Howmet Aerospace HWM
Industrials · Aerospace & Defense · Synthos Deep Dive · 2026-08-04
The Overview
Every jet engine contains a set of blades that spin inside a chamber hotter than the melting point of the metal they are made from. Making those blades — as single crystals of nickel alloy, hollow, with internal cooling passages — is one of the hardest routine manufacturing tasks in the world. Howmet is one of a very small number of companies that can do it, and it also makes the specialised bolts that hold aircraft together and the forged aluminium wheels on heavy trucks.
Business is very good. In the three months to March, sales of engine parts rose 29%, and the profit margin on that division rose from about 33 cents in the dollar to 36.6 cents. Total company sales rose 19%. The company has beaten analysts' profit forecasts five quarters in a row.
The reason this business is attractive over decades rather than years is worth understanding. Engine makers sell the engine itself at roughly break-even and make their money over the next twenty or thirty years selling replacement parts — which, according to an industry analysis in our research base, are sold to airlines at three to four times the price charged to the aircraft manufacturer, at around 60% gross margins, and are worth three to five times the original engine over its life. Howmet supplies parts into both halves of that. Every engine it helps build today is an annuity for decades.
So the company is excellent. Here is the problem.
The shares cost $288.20, which is within half a percent of the highest price they have ever reached, after rising 56% in a year while the wider market rose 24%. At that price you are paying:
- about 70 times last year's profits,
- about 47 times what analysts expect for 2027,
- 13.6 times annual sales,
- and 80 times the spare cash the company actually generated last year — which is another way of saying the cash return on the price you pay is 1.24% a year.
The last figure is the most telling. Howmet is spending heavily to build new factory capacity — capital spending is more than twice its depreciation charge — which is exactly the right thing to do when demand is strong, but it means very little cash comes back to shareholders while it is happening.
Even the analysts think the run is nearly over: their average target is $299.18, only 3.8% above the price, the smallest gap of any company in this batch. Our own estimate is $287 — slightly below the current price.
And the company reports its next results in two days.
- Downside Risk 7/10. A fortress balance sheet, an extreme valuation, and a share price with no drawdown to fall back on.
- Growth Quality 8/10. The best in this batch — broad, accelerating, and improving margins at the same time.
- Exponential Potential 5/10. A genuine multi-decade annuity, growing at the pace two aircraft makers decide.
Putting a number on it: our fair-value estimate is $287 against a current price of $264.85 — real upside if our numbers are right.
Our summary metrics
"Rated 7 — the balance sheet is not the risk; the price is, and there is no cushion under it. Supports: net debt of roughly $2,251 million against trailing EBITDA of approximately $2,544 million — **0.88x net leverage**, the lowest of any levered name in this batch; interest coverage of **15.4x**; cash rising from $742 million to **$2,435 million** in the March quarter; a 17.4% return on invested capital and a 33.1% return on equity; a dividend payout of only 10.7%; and a debt structure actively being improved — the 8-K of 2026-05-28 discloses the early prepayment on 22 May of the remaining ¥29,702 million (approximately $187 million) JPY term loan and a cross-currency swap converting $300 million of 6.750% Notes due 2028 into a Japanese Yen liability at approximately 3.88%, for a stated **$12 million of annual interest saving**. Against that: **69.8x trailing adjusted earnings, 47.5x the 2027 consensus, 13.6x sales, 46.2x enterprise value to EBITDA and 80.4x free cash flow — a 1.24% free-cash-flow yield**; a stock **0.37% below its all-time high** with `max_dd_from_peak` of −0.37%, meaning there is literally no drawdown to fall back on, after a **56.4% twelve-month rise**; a street consensus only **3.8%** above spot with the lowest of 25 targets at $228, some **20.9% below** the price; capital expenditure running at **2.25x depreciation** and 7.5% of trailing revenue, which suppresses free cash flow while capacity is built; and an end market — commercial aircraft build rates — that is cyclical, concentrated in two airframers, and currently at a cyclical high. **A 7: an exceptional company where essentially all of the risk is valuation and timing.**"
"Rated 8 — the best growth profile in this batch, and it is broad-based and margin-accretive rather than bought. Revenue compounded from **$4,972 million (FY2021) to $8,252 million (FY2025) — a 13.5% annual rate** — and accelerated to **+19.1% in the March 2026 quarter** ($2,313 million against $1,942 million). The composition is what makes it an 8: **Engine Products third-party sales grew 29% to $1,253 million with segment adjusted EBITDA up 44% and the margin up 400 basis points to 36.6%**, which the 10-Q attributes to 'growth in the commercial aerospace, defense aerospace, and gas turbines markets, **including engine spares growth**' — the highest-margin revenue in aerospace. Fastening Systems grew 14% with margin up 100 basis points to 31.8%. Company operating margin reached **32.6%** against 25.2% a year earlier, and adjusted earnings have beaten consensus in **five consecutive quarters** by 10.8%, 4.4%, 4.4%, 8.8% and 9.9%. Consensus, with 13-16 analysts, models revenue of $9,752M, $11,039M and $12,282M for 2026-2028 — **+18.2%, +13.2%, +11.3%** — and EPS of $5.06, $6.06 and $7.18, compounding roughly 19% a year. **What holds this to an 8 rather than a 9: Engineered Structures revenue FELL 3% on deliberate product rationalisation, two small acquisitions (Brunner from February 2026 and CAM operations from the second quarter) are inside the Fastening Systems numbers without disclosed contribution, and the entire profile depends on airframer build rates that neither Howmet nor its investors control.**"
"Rated 5 — a genuine annuity buried inside a cyclical, and the knowledge base explains the mechanism better than the company does. Howmet makes investment castings — turbine airfoils, seamless rolled rings, structural parts — for jet engines, plus aerospace fasteners and forged aluminium wheels. The structurally attractive part is the **engine spares** business the 10-Q names as a driver of the 29% Engine Products growth: a knowledge-base claim at conviction 60 explains that engine makers sell original equipment at or near break-even and earn approximately **60% gross margins on aftermarket parts**, and that **aftermarket value is three to five times original-equipment value over an engine's life**; a second at conviction 80 states that aftermarket parts sell to airlines at **three to four times** the price charged to the airframer. Every engine Howmet has ever supplied castings into generates a multi-decade spares stream, and the installed base only ratchets. A third claim, at conviction 70, describes the barrier: jet engines demand 'extraordinary technical performance at extraordinary low cost, creating barriers as hard as semiconductor fabrication', with only three or four primes in existence — and the qualified-supplier list beneath them is shorter still. **What holds this to a 5 rather than higher: Howmet cannot grow faster than the aircraft fleet for long; the pace is set by two airframers and three engine primes; there is no adoption curve; and the same 2019 knowledge-base claim that describes the narrowbody duopoly also notes COMAC's state-backed attempt to break it.**"
What does “fair value” mean?
Fair value is Synthos’s estimate of what one share is worth today, based on our model of the company’s future cash generation and the risks to it. It is not a price target or a prediction of where the stock trades next quarter — it’s the price where we think risk and reward are balanced. Above it, you’re paying for outcomes better than our base case; below it, the market is offering a margin of safety.
The model’s inputs and weightings are proprietary — they’re the product. What we publish is the output, plus an independent, public-math cross-check further down the page (“Check our number”) so you can judge it for yourself.
What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
The Road Ahead
What we expect to matter in each window, and the evidence that would prove us wrong.
Short term 0-6 months
Neutral- Driver
- "An outstanding business at an all-time high, two days from a print, with the street 3.8% above the price. Momentum is unambiguous and positive: $288.20 is **0.37% below the 52-week and all-time high of $289.26** and **68.5% above the low of $171.00**, sits **6.6% above a rising 50-day average of $270.40** and **21.5% above a 200-day of $237.24**; RSI is **57.6** and MACD **+4.02**; and the twelve-month return of **+56.4% against SPY's +24.3%** is a 32-point outperformance, the largest in this batch. Five consecutive earnings beats support it. The problem is entirely arithmetic: **69.8x trailing adjusted earnings and 47.5x the 2027 consensus for a company consensus expects to compound EPS at roughly 19%**, a PEG of 2.4; a **1.24% free-cash-flow yield**; a **street consensus of $299.18, just 3.8% above spot** — the tightest in this batch — with the lowest of 25 targets at **$228, some 20.9% below**; and our own base of **$287, 0.4% BELOW the close**. **`tech.max_dd_from_peak` is −0.37%: there is no drawdown here at all.** Buying two days before earnings at an all-time high with a base fair value below the price is not an entry with an edge in it."
