Aerospace casting scarcity → qualified capacity is allocated to margin, not volume → LTA and metal pass-through fail to route the rent to the buyer → Howmet captures it

Flight-critical casting capacity is qualification-gated with relief dated 2028-2030, so its owner rations it by margin rather than by volume; Howmet's Q2 FY2026 shows 340bps of consolidated EBITDA-margin expansion against LTA-covered business, and — the load-bearing test, added 2026-08-12 — that expansion is concentrated in the *constrained* segment (Engine Products +470bps) rather than spread across the portfolio (Forged Wheels +30bps), which is what an allocation rent looks like and what a rising tide does not.

• Step 3: A Howmet disclosure that decomposes the margin gain and shows it is predominantly aftermarket MIX and operating leverage rather than allocation or price - which would make this operating leverage on a volume recovery, not a scarcity rent.
• Step 4: Howmet EBITDA margin compresses in a quarter where revenue still grows - indicating LTA resets are routing the rent back to the buyer as lta-contract-structure-as-price-insulation predicts.
• Step 2: A second qualified casting supplier reports aero, defense and IGT as separately-served capacity with no cross-market allocation - which would confine Plant's Virginia remark to one facility rather than a structural rule.
• Step 1: Casting or isothermal-forging lead times compress below 12 months before 2028, or a capacity add lands materially earlier than the stated August-2028 floor. (2026-08-12: Howmet says "we already see the need to increase [capex] further in 2027 to support future organic growth expectations" with NO quantified capacity add and no date - relief is being funded but not yet scheduled, which is mildly supportive of the rent persisting past 2027. From 2026-08-12-autoresearch-howmet-q2-margin-decomposition-price-mix-leverage.)
• Step 5: HWM guides FY2027 revenue or EBITDA margin BELOW FY2026 - management stated 2027 revenue will exceed 2026, so a reversal would break the multi-year framing directly.

• Spares grew faster than the company (+37% vs +24%) and now sit at ~22% of first-half revenue. Aftermarket carries higher margin, so an unknown share of the 340bps is MIX, not price or allocation. Howmet disclosed no realized price per unit.
• 21% organic growth would expand margin through operating leverage alone at constant prices. Price, mix and leverage remain unseparated.
• The Q2 source is no longer `partial`. 2026-08-31-howmet-next-checkpoints-into-q3 retrieved the Motley Fool Q2 FY2026 transcript (2026-07-31). Q&A did **not** split price / mix / operating leverage of the +340 bps. The 10-Q says price > pass-through only (directional). Step 4 stays **partial**.
• Spares +37% and ~22% of H1 sit in the Q2 8-K. The $560M vs $460M (sic) first-half-spares figure from the 08-11 ingest is **unresolved** against the 8-K. Do not carry $2.75B Q3 revenue — the spoken figure is a transcription error; written Q3 guide is **$2.575B / $830M / $1.35**. From 2026-08-31-howmet-next-checkpoints-into-q3.
• Carpenter Technology, the purest MELT-tier exposure, reports a FALLING average selling price per pound over a comparable window and declined the shared-capacity claim. That is why this mechanism is written at the casting/forging tier and not the melt tier.
• Plant narrows his own contention quote: 'the majority of our manufacturing equipment ... does not know whether it's building parts for a narrow body aircraft or a wide body aircraft' - fungibility WITHIN aero variants, which is not the cross-market claim.
• The tape ran against the chain in the week it was filed: HONA rebounded +7.7% on 08-07 while HWM fell -2.7%, the opposite of the transfer this family of chains predicts.
• **⚠ Self-correction logged 2026-08-12.** This page's original step-4 arithmetic grossed a segment-level pass-through figure up to a consolidated one, producing a "~700bps underlying" claim that overstated the case *in the direction of the favored narrative* - the second such error in three days (see CALIBRATION #38). The correction is recorded in step 4 rather than silently edited away. Notably, the corrected reading **still supports the mechanism**, via segment distribution rather than gross-up - which is the useful lesson: the wrong argument and the right conclusion are separable, and the process only catches that if the arithmetic is re-derived from the primary rather than from the prior day's summary.
• **The decomposition is now bounded by segment, but price is still not isolated.** Neither the Howmet release nor the available coverage separates realized price from mix or operating leverage. Engine Products' +470bps is itself undecomposed, and gas turbines - the fastest-growing end market at +38% - is plausibly the richest mix, so some of the concentration could be mix-within-segment rather than scarcity rent.
• **Engineered Structures shrank 13% in revenue and still expanded margin 170bps**, via "divestiture of the Savannah disk forging facility and product rationalization." That is a *portfolio* effect and should not be counted as evidence for the scarcity story.

• HWM is the tradeable expression. The rent is at the qualified casting/forging tier, one step downstream of the merchant melt tier where shared-heavy-forging-capacity-to-specialty-alloy-melt-rent originally placed it.
• ATI is the secondary expression: it states the allocation rule most explicitly and holds the tightest lead time in the chain (isothermal forgings >24 months).
• This is the supplier-side mirror of precision-casting-scarcity-to-aftermarket-margin-transfer. The same allocation rule that expands Howmet's margin is what compresses Honeywell Aerospace's - one mechanism, two signs, now observed from both ends in the same week.
• Relief is dated: new commitments deliver from August 2028, IGT capacity adds run 2028-2030, CRS brownfield completes early FY2028. The rent has a roughly two-year runway before capacity lands.
• Because this is an allocation rent rather than a price rent, it should compress FASTER than a price rent once capacity arrives - allocation power evaporates the moment buyers stop being rationed.