Hexcel
It makes the carbon fibre, and the resin-soaked sheets woven from it, that Airbus and Boeing build lighter aircraft out of.
The scores
Growth3/4 ↑
Sales rose 8.9% in H1 2026 but Hexcel does not beat its end market, it is its end market - and FY2025 fell 0.5%.
Business quality3/4 →
Re-qualification genuinely locks in programmes in production, but Boeing prices that at a year and every new programme is open ground.
Financial position3/4 ↑
Debt is real and covered from operations, nothing matures before 2031, but leverage sits above the company's own ceiling so buybacks stopped.
Execution record3/4 ↑
Guidance raised twice and six beats in eight quarters, but FY2025 revenue and adjusted earnings both fell and the 18% target keeps travelling.
Per share capture3/4 →
The count is genuinely shrinking, but the repurchase was borrowed and its interest has so far cost more per share than it saved.
Potential2/4 ↑
Six independent things must land and four are other companies' production schedules; nothing incremental is contracted, and defence is flat today.
Risk profile3/4 →
Airbus and Boeing are 52% of sales and one recovery drives everything - disclosed and survivable, but the Airbus contracts all expire together in 2030.
Valuation2/4 ↑
The price pays for the full ramp and the 18% margin before either arrives, so four flawless years earn an ordinary return.
Evidence quality3/4 →
Figures reconcile to filings and the moat and prize reach Boeing, Airbus and NASA - but the terminal case exists only on a call.
The Verdict
We rate Hexcel unattractive. It makes the carbon fibre material that Airbus and Boeing build lighter aircraft from, and as plane production climbs back its profit is finally rising, so the business is solid. The recovery is real, and the price already owns it: even on management's own end-of-decade targets, the shares look expensive if everything goes right, so it is full, not cheap. The view hinges on margins: if each extra dollar of sales keeps more profit than management guides, the targets arrive sooner and higher. Until that shows for a full year, the price is ahead of the recovery.
The long caseAirbus and Boeing are building more planes, each using more carbon fibre; for Hexcel that recovery must beat what the price already assumes.
The evidence under this - 7 claims
- The trailing figures are struck on a trough. Adjusted operating margin was 9.1% in 2025 against 18% in 2019, so a trailing multiple on a cyclical industrial at the bottom of its margin cycle is the wrong instrument by construction - which is why the frame here is forward.
- The shares are 19.6% below their 52-week high, so the price already reflects some of the doubt: the multiple is lower than it was in July when guidance was raised.
- The company's own median price-to-earnings ratio over prior fiscal years is higher than today's, so on its own history the shares are not expensive.
- If the ramp lands and the shares then trade on a normal industrial multiple of 25 times, the price is roughly $120 against $89.82 today - about 34% over four years, plus a dividend near 0.8% a year. That is a real return, just not a large one for the risk.
- Management says that once peak rates are reached in 2028 and 2029 Hexcel is likely to stay there for a long time because the backlog is close to ten years of production, so the terminal earnings deserve a higher multiple than a cyclical trough figure would.
- The calculation assumes no buyback beyond the 72M share floor and no repricing of the Airbus agreements, which management expects to renegotiate upward when they expire at the end of 2030 - both are upside the arithmetic excludes.
- Management itself guides the run rate down to the mid-30s from the 49% just delivered, and the second-half guidance implies a step down further than that.
- Q2 2026's incremental margin was helped by the absence of the $24.2M Belgian restructuring charge in the base quarter and by the divestiture of lower-margin industrial businesses, neither of which recurs.
- Management says the industry requires running fast to stand still on cost, so the incremental margin is struck before cost inflation it must offset every year.
- 2026-10-21editedQ3 incremental margin of about 24% computed from the release: below the mid-30s run rate management guided.
- 2026-10-22editedthird-quarter call read: management restated the mid-30s run rate for 2026
- Every dollar of it depends on two other companies' production schedules. Airbus built 57 A350s in 2025 against the 144 a year its own rate-12 target implies, and Airbus says a third party's engine deliveries are constraining the ramp.
- The defence and space half is not growing today. Hexcel's Defense, Space & Other sales fell 5.8% in the first half of 2026, and management says Defence and Space alone, excluding the divested industrial businesses, was unchanged year over year in Q2 2026.
- Nothing is contracted for it. The agreements are existing long-term supply agreements whose volume follows the airframers' build rates, and neither the Boeing framework announced in July 2026 nor any Airbus agreement discloses a volume.
The company
Hexcel makes carbon fibre from the raw chemical up, then turns it into resin-soaked sheet, woven cloth, honeycomb core, glues and finished parts. Aircraft makers use these in place of aluminium and titanium: the same stiffness for less weight and far fewer parts, at a higher cost per kilogram.
Each aircraft generation uses more. Composites are about 6% of a Boeing 767 by weight, and 53% of an Airbus A350.
The evidence under this - 2 claims
How it makes money
It is paid per aircraft built. Airbus, Boeing and their suppliers were 52% of 2025 sales, and material delivered four to six months before each plane means sales follow build rates. There is no repair income: the material lasts the life of the plane. Most of the rest is military work, led by the F-35.
