Cleveland-Cliffs

Phase A — Understand the business

Lens 1 · Company Overview

Cleveland-Cliffs is a fully integrated North American steel enterprise — vertically integrated from iron-ore mining and pellet/DRI production, through ferrous-scrap processing, into primary steelmaking and downstream finishing (stamping, tooling, tubing). ~25,000 employees across the US and Canada, headquartered in Cleveland, Ohio.

It is the largest flat-rolled steel producer in the US and the dominant supplier to the domestic automotive industry. The business sells into four end markets, FY2025 revenue mix:

• Direct automotive — 30% ($5,047M; -9% YoY) — cold-rolled, galvanized, aluminized, NOES, stainless
• Infrastructure & manufacturing — 29% ($5,377M; +3%) — hot-rolled, plate, GOES, rail
• Distributors & converters — 28% ($5,195M; -2%) — all grades
• Steel producers — 13% ($2,334M) — slab, scrap, iron ore, HBI, coal, coke

Contract structure: ~35–40% of flat-rolled shipments are under fixed-price contracts (typically annual or multi-year), some with surcharge mechanisms that pass through input-cost changes; the remaining ~60–65% is spot/index-linked. This is the key cyclicality lever — the spot-exposed majority swings hard with HRC pricing, while the contract book lags by a year (management explicitly cited "price realization lags" on the Q1'26 call ).

Reportable segments: Steelmaking (essentially the whole company — $17,953M FY25 revenue) and Other Businesses ($53M Adj EBITDA, stable).

Lens 2 · Supply Chain

CLF is unusual among US steelmakers in being its own upstream — it owns the iron-ore mines (Michigan UP, Minnesota Iron Range), pelletizing, and HBI/DRI capacity, then runs the metal through blast-furnace / basic-oxygen integrated mills (9 references to "blast furnace" in the 10-K vs 1 to "electric arc furnace" ). This matters: CLF is a blast-furnace integrated producer, not a scrap-fed EAF mini-mill like Nucor or Steel Dynamics — higher fixed cost, more carbon-intensive, less flexible to flex down in a downturn.

Map, with named stakeholders:

• Upstream inputs: own iron ore (self-supplied) → metallurgical coal/coke (partly self-supplied via the acquired ArcelorMittal USA coke assets; SunCoke Middletown is a consolidated VIE supplying coke ) → ferrous scrap (Ferrous Processing & Trading, acquired 2021) → natural gas and electricity (the single biggest commodity exposure: a 10% natural-gas price move = ~$52M, vs ~$20M for HRC and ~$10M for electricity ).
• The company: integrated mills across Ohio, Indiana (Burns Harbor, Indiana Harbor — ex-ArcelorMittal), Michigan, Pennsylvania, plus the 2024 Stelco assets in Ontario, Canada.
• End customers: the Detroit Three automakers are the anchor (auto = 30% of revenue) — GM, Ford, Stellantis are the structural buyers, though not named individually in the 10-K; plus service-center distributors and infrastructure/manufacturing OEMs.

Chokepoints / single-source dependencies: (1) Automotive concentration — a US auto-production downturn hits CLF harder than any peer. (2) Energy — natural gas is the dominant input-cost swing factor; the Q1'26 print took an $80M one-time energy hit from extreme cold weather. (3) Blast-furnace inflexibility — CLF idled six facilities (Mar–May 2025) rather than flex production, because integrated mills can't ramp down cheaply.

Lens 3 · Competitive Advantages (moats)

Real moats:

1. Scale + integration in flat-rolled — CLF is the #1 flat-rolled and #1 automotive-steel supplier in North America, and the only domestic producer of certain grades (it is "a leading producer of electrical steels in the U.S." — GOES/NOES for grid transformers, a structurally tight market ). Switching costs in auto are high: qualified steel for exposed body panels takes years to re-source.
2. Domestic, trade-compliant footprint — in a 50% Section-232-tariff world this is the moat. CLF's entire production is US/Canada, USMCA-compliant, exactly the supply the tariffs are designed to protect. The POSCO MoU (Lens 8) explicitly seeks to "leverage Cleveland-Cliffs' unmatched U.S. footprint and trade-compliant operations."
3. Vertical integration — self-supplied iron ore insulates against seaborne ore-price spikes (a relative advantage vs EAF mills that buy scrap on the open market when scrap is tight).