- What we’re watching
- "The 2026-08-06 print against adjusted EPS of $1.24 and revenue of $2,427.9M, and specifically: **Engine Products third-party sales and segment adjusted EBITDA margin**, which ran +29% and 36.6% in the March quarter and are the entire story; whether **engine spares** growth is again named as a driver, since aftermarket is the high-margin annuity; Fastening Systems, now containing both the Brunner acquisition (from February) and the CAM operations (from this quarter) without disclosed contribution; Engineered Structures, which fell 3% on deliberate rationalisation; and the full-year outlook. Capital expenditure, running at 2.25x depreciation and suppressing free cash flow. And whether the stock can hold the all-time high on a beat — the fifth consecutive beat produced a 20.2% three-month move, so a sixth is substantially priced."
- Confidence
- Medium
Medium term 6-24 months
Tailwind- Driver
- "The medium-term case is the commercial-aerospace build-and-spares cycle and it is the strongest end-market backdrop in this batch. The 10-Q states directly: 'In 2026, as compared to 2025, demand in the commercial aerospace, defense aerospace, and gas turbines markets is expected to increase, **including engine spares growth in these markets**. **Capital expenditures are expected to remain elevated, with additional investments in capacity expansions** to support aerospace and gas turbines market growth.' A company that is expanding capacity into demand is a company that believes the cycle has duration. The margin evidence supports it: Engine Products segment adjusted EBITDA margin went from 32.6% to **36.6%** in a single year while absorbing approximately 235 net additional headcount, which means price and mix are outrunning cost inflation. Consensus models revenue of **$11.0 billion in 2027 and $12.3 billion in 2028** — a 49% increase on 2025 — and EPS compounding at 19%. **The knowledge base's sector cluster explains why the margin holds: aftermarket parts sell to airlines at three to four times the price charged to the airframer, and aftermarket value is three to five times original-equipment value over an engine's life.** Howmet sits on the supply side of exactly that economics."
- What we’re watching
- "Airframer build rates, which set the pace and which neither the company nor its investors control. Whether Engine Products margin holds above 35% as capacity additions and 235-plus net new heads absorb into the base. Whether engine spares — the aftermarket component — is disclosed with any granularity, since it is the difference between a cyclical supplier and an annuity. Capital expenditure against the 'remain elevated' guidance, currently 2.25x depreciation and the reason free cash flow yields only 1.24%. Whether Engineered Structures' deliberate revenue rationalisation stops, and at what margin. The integration of the Brunner and CAM acquisitions into Fastening Systems, neither of which has a disclosed price or contribution in this archive. And tariffs, which the 10-Q names repeatedly as an uncertainty affecting both cost and recovery."
- Confidence
- Medium
Long term 2+ years
Tailwind- Driver
- "The long-run structural case rests on a qualified-supplier position in a market with almost no entry. A knowledge-base claim at conviction 70 puts it precisely: jet engines require 'extraordinary technical performance at extraordinary low cost, creating **barriers as hard as semiconductor fabrication**; only 3-4 primes exist, making the industry highly consolidated.' The list of firms qualified to pour single-crystal turbine airfoils that survive above their own melting point is shorter than the list of engine makers, and requalification takes years and a certification programme no customer wants to run. Underneath that sits the aftermarket annuity: every casting Howmet ships into an engine generates spares demand for the twenty-to-thirty-year life of that engine, and the installed base only grows. The financial evidence that the position is real is the margin — a **36.6% segment adjusted EBITDA margin on physically manufactured metal parts**, which is higher than most software-adjacent industrials achieve — combined with a **17.4% return on invested capital** and an ability to raise capital expenditure to 2.25x depreciation without leverage exceeding 0.88x."
- What we’re watching
- "Whether the narrowbody duopoly holds: the only structural bear note in the knowledge base is a 2019 chinatalk claim that COMAC's C919 targets the 737/A320 market 'with massive state backing', while conceding China 'still has a fair way to go'. Whether additive manufacturing or alternative processes ever erode the investment-casting position — no evidence in this file either way. Whether Forged Wheels, the commercial-transportation segment the 10-Q says faces 'low' demand 'with modest recovery', remains in the portfolio; Howmet has already rationalised Engineered Structures and sold a disk-forging facility in Savannah. Chief executive John Plant's succession, on which the file contains no disclosure — he has run this business through its entire post-Arconic history. And whether the current 36.6% Engine Products margin is a cycle peak or a new structural level."
- Confidence
- Medium
Exponential Potential
"Rated 5 — a genuine annuity buried inside a cyclical, and the knowledge base explains the mechanism better than the company does. Howmet makes investment castings — turbine airfoils, seamless rolled rings, structural parts — for jet engines, plus aerospace fasteners and forged aluminium wheels. The structurally attractive part is the **engine spares** business the 10-Q names as a driver of the 29% Engine Products growth: a knowledge-base claim at conviction 60 explains that engine makers sell original equipment at or near break-even and earn approximately **60% gross margins on aftermarket parts**, and that **aftermarket value is three to five times original-equipment value over an engine's life**; a second at conviction 80 states that aftermarket parts sell to airlines at **three to four times** the price charged to the airframer. Every engine Howmet has ever supplied castings into generates a multi-decade spares stream, and the installed base only ratchets. A third claim, at conviction 70, describes the barrier: jet engines demand 'extraordinary technical performance at extraordinary low cost, creating barriers as hard as semiconductor fabrication', with only three or four primes in existence — and the qualified-supplier list beneath them is shorter still. **What holds this to a 5 rather than higher: Howmet cannot grow faster than the aircraft fleet for long; the pace is set by two airframers and three engine primes; there is no adoption curve; and the same 2019 knowledge-base claim that describes the narrowbody duopoly also notes COMAC's state-backed attempt to break it.**"
What could take this further than the base case — and where the market may be underpricing it. Full forward-growth and acceleration math in Deeper analysis, §4 below.
Deeper analysis
Technicals, fundamentals, valuation, and the full expert-claim evidence panel — the detail behind the numbers above.
EARNINGS BANNER — REPORTS IN TWO DAYS
Howmet Aerospace reports second-quarter 2026 results on 2026-08-06, two calendar days after this dive is struck. Vendor consensus is adjusted EPS $1.24 and revenue $2,427.9M, against year-ago actuals of $0.91 and $2,053M — implied growth of +36.3% and +18.3%. Howmet has beaten the adjusted-EPS consensus in each of the last five prints, by 10.8%, 4.4%, 4.4%, 8.8% and 9.9%. The print will be the first to include the CAM operations in Fastening Systems and the second to include Brunner. The verdict below is set with that two-day gap as an explicit input. A dive struck 48 hours before earnings on a stock 0.37% below its all-time high, with a base fair value below the closing price, is not an entry recommendation.
Reference table
| Street consensus | $299.18 (+3.8% — the tightest implied upside in this batch) · median $300 · high $340 · low $228 (−20.9%) · 21 buy / 3 hold / 1 sell across 25 analysts |
| Valuation | 69.8x trailing adjusted EPS · 66.9x trailing GAAP diluted · 56.9x FY2026E · 47.5x FY2027E · 40.1x FY2028E · 13.63x EV/sales · 46.2x EV/EBITDA · 80.4x price/free cash flow — a 1.24% FCF yield |
| The operating numbers | March quarter: revenue $2,313M, +19.1%; operating margin 32.6% against 25.2%. Engine Products third-party sales $1,253M, +29%, segment adjusted EBITDA margin 36.6%, up 400bp. Fastening Systems $471M, +14%, margin 31.8% |
| Balance sheet (filing, 31 Mar 2026) | Cash $2,435M, up from $742M at year-end; long-term debt $4,050M, up from $2,859M; total assets $13,067M; equity $5,522M. Net debt approximately $2,251M — 0.88x EBITDA; interest coverage 15.4x |
| Segment recast, from an 8-K | The 8-K of 2026-05-28 RECASTS the fiscal 2025 10-K to move a titanium alloy location from Engine Products to Engineered Structures. The vendor's seg_prod block did not pick it up — and drops two whole segments besides |
| Corporate actions not in the payload | Brunner acquisition in Fastening Systems from February 2026; CAM operations from the second quarter of 2026 — neither with a disclosed price. JPY term loan of ¥29,702M (~$187M) prepaid 22 May 2026; $300M of 6.750% 2028 Notes swapped into a ~3.88% Yen liability, saving a stated $12M a year |
| Conviction | EMPTY — ZERO KB claims name Howmet or Arconic. 12 sector text hits, ~6 usable as aftermarket-economics context, ~6 discarded including an "aftermarket" homograph meaning post-IPO share performance |
| Technicals | −0.37% from the all-time high of $289.26; +68.5% above the low of $171.00; max_dd_from_peak −0.37% — no drawdown at all; +6.6% above the 50-DMA, +21.5% above the 200-DMA; RSI 57.6; MACD +4.02; 12-month +56.4% vs SPY +24.3% |
What the experts actually said
No independent expert claims in the Synthos knowledge base yet for HWM — this dive is fundamentals- and technicals-driven, not panel-driven.
