AMR-Type Thesis Test: Hated company, below book, real physical assets. Data as of 2026-03-12.
Cleveland-Cliffs is North America's largest flat-rolled steel producer and largest iron ore pellet manufacturer, trading at 0.86x tangible book value ($5.1B market cap vs $3.2B tangible book). The company owns $20B in total assets including $9.5B in net PP&E (steel mills, blast furnaces, iron ore mines), $4.8B in inventory, and $482M in mineral properties. It employs 25,000 people across ~20 steelmaking facilities.
This is NOT an AGI play despite the AGI 7 score. This is a potential AMR-type deep value play: a hated company with real physical assets trading below replacement cost, with the question being whether debt levels ($7.3B) make it a value trap or a coiled spring.
Key difference from AMR: CLF has $7.3B in debt (125% debt/equity) and generated negative free cash flow of -$1.0B in FY2025. AMR was buying back shares aggressively. CLF issued $951M in new equity in 2025. This is the opposite of the AMR playbook. The debt maturity schedule provides runway (no maturities for 2 years), but the company is not compounding per-share value — it is diluting shareholders to stay liquid.
VERDICT: DO NOT DEPLOY — HIGH RISK
CLF fails the AMR-type test on the most critical dimension: capital allocation. AMR was buying back shares at below-book prices, compounding per-share value. CLF is doing the opposite — it issued 75M new shares at $13/share in October 2025 to pay down debt, and shares outstanding went from 494M to 570M (+15.4% dilution). The company is not a buyback compounder; it is a dilution machine trying to deleverage.
The assets are real and substantial, but $7.3B in debt with negative operating cash flow means equity holders sit behind a massive wall of claims. At $8.92/share, the market is pricing in legitimate risk that the equity gets further diluted or wiped out in a prolonged downturn.
| Target | Approach | Assumptions | Price Range |
|---|---|---|---|
| Very Safe Floor | Liquidation of physical assets minus all liabilities | PP&E at 40-50% of book, inventory at 60%, receivables at 80%, minus $13.7B total liabilities | $3.00 - $5.00 |
| Fair Value | Normalized EBITDA x 6-7x EV multiple | Normalized EBITDA $1.5-2.0B (2023 levels), minus net debt $7.2B, divided by 570M shares | $12 - $16 |
| Bull Case | Steel super-cycle + infrastructure boom | $3B EBITDA at 7x = $21B EV, minus $5B net debt (reduced), 500M shares | $30 - $50 |
Current price: $8.92. Below fair value but above liquidation floor. The gap between floor and fair value is uncomfortably large due to the debt load.
Cleveland-Cliffs owns $20.0B in total assets, including some of the most irreplaceable industrial infrastructure in North America. These assets took decades and tens of billions of dollars to build. They cannot be replicated quickly.
| Asset Category | Book Value | Gross Value | Notes |
|---|---|---|---|
| Net PP&E | $9,481M | $15,310M | Steel mills, blast furnaces, coke ovens, finishing lines |
| Land & Improvements | $1,463M | Mill sites, mine sites across US & Canada | |
| Buildings & Improvements | $1,205M | Factory buildings, office space | |
| Machinery & Equipment | $11,987M | Blast furnaces, BOFs, EAFs, hot strip mills, cold mills | |
| Construction in Progress | $655M | Active capital projects | |
| Less: Accumulated Depreciation | ($5,829M) | ~38% depreciated | |
| Mineral Properties (net) | $482M | $783M | Iron ore mines, mineral rights (hugely understated) |
| Inventory | $4,772M | — | $2,401M finished goods + $2,371M raw materials |
| Accounts Receivable | $1,442M | — | Customer receivables |
| Goodwill | $1,814M | — | AK Steel, ArcelorMittal USA, Stelco acquisitions |
| Other Intangibles | $1,135M | $1,310M | Customer relationships, trade names (Stelco) |
| Pension Asset | $469M | — | Overfunded pension plans |