Where the moat is thin:

• Cost-curve position is poor. Blast-furnace integrated steel is higher-cost than scrap-EAF. In FY2025 CLF's gross margin was -5% while Nucor and Steel Dynamics stayed profitable through the same cycle (Lens 7). The moat protects grade and customer access, not unit cost — and unit cost is what kills you at the trough.
• Bargaining power is asymmetric the wrong way. Against the Detroit Three (sophisticated, concentrated buyers) CLF is the price-taker on contract renewals; against energy suppliers it has little leverage. Its pricing power comes from tariffs, i.e. government policy, not from the franchise itself.

Lens 4 · Segments

CLF reports essentially one operating segment (Steelmaking). FY2025 Steelmaking revenue $17,953M vs $18,529M FY2024 (-3%).

By product (shipments, kt):

The tell: shipments rose +4% but revenue fell -3% — a pure price/mix problem. ASP/ton fell from $1,081 → $1,005. The mix also degraded toward lower-value hot-rolled (+16%) and away from cold-rolled (-6%).

Segment Adjusted EBITDA — the decisive number:

The core steel business swung to negative Adjusted EBITDA (-$16M) in 2025 from +$715M in 2024. The only thing keeping consolidated Adj EBITDA marginally positive ($37M) was the small "Other Businesses" line. This is a business that, at the 2025 trough, made no money making steel.

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print: Q1 2026, filed 2026-04-21)

The latest 10-Q shows a clear, price-led inflection off the FY2025 bottom.

vs consensus: Revenue $4.92B beat the ~$4.81B consensus; adjusted loss -$0.40 beat the -$0.42 estimate. Adjusted EBITDA $95M, but inclusive of an $80M one-time energy cost from extreme cold — so a clean run-rate closer to ~$175M.

What drove it: entirely price, not volume. Steelmaking ASP/ton rose $980 → $1,048 (+7%); shipments were roughly flat (-1%). Steelmaking gross margin improved -9% → -2%, and segment Adjusted EBITDA % flipped -4% → +2%. Distributors/converters revenue +19% YoY signals channel restocking.

Guidance/tone: CEO Goncalves guided to sequential quarterly improvement and "healthy positive free cash flow" in Q2 2026 — explicitly framing Q1 as the last ugly quarter.

Balance-sheet flags (the catch): despite the P&L improvement, operating cash flow was still negative -$325M in Q1'26 (vs -$351M Q1'25), dragged by a -$441M accounts-receivable build (seasonal restock). CLF funded the gap by drawing +$507M on the ABL revolver. Free cash flow was ~-$477M. So: earnings improving, cash still bleeding, leaning on the revolver.

Market reaction: despite the double beat, CLF fell ~6% on the print — the market focused on the still-negative cash flow and the energy-cost spike, not the loss-narrowing. Wall Street cut price targets afterward.

Full-year FY2025 context:

In FY2025 COGS ($19,470M) exceeded revenue ($18,610M) — the company lost money at the gross line, before SG&A and interest. That is the trough signature.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on disk (transcripts=0); sentiment is reconstructed from `` call coverage across Q4'24 → Q1'26.

• Tone arc: Goncalves is congenitally promotional — through 2025's worsening losses the messaging stayed defiantly bullish on tariffs ("will support a healthy domestic steel industry for years to come" ). What shifted across 2025 is the addition of a self-help narrative: cost cuts, plant idling, asset sales, the POSCO MoU, and rare-earth optionality — management pivoting from "wait for the cycle" to "we are actively restructuring."
• What they started saying (2025–26): "$300M/year cost savings," "$425M of asset-sale proceeds," "data-center developers," "rare earths," "highly accretive POSCO partnership," "positive free cash flow in Q2."
• What they stopped saying: the acquisition-growth story. After Stelco (Nov 2024) the language flipped from acquire-and-scale to deleverage-and-rationalize. The 2024 buyback talk gave way to dilutive equity issuance.
• Sequential Adj EBITDA trail (the tape behind the tone): Q3'25 $143M → Q1'26 $95M (with the $80M one-timer; ~$175M clean). Lumpy but off the floor.

Read: management credibility is the swing variable. Goncalves has delivered the scale story (Lens 9) but is talking his book hard on the turn — discount the rhetoric, trust the ASP and the segment-EBITDA inflection, which are real.

Lens 7 · Comps

CLF vs the US flat-rolled / steel peer set. Multiples are ``, dated; where I could not source a clean figure I mark n/a rather than fabricate.

Industry benchmark: the US steel industry trades at ~8.0x trailing EV/EBITDA, below the S&P 500's ~15.6x and the materials sector's ~11.5x; the 5-year range is 2.8x–12.9x, median 8.7x.

