Autoresearch: shared melt / forging capacity — the ATI / CRS kill test

Generated by /autoresearch on 2026-08-11. Two rounds (early exit — round 3 would not have materially changed the synthesis; the remaining uncertainty is an IGT-specific disclosure that neither company made). No Grokipedia anchor attempted — this is a live-financials question, not an encyclopedic one. Treat as raw material. Context: vault/projects/stock-market

Summary

Yesterday's shared-heavy-forging-capacity-to-specialty-alloy-melt-rent hypothesis was filed with a deliberate kill test: do commercial aero, IGT and munitions actually compete for the same qualified lines, or only for the same materials category? — and with an explicit flag that ATI and CRS appeared nowhere in this wiki. Both companies have now reported, and the answer is mixed in a way that refines rather than confirms the chain:

• Confirmed at the forging / casting tier. ATI's lead times are ~12 months for nickel alloys, ~20 months for premium-quality titanium, and over 24 months for isothermal forgings — the forging tier is the tightest of the three, exactly as an input-contention story predicts. Independently, Howmet's John Plant states on the 2026-08-06 call that missile programs are "actually competing for the space in our Virginia facility." Two unrelated companies, same tier, same contention.
• Not confirmed at the merchant melt tier. Carpenter's CEO would only concede that space could "potentially compete with aero engine alloys," and immediately discounted it on tonnage. Neither company disclosed an allocation methodology across end markets.
• The price test fails on its own terms — and this is the finding that matters. The hypothesis demanded realized price, not utilisation. Carpenter's CFO states that average selling price per pound declined, attributing it to mix, and adds that "lower-priced products often come with comparable average profit margins." So margin is expanding (SAO hit a record 37.8% adjusted operating margin) while price per pound falls.

The reconciliation is the useful part: the rent is being captured through deliberate allocation of scarce qualified capacity toward higher-margin mix, not through a headline price increase. ATI's Kimberly Fields says so almost explicitly — ATI is "being very deliberate around how it allocates its resources and capacity, with impacts more heavily weighted towards the margin and EBITDA line more so than the revenue line." That is an allocation-rent mechanism, and it is the same mechanism precision-casting-scarcity-to-aftermarket-margin-transfer describes from the buyer's side.

Findings

The forging tier is tighter than the melt tier

ATI's disclosed lead times rank the constraint unambiguously: ~12 months nickel alloys, ~20 months premium-quality titanium, >24 months isothermal forgings (ATI Q2 2026 call coverage). Isothermal forging is the most capital- and qualification-intensive step in the chain and it is the one with the longest queue.

Howmet's CEO independently places the contention at the same tier — missile programs "actually competing for the space in our Virginia facility" — while noting the limits of that contention: "the majority of our manufacturing equipment ... does not know whether it's building parts for a narrow body aircraft or a wide body aircraft" (HWM Q2 FY2026 call). Read carefully, that second quote is a narrowing: equipment is fungible within aero variants, which is not the same as fungible across aero / IGT / munitions. The Virginia quote is the one that speaks to cross-market contention; the narrow/wide-body quote does not.

Merchant melt declines the claim

Carpenter's CEO Tony Thene frames demand, not shared capacity, as the story: "Engine customers indicate demand is less of a concern than the capacity needed to meet that demand" (CRS Q4 FY2026 transcript). Asked about cross-market contention he conceded only that space could "potentially compete with aero engine alloys," adding that space tonnage "is very low compared to engines." No management statement was found at either company asserting that aero, IGT and munitions contend for the same melt slots.

The rent is real, but it is an allocation rent, not a price rent

The margin evidence is strong and consistent across both merchant houses:

• Carpenter: Q4 FY26 operating income $206.9M, +37% y/y; SAO operating income $229.7M, +38%, at a record 37.8% adjusted operating margin; SAO sales ex-surcharge $607.4M (+11% y/y). Adjusted EPS $3.23 vs $3.07 consensus. FY27 operating income guided $850–880M (+21–25%); FY2029 target $1.2–1.3B (CRS Q4 FY26).
• ATI: Q2 2026 sales $1,261.1M, net income $151.0M, diluted EPS $1.09; gross margin 24.6% vs 21.3% a year earlier, attributed to "better pricing and mix." Confirmed order backlog $4.4B vs $3.7B (+18% y/y, +7% sequential) (ATI Q2 2026).

But the price leg specifically fails at Carpenter. CFO Tim Lain: "The decline in average selling price per pound is due to higher proportion of lower-priced products in the mix," with the qualifier that "lower-priced products often come with comparable average profit margins." Thene's pricing claim is forward-looking, not realized: "Price is going to be a big driver for us" (for FY27).