| Finance Lease ROU | $576M | — | Leased equipment and facilities |
| Cash | $57M | — | Almost no cash on hand |
| Other Assets | $1,134M | — | Deferred tax assets, other |
| TOTAL ASSETS | $20,012M | — | |
| Facility | Type | Key Products |
|---|---|---|
| Indiana Harbor (East Chicago, IN) | Integrated (blast furnaces + BOF) | Hot-rolled, cold-rolled, coated |
| Burns Harbor (Burns Harbor, IN) | Integrated (blast furnace + BOF) | Hot-rolled, cold-rolled, coated, plate |
| Dearborn Works (Dearborn, MI) | Integrated (blast furnace + BOF) | Automotive steel, advanced high-strength |
| Cleveland Works (Cleveland, OH) | Integrated (blast furnace + BOF) | Hot-rolled, coated |
| Middletown Works (Middletown, OH) | Integrated (blast furnace + BOF) | Hot-rolled, cold-rolled, electrical steel |
| Butler Works (Butler, PA) | Cold rolling & finishing | Stainless steel, specialty |
| Coatesville (Coatesville, PA) | Plate mill | Carbon & alloy plate |
| Columbus (Columbus, OH) | EAF steelmaking | Stainless, electrical steel (GOES/NOES) |
| Rockport Works (Rockport, IN) | EAF steelmaking | Hot-rolled |
| New Carlisle (New Carlisle, IN) | EAF steelmaking | Hot-rolled |
| Mansfield Works (Mansfield, OH) | Finishing | Tinplate, coated |
| Gary Plate (Gary, IN) | Plate mill | Carbon plate |
| Toledo DR Plant (Toledo, OH) | Direct Reduction | HBI/DRI for EAF feed |
| Zanesville Works (Zanesville, OH) | Finishing | Tubular products |
| Piedmont (Piedmont, SC) | Finishing | Coated products |
| Coshocton Works (Coshocton, OH) | Finishing | Tinplate |
| Stelco — Hamilton (Hamilton, ON) | Integrated (blast furnace + BOF) | Hot-rolled, cold-rolled, coated |
| Stelco — Nanticoke (Nanticoke, ON) | Pickle line & finishing | Pickled & oiled coil |
Cleveland-Cliffs is the largest iron ore pellet producer in North America with capacity of 29 million long tons annually. This is a massively understated asset on the balance sheet at only $482M net. Iron ore mines in Minnesota and Michigan that have been producing for decades.
Gross PP&E is $15.3B. To build an equivalent integrated steel operation from scratch today — mines, pellet plants, blast furnaces, BOFs, hot strip mills, cold mills, coating lines — would likely cost $25-40B and take 10+ years for permitting and construction. The market is valuing the equity (after debt) at $5.1B, implying total enterprise value of $13.2B for assets with replacement cost 2-3x higher.
However, the counterargument is strong: you don't need to replace these assets. The US has overcapacity in steel. Some of these integrated mills are older and less efficient than newer EAF minimills. The replacement cost argument works only if the assets generate adequate returns, which they currently are not.
The market is not stupid. Here are the real reasons CLF trades at 0.86x tangible book:
| Year | Revenue | Net Income | Oper. Cash Flow | Free Cash Flow |
|---|---|---|---|---|
| FY2023 | $22.0B | $385M | $1,460M | $814M |
| FY2024 | $19.2B | ($760M) | $105M | ($590M) |
| FY2025 | $18.6B | ($1,478M) | ($462M) | ($1,023M) |
Revenue has dropped 15% from 2023 while losses have exploded. Operating cash flow went from +$1.5B to -$462M. Free cash flow: -$1.0B. The company is burning cash.
Long-term debt went from $3.1B (end 2023) to $7.3B (end 2025). The Stelco acquisition in 2024 cost $2.5B and was funded largely with debt. This happened right as steel prices crashed. Terrible timing.
Debt-to-equity: 125%. Enterprise value ($13.2B) is 2.6x the equity market cap ($5.1B). Bondholders own more of the company's cash flows than equity holders do.
In October 2025, CLF issued 75 million new shares at $13/share, raising $951M to pay down its credit facility. Shares outstanding jumped from 494M to 570M — a 15.4% dilution. This is the opposite of shareholder-friendly capital allocation. The company is issuing equity at depressed prices to service debt.