Read: CLF is the distressed laggard of its peer group. While Nucor (FY26E EPS ~$15) and Steel Dynamics (Q1'26 net income $403M) are solidly profitable through the same cycle, CLF is loss-making with negative/near-zero LTM EBITDA — so the standard EV/EBITDA and P/E screens are not meaningful for it right now. CLF is valued on normalized/mid-cycle earnings power and the option on the turn, not on trailing multiples. On EV (≈ $7.2B equity + ~$7.3B net debt ≈ ~$14.5B EV ) against a mid-cycle Adj EBITDA in the ~$1.5–1.9B range (2023 was $1,893M), the implied ~7.6–9.7x mid-cycle EV/EBITDA is roughly in line with the industry's ~8x — i.e. not obviously cheap on mid-cycle, and expensive on trough. The torque is operational (margin recovery × the levered balance sheet), not multiple re-rating.

Lens 8 · Stock-Price Catalysts

What has moved CLF >5%, and what it reveals (mostly ``):

• Section 232 tariff escalations (2025). The 25%→50% steel tariff (June 2025) is the dominant macro driver; CLF, as the most domestic name, is the highest-beta tariff trade in the group.
• Rare-earth announcement (Oct 2025): stock +20% in a day. CLF disclosed evidence of rare-earth mineralization at its Michigan UP and Minnesota sites. Pure optionality (pre-feasibility) but it reframed CLF as a "critical-materials" name, not just steel.
• POSCO MoU (Sep 17 2025). A "transformative" strategic partnership with Korea's POSCO, reportedly carrying a potential ~$700M POSCO equity infusion (~10% of the company) to cut debt. Double-edged: deleveraging positive, dilution negative.
• Idled-mills-to-data-centers (2025–26). Selling stranded, grid-connected industrial sites to data-center developers; $425M targeted proceeds, $60M closed, JPMorgan advising on "non-core" assets "worth billions."
• Earnings prints. Q1'26 fell ~6% despite a double beat — the market reacts to cash flow and cost guidance, not headline beats, for this name.
• Credit downgrade (Dec 2025). A ratings downgrade pressured the stock and underscored the leverage overhang.

Pattern: CLF trades on (1) trade policy (tariffs = the beta), (2) balance-sheet de-risking (asset sales, POSCO, downgrades), and (3) optionality surprises (rare earths, data centers). It does not reliably rally on earnings beats while cash flow is negative. The market is pricing a survival-and-normalization thesis, gated on deleveraging.

---

Phase C — Judge people & books

Lens 9 · Management

CEO: Lourenco Goncalves (Chairman, President & CEO since Aug 2014; age 67).

1. Track record — genuinely impressive on scale. Goncalves transformed CLF from a small, near-distressed iron-ore miner into the largest flat-rolled steel producer in the US in ~3 years via four acquisitions: AK Steel (Mar 2020), ArcelorMittal USA (Dec 2020), Ferrous Processing & Trading (Nov 2021), Stelco (Nov 2024). AIST "Steelmaker of the Year" 2021; S&P Global Platts "CEO of the Year" 2021. He is a genuine operator and a ferocious tariff/trade advocate who arguably helped create the policy tailwind CLF now rides.
2. Skin in the game. Long tenure (12 years), founder-like control. Specific insider ownership n/a (no insider-transactions.csv on disk); web notes insider selling alongside the 2025 equity issuance — a yellow flag worth verifying.
3. Capital-allocation history — the double-edged sword. The roll-up created the franchise and ~$8B of debt that has hung over the stock since 2020. ROE/ROIC has now gone deeply negative (FY25 net loss -$1.48B on $6.1B equity ≈ -24% ROE ). Retained earnings flipped to a -$529M deficit. The Stelco deal (Nov 2024, ~$2.5B) added leverage right before the trough. Capital allocation has whipsawed: 2024 buybacks → 2025 dilutive equity raises (shares 493.9M → 569.8M, ~+15% in one year ).
4. Red flags. Nepotism / related-party governance: the CFO, Celso Goncalves (age 37), is the CEO's son — disclosed, but a real governance concern (the two most senior officers are father and son, on a board the CEO chairs). Promotional, combative public style. Combined Chairman/President/CEO roles concentrate power.
5. Archetype: founder-operator / empire-builder, not a caretaker professional manager. Implication for this stage: aggressive, conviction-driven, will press the tariff advantage and the optionality (rare earths, data centers) hard — but has shown he will lever up at the wrong moment and dilute holders to survive. Bet on him to fight, not to be conservative.