The evidence under this - 4 claims
- The 10-K names further penetration of composites in rotorcraft blades including both new and replacement blades as a growth trend, so some replacement demand exists - it is not sized and does not appear as a reported line.
- Engineered Products makes finished parts and structural assemblies which in principle can be spared, and the segment's revenue is not broken down between original equipment and replacement.
- 2026-10-22editedthird-quarter call read: reasons for the fall and the fourth-quarter expectation recorded in C60
Where the money comes from, stream by stream
| Stream | Share of sales (FY2025) | How it earns it |
|---|---|---|
| Commercial Aerospace: Airbus, Boeing and subcontractors | about 47% | Carbon fibre, prepreg, honeycomb and structures for large airliners, paid per aircraft built under long-term agreements. |
| Commercial Aerospace: business jets, regional and other | about 14% | The same materials for business jets, regional aircraft and other commercial applications. |
| Defense, Space & Other | 39.4% | Materials for military aircraft and rotorcraft led by the F-35, space launch structures and the remaining industrial uses. |
Who runs it
Tom Gentile, who ran Spirit AeroSystems, became chief executive in May 2024 and chairman that December; James Coogan took over the finances in May 2026 after months with a stand-in. Insiders own under 2% of the company, the biggest stake a former boss's. The yearly bonus rides on cash generated and operating profit, plus a little for new contracts; the shares ride on return on capital and on earnings growth against similar-sized companies. The last three-year award paid nothing on return on capital.
The evidence under this - 3 claims
The contract-wins modifier lifted a bonus whose free cash flow and adjusted EBIT both missed target, and the ROIC targets for the 2025-2027 cycle are withheld as confidential.
Why it wins, and who could beat it
Why customers come back
An aircraft is certified on one supplier's material from one set of production lines. Switching means redoing the test data and proving the structure again, and the aircraft maker, not Hexcel, pays for that. Boeing's own annual report says qualifying another composite source could take a year or more.
The lock is weakening at the edges. A shared government-backed database now lets new suppliers prove their material matches, cutting certification cost by a factor of ten.
The evidence under this - 3 claims
The threat
The next single-aisle aircraft may not be built the way the A350 was. Toray sells plastic-based laminates that are pressed in minutes and welded together, skipping the slow cure and the glues Hexcel sells. NASA runs a project to build composite aircraft faster, and at 75 planes a month speed, not weight, decides the material.
The evidence under this - 2 claims
The rivals: all private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | How far along | Priced against it |
|---|---|---|---|
| Toray Advanced Composites - Cetex thermoplastic tapes and reinforced thermoplastic laminatesPriced against it: Listed abroad | On sale | Listed abroad | |
| Not listed | NASA's Hi-Rate Composite Aircraft Manufacturing project and the airframers working inside itPriced against it: Not listed | In research | Not listed |
| Not listed | Aluminium-lithium alloy primary structure - the Boeing 737 MAX and Airbus A320neo airframesPriced against it: Not listed | On sale | Not listed |
How big it could get
How big this could get
Hexcel is already on every Airbus and Boeing plane, so it grows only as they build more. Their stated peak rates across the five composite-heavy programmes add up to about 155 aircraft a month. Airbus built 757 aircraft on its three of them in 2025.
Demand is not the constraint. The two had 15,474 aircraft on order at the end of 2025, a 3.8% rise.
The evidence under this - 5 claims
- Airbus delivered 57 A350s in 2025, which is 40% of the 144 a year that rate 12 implies, so the 2028 target requires output to more than double in three years.
- Airbus itself says Pratt & Whitney's failure to commit to the number of engines Airbus ordered is negatively impacting 2026 guidance and the ramp-up trajectory - the binding constraint on the largest programme is a third party's.
- Airbus has missed its own delivery target in several recent years, and cut the 2025 target in December 2025 before reporting 793.
- Boeing's own risk factors say planned rate increases may be delayed or not occur at all if its production health indicators do not support them or it cannot obtain FAA concurrence - the company states the condition itself.
- The 737 rate was below 38 a month as recently as the start of 2025, so 47 is a 24% increase over a line that has been disrupted twice in three years.
- The 787 has been held below ten a month since 2021 by supply chain and quality problems Boeing has not yet closed out.
Who has signed
The Airbus agreements, about 40% of the business, run to the end of 2030, and Airbus has firm orders for at least 80 A350 sets of material through 2026. The new Boeing agreements of July 2026 disclose no volume, term or value.
The evidence under this - 3 claims
- Hexcel's own 10-K risk factors say customers push for cost reduction or improvements in contract terms throughout the supply chain, and that Hexcel may be required to accept increased risk or face margin compression on some products.
- A single expiry covering about 40% of sales is a concentration in itself: the whole Airbus book is renegotiated at one moment rather than 15% to 20% a year, and it falls just outside this thesis's horizon.
- The claim exists only on the earnings call. No filing states the term, the expiry or the share of sales covered.
- Airbus delivered 57 A350s in 2025, so 80 shipsets is Hexcel shipping well ahead of last year's assembly rate - consistent with its four-to-six-month lead time, and equally consistent with the supply chain restocking rather than the aircraft being built.
- No filing carries the order figure. It is management's characterisation of its own order book on a call, and Hexcel's own 10-K says its order backlog is not a meaningful trend indicator.
