Century Aluminum

Phase A — Understand the business

Lens 1 · Company Overview

Century Aluminum is a global producer of primary aluminum — it runs three smelters (Grundartangi, Iceland; Sebree, Kentucky; Mt. Holly, South Carolina) with ~770,000 tpy of nameplate capacity, and produced ~638,000 tonnes in 2025. It is vertically integrating upstream: a 55% JV interest in the Jamalco bauxite mine + alumina refinery in Jamaica (~1.4Mt alumina capacity, ~1.0Mt produced 2025; Government of Jamaica holds the other 45% via CAP), plus a wholly-owned carbon-anode plant at Vlissingen, Netherlands that supplies 93–98% of Grundartangi's anode needs.

The economic engine is brutally simple and almost fully exogenous. Selling price = (i) LME base + (ii) regional premium (Midwest Premium "MWP" for US metal; European Duty Paid Premium "EDPP" for Iceland metal) + (iii) value-added product premium (billet, foundry alloy). Cost of goods is dominated by alumina + electrical power + carbon + labor, which together are >84% of COGS. Because the company sells at published market prices, it cannot pass through cost inflation — margin is the spread between an exogenous price and a largely exogenous cost. One operating segment, one reportable segment (commodity producer). NASDAQ: CENX; Delaware corp, HQ Chicago; 79 holders of record; no dividend.

Customer structure — the defining feature. ~54.0% of FY2025 consolidated net sales went to Glencore and affiliates, who also beneficially own ~36.4% of the stock (per the 10-K; trimmed to ~30.0% by April 2026 — see Lens 9). Glencore buys both US (LME + MWP) and Icelandic (LME + EDPP) metal and also buys Jamalco alumina, and Century has agreements to sell "a substantial portion" of 2026 production to Glencore. The rest of the book is "fairly concentrated among a small number of customers under short-term contracts." This is a related-party customer that is simultaneously the largest shareholder — a structural governance fact, not a footnote.

Contract structure. Alumina: Glencore 500,000 tpy through Dec-2028 (LME-linked); Concord Resources ~540,000 tpy through Dec-2029 (fixed + LME + API-linked). Power: Grundartangi ~70% LME-indexed (a natural hedge) + ~30% fixed/LME through 2026–2036; Sebree market-based (MISO/Kenergy) through May-2028 — the volatile exposure; Mt. Holly Santee Cooper cost-of-service through Dec-2031. Several raw-material contracts carry take-or-pay obligations that bite during curtailments.

Lens 2 · Supply Chain

Upstream → Century → end customer, named at every node:

• Bauxite → mined by Jamalco (Century 55% / Gov. of Jamaica–CAP 45%) in Jamaica; ~84% metallurgical recovery; bauxite transfer price ~$14.43/t.
• Alumina → refined by Jamalco (~1.0Mt 2025; ~31%/~320kt of it consumed internally) + purchased externally from Glencore (500ktpy) and Concord Resources (~540ktpy). Alumina index spiked to $805/t (Dec-2024) on the Guinea export ban vs $343/t avg 2023 — the key swing cost.
• Carbon anodes → Vlissingen (Century, Netherlands) supplies 93–98% of Grundartangi; US smelters self-produce anodes on-site. Coke/pitch/cathodes are single-/limited-source bought-in with no internal production — a named chokepoint.
• Power → Landsvirkjun / OR / HS Orka (Iceland hydro+geothermal), Kenergy/MISO (Sebree, market), Santee Cooper (Mt. Holly, cost-of-service). Power is a chokepoint that has repeatedly forced curtailment: Hawesville fully idled Q3-2022 on MISO power-price spikes; Icelandic hydro curtailments in dry/cold winters 2021/2022/2024.
• Smelting → Grundartangi (320kt cap), Sebree (220kt), Mt. Holly (230kt). Single-point fragility is real: in Sept–Oct 2025 two transformers failed at Grundartangi over a seven-week window, idling the larger potline and cutting that smelter's output ~two-thirds; restart of Potline 2 began 23-Apr-2026, full by end-July.
• Downstream → Glencore (54%) + a few short-term customers; US metal captures MWP, Iceland metal captures EDPP.