Lens 10 · Forensic Red Flags

Acting as a forensic analyst. Label every figure.

Income statement:

• COGS > revenue in FY2025 ($19,470M vs $18,610M) — not an accounting trick, but the starkest possible signal that the business is sub-economic at trough pricing.
• Large income-tax benefit flatters the loss: FY25 tax benefit $581M (29% effective rate, above the 21% statutory), driven by deferred-tax movements and unrecognized tax benefits. Watch for a valuation allowance risk on deferred tax assets if losses persist — deferred income taxes (liability) fell $849M → $375M, consistent with DTAs building against losses.

Balance sheet:

• Goodwill $1,814M + intangibles $1,135M = ~$2.95B of soft assets against $6.1B equity. No goodwill impairment was taken in 2025 despite a -$1.48B loss and a sub-$8B market cap — if the trough extends, an impairment is a live risk (CLF did impair goodwill $125M in 2023, so it's not hypothetical).
• Inventories $4,772M — very large (≈26% of revenue). In a falling-price environment, lower-of-cost-or-market write-downs are a risk; inventory did fall YoY ($5,094M → $4,772M), which is the right direction.
• Cash only $57M against $7,253M long-term debt. The mitigant: ~$3B total liquidity (cash + ABL availability) and no debt maturities until 2029. So leverage is high and coverage is thin, but there is no near-term refinancing wall — this is a survive-the-trough balance sheet, not an imminent-default one.

Cash flow:

• Operating cash flow negative two periods running: FY2025 -$462M (vs +$105M FY24) and Q1'26 -$325M. Earnings are improving faster than cash — the AR build is funding-intensive. This is the single most important watch-item: the bull thesis requires the Q2'26 positive-FCF guide to land.
• DD&A $1,235M — heavy (integrated-mill capital intensity); against ~$561M FY25 capex and a guided ~$800M next-12-months capex, the company is currently under-investing relative to depreciation, sustainable short-term but not indefinitely.
• SBC / non-GAAP: restricted-stock-unit grants were cut to a "nominal level" in Q1'26 — actually a positive governance signal (less dilution from comp), distinct from the share-issuance dilution.

Regulatory findings (required sub-section). Per regulatory/regulatory-findings.md (fetched 2026-06-20, sources SEC EDGAR EFTS LR + AAER):

• SEC Litigation Releases: none naming Cleveland-Cliffs in 2021-06-20 → 2026-06-20.
• AAERs: none in the period.
• Non-SEC enforcement (web search "Cleveland-Cliffs" (FTC OR DOJ OR EPA OR consent decree OR settlement OR fine) enforcement): no material current federal enforcement action surfaced. CLF, as an integrated steelmaker, carries the usual heavy environmental compliance load (the 10-K cites revised National Emission Standards and ongoing environmental/asset-retirement obligations of $682M ), and is routinely party to incidental claims and legal proceedings ("various claims and legal proceedings incidental to our operations" — NOTE 18 Contingencies ), but nothing rising to a disclosed material liability.
• 10-K Item 3 (Legal Proceedings): the 10-K confirms commitments and contingencies are detailed in NOTE 20; no single material proceeding is flagged as reasonably likely to have a material adverse effect.
• Net: No material regulatory or accounting-enforcement findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-K Item 3 / contingencies as of 2026-06-20. The forensic concerns here are cyclical/leverage (cash burn, goodwill, valuation allowance), not fraud or enforcement.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (FY2026 / FY2027 / FY2028 EPS)

Built bottom-up from the latest actuals + guidance. Output ``; inputs labeled. Share count ~570M (FY26 basis). No forecast.ts create is logged — this is the --watchlist loop (forecast logging skipped per SKILL).

Anchors: Q1'26 ran ~-$0.42 EPS but with the loss curve bending up sharply (operating loss -$213M vs -$543M) and ASP +7% YoY; management guides positive FCF from Q2'26; HRC prices +12.7% in Q1'26 and +6.75% in Q2'26; the contract book re-prices upward with a ~1-year lag.

FY2026 — base $(0.30) EPS. H1'26 still loss-making (~-$0.42 Q1 + a smaller Q2 loss as the energy one-timer rolls off and price catches up), H2'26 approaching breakeven as the tariff-supported price floor flows through the contract book. Net: a modest full-year loss, far narrower than FY25's -$2.91.