CLF is the #1 supplier of steel to the US automotive industry. 40% of shipments are hot-rolled, much of it automotive-grade. The CEO cited "persistently weak production levels from the automotive sector throughout the entire year" as a key driver of 2025 weakness. Auto production has not recovered to pre-COVID levels.
Average selling price dropped from $1,081/ton (2024) to $1,005/ton (2025). Hot-rolled coil (HRC) prices have been weak throughout 2025. The gross margin went negative — cost of goods exceeded revenue. This is a cyclical trough.
Gross margin: -4.6%. Operating margin: -6.3%. The company is losing money on every ton of steel it ships. This is unsustainable and reflects both weak pricing and high fixed costs from the integrated mill structure.
Verdict: More overleveraged than AMR. This is the key risk.
| Period | Amount Due | Assessment |
|---|---|---|
| Year 1 (2026) | $0 | No near-term maturities — breathing room |
| Year 2 (2027) | $0 | Another year of runway |
| Year 3 (2028) | $452M | Manageable if cash flow improves |
| Year 4 (2029) | $1,268M | Large maturity wall — needs refinancing or paydown |
| Year 5 (2030) | $750M | Significant but manageable |
| After Year 5 | $4,860M | Long-dated — well pushed out |
| Total Debt | $7,330M | Net of $77M unamortized discount = $7,253M on books |
| Metric | FY2025 | FY2024 | FY2023 |
|---|---|---|---|
| EBITDA (Adjusted) | $37M | $773M | ~$1,650M |
| Interest Expense | $594M | $370M | ~$200M |
| Interest Coverage (EBITDA/Interest) | 0.06x | 2.1x | ~8.3x |
| Cash Interest Paid | $459M | — | — |
Interest coverage of 0.06x in FY2025 is catastrophic. The company cannot service its debt from operating earnings. It is relying on liquidity (ABL facility) and asset base to stay solvent. This is the danger zone.
| Metric | FY2025 | FY2024 | FY2023 | Trend |
|---|---|---|---|---|
| Revenue | $18,610M | $19,185M | $21,996M | Declining (-15% from peak) |
| Cost of Revenue | $19,470M | $19,122M | — | Exceeds revenue in FY25 |
| Gross Profit | ($860M) | $63M | — | Deeply negative |
| Operating Income | ($1,579M) | ($763M) | $659M | Collapsed |
| Interest Expense | $594M | $370M | ~$200M | Doubled (Stelco debt) |
| Net Income | ($1,478M) | ($760M) | $385M | Massive losses |
| EPS (Basic) | ($2.91) | ($1.57) | $0.76 | Deteriorating |
| D&A | $1,235M | $951M | $973M | Increased (Stelco) |
| Steel Shipments | 16.2M tons | 15.6M tons | — | Growing volumes |
| Avg Price/Ton | $1,005 | $1,081 | — | -7% pricing decline |
| CapEx | $561M | $695M | $646M | Reduced (cost control) |
| Quarter | Revenue | Gross Profit | Operating Income | Net Income |
|---|---|---|---|---|
| Q4 2024 | $4,325M | ($273M) | ($437M) | ($447M) |
| Q1 2025 | $4,629M | ($391M) | ($535M) | ($495M) |
| Q2 2025 | $4,934M | ($209M) | ($373M) | ($483M) |
| Q3 2025 | $4,734M | ($46M) | ($201M) | ($251M) |
| Q4 2025 | $4,313M | ($206M) | ($337M) | ($243M) |
Q3 2025 showed the best gross margin (only -$46M loss vs -$391M in Q1), suggesting some stabilization. But Q4 worsened again. No clear recovery trend yet.