- Hexcel's sales guidance for 2026 was raised partly on this A350 upside, so the same claim supports both the order book and the guidance.
- 2026-10-22editedthird-quarter call confirmed the 80 shipsets, said 60 to 65 are pulled and gave the 2027 outlook in C56; the rate wording moved to seven a month
Where the growth is coming from, business by business
Where the growth is coming from
| Business | Share of revenue | Growth: last quarter | Last half-year | Last year |
|---|---|---|---|---|
| Commercial Aerospace | 67% | +20.5% | +19.3% | +12.3% |
| Rest of the company | 33% | -8.7% | -7.9% | -2.3% |
| Whole company | 100% | +8.8% | +8.4% | +6.7% |
The filed split through 2026-09-30. These rates are not carried forward: the reading below says why.
How we read it: The blend. Commercial Aerospace's 18.3% is Airbus and Boeing rebuilding rates, Airbus at 63% of its own targets; the rest's 7.2% fall is the Austrian industrial sale. A recovery toward a ceiling, not a mix shift.
The results
The numbers
Sales grew 7.6% over the twelve months to Sep 2026, faster than the −0.5% in the year to Dec 2025.
Operating margin is 12%, about where it was (12%) in the year to Dec 2023.
Running costs take 13¢ of each sales dollar, down from 15¢ in the year to Dec 2021.
Yearly sales growth has ranged from −12% to +19% over the last 5 years, with sales falling in 2 of them.
It turned 10% of sales into free cash, $210M over the last twelve months.
It owes $857M more than the cash it holds, about 2.9 years of the cash its operations bring in.
The share count fell 9.3% in 4.8 years, and pay in shares costs 1.0% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-21. Fiscal years end in December.
| Dec 2021 | Dec 2022 | Dec 2023 | Dec 2024 | Dec 2025 | 12 mo to Sep 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $1.32B | $1.58B | $1.79B | $1.90B | $1.89B | $2.02B |
| Sales growth | −12% | +19% | +13% | +6.4% | −0.5% | +7.6% |
| Profit | ||||||
| Gross margin | 19% | 23% | 24% | 25% | 23% | 25% |
| Operating margin | 3.9% | 11% | 12% | 9.8% | 9.1% | 12% |
| Net margin | - | - | 5.9% | 6.9% | 5.8% | - |
| What growth costs, per sales dollar | ||||||
| Research and development | 3.4% | 2.9% | 2.9% | 3.0% | 3.0% | 3.2% |
| Selling, general and admin | 10% | 9.4% | 9.2% | 9.3% | 8.9% | 9.1% |
| Pay in shares | 1.4% | 1.3% | 1.2% | 1.2% | 0.8% | 1.0% |
| Capital spending | 2.1% | 4.8% | 6.0% | 4.6% | 3.9% | 4.3% |
| Cash | ||||||
| Cash from operations | $152M | $173M | $257M | $290M | $230M | $297M |
| Free cash flow | $124M | $96.8M | $149M | $203M | $157M | $210M |
| Balance sheet | ||||||
| Cash and short-term investments | $128M | $112M | $227M | $125M | $71.0M | $77.8M |
| Debt | $823M | $724M | $700M | $701M | $993M | $935M |
| Net cash | −$696M | −$612M | −$472M | −$575M | −$922M | −$857M |
| Returns and shares | ||||||
| Return on capital | 1.6% | 6.0% | 8.2% | 6.8% | 6.2% | - |
| Shares (diluted) | 84.6M | 85.0M | 85.5M | 83.0M | 80.0M | 76.7M |
What growth costs
Growing makes Hexcel more profitable, but not because office or research costs shrink: they rise roughly in step with sales. The gain is in the plants, which cost much the same full or half empty, so returning volume lifts the margin. Two flat years recently showed it working in reverse.
The evidence under this - 2 claims
Flat sales in 2024 and 2025 took operating margin down from 12.0% to 9.1%, partly on restructuring, so the leverage runs both ways.
- Management itself guides the run rate down to the mid-30s from the 49% just delivered, and the second-half guidance implies a step down further than that.
- Q2 2026's incremental margin was helped by the absence of the $24.2M Belgian restructuring charge in the base quarter and by the divestiture of lower-margin industrial businesses, neither of which recurs.
- Management says the industry requires running fast to stand still on cost, so the incremental margin is struck before cost inflation it must offset every year.
- 2026-10-21editedQ3 incremental margin of about 24% computed from the release: below the mid-30s run rate management guided.
- 2026-10-22editedthird-quarter call read: management restated the mid-30s run rate for 2026
The last report (29 Jul)
Quarter ended June 2026, filed figures.
The September 2026 quarter showed the recovery is real but uneven. Sales rose 8.8% to $496M and commercial aircraft sales rose 21%, while defence and space fell 9%. Profit margin slipped to 11.0% and each extra sales dollar kept about 24 cents, against 49 cents the quarter before. The yearly forecast was left unchanged and the share-buyback allowance rose to $750M.
On the call management blamed the dip on summer slowness and early hiring (525 people against 400 planned), said about 100 A350 aircraft sets are planned for 2027, 25% more than this year, and spoke of a margin nearer 20% over time.