Price & moving averages 12 months · 50 & 200-day averages · 52-week range
Solid line = price · dashed line = 50-day average · dotted line = 200-day average · the two thin horizontal lines mark the 52-week high and low. Price above both averages is an uptrend.
Data summary: last close $264.85, 5% below the 50-day average ($278), 8% above the 200-day average ($244) — a mixed trend. 9% below the 52-week high of $293, 53% above the 52-week low of $173.
Bollinger Bands 20-day average ± 2 standard deviations
The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.
Data summary: price $264.85 is currently inside the band (band $261–$299).
RSI (14) momentum gauge · 0–100
Above 70 (overbought zone, shaded) = overbought, below 30 (oversold zone, shaded) = oversold. Currently 39.
MACD 12 / 26 / 9 · trend & momentum
The MACD line crossing above the signal line (bars flip to the up color) = momentum turning up; crossing below (bars flip to the down color) = turning down. Bar height = the size of that gap.
Data summary: MACD is currently below its signal line by 2.53, negative momentum.
Relative performance vs S&P 500 & its sector (XLI (sector)), set to 100 a year ago
Solid = HWM · dashed = S&P 500 · dotted = XLI (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.
Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate
Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.
Key stats an RIA wants
1. What the business is, and a segment table the vendor gets badly wrong
Howmet describes itself in the 10-Q as "a global leader in lightweight metals engineering and manufacturing", whose "innovative, multi-material products, which include nickel, titanium, aluminum, and cobalt, are used worldwide in the aerospace (commercial and defense), commercial transportation, gas turbines, and other markets." Chief executive John C. Plant; approximately 25,430 employees; Pittsburgh, Pennsylvania; NYSE-listed since the 2016 Alcoa separation.
Four reportable segments, confirmed by the 10-Q filed 2026-05-07: Engine Products (investment castings including airfoils, and seamless rolled rings, for aircraft engines and gas turbines), Fastening Systems, Engineered Structures, and Forged Wheels. Performance is managed on Segment Adjusted EBITDA, defined as operating income excluding restructuring, depreciation and amortisation, and special items.
The vendor's seg_prod block loses two segments in two years — REJECTED
| Segment | Vendor FY2025 | Vendor FY2024 | Vendor FY2023 | Vendor FY2022 |
|---|---|---|---|---|
| Engine Products | $4,327M | $3,742M | $3,279M | $2,702M |
| Fastening Systems | $1,746M | $1,577M | $1,349M | $1,117M |
| Engineered Structures | ABSENT | $1,075M | $881M | $796M |
| Forged Wheels | ABSENT | ABSENT | $1,147M | $1,058M |
| Vendor sum | $6,073M | $6,394M | $6,656M | $5,673M |
| Reported revenue | $8,252M | $7,430M | $6,640M | $5,663M |
| Gap | −$2,179M (26.4%) | −$1,036M (13.9%) | +$16M | +$10M |
FY2022 and FY2023 reconcile within rounding. FY2024 is short by $1,036M — the Forged Wheels segment. FY2025 is short by $2,179M — Forged Wheels AND Engineered Structures. All four segments still exist and are described in the 10-Q filed 2026-05-07. This is the MTZ vanished-segment failure mode occurring twice in consecutive years on the same file, and the block is rejected in full.
The seg_geo block is also incomplete. FY2025's ten named countries sum to $7,227M against reported revenue of $8,252M — $1,025M, or 12.4%, unaccounted for, presumably in unnamed smaller markets. The United States home market is present at $4,377M (53.0%), so the SYK/DHR failure mode does not occur, but the block does not close and is used only directionally: the United States is roughly half, France ($625M), Japan ($513M), Germany ($439M) and the United Kingdom ($376M) are the largest overseas markets, and China is $130M — 1.6% of revenue and the smallest China exposure in this batch.
The segment recast, from an 8-K the vendor did not read
On 2026-05-28 Howmet filed an 8-K under Item 8.01 recasting its fiscal 2025 Annual Report. The reason:
> "As previously disclosed and as reflected in the Company's Quarterly Report on Form 10-Q for the quarterly period ended March 31, 2026, in the first quarter of 2026, the Company's Chief Operating Decision Maker reorganized Howmet's segments by moving a titanium alloy location from Engine Products to Engineered Structures as it better aligns with the operations of the Engineered Structures segment."
The 8-K restates Item 7 (Management's Discussion — Engine Products and Engineered Structures) and Item 8 (Financial Statements, Note C). The 10-Q adds that "the comparable periods of Engine Products and Engineered Structures have been recast to reflect the new alignment. The recasting had no impact on the Company's consolidated results, financial position or cash flows."
This matters for two reasons. First, it means any Engine Products or Engineered Structures figure predating May 2026 that has not been recast is not comparable — and the vendor's seg_prod block has not been recast. Second, it is a filed, dated, fully documented disclosure change that no vendor field reflects, sitting in an 8-K that most screens would classify as routine.
The segment detail that IS usable, from the 10-Q — March 2026 quarter
| Segment | Q1 2026 sales | Q1 2025 | Growth | Segment Adj. EBITDA | Margin 2026 | Margin 2025 |
|---|---|---|---|---|---|---|
| Engine Products | $1,253M | $974M | +29% | $458M | 36.6% | 32.6% |
| Fastening Systems | $471M | $412M | +14% | $150M | 31.8% | 30.8% |
| Engineered Structures | $294M | $304M | −3% | $66M | 22.4% | 22.0% |
| Forged Wheels (residual) | ~$295M | — | — | — | — | — |
| Total company | $2,313M | $1,942M | +19.1% | — | — | — |
Engine Products is 54.2% of company revenue and grew 29% while expanding its margin 400 basis points. The 10-Q attributes it to "growth in the commercial aerospace, defense aerospace, and gas turbines markets, including engine spares growth" and notes that "the segment absorbed approximately 235 net headcount in the first quarter of 2026 in support of expected revenue increases." Adding 235 people while expanding the margin 400 basis points is the clearest evidence in this file that price and mix, not just volume, are driving the result.
Engineered Structures is deliberately shrinking. The 10-Q: "Third-party sales for the Engineered Structures segment decreased $10, or 3% ... primarily due to further product rationalization. The Engineered Structures segment continues to focus on the optimization of its manufacturing footprint, including the recent sale of its disk forging facility in Savannah, Georgia, and rationalization of product mix in order to maximize profitability. In 2026 ... this is expected to result in lower revenue in the commercial aerospace and defense aerospace markets." A segment being managed for margin rather than growth, and the company says so plainly.
Two acquisitions are inside Fastening Systems and neither has a disclosed price in this archive: "The Brunner acquisition has been included in the operations of the Fastening Systems segment starting in February 2026. The CAM operations will be included in our Fastening Systems segment starting in the second quarter of 2026." Fastening Systems' 14% March-quarter growth therefore contains an unquantified acquired component, and the second-quarter figure will contain more.
2. Growth, margins and the earnings series
| Fiscal year | Revenue | YoY | Gross margin | Operating margin | Net income | Diluted EPS | Capex | Capex % rev |
|---|---|---|---|---|---|---|---|---|
| FY2020 | $5,259M | — | 21.7% | 16.1% | $211M | $0.48 | — | — |
| FY2021 | $4,972M | −5.5% | 22.8% | 17.4% | $258M | $0.60 | — | — |
| FY2022 | $5,663M | +13.9% | 23.9% | 18.3% | $469M | $1.12 | $193M | 3.4% |
| FY2023 | $6,640M | +17.3% | 24.3% | 18.8% | $765M | $1.85 | $219M | 3.3% |
| FY2024 | $7,430M | +11.9% | 27.6% | 22.5% | $1,155M | $2.83 | $321M | 4.3% |
| FY2025 | $8,252M | +11.1% | 30.7% | 25.8% | $1,508M | $3.73 | $453M | 5.5% |
| Q1 2026 | $2,313M | +19.1% | 36.9% | 32.6% | $580M | — | — | 7.5% (TTM) |
Read the operating-margin column: 16.1%, 17.4%, 18.3%, 18.8%, 22.5%, 25.8%, and 32.6% in the latest quarter. That is 1,650 basis points of expansion in six years, and the rate of expansion is accelerating. Revenue over the same period compounded 13.5% a year from the fiscal 2021 trough and net income went from $258M to $1,508M — a 5.8x increase.