| Product | % of Shipments | Notes |
|---|---|---|
| Hot-Rolled | 40% | Commodity product, most price-sensitive |
| Coated (galvanized, etc.) | 28% | Higher margin, automotive & construction |
| Cold-Rolled | 15% | Higher value-add than hot-rolled |
| Plate | 5% | Heavy construction, energy, shipbuilding |
| Stainless & Electrical | 3% | GOES/NOES — the AGI angle (transformers) |
| Other / Slabs | 9% | Semi-finished, tolling |
| Year | Buybacks | Equity Issued | Net Debt Change | Shares Outstanding | Change |
|---|---|---|---|---|---|
| FY2023 | $153M | $0 | — | 505M | -2.3% |
| FY2024 | $737M | $0 | +$3,928M | 494M | -2.2% |
| FY2025 | $0 | $951M | -$188M net | 570M | +15.4% |
The full picture: CLF bought back $737M in shares in 2024 (good!) but then added $4B in debt for the Stelco acquisition (bad timing). When steel prices crashed in 2025, they had to reverse course — issuing 75M new shares at $13 to raise $951M for debt paydown. Net result: shareholders who held through got diluted 15.4%.
Remaining buyback authorization: $1.4B. But the company cannot use it while burning cash and carrying $7.3B in debt. This authorization is theoretical, not practical.
Cleveland-Cliffs scored 7/10 for AGI impact, categorized as a "physical bottleneck" provider. The reasoning:
This is the strongest AGI connection. Cleveland-Cliffs produces grain-oriented electrical steel (GOES) and non-oriented electrical steel (NOES) at its Columbus, OH facility. GOES is used in power transformers — the critical bottleneck for expanding electrical grid capacity.
Bottom line: The electrical steel angle is real but small relative to CLF's total business. It's a nice tailwind, not a thesis-changing catalyst. CLF is fundamentally a commodity steel company that happens to have a small but strategically important electrical steel business.
Data centers use structural steel for buildings, rebar for foundations, and plate steel for equipment. But this demand is a small fraction of total US steel consumption (~100M tons/year). Even a massive data center buildout adds maybe 2-5M tons of incremental steel demand over several years — meaningful at the margin but not transformative for CLF's revenue.
The real AGI play in steel would require a sustained infrastructure supercycle (reshoring, grid expansion, data centers, defense) that lifts steel prices across the board. CLF would benefit as a volume play, not as a targeted AGI beneficiary.
This is the critical question for a below-book value company. If CLF were forced to liquidate today, what would equity holders receive?
| Asset | Book Value | Liquidation % | Liquidation Value | Rationale |
|---|---|---|---|---|
| Cash | $57M | 100% | $57M | Cash is cash |
| Receivables | $1,442M | 80% | $1,154M | Steel customers generally creditworthy |
| Inventory | $4,772M | 55% | $2,625M | Steel inventory saleable but at distressed prices; LIFO reserve adds $158M |
| PP&E (net) | $9,481M | 35-45% | $3,318-4,267M | Some mills valuable (Burns Harbor, Indiana Harbor), others may be worth scrap value. Older integrated mills are hard to sell as going concerns. |
| Mineral Properties | $482M | 100-150% | $482-723M | Iron ore mines are genuinely understated. Real mineral wealth in the ground. |
| Pension Asset | $469M | 80% | $375M | Overfunded plans can be recovered |
| Other Assets | $1,134M | 25% | $284M | Deferred tax assets, ROU assets — limited recovery |
| Goodwill + Intangibles | $2,949M | 0% | $0 | Zero in liquidation |
| TOTAL LIQUIDATION VALUE | $8,295 - $9,485M | |||
| Liabilities Against Assets | Amount |
|---|---|
| Total Liabilities | ($13,689M) |
| Long-term debt | ($7,253M) |
| Current liabilities | ($3,302M) |
| Pension/OPEB liabilities | ($655M) |
| Asset retirement obligations | ($599M) |
| Environmental liabilities | ($112M) |
| Other non-current liabilities | ($1,768M) |
| Equity Recovery (Optimistic) | ($4,204M) to ($5,394M) |
| Per Share (570M shares) | NEGATIVE — ($7.37) to ($9.46) |
In a forced liquidation, equity holders would receive ZERO. Total liabilities ($13.7B) exceed the realistic liquidation value of assets ($8.3-9.5B) by $4.2-5.4B. This is not a company where tangible book value provides a meaningful floor. The assets are real but the claims against them (especially $7.3B in debt) are larger than what the assets would fetch in a fire sale.
This is fundamentally different from AMR, which had limited debt and could return value to shareholders through buybacks. CLF's equity is subordinated to a massive debt stack.