What it did to the thesisThe call strengthens the sales path, especially the A350, but the profit rate stayed below the line we set and the higher margin talk is an aspiration. The price still owns the whole recovery, so the view stays Unattractive.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $0.66 against $0.57 expected.
The evidence under this - 6 claims
- Management itself guides the run rate down to the mid-30s from the 49% just delivered, and the second-half guidance implies a step down further than that.
- Q2 2026's incremental margin was helped by the absence of the $24.2M Belgian restructuring charge in the base quarter and by the divestiture of lower-margin industrial businesses, neither of which recurs.
- Management says the industry requires running fast to stand still on cost, so the incremental margin is struck before cost inflation it must offset every year.
- 2026-10-21editedQ3 incremental margin of about 24% computed from the release: below the mid-30s run rate management guided.
- 2026-10-22editedthird-quarter call read: management restated the mid-30s run rate for 2026
- The free cash flow guidance did not move when the earnings guidance was raised twice, so the extra profit is not yet expected to reach cash in 2026.
- The earnings guidance implies second-half adjusted earnings per share below the first half's, which management attributes to hiring, a carbon fibre line restart, third-quarter European seasonality and caution about the Middle East.
- The shares fell after the raise, and the measured reaction to the release was a 2.6% decline.
- 2026-10-21editedQ3 2026 release reaffirmed the guidance, neither cut nor narrowed.
- 2026-10-22editedthird-quarter call read: guidance language on the fourth quarter added
- Management gives four specific reasons: accelerated hiring of about 400 people, pulling a carbon fibre line restart forward from 2027, the third-quarter European shutdowns and a small provision for the Middle East. All four are investments in 2027 volume rather than deterioration.
- The company has beaten its own earnings estimate in six of the last eight quarters, so a guided step-down is not the same as a delivered one.
- The second half of 2025 carried the same seasonal pattern, so the comparison is half-on-half rather than against the year-ago half the company guides on.
- 2026-10-21editedQ3 delivered $0.50 of the implied $1.05 to $1.15 second half; Q4 implied range derived from the guide and nine-month EPS.
- The 2027 figure is a triangulation of Airbus's long-term schedule, a supplier survey of about 15 sites and early-2027 purchase orders, not a firm order book, and the same company dropped its 2025 guidance twice when Airbus's plan fell from about 98 to about 57 shipsets over that year.
- Hexcel ships six months or more ahead of deliveries and an analyst asked directly about channel stock; management conceded some aircraft are being built without seats and interiors, which delays deliveries without changing Hexcel's pull.
- No price, date or contract is named for the renegotiations, and the thesis's own accepted risk is that the Airbus agreements all reprice only at the end of 2030, outside this horizon, so most of the 'price realisation' is not scheduled to arrive soon.
- The 20% is an aspiration of an executive about six months in the role, described as an opportunity to approach, not a target, and the 18% already assumes the full ramp that the price pays for in C42.
What management did with the money
It bought back $454.3M of shares in 2025, partly with borrowed money, at about $70.95 in the final quarter. Debt is now 2.1 times a year's adjusted profit against a 1.5 to 2.0 target, so buybacks stay paused until the fourth quarter.
The evidence under this - 4 claims
- Not all of the interest increase is the buyback. Hexcel also replaced $400M of 3.95% notes due 2027 with $400M of 4.9% notes due 2031, which adds about $3.8M a year regardless of the repurchase.
- The shares were bought at an average of $70.95 in the fourth quarter of 2025 against $89.82 today, so on price the repurchase was well timed and the comparison above is only the first eighteen months of a permanent share reduction against a temporary debt balance.
- The debt is being repaid - total debt fell $33.6M in the first half of 2026 - so the interest drag shrinks each year while the share reduction does not.
- The ratio falls on the denominator as well as the numerator: at the guided 2026 adjusted earnings and free cash flow above $195M, adjusted cash profit rises and leverage reaches the top of the target range without any extraordinary repayment.
- Nothing matures inside the horizon. The nearest maturity after the 2027 notes were redeemed is the revolving facility and the 2031 notes, both in 2031, so the leverage is a policy constraint rather than a solvency one.
- 2026-10-22editedthird-quarter call gave the leverage the company measures, 2.1 times, down from the 2.3 times in this claim, with a return to its target range expected in the fourth quarter
- The same management took leverage above its own target ten months earlier by borrowing $350M to fund an accelerated repurchase, so the stated priority is recent and follows a decision that broke it.
- The target is a policy the company sets and can reset. Nothing in the credit agreement requires 1.5 to 2.0 times - Note 5 discloses no covenant at that level - so the constraint on buybacks is a choice rather than an obligation.
- The acquisition escape is wide: management says an inorganic opportunity clearing advanced material science, an aerospace and defence focus and a 15% return would be considered, which is a threshold the buyer sets and tests on its own forecast.
The price, and where we disagree
In short. Full. The price already assumes the recovery in aircraft building reaches Hexcel in full and on time, so there is little to gain if it does and a lot to lose if it slips.
The shares are down 12.2% in a month and 19.6% from their high, after gaining 42.9% over the year. They fell 2.6% the day after July's raised forecast.
