services agreement covering "about 2,000 CFM56 and LEAP engines"; net debt of $8.102B at 0.69x EBITDA with interest coverage of 11.5x; the deepest analyst coverage of any name in this batch, with ten analysts still modelling FY2029 — which is itself a statement about how forecastable this revenue is; and a knowledge-base structural claim, corroborated by the 10-Q's own profit bridge, that engines are sold near cost and the money is made on 60%-gross-margin aftermarket worth three to five times the original-equipment value.

What we are declining to pay for: 53.1x trailing adjusted earnings, 44.5x trailing GAAP, 41.4x FY2027 consensus and 109x trailing free cash flow; a base fair value of $395 that is 4.7% above spot and 5.8% BELOW the street's $419.10; a close 0.37% from the 52-week high after a 34.5% three-month run; and above all an operating cash flow conversion of 0.555x net income against 0.98x last year — roughly $4 billion a year of reported profit not appearing as cash — which we cannot explain because the cash-flow statement is not extractable. Add a knowledge-base lane that is six claims from effectively two sources, with nine of the twenty-three raw hits belonging to companies GE spun off.

The distinction that matters. We are not saying GE Aerospace is expensive relative to what it is. An 80,000-engine certified installed base compounding aftermarket revenue at 30% is worth a premium and always has been. We are saying that we cannot currently reconcile the earnings this premium is paid on with the cash the company is generating, that our own fair value sits below the street's, and that the entry price is a 52-week high. In the Synthos frame, an outstanding business with an unexplained cash gap and no valuation cushion is a Hold — own the annuity, do not add into the high, and let the October cash-flow statement decide.

Pre-registered KILL criteria — what would take this to Avoid:

• A third consecutive quarter of operating cash flow converting below 0.6x net income. At 109x free cash flow this is the thesis, and two quarters is a working-capital story while three is an earnings-quality story.
• An impairment of the Boeing 777X contract fulfillment costs, which the 10-Q identifies as "the most significant program" in that category and which amortise only "once entered into service."
• Middle East conflict effects becoming material to utilisation, shop visits or spare parts, which the filing explicitly warns could produce "lower volume related to shop visits, spare parts and spare engines and lower profitability of our long term contracts."
• A required capital contribution to the run-off long-term-care insurance operation, which the 10-Q lists among its forward-looking risks and which sits behind $38.788 billion of investments.
• Internal shop visit revenue growth falling below roughly 15%, halving the current rate and removing the operating leverage the margin depends on.
• The effective tax rate normalising toward statutory without a compensating operating improvement, which would remove roughly a fifth of GAAP net income.

Pre-registered FLIP conditions — what would take this to Buy — Tactical:

• Operating cash flow converting at 0.9x or better in the 2026-10-20 print. This is the single most likely path to an upgrade and it resolves in eleven weeks. At 0.9x conversion the free cash flow multiple falls from 109x to roughly 65x and our principal objection disappears.
• Price below approximately $320 — roughly 31x FY2028E and close to the 200-day average of $315.82 — with no deterioration in the operating picture. At $320 the base case is +23% and the risk/reward turns clearly asymmetric. That is 15.2% below today's close.
• Departures turning up while shop-visit growth holds above 25%, which would add a volume leg to what is currently a workscope-and-price story and would make the FY2029 consensus look conservative rather than heroic.
• A verified remaining-performance-obligation figure at or above the $175 billion the knowledge base claims, which would substantiate the multi-year visibility argument that justifies the multiple. We could not verify it and its verification would matter.
• A second, independent name-level knowledge-base source appearing. Five of six claims come from one channel; genuine breadth would be worth roughly one tier.

Where GE fits in the Synthos Framework Portfolio. The aerospace aftermarket sleeve, core compounder expression. On batch overlap: GE and CAT are both hundred-year-old industrials that have re-rated hard and both are in this batch — but unlike CAT, GE's driver is aviation aftermarket, not datacenter power, so the two are genuinely different exposures despite superficial similarity. The relevant overlap for GE is with TransDigm, which is not in this batch but is the correct benchmark for what an aerospace aftermarket annuity is worth. Logged as a tracked Synthos call (Hold) as of 2026-08-04 at $377.28.