The liquidation analysis is deliberately pessimistic. As a going concern, CLF's assets are worth far more because they generate revenue. The real question is not "what are the assets worth in liquidation?" but "can the company survive long enough for the steel cycle to turn?"
The AMR thesis requires: (1) hated company, (2) below book value, (3) real physical assets, (4) aggressive buybacks compounding per-share value, (5) manageable debt, (6) positive free cash flow. How does CLF stack up?
| Criterion | AMR Ideal | CLF Reality | Pass/Fail |
|---|---|---|---|
| Hated by the market? | Yes | Yes — 16% short interest, down 57% from 2024 high | PASS |
| Below tangible book? | Yes | Yes — 0.86x tangible book | PASS |
| Real physical assets? | Yes | Yes — $9.5B PP&E, mines, steel mills | PASS |
| Buying back shares? | Yes, aggressively | NO — Issued 75M shares in 2025 (+15.4% dilution) | FAIL |
| Manageable debt? | Low leverage | $7.3B debt, 126% D/E, interest coverage 0.06x | FAIL |
| Positive free cash flow? | Yes | -$1.0B FCF in FY2025 | FAIL |
| Durable business model? | Yes | Cyclical — steel demand swings wildly | MIXED |
| Secular tailwind? | Helpful | Infrastructure spending is a tailwind but offset by imports/overcapacity | MIXED |
| Overall AMR Score | 3/8 — FAILS the AMR test | ||
CLF fails the AMR test on the three most important dimensions: It is not buying back shares, it has excessive debt, and it is burning cash. The first two criteria (hated + below book) are necessary but not sufficient. The AMR thesis only works when the company is using its cash flow to retire shares, compounding per-share value while the market is looking the other way. CLF is doing the exact opposite — it is issuing shares to delever.
| Catalyst | Probability | Impact | Timeline |
|---|---|---|---|
| POSCO definitive agreement — Strategic partnership with equity investment | Medium | High — could cut debt by $2-3B | H1 2026 (targeted) |
| Steel price recovery — HRC prices return to $800-1,000/ton | High | High — every $100/ton = ~$1.6B revenue | 2026-2027 |
| Infrastructure spending — IIJA, CHIPS Act procurement | High | Medium | Ongoing through 2028 |
| Auto production recovery — Return to ~16M SAAR | Medium | Medium | 2026-2027 |
| Grid expansion / transformer demand — GOES shortage drives pricing power | High | Low-Medium (3% of volume) | 2025-2030 |
| Resumption of buybacks — At current prices would be highly accretive | Low | High — $1.4B authorization available | Not until FCF positive |
| Expiration of unfavorable slab contract — CEO mentioned this drag ending | High | Medium — improves margins | 2026 |
| Item | Dec 2025 | Dec 2024 | Change |
|---|---|---|---|
| Cash | $57M | $54M | +$3M |
| Accounts Receivable | $1,442M | $1,576M | -$134M |
| Inventory | $4,772M | $5,094M | -$322M (destocking) |
| Current Assets | $6,435M | $6,907M | -$472M |
| Net PP&E | $9,481M | $9,942M | -$461M (depreciation > capex) |
| Goodwill | $1,814M | $1,768M | +$46M |
| Other Intangibles | $1,135M | $1,170M | -$35M (amortization) |
| Pension Asset | $469M | $427M | +$42M |
| Total Assets | $20,012M | $20,947M | -$935M |
| Accounts Payable | $1,893M | $2,008M | -$115M |
| Current Liabilities | $3,302M | $3,361M | -$59M |
| Long-Term Debt | $7,253M | $7,065M | +$188M |
| Pension/OPEB Liabilities | $655M | $751M | -$96M |
| Asset Retirement Obligations | $599M | $601M | -$2M |
| Total Liabilities | $13,689M | $14,080M | -$391M |
| Common Stock Equity | $6,116M | $6,634M | -$518M |
| Minority Interest | $207M | $233M | -$26M |