The fourth-quarter results in late January are the next test. Management says the hiring cost fades through 2027 as volumes rise; a margin that climbs above Q3's 11.0% would back that.
The evidence under this - 4 claims
- Management gives four specific reasons: accelerated hiring of about 400 people, pulling a carbon fibre line restart forward from 2027, the third-quarter European shutdowns and a small provision for the Middle East. All four are investments in 2027 volume rather than deterioration.
- The company has beaten its own earnings estimate in six of the last eight quarters, so a guided step-down is not the same as a delivered one.
- The second half of 2025 carried the same seasonal pattern, so the comparison is half-on-half rather than against the year-ago half the company guides on.
- 2026-10-21editedQ3 delivered $0.50 of the implied $1.05 to $1.15 second half; Q4 implied range derived from the guide and nine-month EPS.
- Management itself guides the run rate down to the mid-30s from the 49% just delivered, and the second-half guidance implies a step down further than that.
- Q2 2026's incremental margin was helped by the absence of the $24.2M Belgian restructuring charge in the base quarter and by the divestiture of lower-margin industrial businesses, neither of which recurs.
- Management says the industry requires running fast to stand still on cost, so the incremental margin is struck before cost inflation it must offset every year.
- 2026-10-21editedQ3 incremental margin of about 24% computed from the release: below the mid-30s run rate management guided.
- 2026-10-22editedthird-quarter call read: management restated the mid-30s run rate for 2026
- 'About 13%' for the year is the chairman's rounding of a guide whose own nine-month figure is 12.8% and whose third quarter was 11.0%, so it requires a fourth-quarter margin near 14%, a step that has no precedent in the quarters reported this year.
- The 'mid-30s' incremental is the same figure the company gave in July, when the second quarter ran at 49%; this quarter ran at about 24% (C30), so management is restating a run rate the last quarter did not deliver.
How we read the multiple
A buyer pays about $45 for each dollar of last year's profit, or $38 on this year's forecast adjusted profit. Management's plan for the end of the decade:
- About $500M more sales a year once Airbus and Boeing build at their planned peak.
- About $200M more from military and space.
- Profit of 18 cents on each dollar of sales.
- Even then, today's price is about $18 for each dollar it would earn, so four flawless years would leave a buyer little.
Against its own history. Cheaper than its own normal on profit, dearer on sales and on what it owns, while the business is still only partway back to its best year.
The evidence under this - 5 claims
- The trailing figures are struck on a trough. Adjusted operating margin was 9.1% in 2025 against 18% in 2019, so a trailing multiple on a cyclical industrial at the bottom of its margin cycle is the wrong instrument by construction - which is why the frame here is forward.
- The shares are 19.6% below their 52-week high, so the price already reflects some of the doubt: the multiple is lower than it was in July when guidance was raised.
- The company's own median price-to-earnings ratio over prior fiscal years is higher than today's, so on its own history the shares are not expensive.
- Every dollar of it depends on two other companies' production schedules. Airbus built 57 A350s in 2025 against the 144 a year its own rate-12 target implies, and Airbus says a third party's engine deliveries are constraining the ramp.
- The defence and space half is not growing today. Hexcel's Defense, Space & Other sales fell 5.8% in the first half of 2026, and management says Defence and Space alone, excluding the divested industrial businesses, was unchanged year over year in Q2 2026.
- Nothing is contracted for it. The agreements are existing long-term supply agreements whose volume follows the airframers' build rates, and neither the Boeing framework announced in July 2026 nor any Airbus agreement discloses a volume.
- If the ramp lands and the shares then trade on a normal industrial multiple of 25 times, the price is roughly $120 against $89.82 today - about 34% over four years, plus a dividend near 0.8% a year. That is a real return, just not a large one for the risk.
- Management says that once peak rates are reached in 2028 and 2029 Hexcel is likely to stay there for a long time because the backlog is close to ten years of production, so the terminal earnings deserve a higher multiple than a cyclical trough figure would.
- The calculation assumes no buyback beyond the 72M share floor and no repricing of the Airbus agreements, which management expects to renegotiate upward when they expire at the end of 2030 - both are upside the arithmetic excludes.
- It is an airframer naming its supplier as the reason its own targets are at risk, which is exactly the statement a company under delivery pressure has an incentive to make. RTX has not conceded the shortfall in the same terms.
- Engine supply constrains the A320 family, a narrowbody programme where Hexcel's content is concentrated in engines and nacelles rather than the airframe - so the constraint bites hardest on the part of the pond Hexcel earns least from.
- The A350, the programme this thesis actually turns on, is powered by Rolls-Royce rather than Pratt & Whitney, so this constraint does not explain the 57 deliveries in 2025.
- The 2019 figures appear in no document this research could open. They are stated on a call about a year Hexcel no longer reports, and the 24.8% margin implies revenue of about $2.36B, which is reconciled nowhere in the current filings.
- The company is not the 2019 company. It has since sold its Austrian business and a Hartford 3D printing operation, closed Welkenraedt and shut Leicester's industrial lines, so the perimeter meant to earn the target margin is smaller than the one that earned 24.8%.
- Adjusted EBITDA is the company's own measure on both dates and the add-backs differ: 2025 and 2026 carry restructuring charges from those exits that 2019 did not, so the gap between $390M and $585M is not measured on one definition.