The mechanism that does have direct management support is allocation. ATI's Kimberly Fields: ATI is "being very deliberate around how it allocates its resources and capacity, with impacts more heavily weighted towards the margin and EBITDA line more so than the revenue line as the company continues to pivot its mix." And Thene describes buyers bidding for allocation rather than haggling on price: "Every contract we're working on, they ask for more material ... whatever percent of their needs, they want that percentage to be higher with each contract."

Defense is the fastest-growing claimant on that capacity

ATI's defense revenue hit an all-time high, +36% y/y, with defense sales within the A&D segment up 90%, driven by naval nuclear, missile and missile-defense applications; A&D segment revenue overall rose 34%. A new five-year naval nuclear contract worth roughly $1 billion — about double the prior contract — was disclosed. Carpenter's A&D sales rose a more modest 17%, against a Boeing/Airbus backlog of roughly 16,000 aircraft.

The asymmetry is itself informative: the munitions/naval claimant is growing several times faster than the aero claimant at the company with the tighter forging position.

Relief is dated, and it is late

Carpenter's brownfield expansion is "scheduled to be completed by the start of fiscal year 2028," with the CFO confirming it is "on budget and on schedule for early fiscal year 2028 completion." ATI is investing in its nickel melt system and adding vacuum induction melting capacity. Howmet's own framing is the sharpest: a brand-new commitment made in August 2026 delivers "earliest August of 2028," with IGT capacity additions running 2028–2030. Every relief date found sits in 2028–2030, consistent with the hypothesis's original dating (Ecotitanium 2028, Safran Gennevilliers 2029).

Both are merchant, not captive

The hypothesis asked whether the capacity is "listed rather than captive." Both ATI and Carpenter are independent NYSE-listed merchant suppliers selling into the primes and engine OEMs rather than captive units, and both disclose confirmed order backlogs ($4.4B at ATI) and LTA negotiations with external customers — which is what makes the rent, if it exists, accrue to a public equity. That sub-question resolves cleanly in the hypothesis's favour.

Contradictions and open questions

• The central contradiction: ATI attributes margin gains partly to "better pricing" while Carpenter reports a falling average selling price per pound over a comparable window. Both expanded margins. Either the two occupy different positions on the qualification ladder, or "price" and "mix" are being used inconsistently between the two management teams. Not resolvable from public disclosure alone.
• No IGT-specific disclosure from either merchant house. Howmet is the only company in this research that quantified IGT (revenue +38%, >50% global share of IGT blades). Whether ATI's or Carpenter's melt is being pulled by power-turbine demand is unanswered — this is the single remaining gap, and it is the leg that would make the chain a genuinely four-demand-shock story rather than an aero-plus-defense story.
• Carpenter's silence is weak evidence, not counter-evidence. Management not volunteering cross-market contention on an earnings call is not the same as denying it. But the hypothesis set a kill bar, and silence does not clear a kill bar.
• Does the allocation rent survive the 2028 capacity? Every relief date is 2028–2030. If the rent is allocation-driven rather than price-driven, it should compress faster than a price rent once capacity lands, because allocation power evaporates the moment buyers stop being rationed.

Provenance

Rounds run: 2 of 3 (early exit — the one remaining open question, IGT pull on merchant melt, was not disclosed by either company, so a third round would have re-searched the same absent disclosure).

Sub-questions by round:

Round 1 (broad survey):

1. What did ATI report for Q2 2026 on melt capacity, lead times, pricing, and defense demand?
2. What did Carpenter Technology report for Q4 FY2026 on backlog, pricing, and aerospace/defense?

Round 2 (drill-down):

1. Does management at either company state that aero / defense / IGT compete for the same constrained capacity? — targeting the kill test directly.
2. Is margin expansion attributable to realized price or to mix/volume? — targeting the hypothesis's explicit "price, not utilisation" graduation bar.

URLs fetched (2 successful, 0 failed; 2 further sources via search-result extraction):

Round 1:

• ATI Q2 2026 8-K / results — company filing — Q2 financials, backlog, defense growth.
• Carpenter Q4 FY26 slides coverage — news — record SAO margin, FY27/FY29 guidance.

Round 2:

• Carpenter Technology (CRS) Q4 2026 earnings call transcript — primary transcript — Thene and Lain quotes on capacity, ASP, brownfield timing.
• ATI Q2 2026 earnings call transcript — primary transcript — lead times by product, Fields on capacity allocation, naval nuclear contract.

Cross-referenced from today's earnings vein (not a web fetch):

• 2026-08-06-earnings-hwm-q2-fy2026 — Plant's Virginia missile-contention quote and the August-2028 capacity-lead-time quote.

Tools used: WebSearch, WebFetch. Generated: 2026-08-11.