Chokepoints, ranked: (1) single-source power per site + market exposure at Sebree; (2) high-voltage equipment (the Grundartangi transformer failure proves a partial HV fault can halve a smelter); (3) bought-in coke/pitch/cathodes; (4) alumina index (Guinea/Guinea-style supply shocks); (5) Jamaica weather (Hurricane Beryl 2024, Hurricane Melissa delayed Jamalco's return to full production).

Lens 3 · Competitive Advantages (moats)

For a price-taking commodity producer, "moat" = cost position + tariff-protected geography + duty-free market access, not brand. Century's real, durable edges:

1. Inside-the-wall location. The US and EU are the #2 and #3 aluminum-consuming regions but are structurally short domestic primary metal. Century's US smelters capture the Midwest Premium and its Icelandic smelter accesses the EU duty-free (Iceland is in the EEA) while non-EEA competitors pay EU tariffs, and US imports pay a 50% Section 232 tariff. This is the single biggest moat and it is policy-granted (see Lens 8/10).
2. Low-carbon Icelandic metal (Natur-Al™, ASI-certified, hydro/geothermal) — one of the industry's lowest carbon footprints, hedging future carbon regulation and commanding green premiums in Europe.
3. Partial vertical integration (Jamalco alumina + Vlissingen anodes) — controls its two largest input categories more than a pure smelter would, and ~70% of Grundartangi power is LME-indexed (a natural margin hedge).

Bargaining power is weak, and asymmetric. Century needs Glencore (54% of sales + 30–36% of the stock + an alumina supplier) more than Glencore needs Century. Inputs like coke/pitch/cathodes are limited-source. The honest read: Century's moat is geographic + regulatory, not commercial — it has little pricing power over either customers or suppliers, and its largest counterparty sits on both sides of the table and the board. Many competitors (Alcoa, RUSAL, the Chinese majors) are larger, vertically integrated, and have superior cost positions.

Lens 4 · Segments

The company reports one segment (primary aluminum, commodity), so there is no product-line EBITDA split in the filing — the meaningful disaggregation is geography (US vs Iceland) and price-vs-volume. segments.csv is header-only, so all figures below are from the filing tables:

The trend, decoded: volume is decelerating (FY25 total shipments −4.5% YoY; Q1-26 total −27% YoY) — Iceland volume collapsed (Q1-26 29kt vs 74kt Q1-25) because of the Grundartangi transformer failure. Yet US revenue is accelerating violently on price (Q1-26 US revenue $494.3M on 93.7kt vs $306.6M on 94.6kt a year earlier — i.e., ~flat tonnes, +61% revenue). The entire revenue story is price (MWP), not volume, and it is concentrated in the US footprint. Iceland is currently a drag (down on volume) but is the green-premium / EDPP optionality once Potline 2 is back.

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print = Q1 2026, reported 7-May-2026)

Income statement, Q1-2026 vs Q1-2025 (As Restated):