Priced against its rivals
Its rivals have not yet been checked against their own filings, so there is no group to price it against; the price is judged against its own history instead.
Where we disagree with the market
A late-cycle recovery whose good news is out: analysts split evenly between buy and hold.
The profit recovery is real and arriving, and the price already owns all of it, delivered by four other companies.
Nobody runs the end-of-decade sums out loud. Done with management's own targets, they leave the shares expensive even if everything goes right.
The case against us is the margin. Each extra dollar of sales kept 49 cents of profit in June against the mid-30s guided; if that is the real rate, the 18% target arrives sooner and higher. Airbus passing 10 A350s a month in 2027 would put the ramp ahead of us.
What analysts say, and the evidence
The evidence under this - 5 claims
- If the ramp lands and the shares then trade on a normal industrial multiple of 25 times, the price is roughly $120 against $89.82 today - about 34% over four years, plus a dividend near 0.8% a year. That is a real return, just not a large one for the risk.
- Management says that once peak rates are reached in 2028 and 2029 Hexcel is likely to stay there for a long time because the backlog is close to ten years of production, so the terminal earnings deserve a higher multiple than a cyclical trough figure would.
- The calculation assumes no buyback beyond the 72M share floor and no repricing of the Airbus agreements, which management expects to renegotiate upward when they expire at the end of 2030 - both are upside the arithmetic excludes.
- Management itself guides the run rate down to the mid-30s from the 49% just delivered, and the second-half guidance implies a step down further than that.
- Q2 2026's incremental margin was helped by the absence of the $24.2M Belgian restructuring charge in the base quarter and by the divestiture of lower-margin industrial businesses, neither of which recurs.
- Management says the industry requires running fast to stand still on cost, so the incremental margin is struck before cost inflation it must offset every year.
- 2026-10-21editedQ3 incremental margin of about 24% computed from the release: below the mid-30s run rate management guided.
- 2026-10-22editedthird-quarter call read: management restated the mid-30s run rate for 2026
- Every dollar of it depends on two other companies' production schedules. Airbus built 57 A350s in 2025 against the 144 a year its own rate-12 target implies, and Airbus says a third party's engine deliveries are constraining the ramp.
- The defence and space half is not growing today. Hexcel's Defense, Space & Other sales fell 5.8% in the first half of 2026, and management says Defence and Space alone, excluding the divested industrial businesses, was unchanged year over year in Q2 2026.
- Nothing is contracted for it. The agreements are existing long-term supply agreements whose volume follows the airframers' build rates, and neither the Boeing framework announced in July 2026 nor any Airbus agreement discloses a volume.
- Airbus delivered 57 A350s in 2025, which is 40% of the 144 a year that rate 12 implies, so the 2028 target requires output to more than double in three years.
- Airbus itself says Pratt & Whitney's failure to commit to the number of engines Airbus ordered is negatively impacting 2026 guidance and the ramp-up trajectory - the binding constraint on the largest programme is a third party's.
- Airbus has missed its own delivery target in several recent years, and cut the 2025 target in December 2025 before reporting 793.
What could go wrong, and how we would know
The risks
- Two customers. Airbus and Boeing are 52% of sales, and every slip in their build rates reaches Hexcel months later.
- Others set the pace. Airbus says an engine maker's failure to deliver is already hurting its 2026 deliveries and ramp.
- Debt above target. Net debt is about 2.3 times yearly cash profit, above the company's own 1.5 to 2.0, so cash goes to lenders.
- A new way to build planes. The next single-aisle may use faster plastic-based materials that skip the curing and glues Hexcel sells.
The evidence under this - 6 claims
- It is an airframer naming its supplier as the reason its own targets are at risk, which is exactly the statement a company under delivery pressure has an incentive to make. RTX has not conceded the shortfall in the same terms.
- Engine supply constrains the A320 family, a narrowbody programme where Hexcel's content is concentrated in engines and nacelles rather than the airframe - so the constraint bites hardest on the part of the pond Hexcel earns least from.
- The A350, the programme this thesis actually turns on, is powered by Rolls-Royce rather than Pratt & Whitney, so this constraint does not explain the 57 deliveries in 2025.
- The ratio falls on the denominator as well as the numerator: at the guided 2026 adjusted earnings and free cash flow above $195M, adjusted cash profit rises and leverage reaches the top of the target range without any extraordinary repayment.
- Nothing matures inside the horizon. The nearest maturity after the 2027 notes were redeemed is the revolving facility and the 2031 notes, both in 2031, so the leverage is a policy constraint rather than a solvency one.
- 2026-10-22editedthird-quarter call gave the leverage the company measures, 2.1 times, down from the 2.3 times in this claim, with a return to its target range expected in the fourth quarter
What we still do not know
- How much Hexcel earns from each aircraft. It has never stated its content per plane, so the $500M target cannot be checked.
- Whether 49 cents of profit on each extra dollar is real. Q3 was 24; management still says mid-30s.
The evidence under this - 3 claims
- Every dollar of it depends on two other companies' production schedules. Airbus built 57 A350s in 2025 against the 144 a year its own rate-12 target implies, and Airbus says a third party's engine deliveries are constraining the ramp.