Single biggest risk: the cash conversion. GE Aerospace reported $8.971 billion of trailing net income and generated approximately $5.0 billion of operating cash flow — a 0.555x conversion against 0.98x for fiscal 2025 — leaving roughly $3.6 billion of free cash flow against a $391 billion market capitalisation, a 0.92% free cash flow yield at 109 times free cash flow. A company priced at 53 times adjusted earnings is being valued on an earnings number its own cash statement is not corroborating, and we cannot decompose the gap because the cash-flow statement is presented in tables this extraction strips. The plausible explanations — a 21.6% inventory build as deliveries ramp 37%, long-term contract billing timing, tax — are all benign and all temporary. But we do not know that they are the explanation, and at this multiple the burden of proof sits with the company.

Most fragile assumption in the price: that reported earnings and cash converge. Every other assumption on this page concerns volumes and margins that are visible quarter by quarter and have been beaten five times running. This one is about whether the earnings are real in the sense that matters. If conversion returns to 0.95x, free cash flow is roughly $7.5 billion, the multiple is 52x free cash flow, and the stock is defensible at 41x forward earnings. If it stays near 0.55x, GE Aerospace is a 109x-free-cash-flow security and the bear case at $270 is the honest number. The 2026-10-20 cash-flow statement decides it, and that is why the verdict is Hold rather than anything more committed.