| Tangible Book Value | $3,167M | $3,696M | -$529M |
| Tangible BV / Share | $5.56 | $7.49 | -$1.93 (loss + dilution) |
| Shares Outstanding | 569.8M | 493.9M | +75.9M (+15.4%) |
| Retained Earnings | ($529M) | $949M | -$1,478M (entire 2025 loss) |
| Item | FY2025 | FY2024 |
|---|---|---|
| Net Income | ($1,428M) | ($714M) |
| D&A | $1,235M | $951M |
| Deferred Tax | ($506M) | ($196M) |
| Working Capital Changes | $366M | ($15M) |
| Operating Cash Flow | ($462M) | $105M |
| CapEx | ($561M) | ($695M) |
| Acquisitions | $0 | ($2,512M) |
| Free Cash Flow | ($1,023M) | ($590M) |
| Debt Issued | $1,983M | $4,781M |
| Debt Repaid | ($1,794M) | ($845M) |
| Equity Issued | $951M | $0 |
| Share Buybacks | $0 | ($733M) |
| Interest Paid (Cash) | $459M | — |
| Category | Percentage |
|---|---|
| Institutional Investors | 81.6% |
| Retail / Other | 17.0% |
| Insiders | 1.4% |
| # | Holder | Type | Notable? |
|---|---|---|---|
| 1 | Vanguard Group | Index fund | Passive |
| 2 | BlackRock | Index fund | Passive |
| 3 | State Street | Index fund | Passive |
| 4 | Slate Path Capital | Hedge fund | Active — large position |
| 5 | Castle Hook Partners | Hedge fund | Active — 5.7x increase in position |
| 6 | Fairfax Financial | Insurance/Value investor | Prem Watsa's vehicle — deep value |
| 7 | Dimensional Fund Advisors | Factor-based | Systematic value |
| 8 | Maple Rock Capital | Hedge fund | Active — doubled position |
| 9 | D.E. Shaw | Quant | Algorithmic |
| 10 | Geode Capital | Index fund | Passive |
Notable: Fairfax Financial (Prem Watsa) is a known deep-value investor with a Buffett-like approach. Castle Hook Partners nearly 6x'd their position recently. Several active hedge funds are loading up — this is a contested stock, not a forgotten one.
| Period | Price | Context |
|---|---|---|
| Mar 2022 (Peak) | $32.21 | Post-Ukraine steel spike |
| Dec 2023 | $20.42 | Steel prices normalizing |
| Mar 2024 | $22.74 | Pre-Stelco, optimism |
| Dec 2024 | $9.40 | Stelco integration + steel crash |
| May 2025 (Trough) | $5.83 | Maximum pessimism |
| Oct 2025 (Equity raise) | $12.43 | 75M shares issued at $13 |
| Dec 2025 | $13.28 | Some recovery |
| Mar 2026 (Today) | $8.92 | Down 33% from Dec — new leg down |
52-week low: $5.63. 52-week high: $16.70. Current price is in the lower quartile of its range.
| Metric | 2026 Guidance | 2025 Actual |
|---|---|---|
| Steel Shipments | 16.5-17.0M tons | 16.2M tons |
| CapEx | ~$700M | $561M |
| SG&A | ~$575M | $543M |
| D&A | ~$1,100M | $1,235M |
| Pension/OPEB Payments | ~$125M | $154M |
| Unit Cost Reduction | ~$10/ton improvement | — |
Cleveland-Cliffs is NOT an AMR-type play. It fails the test on the three most critical dimensions: buybacks (diluting instead), debt (overleveraged), and free cash flow (deeply negative).
What CLF IS: A deeply cyclical, overleveraged steel company at a cyclical trough. The assets are real and valuable as a going concern — but the $7.3B debt wall between you and those assets means equity is a levered bet on steel price recovery. If steel prices recover to 2023 levels, CLF could easily trade at $20-30. If they don't, further dilution or even debt restructuring is possible.
What would change our mind:
Decision: DO NOT DEPLOY. Monitor for POSCO deal and steel price recovery. Revisit if debt/equity ratio falls below 80% and FCF turns positive.
Analysis generated 2026-03-12. Data sources: SEC EDGAR XBRL (CIK 764065), yfinance, Cleveland-Cliffs press releases. Saved to /tmp/clf_analysis/. Not investment advice.