• Beat on adjusted, miss on GAAP optics. Adjusted EPS $1.63 vs $1.56 consensus (beat); revenue $649.2M vs $633.85M (beat). But the GAAP headline was noisy: the $287.9M Hawesville gain + $33.0M insurance gain inflated GAAP EPS to $3.23, and at least one tape flagged a GAAP "miss" of $1.84. Adjusted EBITDA was $231.4M, +$60.8M sequentially.
• The driver is unambiguous: regional premium. MWP averaged $2,294/t in Q1-2026 vs $729/t in Q1-2025 (+215%) and LME $3,191 vs $2,631. Gross profit doubled YoY on +$193.7M favorable metal-price realization, partly offset by −$47.2M unfavorable volume/mix (Grundartangi down), −$41.1M raw-material, and −$39.5M higher US power (extreme winter). This is the thesis in one line: a ~flat-volume quarter still doubled gross profit because the tariff lifted the price the company sells into.
• Guidance stepped up hard. Management guided Q2-2026 Adjusted EBITDA $315–335M — embedding higher realized premiums + incremental Mt. Holly/Grundartangi tons not yet at run-rate. Annualized off the Q2 midpoint (~$325M × 4) that is ~$1.3B EBITDA run-rate.
• Balance-sheet flags — net all positive this quarter. Cash jumped to $244.1M + restricted $89.5M (Hawesville proceeds, $88.1M ring-fenced for capex); liquidity $611.0M; total debt ~$545.9M ($66.1M current + $479.8M LT) → net debt ~$302M, or ~near-zero counting restricted cash. Century equity rose to $1,150.9M (the gain shrank accumulated deficit to −$1,288.1M). The one red flag: derivative liabilities ballooned to $104.0M (from $58.2M at YE25 and $4.4M a year before) — the hedge book is a growing negative mark against the rising MWP (see Lens 10/13).
• Market reaction: sold the news. Despite the adjusted beat and the EBITDA step-up, the stock fell ~5.9% after-hours to ~$58.91. That tells you a record premium and a strong guide were already priced after a huge run — the bar is now high (CENX +~150%-class move over the prior year, mirroring Kaiser's +152% ).

FY2025 full-year anchor for context: net sales $2,527.9M, gross profit $256.4M (~10.1% margin), operating income $158.1M, net income to Century $41.8M, diluted EPS $0.42, operating cash flow $185.0M, capex $100.2M → FCF ~+$84.8M. FY2024 was distorted by a one-time $245.9M Jamalco bargain-purchase gain (EPS $3.27 diluted, almost entirely non-operating).

Lens 6 · Earnings Calls (sentiment trend)

Transcripts dir is empty in the research layer, so this is ``-grounded. Management tone has shifted from defensive/survival (2022–23: curtailments, energy crisis, Hawesville idle) to aggressively expansionary and overtly political (2025–26). On the Q1-2026 call management framed it as "one of the most dynamic markets for aluminum in recent memory," citing demand from lightweighting, electrification, power/data-center build-out, commercial aviation, and defense/rearmament. The recurring, almost unmistakable phrases now: "Section 232," "America First," "no exemptions and no exceptions," "first new US smelter in ~50 years." Century has gone further than peers — it publicly issued a statement "proudly stand[ing] with President Trump" on the IEEPA ruling. What they've stopped saying: the 2022–23 language of curtailment economics and balance-sheet survival. The risk in this tone shift: the bull case is now explicitly staked on a single politician's trade policy persisting — management has made the political dependency a feature, which is also the bear's single point of failure.

Lens 7 · Comps

Peer table — multiples are `` with source/date or n/a. The research-layer _index.json lists no other critical-materials peers, so peers are pulled from the obvious primary-aluminum set.

Read: On trailing optics CENX looks expensive (P/E n/m, EV/EBITDA mid-teens-plus) because 2023–25 earnings were trough/restructuring. On the forward Q2-run-rate (~$1.3B EBITDA), EV/EBITDA collapses to ~5.5x — below Alcoa (~9–10x) and the metal-fabrication group's ~6.8x forward. That is the entire quantitative bull case: if the MWP holds, CENX is the cheapest, highest-beta primary-aluminum equity in the comp set. The catch is the denominator — at a record premium, EBITDA is at cycle-peak, so a "cheap" forward multiple on peak earnings is exactly what a commodity top looks like.