Provenance & disclosures

• Traceability: 23 raw knowledge-base hits on GE, General Electric, GE Aerospace, Larry Culp across 51,928 distilled claims (20 entity matches, 3 text matches). 6 used as name-level conviction; 17 discarded. The dominant discard class — NINE claims — is a corporate-structure failure: claims about GE Vernova, GE HealthCare and the GE-Hitachi nuclear joint venture, all separate registrants since the 2023-2024 breakup, remain tagged to the GE ticker. All three claims from Jordi Visser, the only skill-2.0 speaker in this lane, fall in that group and all three are BEARISH (GE Vernova 'sold out ~5 years with high PEs', 'already priced in', datacenter-power names 'at 30-50x already price in three years of demand'); a further discard, dwarkesh 2025-08-15 on Brayton-cycle turbine economics, is likewise a GE Vernova claim tagged to GE. Had these been counted, a different company's valuation problem would have driven this name's net conviction negative. A further five discards are pre-breakup claims about the old conglomerate (GE Capital's earnings quality, the 2018 junk-market downgrade, buybacks-over-innovation, debt-laden old-economy corporates) and three are GE Vernova sub-business or management-principle claims. Of the six survivors, FIVE come from a single channel (business_breakdowns) — three from a May 2023 episode and two from a February 2026 episode with named speaker Romesh Naranaswami — and the sixth is a one-line watchlist mention at conviction 30. Effective breadth is 2 and that is stated as a finding. The 2023 claim that a "breakup re-rates the stock closer to peers" has been fully realised. The 2026-02-01 claim of an "~$175B backlog" COULD NOT BE VERIFIED — remaining performance obligation is disclosed in the 10-Q only in numeric tables, which this extraction strips — and is retained as unverified rather than treated as confirmed. All quotes are verbatim from the stored claim text.
• Data as-of: annual fundamentals to 2025-12-31; quarterly fundamentals through 2026-06-30; most recent filing-verified financials 2026-06-30, from the 10-Q filed 2026-07-16 · estimates 2026-08-04 · prices 2026-08-04, quote timestamp 1785873603 = 2026-08-04T20:00:03Z ($377.28, +2.26%; 50-DMA $348.05; 200-DMA $315.82; RSI 59.6; MACD +4.47) · knowledge-base claims 2026-08-04. All figures come from the Synthos vendor data file for GE or from the SEC filings in the GE archive; no figure comes from memory, recall or external retrieval.
• Filing archive contents: 10-K filed 2026-01-29 (fiscal 2025); 10-Q filed 2026-04-21 (March 2026 quarter); 10-Q filed 2026-07-16 (June 2026 quarter — the primary filing source for this dive); 8-K filed 2026-06-11; 8-K filed 2026-06-25; 8-K filed 2026-07-16 (second-quarter results). The 8-Ks are cover documents; the earnings release sits in an unextracted Exhibit 99. Numeric tables are stripped from all extracted filings, so the cash-flow statement, remaining performance obligation, segment revenue and profit tables, and insurance reserve balances are unavailable; every figure used is either prose-stated in the filing or explicitly labelled as derived.
• Where the filings contradicted or corrected the vendor (detailed in Section 7): seg_geo FY2025 double-counts $27.661B by listing "Non-US" alongside its own four components, with the corrected total of $45.855B matching reported revenue exactly; the FY2023 seg_geo row is a byte-identical duplicate of FY2024 and is rejected; the FY2023 seg_prod row is on the pre-breakup basis ($67.954B including a $17.731B "Power Segment" that is now GE Vernova) against reported revenue of $35.348B, and is rejected; inc_a before FY2024 describes a pre-breakup company and no trend in this dive crosses that boundary; the vendor mixes GAAP and adjusted revenue and EPS without labelling them (Q2 2026 revenue $13.349B GAAP against $12.634B adjusted; trailing EPS $8.475 GAAP against $7.11 adjusted, with GAAP EXCEEDING adjusted because of $1.230B of non-operating income and a 13.44% effective tax rate); the June-quarter D&A field of $0.0 is a plain error replaced with a derived ≈$1.23B; trailing EBITDA recomputed from $12.207B to $11.765B because the vendor's "EBIT" is pre-tax income plus interest; est.ebitdaAvg/ebitAvg trip the impossibility signature in FY2023 ($652.5M EBITDA against $3.460B EBIT) and the block is not used; bal_a is two quarters stale and misses a $3.1B cash decline the 10-Q documents ($12.392B at 2025-12-31 against $9.3B at 2026-06-30); and quote.yearHigh/yearLow ($382.97/$263.80) are rejected in favour of tech.hi52/lo52 ($378.68/$265.93). The $38.788B of long-term investments is deliberately NOT netted against debt — it backs run-off insurance reserves — and the standard correction of this batch is therefore inverted here, with the reasoning stated. Where vendor and filing AGREED — worth recording: enterpriseValueTTM of $401.264B against our rebuilt $399.55B, a 0.4% difference and the cleanest EV figure in this batch; the FY2024 and FY2025 seg_prod rows tying exactly to reported revenue; the quarterly revenue series reconciling to earn_cal on the adjusted basis; the employee count, chief executive and two-segment structure; and interestCoverageRatioTTM of 11.45x.
• Basis note: analyst estimates are adjusted; reported diluted EPS is GAAP. Unusually, GAAP EXCEEDS adjusted here — trailing $8.475 against $7.11 — so the adjusted multiple of 53.1x is the CONSERVATIVE one and is what this dive leads with. Both trailing multiples are shown. Forward multiples use consensus adjusted EPS. Consensus revenue is also on the adjusted basis and is roughly $2.3-4.1 billion below GAAP revenue, the difference being approximately the run-off insurance segment.
• Estimate coverage: 12-15 analysts on FY2026-FY2028 revenue; 13, 11 and 12 on FY2026, FY2027 and FY2028 EPS, and TEN on FY2029 — the deepest out-year coverage in this batch, which is why FY2029 is used as a cross-check here where equivalent rows are rejected on other names. FY2030 rests on 5 analysts and is shown but not used.
• Peer note: the vendor peer set is genuinely relevant — RTX ($293.7B), Boeing ($187.4B), Lockheed Martin ($136.0B), General Dynamics ($104.4B), Northrop Grumman ($78.4B), TransDigm ($71.3B), L3Harris ($53.1B), Woodward ($22.3B), Leonardo DRS ($12.1B) — plus Caterpillar, which is in this batch and is not an aerospace comparable. TransDigm is the closest business-model analogue and the correct benchmark for an aftermarket annuity multiple. No peer multiples are supplied in the file, so no peer-multiple comparison is drawn, which is a real limitation on a name whose entire valuation question is what such an annuity is worth.
• Fair-value caveat: the $270 / $395 / $500 anchors are multiples of the FY2028 consensus adjusted EPS distribution — 29.9x the low of $9.017, 38.1x the mean of $10.372, and 43.9x the high of $11.402 — cross-checked against FY2027E and FY2029E. Stated arithmetic, not a discounted cash flow. The base is sensitivity-disclosed: 36x gives $373 (essentially spot); 41x gives $425 (essentially the street). The whole answer lives in a 36-41x band. The base embeds an 8% de-rating from today's 41.4x forward multiple; holding the multiple would give $429.
• Timing: second-quarter results were released 2026-07-16, nineteen days before this dive, with the 10-Q filed the same day. The next print is 2026-10-20, 77 days away — no near-term company-specific catalyst, which is an explicit input to the Hold verdict. The 2026-10-20 cash-flow statement is the single most important scheduled disclosure for this thesis.
• 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.