• Bull $0.40: HRC holds >$1.05/kg, contract re-pricing + $300M cost savings + asset-sale proceeds + a clean (no one-timer) cost base push H2 solidly profitable.
• Bear $(1.20): auto demand rolls over, HRC fades back toward $0.85/kg, energy/cost pressure persists; another ugly year.

FY2027 — base $1.10 EPS. First full year of mid-cycle pricing with the tariff floor intact + full run-rate cost savings + lower idled-facility drag. Implies $1.5B+ Adj EBITDA (toward the low end of the 2023 mid-cycle level), with interest expense ($594M run-rate) and DD&A as the drags.

• Bull $2.20: POSCO deal closes and deleverages (cutting interest); rare-earth/data-center proceeds fund debt paydown; HRC stays elevated; CLF earns toward its 2021–22 power.
• Bear $0.00: leverage + a soft cycle keep it stuck at breakeven.

FY2028 — base $1.60 EPS. Normalized mid-cycle, assuming the tariff regime persists, debt is meaningfully reduced (via FCF + asset sales + POSCO), and steel demand is supported by reshoring/grid/infrastructure. The swing factor is deleveraging: every $1B of debt retired at ~7% saves ~$70M pretax ≈ ~$0.10 EPS.

Caveat: CLF EPS is extremely operationally geared — small HRC-price and volume moves swing EPS by dollars because of the levered balance sheet and the spot-exposed contract majority. These are scenario midpoints, not point forecasts; the bear-to-bull FY27 spread ($0.00–$2.20) is the honest uncertainty.

Lens 12 · Bull vs Bear

Bull case. CLF is the single most tariff-levered, most domestic name in US steel, hitting the trough exactly as the 50% Section 232 wall + reshoring + a record-old US vehicle fleet drive a multi-year domestic-pricing up-cycle. The Q1'26 inflection is already visible (operating loss -61% YoY, ASP +7%, segment Adj EBITDA flipped positive). On top of the cyclical recovery sit three free options the market is only starting to price: (1) the POSCO partnership (~$700M equity + strategic alignment, deleveraging); (2) idled-mills-to-data-centers ($425M+ proceeds, "non-core" assets "worth billions," selling stranded grid-connected sites into the most acute power bottleneck of the AI era); (3) rare earths at the Michigan/Minnesota mines (a literal "critical-materials" call option that already moved the stock +20% once). Management has proven it can build scale and is now executing a credible $300M/yr self-help program. At ~$12.68 / $7.2B market cap, mid-cycle earnings power of $1.50–2.00 EPS implies a single-digit normalized P/E.

Bear case (permanent-impairment risks).

1. The balance sheet is the business's master. $7.3B+ debt, ~$594M/yr interest, negative operating cash flow two periods running, ~$57M cash. The thesis requires the cycle and the self-help to both work; if HRC fades and FCF stays negative, the equity is a thin sliver under a large, senior debt stack — and CLF has shown it will issue equity to survive, diluting holders (already +15% shares in 2025).
2. Structurally high-cost blast-furnace assets. CLF stayed loss-making at the gross line in FY25 while EAF peers (Nucor, STLD) stayed profitable. Its margin advantage is policy (tariffs), not cost — and policy can reverse (a future administration, a USMCA renegotiation, retaliatory tariffs).
3. Automotive concentration into a possible auto-demand air-pocket (30% of revenue) — a US new-vehicle downturn hits CLF harder than any peer.

Pre-mortem (18 months out, thesis broke): HRC rolled back toward $0.85/kg as tariff-pulled domestic capacity restarted and demand disappointed; the Q2'26 "positive FCF" guide slipped; CLF tapped the equity market again, diluting ~10–15%; a goodwill impairment hit; the POSCO deal stalled. The stock round-tripped to the $7 low.

Are multiples too high? On trough/trailing numbers CLF is not cheap (negative EBITDA, n/m P/E). On mid-cycle, ~7.6–9.7x EV/EBITDA is roughly fair vs the ~8x industry — so this is not a deep-value multiple story; it's an operational-recovery + optionality story. You're paying a fair mid-cycle price for the turn plus three free call options.

Contrarian view (what the market refuses to see): the consensus is anchored on "still losing money, too much debt" and a Hold rating — pricing the trailing P&L. What it under-weights is that CLF is structurally the prime beneficiary of two converging mega-trends — steel protectionism and the AI power buildout (via its stranded grid-connected sites) — and that the operating inflection is already in the tape, not a hope. The asymmetry skews up if deleveraging lands.

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

Dismantling the bull case.