The earnings series is a clean run of beats:
| Report date | Quarter | Revenue | Adj. EPS | vs est. | Surprise |
|---|---|---|---|---|---|
| 2025-05-01 | Q1 2025 | $1,942M | $0.86 | vs $0.776 | +10.8% |
| 2025-07-31 | Q2 2025 | $2,053M | $0.91 | vs $0.872 | +4.4% |
| 2025-10-30 | Q3 2025 | $2,089M | $0.95 | vs $0.910 | +4.4% |
| 2026-02-12 | Q4 2025 | $2,168M | $1.05 | vs $0.965 | +8.8% |
| 2026-05-07 | Q1 2026 | $2,313M | $1.22 | vs $1.110 | +9.9% |
| 2026-08-06 | Q2 2026 | est $2,427.9M | est $1.24 | — | — |
Five for five, and revenue rose sequentially in every one of the five quarters. The earn_cal basis is consistent across all rows, so the series is like-for-like.
One basis observation worth recording: trailing GAAP diluted EPS of approximately $4.31 is HIGHER than the trailing adjusted figure of $4.13. That is unusual — for most companies adjusted exceeds GAAP. Here the difference is driven by low effective tax rates in individual quarters (13.2% in Q2 2025, 13.5% in Q4 2025) and by items the company adjusts out of a favourable GAAP result. We use adjusted EPS for forward multiples to match the estimate basis, and note that the trailing GAAP multiple of 66.9x is slightly lower than the adjusted 69.8x.
3. Balance sheet, cash flow and the free-cash-flow problem
Filing-verified, 31 March 2026 (10-Q filed 2026-05-07), against the 31 December 2025 year-end:
| (millions) | 31 Mar 2026 | 31 Dec 2025 | Change |
|---|---|---|---|
| Cash and cash equivalents | $2,435 | $742 | +$1,693 |
| Total assets | $13,067 | $11,179 | +$1,888 |
| Long-term debt, less current | $4,050 | $2,859 | +$1,191 |
| Total equity | $5,522 | $5,353 | +$169 |
Howmet raised roughly $1.2 billion of long-term debt and more than tripled its cash balance in a single quarter. The vendor's enterprise value implies net debt of $2,251 million ($115,311M of market capitalisation against $117,562M of enterprise value), which reconciles as approximately $4,686M of total debt less $2,435M of cash — arithmetically consistent with the March balance sheet, and we accept it. Net debt of $2,251M against trailing EBITDA of approximately $2,544M is 0.88x — the lowest leverage of any indebted company in this batch — and interest coverage is 15.4x.
And the debt is being actively improved. From the 8-K of 2026-05-28:
> "On May 22, 2026, the Company completed the early prepayment of the remaining outstanding principal amount under its JPY Term Loan Facility of ¥29,702 million (approximately $187 million) with cash on hand. The Company also entered into a cross-currency swap to synthetically convert the outstanding $300 million aggregate principal amount of its 6.750% Notes due 2028 ... into a Japanese Yen liability of approximately ¥47,760 million. The fixed interest rate on the Japanese Yen liability is approximately 3.88% per annum. ... As a result of these actions, the Company expects to achieve a reduction in annual interest expense of approximately $12 million."
Swapping a 6.750% dollar obligation into a 3.88% yen one is a genuine and well-executed piece of treasury work, and it appears in no vendor field.
Cash flow — where the valuation problem lives
| Fiscal year | Operating cash flow | Capex | Free cash flow | FCF margin | Buyback | Dividends | Capex / D&A |
|---|---|---|---|---|---|---|---|
| FY2022 | $733M | $193M | $540M | 9.5% | $400M | $42M | 0.73x |
| FY2023 | $901M | $219M | $682M | 10.3% | $250M | $71M | 0.81x |
| FY2024 | $1,298M | $321M | $977M | 13.1% | $500M | $107M | 1.16x |
| FY2025 | $1,884M | $453M | $1,431M | 17.3% | $755M | $181M | 1.60x |
| TTM | — | — | — | — | — | — | 2.25x |
Free cash flow has compounded 38.4% a year since fiscal 2022 and the margin has expanded from 9.5% to 17.3% — which is excellent. And at $288.20 the market capitalisation is $115.31 billion, so $1,431 million of free cash flow is a 1.24% yield and an 80.4x price-to-free-cash-flow multiple.
The reason cash conversion lags earnings is disclosed and is the right kind of reason. capexToDepreciationTTM is 2.25x — Howmet is spending more than twice its depreciation charge — and the 10-Q says explicitly: "Capital expenditures are expected to remain elevated, with additional investments in capacity expansions to support aerospace and gas turbines market growth." A company building capacity into visible demand at a 36.6% segment margin is doing exactly what it should. But a buyer at today's price is paying 80 times the cash that comes out while it happens.
Capital returns are modest relative to the multiple. The dividend is $0.48 a year — a 0.17% yield on a 10.7% payout ratio; the buyback was $755M in fiscal 2025, 0.65% of market capitalisation. Total shareholder yield is roughly 0.8%. Diluted shares have fallen from 421M (FY2022) to 403M in the March quarter — a 4.3% reduction over four years, or roughly 1% a year.
4. Valuation — priced in or room?
At $288.20 (market cap $115.31B, ~400.1M shares, enterprise value $117.56B):
| Trailing (TTM) | FY2026E | FY2027E | FY2028E | FY2029E | |
|---|---|---|---|---|---|
| Revenue | $8,623M | $9,752M (13) | $11,039M (14) | $12,282M (16) | $13,445M (15) |
| Revenue growth | — | +18.2% (vs FY2025) | +13.2% | +11.3% | +9.5% |
| Adjusted EPS | $4.13 (four earn_cal actuals) | $5.065 (13) | $6.062 (13) | $7.184 (12) | $8.172 (10) |
| EPS growth | — | +36.8% | +19.7% | +18.5% | +13.8% |
| GAAP diluted EPS | ~$4.31 | — | — | — | — |
| P/E on adjusted EPS | 69.8x | 56.9x | 47.5x | 40.1x | 35.3x |
| P/E on trailing GAAP | 66.9x | — | — | — | — |
| EV/Sales | 13.63x | 12.06x | 10.65x | 9.57x | 8.74x |
| EV/EBITDA | 46.2x | — | — | — | — |
| Price / free cash flow | 80.4x | — | — | — | — |
| Free cash flow yield | 1.24% | — | — | — | — |
| Price/book | 20.93x | — | — | — | — |
Estimate coverage is consistent and adequate: 12-16 analysts on every line from FY2026 to FY2029, and 15 and 5 on FY2030. The FY2030 EPS row, on five analysts, is not used for any conclusion. This is the most evenly covered estimate block in this batch and it is worth saying so.
est.ebitdaAvg and est.ebitAvg are fixed-ratio fabrications and are REJECTED. ebitdaAvg is exactly 38.80% of revenueAvg in every year FY2026 to FY2030; ebitAvg is exactly 37.25%. Howmet's actual fiscal 2025 EBITDA margin was 27.6%. The FY2026 ebitdaAvg of $3,784M overstates a realistic figure by roughly 40%. All forward valuation runs on epsAvg.
est.epsAvg validates cleanly. The FY2025 row shows $3.70265 against the reported $3.73 — a 0.7% difference. The EPS row is reliable.
Now look at what the multiples say against the growth. At 47.5x the FY2027 consensus for EPS growth consensus itself puts at 19.7%, the forward PEG is 2.4. For comparison, Parker-Hannifin in this batch trades at 29.1x for 8.8% growth (PEG 3.3) and Fortinet at 45.0x for 9.4% (PEG 4.8) — so Howmet is not the worst on that measure. But on cash it is the most expensive name here without qualification: a 1.24% free-cash-flow yield against Parker's 2.94%, Lowe's 6.23%, Progressive's 13.0% and S&P Global's 4.6%.
