GE Aerospace: Full Throttle [Business Breakdowns Episode 235]

Summary

• GE Aerospace is now a pure-play aerospace business dominated by an unusually powerful engine franchise: ~70,000 engines in service (45,000 commercial, 25,000 military), powering “three out of four commercial takeoffs pretty much every day,” with ~70% narrow-body share via the CFM56 and LEAP, including all Boeing 737s and sole-source positions on the 737 MAX and COMAC C919. Widebody share is ~50% of fleet and backlog, with the GEnx and GE90 at ~80% of their programs and the GE9X sole-source on the 777X when it enters service.
• The economics are extreme razor-and-blade: LEAP engines list at $20–22M but sell to airframers at discounts “up to 70 or 80%” — realized revenue ~$6M per engine, at a loss — while aftermarket spare parts carry ~60% gross margins and plausibly 40%+ operating margins. Over a 25-year-plus engine life, “the aftermarket can be three to four to five times the OE sale,” and 70% of GE’s revenue is already services.
• Visibility is the thesis: a $175B backlog is 4.5 years of revenue on a headline basis, ~6 years for commercial alone, and closer to ~7 years when the services element is excluded from the commercial-revenue denominator — with Airbus and Boeing themselves running near a decade of backlog. Ramesh Narayanaswamy sees commercial services growing at 8–10% “very predictably” for at least five years on installed-fleet growth plus mid-single-digit aftermarket pricing.
• Barriers to entry are exceptionally high: making a jet engine at scale is “one of humanity’s toughest technical challenges right up there with semiconductor fabrication,” Pratt’s geared turbofan reportedly cost ~$10B, Rolls-Royce was nationalized in the 1970s developing the RB211, and even China’s COMAC chose a CFM engine. A new entrant must also fund years of OE losses before the first spare-parts profit arrives — “you’re unlikely to wake up tomorrow and read that an AI startup has launched a new jet engine.”
• Latent growth is baked in: the LEAP fleet is expected to reach twice the size of the CFM56 it replaces, ~40% of CFM56 engines haven’t yet had their first shop visit, and 2025 shop visits are only roughly at 2019 levels. LEAP OE losses should shrink toward breakeven over five years while the slower-than-expected CFM56 retirement provides a “cash cushion.”
• The long-cycle risk cuts both ways: “even the mighty fall” — Pratt & Whitney went from ~60% of the commercial fleet in 1995 to under 20% — and GE has “gone all in” on open-fan architecture, at least a decade away, with bypass ratios of 40–50x versus today’s 10–12x. GE believes the architecture may be the only path to a 20% fuel-burn gain, making it a higher-risk/higher-upside bet versus competitors’ geared variants. Boeing could also dual-source its next narrow body, in which case “GE’s 100% market share can only go down.”
• Valuation already prices the quality: at ~40x free cash flow, you need low-double-digit revenue and low-to-mid-teens EPS/FCF-per-share growth to be “buying it on 10 times earnings in 10 years’ time” — “there’s quite a bit of optimism about the future in the share price today.” Historically, entry points came in crises, when the installed base could be valued “like a bond, an inflation-protected bond” — sometimes below liquidation value.

Deep dive

1. A pure-play aerospace business with dominant engine positions

• Ramesh Narayanaswamy (Tubian Partners, as heard) lays out the franchise for host Matt Russell: ~70,000 engines in service — 45,000 commercial, 25,000 military — with GE powering “three out of four commercial takeoffs pretty much every day.” In narrow body (~70% share), the legacy CFM56 powers all Boeing 737s and just over half the A320 family; the LEAP is sole-source on the 737 MAX and COMAC C919 and ~60% of the A320neo.
• Widebody is ~50% of fleet and backlog: GEnx at ~80% of 787 backlog, GE90 ~80% on the 777, and the GE9X will be sole-source on the 777X when it enters service. Profitability between the two is not meaningfully different — ~20% margins across programs — though narrow body “might be more profitable by a few percentage points.”
• Defense is better judged by program participation than share: nearly two-thirds of U.S. military aircraft including helicopters, but GE “lost out on the F-35” to Pratt. Overall: ~$40B revenue this year, 75% commercial engines and services at ~25% operating margins, 25% defense and propulsion at 11–12%, plus a mid-single-digit legacy insurance runoff — “in terms of earnings contribution, you’re really looking at a commercial engine franchise.”

2. The backlog: unusually long visibility

• The headline $175B backlog is ~4.5 years of revenue, but stripping to commercial only it’s closer to six years, and removing the services element from that commercial-revenue denominator brings it closer to seven — “unusually long.” Context: Airbus and Boeing at current production rates run “close to a decade’s worth of backlog.”
• Services (~70% of backlog) should burn into revenue steadily; OE is lumpier because it ties to airframer production rates — but with Boeing’s issues fixed and Airbus in rhythm, both look set for “a fairly steady cadence over the next 5 to 10 years,” pandemics and recessions aside. Backlogs are moderately above historical averages, helped by the rising prevalence of long-term service agreements.

3. Culp’s deconglomeration: “common sense vigorously applied”

• Narayanaswamy’s history: under Welch and Immelt, GE was defined by acquisitions, “earnings per share management,” and GE Capital leverage, with “no true business rationale holding them together” — and “there are very few things as reliable as a cycle of conglomeration and deconglomeration.” Larry Culp, GE’s first outsider CEO from October 2018, brought Danaher’s Kaizen/lean playbook — “walked the gemba,” fixed the “don’t shoot the messenger” culture — then spun off GE HealthCare and GE Vernova and pursued debt reduction.
• On incentives: the 2020 one-time grant (~$200M in shares tied to stock targets) was “a little bit controversial” given the depressed pandemic share price but aligned outcomes; a smaller 2024 grant tied to operating metrics followed. Why Culp kept aerospace: familiarity with industrials, and “perhaps it was an implicit indicator of how strong aerospace was… he wanted to be the CEO of the crown jewel asset.”