• What structurally breaks the money-machine: CLF makes money only when domestic HRC clears comfortably above its blast-furnace cash cost. That spread is policy-dependent (50% tariffs) and demand-dependent (auto). Remove either and CLF is structurally loss-making — FY2025 proved it can lose $1.5B in a single down year even with tariffs rising.
• Revenue concentration: 30% automotive into the Detroit Three. If US auto production drops 10–15% (rates, affordability, EV-transition disruption), CLF's highest-margin volume evaporates and the spot-exposed 60% of the book gets crushed on price simultaneously.
• Why the moat is weaker than bulls think: it is a grade-and-access moat, not a cost moat. EAF mini-mills (Nucor, STLD, and STLD's new Texas flat-rolled EAF) are structurally lower-cost and are adding flat-rolled capacity — encroaching on CLF's core just as tariffs invite new domestic supply. The most dangerous competitor bulls underestimate is Steel Dynamics, ramping low-cost EAF flat-rolled directly into CLF's market.
• Worst capital-allocation moves: levering up for Stelco (Nov 2024, ~$2.5B) at the cycle top, right before a -$1.48B loss year; whipsawing from 2024 buybacks to 2025 dilutive equity; a father-son CEO/CFO governance structure on a board the CEO chairs.
• Assumptions that must hold for $12.68: tariffs persist at 50%; HRC stays >$1.00/kg; Q2'26 FCF turns positive and stays positive; no further equity dilution; POSCO closes; no goodwill impairment. That's a lot of "ands."
• Valuation if growth disappoints 20–30%: knock 25% off the mid-cycle Adj EBITDA assumption ($1.5B → $1.1B) and at ~8x EV/EBITDA the EV is ~$8.9B; net out ~$7.3B debt and the equity is ~$1.6B vs the current $7.2B — i.e. ~75% downside in a no-recovery scenario. The leverage cuts both ways, hard.
• Single scenario that permanently impairs: a prolonged sub-mid-cycle steel market (2 more years) forces repeated equity raises at depressed prices, permanently diluting holders and transferring value to debt — the classic levered-cyclical death spiral. Plausibility: moderate — gated by the ~$3B liquidity and no-maturities-to-2029 cushion, which buys time, but not by profitability.

Lens 14 · Management Questions (ordered by information value)

1. Q1'26 operating cash flow was -$325M and you guided to positive FCF in Q2'26 — walk me through the bridge: how much is price realization catching up to spot vs. the AR build reversing vs. cost savings, and what HRC price does that guide assume?
2. At what sustained HRC price does the integrated business generate positive free cash flow through a full year, after interest and sustaining capex?
3. On the POSCO MoU — is the reported ~$700M equity infusion accurate, what ownership does it imply, what's the use of proceeds (debt paydown vs. growth), and what's the timeline and closing risk?
4. You're carrying ~$2.95B of goodwill + intangibles after a -$1.48B year — what impairment-test assumptions (HRC price, discount rate) are you using, and at what point does an impairment become unavoidable?
5. The idled-mills-to-data-centers proceeds — $425M targeted, $60M closed: what's the realistic total, the timeline, and are these outright sales or power/JV structures that retain upside?
6. What is the path and timeline to get net debt below 2x mid-cycle EBITDA, and does it depend on asset sales and POSCO, or can organic FCF do it alone?
7. Will you commit to not issuing equity below a stated price, given holders absorbed ~15% dilution in 2025?
8. On rare earths (Michigan/Minnesota) — where are you in feasibility, what capital would commercial production require, and is this a real business line or optionality you're flagging?
9. Your cost curve sits above the EAF mini-mills; as Nucor and Steel Dynamics add flat-rolled EAF capacity, how does CLF defend share and margin without relying on tariffs?
10. What happens to your contract book and spot exposure if the 50% Section 232 tariff is reduced or USMCA is renegotiated — how much of your margin is policy-dependent?
11. Auto is 30% of revenue — what's your demand assumption for US vehicle production in 2026–27, and how exposed are you to an auto air-pocket?
12. The $300M/year cost-savings program — how much is in the run-rate today, how much is structural (idled capacity) vs. cyclical, and what's the timeline to full realization?
13. Capex is guided ~$800M vs. ~$1.2B depreciation — how long can you under-invest before it compromises the asset base, and what's true maintenance capex?
14. How do you think about decarbonization capital for your blast-furnace fleet given tightening emissions standards — is there a multi-billion-dollar capital cliff we should be modeling?
15. Succession and governance: the CEO/Chairman and CFO are father and son — what's the board's independent oversight of related-party governance, and what is the succession plan?

---