Peer context. The vendor peer set is a reasonable industrials cohort — Cummins, Emerson, Illinois Tool Works, Johnson Controls, 3M, Northrop Grumman, TransDigm, UPS, Waste Management, and Parker-Hannifin, which is also in this batch and shares the aerospace-aftermarket driver. TransDigm is the most instructive comparison and is the subject of two 2026 knowledge-base claims on sole-source aerospace pricing power. No peer multiples are supplied in the file, so no peer-multiple comparison is drawn.
4a. What today's price assumes (the inversion)
At $288.20 — 47.5x FY2027 consensus and 40.1x FY2028 — the price embeds:
- Revenue reaches $11.04 billion in 2027 and $12.28 billion in 2028 — +13.2% then +11.3%, a 49% increase on fiscal 2025's $8.25 billion in three years. (Consensus; 14 and 16 analysts.) The nearest check is the 2026-08-06 print at a consensus $2,427.9M, which would be +18.3% year on year.
- Adjusted EPS reaches $6.06 in 2027 and $7.18 in 2028 — +19.7% then +18.5% — (consensus; 13 and 12 analysts), which requires margins to keep expanding on top of the 1,650 basis points already delivered since fiscal 2020.
- Engine Products holds a segment adjusted EBITDA margin near or above 36%. (Filing-confirmed at 36.6% in the March quarter, up 400 basis points.) This is the most fragile assumption in the price — not because there is evidence against it, but because 36.6% is the highest margin this segment has recorded, it was achieved while absorbing 235 net new heads, and a peak margin extrapolated is the standard way cyclical suppliers are mis-valued.
- Commercial aircraft build rates keep rising. (Our number.) Howmet's growth is a derivative of two airframers' production schedules and three engine primes' spares demand. A knowledge-base claim at conviction 70 notes that only three or four engine primes exist — which is the moat and simultaneously the concentration.
- The market keeps paying 40-48x forward earnings and accepts a 1.24% free-cash-flow yield. (Our number.) At 34x FY2028E the stock is $244; at 46x it is $330. The multiple, not the earnings, is the variance — and it is being paid at an all-time high with no drawdown.
4b. The return bridge (why the multiple moves)
Expected return over the next twelve months decomposes as: EPS growth (+18.5%, from FY2027E $6.062 to FY2028E $7.184) + multiple drift (COMPRESSION, from 47.5x to roughly 40x on the respective forward year, −16%) + shareholder yield (+0.8%, being a 0.17% dividend and roughly 0.65% of buyback) ≈ +2% to +4%.
Our base assumes meaningful compression, and the reason is the growth path itself rather than any view on the business. Consensus has EPS growth decelerating from +36.8% (2026) to +19.7% (2027) to +18.5% (2028) to +13.8% (2029). A 47.5x multiple is appropriate to 37% growth; it is not appropriate to 14%, and the market normally begins discounting the deceleration a year ahead of it. This is not a rolldown artefact: we are assuming the market pays 40x on the then-forward year against 47.5x today, a genuine de-rating of 16%.
The shareholder-yield leg is almost nothing — 0.8% — which is unusual for an industrial and is a direct consequence of capital expenditure running at 2.25x depreciation. If the multiple simply held at 47.5x on FY2028E the price would be $341 (+18.4%); at 34x it would be $244 (−15.3%). That 34-point spread on a multiple assumption, against a base case of −0.4%, is the entire argument for waiting.
4c. Variant perception (where we differ, what would surprise)
- We have essentially no variant perception against the street, and that is the verdict input. Consensus is $299.18, just 3.8% above spot — the tightest implied upside in this batch — and our base of $287 is 0.4% below. When the crowd and our own arithmetic both say the stock is at fair value, and the knowledge base contributes nothing at all, there is no edge to act on. No-edge names default toward Watch.
- Where we do differ is on the DATA, and in three places. The vendor's
seg_prodblock loses two entire segments and reconciles $2,179 million short of fiscal 2025 revenue; the 8-K of 2026-05-28 recast the fiscal 2025 segment presentation and no vendor field reflects it; and two acquisitions (Brunner from February 2026, CAM operations from the second quarter) are inside the Fastening Systems figures with no disclosed price or contribution. A reader relying on screens is looking at a $6.1 billion, two-segment company. Howmet is an $8.3 billion, four-segment company that has just reorganised and made two acquisitions. - We think the free-cash-flow yield is the number the market is under-weighting, and we say why it is defensible anyway. 1.24% is extraordinary for an industrial. The defence is that capital expenditure at 2.25x depreciation is capacity expansion into visible demand at a 36.6% segment margin, which is the highest-return use of capital available to this company. We accept the defence and still note that a buyer today pays 80 times current cash generation for the privilege of funding it.
- Positive surprise that would force a re-rate: Engine Products segment adjusted EBITDA margin above 38% on 2026-08-06; revenue above the $2,427.9M consensus with an increased full-year outlook; explicit disclosure of engine-spares growth or aftermarket mix, which would let the market re-rate the revenue quality rather than the growth rate; or capital expenditure guidance falling, which would immediately lift the free-cash-flow yield.
- Negative surprise that would break the thesis: Engine Products margin below 34%, which would suggest 36.6% was a peak; any airframer build-rate reduction; revenue below roughly $2,350M in the second quarter; tariff costs proving unrecoverable — the 10-Q names tariffs four separate times as an uncertainty in both directions; or a sixth consecutive beat failing to hold the all-time high, which would be a clear signal the good news is fully in the price.
Synthos fair values
All three anchors are multiples of the FY2028 consensus adjusted-EPS distribution — mean $7.184, low $6.348, high $8.598 (12 analysts) — cross-checked against FY2027E of $6.062 (13 analysts). FY2028 is used because it is the first year in which the current capacity expansion should be substantially converting to revenue, and because the coverage is even across both years.
- Bear ~$178 — 28.0x the FY2028 consensus LOW of $6.348. Cross-check: 30.3x the FY2027 consensus low of $5.869; roughly 4% above the 52-week low of $171.00. The scenario: airframer build rates flatten or step down; Engine Products margin proves to have peaked at 36.6% and normalises toward 32%; the capacity added at 2.25x depreciation arrives into softer demand; and a cyclical aerospace supplier at 1.2% free-cash-flow yield de-rates to a high-twenties multiple. −38.2%.
- Base ~$287 — 40.0x the FY2028 consensus MEAN of $7.184. Cross-check: 47.3x FY2027E, essentially the current multiple; EV/sales of 9.6x on FY2028E revenue. Sensitivity, stated openly: 36x gives $259 and 44x gives $316. The whole answer lives in a 36-44x band and the current price sits in the middle of it. The scenario: revenue and EPS land on consensus, Engine Products holds its margin, capital expenditure stays elevated and the free-cash-flow yield stays low, and the multiple compresses modestly as growth decelerates from 37% to 14%. −0.4%, plus roughly 0.8% of shareholder yield.
- Bull ~$385 — 44.8x the FY2028 consensus HIGH of $8.598. Cross-check: 63.5x FY2027E; 33.1% above the all-time high and 13.2% above the street's high target of $340. The scenario: the aerospace cycle extends further than consensus allows, engine spares grow faster than original equipment and lift the revenue quality, Engine Products margin pushes toward 40%, the Brunner and CAM acquisitions prove accretive, capital expenditure normalises and the free-cash-flow yield doubles, and the market pays a full aftermarket-annuity multiple. +33.6%.
Base is 0.4% BELOW spot; asymmetry roughly 0.88:1 — more downside than upside on our own numbers. That, plus a street target only 3.8% above, an empty conviction lane, an all-time high with zero drawdown, and earnings in 48 hours, is why this is a Watch and not a close call.
5. Knowledge base — an empty name lane and a genuinely useful sector cluster
Raw entity hits naming Howmet Aerospace or Arconic: ZERO. Sector text sweep: 12 hits, none naming Howmet. Used as labelled sector context: 6. Discarded: 6.
The primary sweep covered entity matches on HWM, Howmet and Arconic — Howmet's former name — across all 51,928 distilled claims. It returned nothing. A supplementary text sweep on Howmet, aerospace supply chain, jet engine, aircraft build rate, narrowbody, GE Aerospace, TransDigm and aftermarket returned 12 hits, not one of which names Howmet. The lane is empty at the name level and is reported as empty.