4. Why almost nobody can make a jet engine

• The barrier is “extraordinary technical performance at extraordinarily low cost” — hot-section temperatures exceed the melting point of the alloys, and “atomic-scale defects can be catastrophic”: Pratt’s GTF fleet was grounded worldwide after “a microscopic contaminant in the manufacturing process.” Costs haven’t reliably fallen with scale or maturity because engines are “constantly pushing at the leading edge of materials technology”; the GTF reportedly cost Pratt ~$10B.
• Then you must convince Airbus/Boeing you can supply at scale, convince airlines on decades of reliability and total cost of ownership — “and oh by the way, you also need to sell at a loss to Airbus and Boeing before you can see the first profit from spare parts in five or 10 years’ time.” Rolls-Royce collapsed and was nationalized building the RB211; COMAC, despite China’s manufacturing muscle, chose a CFM engine.
• Hence risk-sharing JVs: CFM International, GE’s ~50-year 50/50 venture with Safran (GE historically the hot side, Safran the cold side), “one of the most successful aviation franchises in history.”

5. Two customers, one profit pool: the bifurcation that makes the moat

• The core structural insight: OE buyers (Airbus/Boeing) are consolidated and powerful, so engines sell at discounts “up to 70 or 80%” off a ~$20–22M LEAP list price — GE’s revenue per engine is ~$6M, at negative margins until program maturity, then breakeven at best. The aftermarket customer — hundreds of fragmented airlines subject to mandated shop visits every 5–8 years and warranty-voiding rules on non-OE parts — delivers ~60% gross margins, plausibly 40%+ operating. Mix in commercial engines and services: 75% services / 25% OE.
• Two aftermarket models: time-and-materials (airlines bear reliability risk, typical for mature engines) versus power-by-the-hour subscriptions where “GE takes more risk — it’s like selling insurance contracts.” ~60% of LEAP engines sit under long-term contracts (30% revenue-per-flight-hour); widebody runs 60–70% LTAs with 60–80% pay-by-the-hour. Notably, LTA mix has been “reducing meaningfully” and life-limited parts are often out of scope.
• His generalizable lesson: buyer/user bifurcation “is a recurring pattern in many enduring businesses” — a new entrant must solve “a 3D puzzle.”

6. Growth, resilience, and healthy-but-not-spectacular returns

• Demand: RPKs grow mid-single digits, reasonably ~1.5x GDP (3x in emerging markets). GE’s extra levers: LEAP fleet doubling versus the CFM56, 40% of CFM56s yet to see a first shop visit, 2025 shop visits only matching 2019, and aftermarket pricing staying at least mid-single digits through decade-end — summing to 8–10% services growth with high visibility. Defense should grow in line with budgets plus program mix, at roughly 4%.
• Production is past the worst of the LEAP ramp-up: OE losses should reduce substantially toward breakeven over five years, while slower CFM56 retirements provide a “cash cushion.”
• Downturn insulation despite discretionary end-demand and load factors “maxed out at low 80s”: engine maintenance is mission-critical and regulation-mandated, spares are a small share of airline opex versus fuel, and price increases went through during the pandemic — Safran, the closest comparable, “reported positive free cash flow even during 2020 when air traffic practically ground to a halt.”
• On margins, his correction to the headline narrative: reported expansion was “largely driven by portfolio rationalization” (lossmaking Vernova depressed the old group), with commercial engine margins ~20% even pre-Culp; still, roughly 500bps of genuine improvement over the last few years, from ~20% to ~25%, came from pricing and efficiencies. Capex is under 3% of revenue with ~100% FCF conversion, but that reflects a “harvest phase” — adjusted return on tangible operating capital is ~20–25% on a cash basis. “What you lose in ultra-high returns on capital you make up for in durability and visibility.” Capital returns: ~70% of excess cash to shareholders, ~30% dividend payout, rest buybacks.

7. Decade-scale risks, relative PMA protection, and a full price

• “Even the mighty fall”: Pratt went from ~60% of the commercial fleet in 1995 to under 20%. GE’s open-fan bet for next-generation aircraft, potentially at least a decade away — bypass ratios of 40–50x versus today’s 10–12x — is “higher risk and higher upside” than the geared variants Pratt/MTU/Rolls pursue. GE believes it may be the only route to a 20% fuel-burn improvement; “it could be a game-changer for GE if they get it right,” but it’s “a bit too early to tell.” Boeing could also dual-source next time, and GE’s GTF-driven share gains “might revert” once Pratt fixes its issues.
• On PMA parts (the HEICO model): engines are relatively protected — Pratt itself tried ~20 PMA life-limited parts on the CFM56, reportedly costing several billion dollars, and “failed spectacularly.” Lessors (over half the market) oppose PMA, warranties void, and reliability data only exists late in a program’s life — “the true addressable market for a PMA-parts maker is much smaller than it initially appears.”
• Valuation, his sharpest framing: the method the market uses “tells you more about where we are in the cycle than anything about the value itself.” In crises, the installed base can be valued engine-by-engine “like a bond, an inflation-protected bond,” sometimes below liquidation value (MTU fell to 10–11x earnings on the GTF powder-metal recall). Today, at ~40x FCF, low-double-digit revenue growth and low-to-mid-teens EPS and FCF-per-share growth gets you “10 times earnings in 10 years’ time” — “quite a bit of optimism about the future in the share price today.”