Lens 8 · Stock-Price Catalysts (last ~5 years)

What actually moves CENX, ranked by what the tape rewards/punishes:

1. Section 232 tariff actions (the dominant lever). 10%→25%→50% on primary aluminum (25% eff. Mar-12-2025, 50% eff. Jun-4-2025, no exemptions) drove the MWP from ~$427/t (2024 avg) to ~$1,295/t (2025 avg) to a record ~$2,529/t by early May-2026. Each escalation is a step-change up in the stock.
2. Midwest Premium prints — the MWP is now >40% of US all-in cost (~$5,340/t all-in vs ~$3,300 LME). CENX trades as a levered MWP call.
3. The Feb-2026 Supreme Court IEEPA ruling — struck IEEPA tariffs but explicitly left Section 232 intact; Century's stock and the thesis survived a genuine legal kill-shot. A de-risking event, not a catalyst down.
4. Operational shocks — Hawesville full curtailment (Q3-2022, energy crisis) was a major down-leg; the Grundartangi transformer failure (Oct-2025) cut a smelter by two-thirds; restarts (Mt. Holly to 100%, Grundartangi Potline 2) are up-catalysts.
5. Strategic / capital events — DOE $500M award (Jan-2025), the EGA JV new-smelter announcement (Jan-26-2026), the Hawesville sale to TeraWulf ($200M + 6.8% data-center stake, Feb-2-2026), and Glencore's 6.3M-share block sale at $51.75 (Mar-4-2026) — the last gapped the stock down on supply.
6. Earnings — note Q1-2026 fell ~6% on a beat — the pattern is "buy the tariff, sell the print" once premium is in the number.

Pattern verdict: the market reacts to US trade policy and the MWP far more than to Century's own operations. This is a macro/policy instrument with an equity wrapper.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO: Jesse Gary — President & CEO since 1-Jul-2021 (~5y tenure); joined Century 2010, prior EVP/COO/General Counsel; before that a lawyer at Wachtell, Lipton, Rosen & Katz. A lawyer-operator, not a metallurgist — fitting for a company whose core skill is now trade-policy navigation, JV structuring, and capital markets rather than smelting innovation.
• Track record: landed the $500M DOE award (Jan-2025), structured the EGA JV (first new US smelter in ~50 years), monetized the dead Hawesville asset for $200M cash + a 6.8% data-center stake (a genuinely clever AI-adjacent value-unlock of a stranded site), drove Sebree to a 5-year production high, and completed the Iceland low-carbon casthouse. Real, quantified value creation in the macro window.
• Skin in the game: comp ~$7.17M (FY), 83% at-risk, LTI tilted 60% PSUs on relative TSR — well-structured alignment. But: Gary directly owns only ~0.14% of shares, and sold 150,000 shares for ~$7.2M on 23-Jan-2026 — ~21% of his holdings, his largest-ever sale. SVP Strategy Matt Aboud also sold. Flag: the CEO cut a fifth of his stake three days before the EGA JV announcement (Jan-26) — insiders are taking chips off at the top of the tariff-driven run.
• Capital allocation: disciplined balance-sheet repair in 2025 — refinanced into $400M 6.875% 2032 Notes, redeemed the 2028 Notes, paid off the Grundartangi casthouse facility, no buybacks despite $43.7M authorization, no dividend, reinvesting Hawesville proceeds into Mt. Holly + the new smelter. ROE/ROIC is still n/m (accumulated deficit −$1.29B) — the franchise is only now earning its cost of capital, and only because of the tariff.
• Archetype: professional manager / dealmaker, not founder. Implication: excellent at harvesting the policy window and structuring JVs; the open question is whether this team can execute a multi-billion greenfield smelter (the EGA project) — a capability the company has not demonstrated in 45 years.

Lens 10 · Forensic Red Flags

Ground in financials + filings; every figure labeled.