Six sector claims are usable and, unusually for a sector lane, they explain the economics of this business better than any company disclosure in the archive.
> 2026-02-01 · bullish · conviction 60 · entity: GE · speaker: Romesh Naranaswami · channel: 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."
> 2024-03-31 · bullish · conviction 80 · horizon: principle · channel: business_breakdowns · no named speaker
> "Aftermarket parts/repair is a ~$100B market carrying far more profit than OEM sales; parts sell to airlines at 3-4x the price sold to Boeing."
> 2026-02-01 · neutral · conviction 70 · horizon: principle · channel: business_breakdowns
> "Jet engines require extraordinary technical performance at extraordinary low cost, creating barriers as hard as semiconductor fabrication; only 3-4 primes exist, making the industry highly consolidated."
> 2026-02-01 · bullish · conviction 62 · entity: GE · speaker: Romesh Naranaswami
> "GE Aerospace has ~$175B backlog (~6-7yrs commercial services), 70% predictable high-margin aftermarket revenue and dominant engine share, supporting steady growth."
> 2026-05-29 · bullish · conviction 55 · entity: TDG · speaker: Colin · channel: invest_like_the_best
> "TransDigm a buy today — operates like public private-equity, buying niche proprietary aerospace parts with pricing power and expanding margins; but small <1% position and slower cyclical with rate-sensitive debt."
> 2019-05-01 · neutral · conviction 62 · entities: Boeing, Airbus, COMAC · channel: chinatalk
> "COMAC's C919 targets the 737/A320 narrowbody market with massive state backing, but China still has a 'fair way to go' to break the duopoly."
What this set is worth, stated carefully. None of these claims is about Howmet, and none contributes name-level conviction. But three of them describe the mechanism that makes Howmet's 36.6% Engine Products margin possible and durable: aftermarket parts priced at three to four times the original-equipment price, at roughly 60% gross margins, worth three to five times the engine's original value over its life, in an industry with three or four primes and semiconductor-grade qualification barriers. The 10-Q's phrase "including engine spares growth" is the only place Howmet itself points at this, and the knowledge base quantifies what the company leaves implicit. A fourth (GE Aerospace's ~$175 billion backlog covering six to seven years of commercial services) is the demand visibility Howmet's own capacity expansion is built against. The COMAC claim is the only structural bear note available and it is seven years old.
Six discards, and one is a genuine homograph.
| Discard reason | Claims |
|---|---|
| "Aftermarket" meaning post-IPO share-price performance | 1 — biotech_hangout, 2025-10-24: "deals now see positive aftermarket performance, creating FOMO". A financing-market term, not an aerospace one. |
| Aftermarket economics in NON-aerospace industrials | 3 — Ametek (through-cycle growth, ~30% recurring aftermarket), Graco (corrosive-fluid wear, aftermarket at 5-6x system cost), Diploma ("Amazon of seals") |
| TransDigm duplicate | 1 — a second brent_johnson claim making the same sole-source-pricing point |
| Trane aftermarket mix | 1 — HVAC service mix, unrelated end market |
Conclusion. The Synthos knowledge base has no view on Howmet Aerospace, and the entity sweep is genuinely empty rather than thin. What it has is a well-evidenced, mechanistic account of why aerospace aftermarket economics are attractive, which supports the exponential-potential score of 5 and the qualitative case in Section 1 — and contributes nothing to the fair value, nothing to the verdict, and no reason to prefer this name over any other aerospace supplier. Conviction rating: none at the name level. Breadth 3 (business_breakdowns, invest_like_the_best, chinatalk). Claim count at the name level: ZERO.
6. Data integrity — what we rejected and why
Eight findings.
1. seg_prod loses TWO entire segments in consecutive years — REJECTED in full. The block omits Forged Wheels from FY2024 onward and both Forged Wheels and Engineered Structures from FY2025, leaving four lines that sum to $6,073M against reported fiscal 2025 revenue of $8,252M — a 26.4% shortfall. FY2022 and FY2023 reconcile within rounding. All four segments exist and are described in the 10-Q filed 2026-05-07. This is the MTZ vanished-segment failure mode occurring twice on one file. The 10-Q's segment tables are used instead.
2. A SEGMENT RECAST filed in an 8-K appears in no vendor field. The 8-K of 2026-05-28, filed under Item 8.01, recasts the fiscal 2025 Annual Report's Item 7 and Item 8 because "in the first quarter of 2026, the Company's Chief Operating Decision Maker reorganized Howmet's segments by moving a titanium alloy location from Engine Products to Engineered Structures." Any pre-recast Engine Products or Engineered Structures figure is not comparable to a post-recast one, and the vendor's block is pre-recast.
3. TWO acquisitions are inside the reported numbers with no disclosed price — flagged, not quantified. The 10-Q states that "the Brunner acquisition has been included in the operations of the Fastening Systems segment starting in February 2026" and that "the CAM operations will be included in our Fastening Systems segment starting in the second quarter of 2026." Neither purchase price nor revenue contribution is disclosed in this archive. Fastening Systems' 14% March-quarter growth therefore contains an unquantified acquired component, and the 2026-08-06 print will contain more. cf_a.acquisitionsNet of +$9M for fiscal 2025 predates both.
4. seg_geo does not close — used directionally only. The ten named countries sum to $7,227M against $8,252M of reported fiscal 2025 revenue — $1,025M (12.4%) unaccounted for. The United States home market IS present at $4,377M (53.0%), so the SYK/DHR failure mode does not occur, but no percentage from this block is treated as precise. The one figure worth recording is China at $130M — 1.6% of revenue, the smallest China exposure in this batch.
5. est.ebitdaAvg and est.ebitAvg are fixed-ratio fabrications — REJECTED. ebitdaAvg is exactly 38.80% of revenueAvg and ebitAvg exactly 37.25% in every year FY2026 to FY2030, against an actual fiscal 2025 EBITDA margin of 27.6%. The FY2026 figure overstates by roughly 40%.
6. est.epsAvg VALIDATES — and carries the valuation. The FY2025 row shows $3.70265 against a reported $3.73, a 0.7% difference. Coverage is also unusually even — 12 to 16 analysts on every line from FY2026 to FY2029 — the most consistent estimate block in this batch. Only the FY2030 EPS row (5 analysts) is excluded.
7. Trailing GAAP EPS EXCEEDS trailing adjusted EPS — noted rather than corrected. Trailing GAAP diluted is approximately $4.31 against a trailing adjusted total of $4.13 from earn_cal. For most companies the relationship is reversed. The cause is favourable items inside GAAP — notably low quarterly effective tax rates of 13.2% (Q2 2025) and 13.5% (Q4 2025) — that the company adjusts out. We use adjusted EPS for forward multiples to match the estimate basis and show both trailing figures.
8. returnOnEquityTTM is null in ratios_ttm against 33.12% in km_ttm — the same two-block disagreement seen on LOW, NOW, SPGI, VRTX and SBUX in this batch. We use the km_ttm figure. returnOnInvestedCapitalTTM of 17.4% is quoted alongside and is the more robust measure given $4,479M of goodwill and intangibles (34.7% of assets).
Not defects, correctly reported and independently confirmed: km_ttm.enterpriseValueTTM of $117,562M implies net debt of $2,251M, which reconciles to the 31 March 2026 balance sheet's cash of $2,435M against approximately $4,686M of total debt — the vendor EV is correct and is accepted; inc_q March-quarter revenue of $2,313M matches the earn_cal actual exactly; capexToDepreciationTTM of 2.25x is consistent with the 10-Q's statement that capital expenditure will "remain elevated"; interestCoverageRatioTTM of 15.4x is consistent with the debt and income figures; dividendPayoutRatioTTM of 10.7% reconciles to $0.48 on trailing GAAP EPS; profile.currency and inc_a[0].reportedCurrency are both USD — no currency mixing; and earn_cal uses a consistent adjusted basis across all five reported rows, so the five-beat series is like-for-like.
quote.yearHigh/yearLow versus tech.hi52/lo52: quote $295.28 / $169.45; tech $289.26 / $171.00. Intraday versus close. We use tech throughout. Note that tech.pct_from_hi and tech.max_dd_from_peak are BOTH −0.37% — identical — which means the 252-day high is also the six-year peak and the stock is at an all-time closing high. That is the legitimate coincidence, not a defect, and it is the most important single technical fact in this dive.
