Price $217.50 · Base $83.59 · Price is +160% above base
What they do
Howmet makes the hardest-to-make metal parts inside jet engines and airframes. Its biggest business casts the single-crystal turbine blades and vanes that sit in the hottest part of an engine, plus seamless rolled rings. It sells them to the three big engine makers (GE/CFM, Pratt & Whitney/RTX, Rolls-Royce) and to industrial gas-turbine makers for power plants. Howmet says it holds more than half the global market for gas-turbine blades.
Segments (Q2'26 revenue, segment adj. EBITDA margin): Engine Products $1.37B (54%, 37.7%) · Fastening Systems $0.59B (23%, 30.1%; includes CAM, bought from Stanley Black & Decker for ~$1.8B in April 2026) · Engineered Structures $0.27B (11%, 23.8%; titanium forgings and extrusions) · Forged Wheels $0.32B (12%, 27.8%; aluminum wheels for heavy trucks). By end market: commercial aerospace about half, then defense, gas turbines and commercial transportation.
It makes money three ways. It sells parts for new engines, often at thin initial margins. It sells replacement blades and parts for the installed fleet ("spares"), which carry much higher prices. And it uses multi-year contracts with price escalators. Key sensitivities: aircraft build rates (Boeing/Airbus), shop visits on the LEAP/GTF fleets, gas-turbine orders driven by AI and data-center power demand, nickel/titanium pass-through, and whether customers like GE bring castings in-house.
Price (10/15 intraday)
$217.50
Base intrinsic value
$83.59
Bear → Blue-sky · prob-wtd
$41 – $131 · $84
Key driver
Discount rate: price implies a 5.6% WACC
The Take — Best-in-class franchise, priced as if capital were free. The stock implies a 5.6% cost of capital.
The business is as good as the bulls say. Q2'26 revenue rose 24% (21% organic) to $2.55B. Adjusted EBITDA margin hit 32.1%, up 340bp. Engine Products grew 32% at a 37.7% segment margin, and gas-turbine revenue grew 38%. FY26 guidance was raised to $10.05B revenue, $3.23B adjusted EBITDA, $5.27 EPS and ~$1.9B free cash flow. Clean EBIT margin (SBC expensed; restructuring and acquired amortization added back) climbed from 17.6% in FY21 to 26.2% in FY25 and is tracking ~29% in FY26 (estimate from guidance). After-tax ROIC is about 24% on book capital that includes ~$4B of goodwill. Note: the record margin is 32% on EBITDA, not 28–30%. On EBIT it is ~29%, and that is the figure this model uses.
A generous base case is still worth only $83.59. Revenue grows 11/10/9/8/7% to $15.5B by FY31, 54% above the FY26 guide. Margin holds at ~29% for two years, then fades only to 26%, which is still 5pt above the 5-year normalized 20.7%. Terminal ROIC is 14% and WACC is 9.5% gliding to 9.2%. Holding 29% forever adds just $9 (to $93). The bull case (1.2× growth, 29% margin, 16% terminal ROIC) is $104. Blue-sky (1.4× growth, 32% margin, 20% ROIC) is $131. Every cell of the margin × growth grid sits below the price.
Reverse DCF: the price is a bet on the discount rate, not the business. At the base growth path, $217.50 needs a 65% steady-state EBIT margin. At a 26% margin it needs 3.2× the base growth path, which means roughly 35% growth in year one. Holding the operating case fixed, it needs a 5.6% WACC, an equity risk premium of about 0.4% over 5.2% Treasuries. On EV of ~$91B the FY27 FCFF yield is ~2.3%. The stock trades at 41× FY26 guided EPS.
The market has already started to test the moat. HWM is down 30% from its $310 high on Aug 6. The late-August leg came after Elon Musk said SpaceX plans to make turbine blades in-house. The Sept 8 leg (−10.7%) came after GE Aerospace, one of Howmet's largest customers, agreed to buy rival caster CPP for $11.75B. Neither changes the next two years of earnings. Both go straight at the one assumption the price cannot do without: that a 30%-margin, 50%-share position in castings is permanent. Machine-tool lead times protect Howmet until about 2028. The terminal value runs far past that.
Call: AVOID new money. Accumulate only below ~$130 (blue-sky value); a starter is defensible at ≤$160 (blue-sky at a 100bp lower WACC). Holders: this is a quality compounder bought at a price that already assumes the bull case. Trim to a core position and fund the trim into strength. Covered calls at the $240–250 strikes, 30–45 DTE after the print, or a zero-cost collar, get paid for the volatility. Q3 print: Oct 29, before the open. Management has promised a first look at 2027 revenue around then, and a strong guide can gap the stock 8–12%. If you want short exposure, use a defined-risk bear call spread above $250–260 after the print. Do not short it naked: buybacks ($800M year to date through July), momentum and a 22-Buy Street make squeezes likely. No prior valuation on file.
Risk Profile — MEDIUM-HIGH
HIGH
Valuation / discount rate (long falling real rates). The price implies a ~5.6% WACC; ±100bp of WACC moves base value about ±$15. Terminal value is 54% of base value. Multiple compression to a 25–30× aerospace multiple is a 27–39% drawdown even if every earnings target is met.