• THE headline: a restatement + an unremediated material weakness + a Deloitte ADVERSE ICFR opinion. Century filed a Comprehensive 10-K restating FY2024, FY2023, and all three 2025 quarters because it had used proportionate (not full) consolidation for Jamalco's net assets. The restatement did not change net income attributable to Century — it was a consolidation/geography error, not an earnings fraud. But Deloitte issued an ADVERSE opinion on internal control over financial reporting for a material weakness in Jamalco business-process + consolidation controls that remains unremediated at 31-Dec-2025. Crucial nuance for the bull: Deloitte's opinion on the financial statements themselves is UNQUALIFIED (clean) — the numbers are audited-clean; the control environment producing them is not. The IT-general-controls weakness was remediated in 2025; the business-process/consolidation one was not. This is a real but bounded governance risk (Jamalco-specific, no income impact), and the single biggest reason to size any position conservatively.
• Earnings quality is muddied by two large non-cash / one-off items. (1) The Section 45X advanced-manufacturing credit is booked as a receivable + an offset to COGS — manufacturing credit receivable jumped to $172.6M (YE25) / $197.5M+ across current+LT (Q1-26 ~$173.3M + $24.9M) from $81.5M. This flatters gross profit with a credit that the One Big Beautiful Bill Act now phases out (−25%/yr from 2031, zero by 2034) — a known cliff. (2) FY2024's $245.9M Jamalco bargain-purchase gain and Q1-2026's $287.9M Hawesville gain + $33.0M insurance gain make GAAP net income a poor guide to run-rate earnings — use adjusted EBITDA.
• The hedge book is bleeding against the thesis. Derivative liabilities ran $4.4M → $58.2M → $104.0M (Q1-25 → YE25 → Q1-26); the income statement took −$94.7M (FY25) and −$65.3M (Q1-26) of net derivative losses, with a $62.8M unrealized derivative loss add-back in FY25 cash flow. Translation: Century hedged MWP/LME and the premium then exploded, so the hedges are deep underwater — a self-inflicted cap on upside and a potential margin-call liquidity draw in a further spike (flagged as a risk factor).
• Customer + related-party concentration: 54% of sales to Glencore, which is also a ~30–36% owner and an alumina supplier; FY25 COGS includes $294.0M of purchases from this related party. Any number of related-party transactions clear through a counterparty that controls the cap table.
• Going-concern / leverage: NOT a going-concern risk — liquidity $611.0M, net debt ~$302M, in compliance with all covenants, springing FCCR only at <$25M availability. Convertible Notes ($86.3M due 2028) + capped calls add dilution/accounting complexity (if-converted diluted share count). Valuation allowance against US + Jamaican + part of Icelandic DTAs is the critical audit matter (Deloitte) — management judges US DTAs more-likely-than-not NOT realizable, i.e., it does not yet trust its own US taxable-income durability.

Regulatory findings (required sub-section).

• SEC Litigation Releases / AAERs: None. Verified via SEC EDGAR EFTS (LR + AAER), period 2021-06-18→2026-06-18, total SEC findings = 0.
• Item 3 (Legal Proceedings), FY2025 10-K: management states ordinary-course litigation only, "neither individually nor in the aggregate... likely to result in a material adverse effect," pointing to Note 17.
• Non-SEC / civil enforcement (web): (1) a securities class action (Rule 10b-5), Lead Plaintiff Howard M. Rensin, Levi & Korsinsky lead counsel (appointed Jul-2023), with a Stipulation of Settlement dated 25-Apr-2025 — i.e., settled, not an open existential suit. (2) A ~$944K class-action settlement + a six-figure civil fine over alumina-dust emissions at Mt. Holly (Sept-2023 incidents, settled ~Mar-2025). Both are immaterial to a $6.7B-cap company.
• Net: No SEC enforcement; two settled civil matters (one securities, one environmental), both small. The governance issue that matters is the ICFR material weakness, not litigation.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (EPS, next three fiscal years: FY2026 / FY2027 / FY2028)

Built bottom-up from the latest actuals + guidance; every input labeled; outputs ``. The swing variable is the MWP, so the scenarios fan on the premium, not on volume.

Anchors: Q1-26 adj EBITDA $231.4M; Q2-26 guide $315–335M; FY shares ~99M basic / ~105M diluted; D&A ~$92M, interest ~$45M, ~near-zero cash tax (valuation allowance / NOLs). Full-year capacity ramping toward ~770kt as Mt. Holly hits 230kt and Grundartangi returns by end-July.