Non-equity tripwire — checked and passed. HWM is common stock, NYSE-listed. The $288.20 price is not par-like; beta is 1.189; the dividend is small (0.17%) and variable rather than fixed; volume was 1.94M shares (~$559M of turnover); and the 52-week band of $171.00 to $289.26 is a 69% range. This is common equity. Note that bal_a shows $55M of preferred stock in FY2023 and FY2024 and $0 in FY2025, consistent with a small legacy preferred issue having been retired — immaterial to the equity analysis and recorded for completeness.
7. Technicals
- Price $288.20, up 0.65%. −0.37% from the 52-week high of $289.26, which is also the six-year peak:
pct_from_hiandmax_dd_from_peakare both −0.37%. There is no drawdown. +68.5% above the 52-week low of $171.00. Position within the annual range: 99.0th percentile. - Moving averages, both far below and both rising: +6.6% above a 50-day of $270.40; +21.5% above a 200-day of $237.24.
- RSI 57.6 — firm, not overbought. MACD +4.02 — the most positive absolute reading in this batch.
- Relative performance, uniformly strong: 3-month +20.2% vs SPY +7.6% and QQQ +7.7%; 6-month +39.1% vs SPY +11.1% and QQQ +15.7%; 12-month +56.4% vs SPY +24.3% and QQQ +30.8%. A 32.2-point outperformance of the S&P over twelve months — the largest in this batch.
- Sentiment: 21 buy, 3 hold, 1 sell across 25 analysts. Consensus $299.18 (+3.8%), median $300 (+4.1%), high $340 (+18.0%), low $228 (−20.9%).
What the technical picture does to the entry. This is a stock doing everything right and priced accordingly. RSI at 57.6 with MACD at +4.02 says the trend is intact and not yet extended; the 50-day and 200-day averages are 6.6% and 21.5% below. What it does not offer is any margin for error: at the 99th percentile of its annual range, with zero drawdown from the multi-year peak, the first technical support is 6.6% down and the second is 21.5% down. Combined with a consensus target 3.8% above and our base 0.4% below, and with earnings in two days, the honest position is that the chart is excellent and the entry is not.
8. Insiders
The insider block for Howmet contains no transactions that support inference in the records available. Reported as an absence of data rather than as a signal.
Corporate activity is the substitute: buybacks of $755 million in fiscal 2025, up from $500 million in fiscal 2024, $250 million in fiscal 2023 and $400 million in fiscal 2022, taking diluted shares from 421 million to 403 million — a 4.3% reduction over four years, roughly 1% a year. At $288.20 the fiscal 2025 programme is 0.65% of market capitalisation, and combined with a 0.17% dividend yield gives total shareholder yield of approximately 0.8% — the lowest in this batch. A company generating $1,431 million of free cash flow and returning $936 million of it is retaining the rest for the capacity expansion the 10-Q describes, which is the correct choice and which is why the yield is what it is.
9. Verdict, kill-criteria and flip conditions
Watch.
What is genuinely exceptional — and this is the best operating record in the batch: Engine Products third-party sales up 29% with segment adjusted EBITDA up 44% and the margin up 400 basis points to 36.6%, achieved while absorbing 235 net new heads; company revenue up 19.1% and operating margin at 32.6% against 25.2%; 1,650 basis points of operating-margin expansion in six years, accelerating; five consecutive earnings beats of 4.4% to 10.8%; free cash flow compounding 38.4% a year since fiscal 2022 with the margin widening from 9.5% to 17.3%; 0.88x net leverage, 15.4x interest coverage, and cash up from $742M to $2,435M in one quarter; a 17.4% return on invested capital; and treasury execution that swapped a 6.750% dollar liability into a 3.88% yen one for a stated $12 million of annual saving.
What we are declining to pay for: 69.8x trailing adjusted earnings, 47.5x the 2027 consensus, 13.6x sales, 46.2x enterprise value to EBITDA, and 80.4x free cash flow — a 1.24% free-cash-flow yield, less than half the next-lowest in this batch; a stock 0.37% below its all-time high with literally no drawdown after +56.4% in twelve months; a street consensus 3.8% above spot — the tightest here — with its lowest target 20.9% below; a base fair value of $287 that is 0.4% BELOW the close; a 0.8% total shareholder yield; a completely empty knowledge-base lane; and a print in two days.
The distinction that matters. There is nothing wrong with Howmet Aerospace, and this dive should not be read as if there were. It is, on the evidence, the highest-quality operating performance in these twelve names and the aftermarket economics the knowledge base describes are genuinely durable. The objection is entirely that the security has already been priced for it — our number, the crowd's number and the current price all sit within 4% of one another, which is the definition of a market that has finished discounting the available information. When there is no edge, no cushion, no drawdown and no conviction lane, and results land in 48 hours, Watch is the answer by construction.
Pre-registered KILL criteria — what would take this to Avoid:
- Engine Products segment adjusted EBITDA margin below 34%, which would mark 36.6% as a cycle peak and remove the basis for the entire multiple.
- Any disclosed reduction in airframer build rates or engine-programme schedules, which is the exogenous driver Howmet does not control.
- Revenue below approximately $2,350M in the second quarter or a reduced full-year outlook, breaking a five-quarter run of sequential growth and beats.
- Capital expenditure rising further without a corresponding revenue step, which would push the free-cash-flow yield below 1% and make the 80x price-to-cash multiple indefensible.
- Tariff costs proving unrecoverable — the 10-Q names tariffs four separate times as a two-sided uncertainty for both cost and recovery.
- A break below the 200-day average of $237.24, which given the absence of any drawdown would represent an 18% fall from here.
Pre-registered FLIP conditions — what would take this to Buy — Tactical:
- Price below approximately $230 (32.0x FY2028E, 37.9x FY2027E) with no deterioration in Engine Products margin. At $230 the base case is +24.8% against a bear case only 23% lower. That is 20.2% below today's close and roughly at the street's low target.
- Engine Products segment adjusted EBITDA margin above 38% on 2026-08-06, which would establish 36.6% as a waypoint rather than a peak and justify the multiple on quality rather than on cycle.
- Explicit disclosure of engine-spares revenue or aftermarket mix, which would allow the market to value the annuity separately from the build-rate cyclical — the single most valuable disclosure this company could make, and the one the knowledge-base sector cluster suggests is worth the most.
- Capital expenditure guidance normalising toward 1.2x depreciation, which would roughly double the free-cash-flow yield without any change to earnings.
- Any name-level knowledge-base conviction appearing. At present the lane is completely empty; a single credible independent claim about Howmet specifically would be worth revisiting.
Where HWM fits in the Synthos Framework Portfolio. The aerospace-supply-chain sleeve, at 0% today with a 1.5-2% target on a fill near $230 or on Engine Products margin clearing 38%. Batch overlap note, and it is material: Parker-Hannifin (PH) is in this batch, appears in Howmet's own vendor peer set, and Howmet appears in Parker's — and both are levered to the same commercial-aerospace build-and-aftermarket cycle. They are not the same trade: Parker is 33% aerospace with a two-thirds industrial business and trades at 29.1x forward; Howmet is essentially pure aerospace at 47.5x. Holding both doubles the build-rate exposure at very different prices for it. On the specific question of which expresses the aerospace cycle better, Howmet is the higher-quality and far more expensive answer, and Parker the more diversified and cheaper one — and both report within two days of this dive. Logged as a tracked Synthos call (Watch) as of 2026-08-04 at $288.20.
Single biggest risk: the 1.24% free-cash-flow yield, and what it implies if the cycle turns. Howmet generated $1,431 million of free cash flow in fiscal 2025 against a market capitalisation of $115.31 billion — a 1.24% yield and an 80.4x multiple, the most expensive on that measure in this batch by a factor of more than two. The reason is defensible: capital expenditure at 2.25x depreciation is capacity expansion into demand the company can see, at a segment margin of 36.6%, and it is the right use of the money. But it means the entire investment case is the forward earnings stream and none of it is current cash. If commercial build rates flatten — a decision made by two airframers, not by Howmet — the company will own newly built capacity, a decelerating revenue line, a peak segment margin to defend, and a share price at 47.5x forward earnings with no drawdown whatsoever to absorb the adjustment. The bear case of $178 is not a forecast; it is what a 28x multiple on the consensus low looks like, and it is 38% below today's price.