HIGH
Customer insourcing / pricing power (short the moat). GE is buying CPP ($11.75B, closing 2H27) and SpaceX plans in-house blades. Revenue is concentrated in GE/CFM, RTX/Pratt and Rolls-Royce, who negotiate long-term agreements with real leverage.
MED
Aero cycle / spares normalization (long air traffic and shop visits). In 2020 revenue fell 26% and clean margin dropped to 16%. The current spares boom (LEAP/GTF durability fixes, old fleets flying longer) is a cycle peak, not a new floor.
MED
Gas turbines / AI power demand (long). IGT revenue grew 38% in Q2 on data-center power orders. It is a real tailwind, but it is the newest and most momentum-dependent part of the story.
LOW
Balance sheet. Debt rose from $3.05B to $4.50B with the CAM deal. Net debt is ~$3.9B, about 1.2× FY26 EBITDA. The investment-grade rating is intact and the pension deficit is ~$0.6B.
LOW
Truck wheels (long the Class 8 cycle). 12% of revenue; Q2 volumes were down 8% Y/Y. Cyclical, but small.
The real bet: you are not buying castings capacity. You are buying the assumption that 29%+ margins and 10%+ growth last for a decade and that investors keep accepting a sub-6% expected return on a cyclical aerospace supplier. The tornado shows the discount rate as the widest bar, ahead of margin and growth. The 5-yr normalized margin (20.7%) gives about $63/share.
Scenario values per share vs price
Bear: aero downturn, with FY28 revenue −10% and FY29 −4%, and margin reverting to the 21% normalized level; 12% terminal ROIC, g 2.5%. Base: 11→3.5% growth, 29% margin fading to 26%, 14% ROIC. Bull: 1.2× growth path, 29% held, 16% ROIC. Blue-sky: 1.4× growth (FY31 ~$18.2B), 32% margin, 20% ROIC. Prob-weighted 20/50/20/10 = $83.87.
Driver tornado: Δ value/share from base $83.59
No single swing gets within $115 of the price. Margin ±3pt applies to the steady-state margin (23% / 29%).
Green = above price $217.50; yellow = within ±5%; orange = up to 30% below; red = more than 30% below. Every cell is red. Even 1.6× the base growth path (FY31 revenue ~$19.7B, almost double the FY26 guide) at a 32% margin forever is worth $136. Normalized margin (FY21–25 average) = 20.7%; FY26E ≈ 29%.
Sensitivity: value/share by WACC (high-growth → stable) × terminal growth
WACC ↓ / g →
2.0%
2.5%
3.0%
3.5%
5.5% → 5.2%
$197
$219
$251
$301
6.5% → 6.2%
$147
$157
$170
$188
7.5% → 7.2%
$116
$121
$128
$136
8.5% → 8.2%
$95
$98
$101
$106
9.5% → 9.2%
$80
$82
$84
$86
10.5% → 10.2%
$69
$70
$71
$72
Base row is 9.5% → 9.2%. The price only appears around a 5.5% WACC, barely above the 5.2% risk-free rate.
FY26E (faded) = guidance midpoint revenue $10.05B; EBIT is an estimate (adjusted EBITDA $3.23B less ~$320M D&A). Margin line plotted on a 14–30% scale.
Clean EBIT = GAAP operating income + "restructuring and other charges" + acquired-intangible amortization; SBC stays expensed. *TTM restructuring includes a $93M gain on the Savannah disk-forging sale (H1'26), which is removed, plus $22M of CAM acquisition costs added back; amortization ~$34M is an estimate. FY20 includes Arconic-separation costs. Base-case choice: this is a cyclical end market (FY20 −26% revenue), so the base does not capitalize the 29% peak forever. It fades to 26%, which credits structural pricing and mix gains (+5pt over normalized) but not the full spares-cycle peak.
Base year FY26E = guidance ($10.05B revenue). Reinvestment = Δrevenue ÷ sales-to-capital 2.5 (FY25 net capex $170M plus working capital on $822M of growth implies ~4; 2.5 allows for the announced casting-capacity build). Tax is 20% (FY25 effective 18%), gliding to 23%. Terminal growth of 3% is the cap for a secular aero-traffic grower; TV share is 54%, under the 70% flag.
− Unfunded pension + OPEB ($546 + $42 at 12/31/25), after 21% tax
($465)
− Noncontrolling interest
$0
Equity value
$33,521
÷ Diluted shares (398.8M basic + ~2M equity awards; Q2 diluted avg 402M before $500M of buybacks)
401.0M
Value per share
$83.59
Street: 22 Buy / 4 Hold / 0 Sell, mean target $327 (range $255–377)
Base is −74% vs mean target
Sources: SEC XBRL via Robinhood (10-Ks filed 2/2022, 2/2024, 2/2026; 10-Q filed 8/6/2026); Howmet Q2 2026 earnings release (8/6/26) for segment data, guidance, buybacks, CAM/Savannah; TIKR blog (9/2026) on the GE–CPP deal and Sept 8 move; Trefis (9/9/26) on the SpaceX blade comment; Robinhood quote, fundamentals, analyst ratings and earnings calendar (10/15/26). Street FY27+ revenue consensus was not reliably available (sources found were stale), so growth paths are my estimates. FY26E EBIT and TTM amortization are estimates. Q3 2026 earnings Oct 29, before the open. Educational analysis, not investment advice.