• FY2026 base. H1 ≈ Q1 $231.4M + Q2 ~$325M = ~$556M; H2 at full run-rate but assume MWP normalizes off the record ~$2,529 toward ~$1,800–2,000 → H2 ~$450–500M. FY26 adj EBITDA ~$1.0–1.05B. Less D&A ~$92M, interest ~$45M, minimal tax, less NCI → adj net income ~$0.75–0.85B → adj diluted EPS ~$7.20–8.10. (GAAP EPS far higher due to the $287.9M Hawesville gain — ignore for run-rate.)
• FY2027 base. MWP mean-reverts to a tariff-supported but off-peak ~$1,400–1,700/t; full-year full-capacity volume (~770kt) partly offsets lower premium; 45X credit still intact. FY27 adj EBITDA ~$700–850M → adj diluted EPS ~$4.50–6.00.
• FY2028 base. MWP ~$1,200–1,500/t (durable wall, no peak); steady-state. FY28 adj EBITDA ~$650–800M → adj diluted EPS ~$4.00–5.50.

All ``; arithmetic: EBITDA scenario − ~$92M D&A − ~$45M interest − minimal tax − NCI, ÷ ~105M diluted shares. Sensitivity is enormous — a 20–30% growth/premium disappointment roughly halves FY27 EPS, exactly the commodity-peak fragility. (No forecast.ts create — breadth/watchlist loop.)

Lens 12 · Bull vs Bear

Bull case. Century is the purest, most-levered equity expression of US aluminum protectionism, and the protection just survived its biggest legal test. The Section 232 50% wall is intact post-IEEPA-ruling; the MWP is at a record ~$2,529/t (>40% of all-in US price); demand has a genuine secular leg (electrification, grid/data-center build-out, defense/rearmament, lightweighting). On the Q2 run-rate the stock is ~5.5x forward EV/EBITDA — below Alcoa (~9–10x) and the fab group (~6.8x). The balance sheet was repaired (net debt ~$302M, liquidity $611M), Hawesville was monetized cleverly into $200M cash + AI-data-center optionality, and there are three free call options: Mt. Holly to 100%, Grundartangi back to full, and the EGA/Inola greenfield (40% of a 750kt smelter, $500M DOE-backed) that would roughly double US primary capacity. If the wall holds, earnings power is a multiple of the trough the multiple is anchored on.

Bear case (permanent-impairment risks). (1) The thesis is one policy decision wide. Earnings are explicitly staked on a 50% tariff and a record premium that a future administration — or a trade deal, or a Canada/Mexico carve-out — could compress overnight; management has made the political dependency the headline. A return of the MWP to ~$800 takes FY27 EPS toward ~$2.25. (2) Cycle-peak optics masquerading as cheap. ~5.5x forward is on peak-of-cycle EBITDA at a record premium — the textbook commodity value trap. (3) Self-inflicted hedge drag + the 45X cliff — a $104M underwater derivative book caps the upside the bull is paying for, and the 45X COGS credit (flattering gross profit) phases to zero by 2034. Pre-mortem (18 months out, thesis broke): the tariff was trimmed or a strong-dollar/soft-LME quarter hit just as the MWP mean-reverted; Grundartangi's restart slipped or a second HV failure recurred; the EGA power negotiation with PSO stalled and the greenfield was deferred/impaired; the hedge book forced a margin-call liquidity draw — and the stock, having sold off on a beat at the peak, de-rated as "peak earnings × peak multiple." Are multiples too high? Not on forward EBITDA — but the forward EBITDA is the risk, not the multiple. Contrarian view the market is missing: the bull and bear are the same fact — this is not an "aluminum company," it's a leveraged, policy-contingent claim on the Midwest Premium; treat it as a thematic trade with a hard stop, not a compounder.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case:

• Revenue concentration is a loaded gun. 54% of sales sit with Glencore — which also owns ~30% of the stock and just sold a 6.3M-share block at $51.75 (Mar-2026). Your largest customer/owner is distributing into the rally — the best-informed insider on both the cap table and the metal flow is reducing exposure. What do they see that the multiple doesn't?
• The moat is rented, not owned. Strip out Section 232 and the MWP and Century is a sub-scale, high-cost smelter with weak bargaining power, single-source power per site, bought-in coke/pitch/cathodes, and a negative accumulated-deficit equity story (−$1.29B). The "moat" is a Presidential proclamation.
• The accounting tells you management doesn't trust its own durability: it keeps a valuation allowance against US deferred tax assets — i.e., it does not believe US taxable income is more-likely-than-not durable. And it carries an unremediated material weakness with a Deloitte adverse ICFR opinion — the controls that produce these blow-out numbers failed an audit.
• Most dangerous competitor bulls underestimate: not RUSAL or the Chinese majors — it's secondary/recycled aluminum + Canadian primary metal flooding in around the tariff, plus domestic restarts (including Century's own) that erode the very premium the thesis capitalizes. Add more US supply and you shrink the MWP.
• Hedge book + margin calls: a $104M derivative liability that grows as the thesis works is a structural short-seller gift — it caps upside and threatens liquidity on a further spike.
• What must hold for today's ~$68 price: MWP stays near record, the tariff persists through at least one more administration's posture, Grundartangi restarts cleanly with no recurrence, and the greenfield doesn't become a cash sink. Break any one and a −30% growth/premium miss roughly halves FY27 EPS.
• Single scenario that permanently impairs: a negotiated rollback/exemption of the aluminum Section 232 tariff (trade deal, court, or successor administration) collapses the MWP back toward global parity — Century reverts to a marginal-cost producer carrying greenfield capex commitments. Plausibility: low-to-moderate over 12–24 months given current politics, but it is the entire risk, and it is binary.

Lens 14 · Management Questions (ordered by information value)

1. What is your explicit base-case assumption for the Midwest Premium in your Q2 and full-year EBITDA guidance, and at what MWP level does the EGA/Inola greenfield stop clearing its cost of capital?
2. Quantify the mark-to-market and cash-collateral exposure of the derivative book at quarter-end MWP/LME — at what premium level do margin calls become a material liquidity draw?
3. What is the specific, dated remediation plan for the Jamalco consolidation material weakness, and when do you expect Deloitte to reverse the adverse ICFR opinion?
4. Why did the CEO sell ~21% of his holdings three days before the EGA announcement, and what is the board's view on insider selling into the tariff-driven rally?
5. What is Glencore's intent with its remaining ~30% stake and its 2028 alumina supply agreement — and is the 6.3M-share block sale the start of a continued distribution?
6. Walk through the EGA JV capital plan: Century's total cash commitment, the PSO power-price threshold required for FID, and the conditions under which you'd defer or exit.
7. How do domestic restarts (yours and competitors') + Canadian/recycled inflows affect your own MWP assumption — are you cannibalizing the premium you're capitalizing?
8. What replaces the Section 45X credit's contribution to gross profit as it phases out (2031–2034)?
9. What is the probability and timeline of returning Grundartangi to 100%, and what hardening prevents a recurrence of the transformer failure?
10. Under what LME/MWP path do you reverse the US deferred-tax valuation allowance — i.e., when will you certify US earnings durability?
11. What is your through-cycle capital-allocation priority once Hawesville proceeds are spent — buybacks (the $43.7M authorization), debt reduction, or growth?
12. How concentrated is the non-Glencore book, and what is the renewal risk on those short-term contracts in 2027?
13. What is the all-in delivered cost per tonne at each smelter today vs the global cost curve, ex-tariff?
14. How exposed is Jamalco to Jamaican weather + the CAP joint-venture governance rights, and what is the normalized alumina self-sufficiency target?
15. What scenario would cause you to curtail again, and what is the trigger LME/power level by site?

---