Provenance & disclosures
- Traceability: ZERO tagged knowledge-base claims name Howmet Aerospace (entity sweep on HWM, Howmet and Arconic — the former name — across all 51,928 distilled claims returned nothing; name-level conviction: none). This is an empty lane and is reported as empty. A supplementary sector sweep on Howmet, aerospace supply chain, jet engine, aircraft build rate, narrowbody, GE Aerospace, TransDigm and aftermarket returned 12 hits, none naming Howmet, of which six are used as labelled sector context and six are discarded. The usable set is unusually mechanistic: Romesh Naranaswami (business_breakdowns, 2026-02-01, named speaker, conviction 60 and 62) on engine makers selling original equipment at break-even and earning ~60% gross margins on aftermarket worth 3-5x original-equipment value over an engine's life, and on GE Aerospace's ~$175B backlog covering six to seven years of commercial services; an unattributed business_breakdowns claim (2024-03-31, conviction 80) that aftermarket parts sell to airlines at 3-4x the price charged to Boeing in a ~$100B market; a business_breakdowns principle claim (2026-02-01, conviction 70) that jet engines create barriers 'as hard as semiconductor fabrication' with only three or four primes; Colin (invest_like_the_best, 2026-05-29, named speaker, conviction 55) on TransDigm's sole-source pricing power — TransDigm sits in Howmet's own vendor peer set; and a chinatalk claim (2019-05-01, neutral 62) on COMAC's state-backed C919 targeting the narrowbody duopoly, the only structural bear note available and seven years old. Six discards, including a genuine homograph: "aftermarket" meaning post-IPO share-price performance in a biotech_hangout claim about the financing market, plus three non-aerospace industrial aftermarket principles (Ametek, Graco, Diploma), a duplicate TransDigm claim, and a Trane HVAC service-mix claim. No claim in any set carries a management
speaker_role, so the half-weight management discount does not apply. No concentration test was required at the name level because the lane is empty; within the sector set business_breakdowns supplies four of six, which is disclosed. All quotes are verbatim from the stored claim text. - Data as-of: most recent filing-verified financials 2026-03-31, from the 10-Q filed 2026-05-07 (balance sheet, segment third-party sales and segment adjusted EBITDA by division, the segment reorganisation note, and the acquisition references) · annual statements fiscal 2025 ended 2025-12-31 from
inc_a/bal_a/cf_a, corroborated by the 10-K filed 2026-02-12 as recast by the 8-K of 2026-05-28 · estimates 2026-08-04 · prices 2026-08-04, quote timestamp 1785873603 = 2026-08-04T20:00:03Z ($288.20, +0.65%; 50-DMA $270.40; 200-DMA $237.24; RSI 57.6; MACD +4.02; beta 1.189) · knowledge-base claims 2026-08-04. Fiscal note: Howmet's fiscal year is the calendar year. The second quarter of 2026 has NOT yet been reported — it is released 2026-08-06, two days after this dive. All figures come from the Synthos vendor data file for HWM or from the SEC filings in the HWM archive; no figure comes from memory, recall or external retrieval. - Filing archive contents: 10-K filed 2026-02-12 (fiscal 2025); 10-Q filed 2025-10-30 (September 2025 quarter); 10-Q filed 2026-05-07 (March 2026 quarter, the source of all segment detail used here); 8-K filed 2026-05-07 (Q1 2026 results, 41,129 characters — the exhibit was extracted); 8-K filed 2026-05-26 (Item 5.07, annual meeting voting results); 8-K filed 2026-05-28 (Item 8.01, 70,576 characters — a RECAST of the fiscal 2025 Annual Report's Item 7 and Item 8 for the segment reorganisation, plus disclosure of the JPY term-loan prepayment and cross-currency swap; the most consequential 8-K in this dive and one a routine scan would dismiss).
- Where the filings contradicted or corrected the vendor (detailed in Section 6):
seg_prodrejected in full — it omits Forged Wheels from FY2024 onward and both Forged Wheels and Engineered Structures from FY2025, summing $2,179M (26.4%) short of reported fiscal 2025 revenue, while all four segments are described in the 10-Q; the 8-K of 2026-05-28 recast the fiscal 2025 segment presentation (a titanium alloy location moved from Engine Products to Engineered Structures) and no vendor field reflects it, so pre- and post-recast segment figures are not comparable; two acquisitions — Brunner from February 2026 and the CAM operations from the second quarter of 2026 — are inside the Fastening Systems numbers with no disclosed price or contribution, so the segment's 14% March-quarter growth contains an unquantified acquired component; the JPY term-loan prepayment of ¥29,702M (~$187M) on 2026-05-22 and the cross-currency swap of $300M of 6.750% 2028 Notes into a ~3.88% Yen liability, saving a stated $12M a year, appear in no vendor field;seg_geonoted as failing to close by $1,025M (12.4%) and used directionally only;est.ebitdaAvgandest.ebitAvgrejected as fixed-ratio fabrications at exactly 38.80% and 37.25% of revenue in every year against an actual fiscal 2025 EBITDA margin of 27.6%; trailing GAAP EPS noted as EXCEEDING trailing adjusted EPS, the reverse of the usual relationship;returnOnEquityTTMagain null inratios_ttmagainst 33.12% inkm_ttm;quote.yearHigh/yearLowrejected in favour oftech.hi52/lo52. Where vendor and filing AGREED — worth recording:km_ttm.enterpriseValueTTMimplies net debt of $2,251M, which reconciles to the 31 March 2026 balance sheet's $2,435M of cash against approximately $4,686M of total debt — the vendor EV is correct and is accepted without adjustment, which is rare in this batch;inc_qMarch-quarter revenue matches theearn_calactual exactly;est.epsAvgfor FY2025 ($3.70265) is within 0.7% of the reported $3.73; the United States home market is present inseg_geoat 53.0%. - Basis note: analyst estimates and
earn_calactuals are adjusted (non-GAAP); reported diluted EPS is GAAP. Unusually, trailing GAAP diluted EPS of approximately $4.31 EXCEEDS the trailing adjusted total of $4.13 — the reverse of the normal relationship — because GAAP contains favourable items, notably quarterly effective tax rates of 13.2% and 13.5%, that the company adjusts out. Theearn_calbasis is consistent across all five reported rows, so the five-beat series (+10.8%, +4.4%, +4.4%, +8.8%, +9.9%) is like-for-like. Forward multiples use consensus adjusted EPS; the trailing multiple is shown on both bases (69.8x adjusted, 66.9x GAAP diluted). - Estimate coverage: 12 to 16 analysts on every revenue and EPS line from FY2026 through FY2029 — the most consistently covered estimate block in this batch, with no thin year until FY2030 (15 analysts on revenue but only 5 on EPS, which is excluded from every conclusion). The fair-value anchors rest on the FY2028 row (12 analysts on EPS), cross-checked against FY2027 (13).
- Peer note: the vendor peer set is a genuine industrials cohort — Cummins, Emerson, Illinois Tool Works, Johnson Controls, 3M, Northrop Grumman, TransDigm, UPS, Waste Management and Parker-Hannifin, which is also in this batch and shares the commercial-aerospace driver. TransDigm is the most instructive comparison and is the subject of two 2026 knowledge-base claims on sole-source aerospace pricing power. No peer multiples are supplied in the file, so no peer-multiple comparison is drawn.
- Fair-value caveat: the $178 / $287 / $385 anchors are multiples of the FY2028 consensus adjusted-EPS distribution — 28.0x the low of $6.348, 40.0x the mean of $7.184, and 44.8x the high of $8.598 — cross-checked against FY2027E of $6.062. FY2028 is used because it is the first year in which the current capacity expansion should be substantially converting to revenue. Stated arithmetic, not a discounted cash flow. The base is sensitivity-disclosed: 36x gives $259 and 44x gives $316, and the current price sits in the middle of that band. The base assumes multiple compression from 47.5x to roughly 40x, and Section 4b states plainly that this is driven by the consensus EPS growth path decelerating from +36.8% to +13.8%, not by any view on the business.
- Timing: second-quarter 2026 results are released 2026-08-06, TWO DAYS after this dive. Consensus is adjusted EPS $1.24 and revenue $2,427.9M, and the print will be the first to include the CAM operations and the second to include Brunner. The verdict is set with that timing as an explicit input, and the dive states plainly that entering 48 hours ahead of a print, at an all-time closing high with zero drawdown and a base fair value below the current price, is not an entry with an edge in it. 2026-08-04 was a modestly positive day for the stock (+0.65%); no company-specific news for that date appears in this file.
- 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.