- The defence and space half is not growing today. Hexcel's Defense, Space & Other sales fell 5.8% in the first half of 2026, and management says Defence and Space alone, excluding the divested industrial businesses, was unchanged year over year in Q2 2026.
- Nothing is contracted for it. The agreements are existing long-term supply agreements whose volume follows the airframers' build rates, and neither the Boeing framework announced in July 2026 nor any Airbus agreement discloses a volume.
- Management itself guides the run rate down to the mid-30s from the 49% just delivered, and the second-half guidance implies a step down further than that.
- Q2 2026's incremental margin was helped by the absence of the $24.2M Belgian restructuring charge in the base quarter and by the divestiture of lower-margin industrial businesses, neither of which recurs.
- Management says the industry requires running fast to stand still on cost, so the incremental margin is struck before cost inflation it must offset every year.
- 2026-10-21editedQ3 incremental margin of about 24% computed from the release: below the mid-30s run rate management guided.
- 2026-10-22editedthird-quarter call read: management restated the mid-30s run rate for 2026
- 'About 13%' for the year is the chairman's rounding of a guide whose own nine-month figure is 12.8% and whose third quarter was 11.0%, so it requires a fourth-quarter margin near 14%, a step that has no precedent in the quarters reported this year.
- The 'mid-30s' incremental is the same figure the company gave in July, when the second quarter ran at 49%; this quarter ran at about 24% (C30), so management is restating a run rate the last quarter did not deliver.
What would change our view
- Unattractive→AttractiveThe margin arrives and the balance sheet is repaired, at a lower multiple
It needs all of:
- ○A full year with adjusted operating margin above 15%, and each extra dollar of sales keeping at least 25 cents for four rising quarters.checked at the next results, due by 27 Jan
- ○Debt falls back inside the company's own target and buybacks restart.watched as it happens
- ○The whole company is priced at about 13 times yearly adjusted cash profit.about ≈$59.27 today, 34% below the last close
- Reopens the researchThe shipset content is finally disclosed
It needs:
- ○Hexcel says how much material it sells per aircraft on any programme, so the $500M target can be checked.checked at the next results, due by 27 Jan
We would also be wrong if
- Airbus builds 10 A350s a month during 2027 and Boeing is cleared to build 47 of its 737s a month.
- For four quarters, each extra dollar of sales keeps nearer the 49 cents it kept in the June 2026 quarter than the mid-30 cents management expects.
- Hexcel raises its end-of-decade goals above $500M of extra aircraft sales or an 18% margin.
What is coming
- 21 October 2026 Q3 2026 results: Whether adjusted operating margin holds near 12.5% through the guided second-half step-down, and whether the A350 guide of at least 80 shipsets survives.
Every dated event ahead
- 21 October 2026 Q3 2026 results: Whether adjusted operating margin holds near 12.5% through the guided second-half step-down, and whether the A350 guide of at least 80 shipsets survives.
- December 2026 Boeing reaches 47 737s a month: Whether the second airframer's ramp arrives on Boeing's own schedule, which Hexcel ships four to six months ahead of.
- January 2027 Q4 2026 results and 2027 guidance: Whether 2027 guidance prices incremental margin nearer Q2's 49% or the mid-30s, and whether buybacks resume as net debt nears the 2.0 times ceiling.
- April 2027 2027 proxy: 2024-2026 performance-share payout: Whether return on invested capital cleared the threshold the 2023-2025 cycle missed, the pay plan's own test of the capital record.
Since we wrote this
- 22 OctEarningsThe call lifted the A350 outlook but not the marginManagement plans about 100 A350 shipsets in 2027 and talks of a margin near 20%, but the third-quarter profit rate stayed low and the price already owns the recovery.
- 21 OctEarningsQ3 margin fell to 11.0%; guidance heldMargin landed under the 11.5% line we set, but guidance was reaffirmed, so the view held.
- 5 OctThesis reviewThe long case now sits under the verdictAfter the five-name pilot, the owner asked for the long case on every thesis.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and filings.
- 26 SepThesis reviewWhat would prove us wrong now points up. View heldEvery test of the view described the recovery disappointing, which could only confirm a view that the price already owns it. They now name a faster ramp, a higher profit rate and bigger targets.
- 11 SepThesis writtenUnattractive, medium convictionSix weeks after a raised forecast that the shares fell on, with the recovery visible in the margins and already in the price.
Also checked 1 time with nothing new to act on.
Sources 21
Sources - 21
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-02-11the product and process description, customer concentration, the market split, the composite-content-by-generation table, the competition section, the barriers-to-entry language and the 2025 buyback detailbacksC01C02C03C46C05C10C15C16C18C23C24C48C45C35C49C53C55
- S02primarycompany filing2026-07-29the income statement, the debt note, the share-count reconciliation and the effective tax ratebacksC34C36C38C53C55
- S03primarycompany filing2026-07-29the raised 2026 guidance, the market split, free cash flow, the notes refinancing and the buyback pausebacksC25C28C45C35
- S04primarytranscript2026-07-30Hexcel Q2 2026 earnings call, held 30 July 2026management's $500M and $200M incremental-sales figures, the 18% margin target, incremental margins, the carbon fibre line restarts, the long-term agreement structure and the 2030 Airbus expiry; and eight analysts' questions, including RBC's on the second-half margin step-down
When Airbus and Boeing ultimately achieve their publicly stated peak build rates on the major commercial programs, that will generate $500 million in incremental annual sales for Hexcel from our existing contracts. In addition, we expect about $200 million in incremental sales in Defense & Space before the end of the decade. … our agreements tend to be for about 5 to 7 years. So every year, 15% to 20% of them come due. Now obviously, the Airbus ones, which represent 40% of our business expire at the end of 2030. … we now have firm orders from Airbus through the end of 2026 that confirm at least the 80 shipsets for the A350. … in our peak year of 2019, adjusted EBITDA totaled $585 million with an adjusted EBITDA margin of 24.8%.
backsC07C19C20C21C26C27C30C31C32C33C45C39 - S05primarycompany filing2026-01-28the full-year 2025 figures, the original 2026 guidance and the first statement of the $500M incremental-revenue figurebacksC23C24C31
- S06primarycompany filing2026-07-29the quarterly sales split by segment and market - and, read against the release, confirmation that Hexcel publishes no aftermarket or spares lineread, but no claim rests on it
- S07primarycompany filing2026-01-30The Boeing Company 10-K for the year ended 31 December 2025 The Boeing Companya customer's own account of what it costs to qualify a composite source, and its stated 737 and 787 production rates - the same market described by a party with the opposite incentivebacksC04C12C12b
- S08primarycompany release2026-01-12the 2025 delivery count and the per-programme split - 607 A320 Family, 93 A220, 57 A350, 36 A330 - and a year-end backlog of 8,754 aircraft
Airbus delivered 793 commercial aircraft to 91 customers globally in 2025 … The backlog at the end of December 2025 increased to a new year-end record of 8,754 aircraft
backsC13 - S09primarycompany release2026-02-18the A320, A350 and A220 rate targets, the 2026 delivery guidance, and Airbus's own statement that Pratt & Whitney engine commitments are constraining the ramp
expects to reach a rate of between 70 and 75 aircraft a month by the end of 2027, stabilising at rate 75 thereafter … continues to target rate 12 for the A350 programme in 2028 … targeting a rate of 13 aircraft a month for the A220 programme in 2028 … around 870 commercial aircraft deliveries
backsC11C22 - S10primaryindustry data2026-09-11NCAMP - the National Center for Advanced Materials Performance, Wichita State University NIAR Wichita State University NIARwhat a composite qualification costs and how long it takes, measured by the body that exists to reduce it, and the fact that the FAA and EASA both accept its design values
the time required for certification of new composite materials by a factor of four and the cost of certification by a factor of 10
backsC06 - S11primaryindustry data2026-09-11Toray Advanced Composites - thermoplastic composites product documentation Toray Advanced Compositesthe attacker's own account of what its material does that Hexcel's thermoset prepreg does not, including welding as a replacement for adhesive bonding
can be press formed in minutes and overmolded for one shot processing
backsC08 - S12primarygovernment data2026-09-11a government programme, outside every party's marketing, aimed at the manufacturing-rate problem that decides which material the next single-aisle aircraft is built from
addresses an aviation industry need for more rapid production of composite aircraft to meet increasing global demand for lightweight transport aircraft
backsC09 - S13primarycompany release2026-07-20the programmes covered - and the absence of any volume, term or value
a broad enterprise-level composite material supply framework
backsC21 - S14primarycompany release2026-07-21Hexcel putting one of its own systems into the public shared database, and its stated reason: making the performance data readily available to new aerospace and defence entrantsbacksC07
- S15secondaryvendor data2026-09-11Research pack, frozen at content/HXL/research-pack.jsonprice, enterprise value, the return series, the analyst split, the FINRA short-interest settlement, the insider Form 4 transcription and the segment split from the filing's own XBRL instancebacksC43C44
- S16primarycompany filing2026-04-01beneficial ownership at 17 March 2026, director and officer biographies, the CD&A's bonus and performance-share measures, targets and 2025 payouts, and the 2023-2025 PSA vestingbacksC50C51C52
- S17primarycompany filing2026-03-13Coogan's appointment effective 1 May 2026, his prior roles at Axcelis and Kaman, and his offer letterbacksC52
- S18primarycompany filing2023-02-08the 2020 to 2022 income statements: sales, cost of sales, selling and admin, research and technology and operating incomebacksC55
- S19primarycompany filing2026-10-21Q3 sales, adjusted operating margin, market split, unchanged 2026 guidance and the $750M buyback authorization.backsC28C30
- S20primarycompany filing2026-10-21Total debt, cash and the financing section at 30 September 2026.backsC34
- S21primarytranscript2026-10-22management's prepared remarks on A350 purchase orders and the 2027 outlook, defence and space softness, hiring and margin, leverage and the buyback; and twelve analysts' questions on the A350 and channel stock, hiring cost, price realisation, capital spending, competition and the 2027 framework. The vendor's text prints the year as 2020 in many places and garbles names; only figures that reproduce the 21 October release or are stated whole are used
Current demand signals indicate around 100 A350 ship sets.
backsC20C32C56C57C58C59C60