SXC — SunCoke Energy, Inc. · 2026-09-06 · Verdict: WATCH · Conviction 3
Price $10.33 (screen row dated 2026-09-05; no live prices available) · Mkt cap $877M · EV ~$1,495M · EV/EBIT n/m (FY25 EBIT -$44.4M), EV/2026E adj. EBITDA 5.8x · FCF yield 17.7% on 2026 guidance (4.8% on CY2025 actual) · Net debt $617.8M · ADV ~$11.3M Sources read: 10-K 2026-02-20 (Items 1, 1A, 7), 10-Q 2026-07-30, DEF 14A 2026-04-01, 8-Ks 2026-04-30 and 2026-07-30 including EX-99 slides, 8-K 2026-05-19, Form 4s (trailing 12m). No transcript was in the bundle, so quotes come from the earnings releases and slides.
1. What the business actually does
SunCoke is the largest independent producer of blast furnace and foundry coke in the Americas, running five US heat-recovery cokemaking plants with about 3,690 thousand tons of nameplate capacity. Most output is sold under long-term take-or-pay agreements: coal, freight, taxes and regulatory costs pass through, and SunCoke earns a fixed fee per ton plus or minus its coal-to-coke yield performance (10-K 2025, Item 1). Industrial Services is three bulk terminals plus fifteen molten slag and mill-services sites in four countries, most acquired with Phoenix Global on August 1, 2025 for $295.8 million (10-K 2025, Item 7).
2. Why it is mispriced — the edge case
Honestly, the edge case named at triage has largely closed. FY2025 shows a $44.4 million operating loss and negative ROIC, but that is a $90.3 million Haverhill I impairment triggered by Algoma Steel's breach, against segment Adjusted EBITDA of $219.2 million (10-K 2025, Item 7). A screen reading GAAP EBIT rejects the company; the market has not. The stock is up roughly 79% in six months and sits about 2% below its 52-week high (screen row, 2026-09-05); the three insiders who bought 27,500 shares in late February 2026 at $5.53 to $5.84, with no sales (Form 4s), have since seen the price roughly double. What is left is a question, not a mispricing: whether the 2026 free cash flow guide of $150 to $160 million (Q2 slides, 2026-07-30) is durable, or is partly the harvesting of a coke fleet that depreciates $171 million a year against $90 to $100 million of capex.
3. Unit economics and growth
Q2 Adjusted EBITDA was $69.6 million versus $43.6 million, with Domestic Coke capacity utilization at 100% versus 95% and favorable coal-to-coke yields, offset by the Haverhill I shutdown and the Middletown turbine failure (10-Q Q2 2026, Item 2). Management raised FY2026 consolidated Adjusted EBITDA guidance from $230-250 million to $250-265 million, Domestic Coke from $162-168 million to $172-178 million, and Industrial Services from $90-100 million to $110-115 million (Q2 slides, 2026-07-30). Industrial Services is now roughly 43% of segment EBITDA against $50.4 million in all of 2024 before Phoenix (10-K 2025, Item 7), so the roughly $60 million increment against $295.8 million paid is near 5x, the best capital allocation decision here.
Customer concentration is extreme and disclosed as such (10-K 2025, Item 1A). The contract book is Indiana Harbor 1,220 kt to Cliffs through September 2035, Middletown 550 kt to Cliffs through December 2032, Haverhill II 500 kt to Cliffs through December 2028, and Granite City 650 kt to U.S. Steel expiring December 2026 while "operating in a turn-down mode in 2026 as part of the contract extension" (Q2 slides, 2026-07-30). Returns are mediocre: 2026E EBIT near $86 million on $1.23 billion of invested capital is about 7% pre-tax.
4. Balance sheet and capital allocation
Gross debt is $660.5 million: $500 million of senior notes due 2029 and $160.5 million drawn on a revolver due 2030, with $207.2 million of liquidity. Net debt is $617.8 million, 2.55x LTM Adjusted EBITDA, guided to 1.83-1.98x by year end. Q2 operating cash flow was negative $27.2 million because roughly $65 million of receipts landed in early July (Q2 slides, 2026-07-30). There is no buyback: treasury stock was unchanged at 15,404,482 shares and issued shares rose slightly on vesting (Q2 balance sheet). The only return is the $0.12 quarterly dividend, the 28th consecutive one, about $41 million a year. Incentives key off Adjusted EBITDA, operating cash flow, pre-tax ROIC and a TSR modifier, with 96.0% say-on-pay support (DEF 14A 2026).
5. Management: said versus did
In April the CEO said Middletown power would resume "late in the second quarter" and that the company would "make up coke production tons during the balance of the year" (Q1 release, 2026-04-30); the turbine returned in May and Q2 utilization hit 100%. The longer record is weaker: the 2025 annual incentive scored Adjusted EBITDA of $177.4 million against a $204.2 million target and operating cash flow of $149.5 million against $157.0 million, and the 2023-2025 PSU paid 82.0% before an 87.3% TSR modifier (DEF 14A 2026). By their own scorecards they missed.
6. Valuation
At $10.33 the EV is about $1.495 billion, or 5.8x the midpoint of 2026 guided Adjusted EBITDA, with a 17.7% FCF yield on the $155 million guide. That guide flatters: it includes an $8-12 million net cash tax receipt and $16 million of favorable non-cash and working capital items (Q2 slides). Normalized unlevered free cash flow is closer to $140 million after roughly $95 million of capex, $15 million of cash taxes and the 14.8% Indiana Harbor minority.
Reverse DCF: at a 10% discount rate the current EV implies that roughly $140 million of unlevered free cash flow simply persists forever with no growth, neither a bargain nor a bubble for a business whose end market is blast furnace steel.
Base (55%): Granite City renewed or its tons redeployed; 2027-28 EBITDA $245 million at 5.5x, net debt $320 million by end-2028. Equity $12.10 plus $1.00 of dividends, about $13.10. Bear (25%): Granite City not renewed and the ovens cannot be cheaply idled (10-K 2025, Item 1A); EBITDA $215 million at 4.5x, about $7.90 including dividends. Bull (20%): renewal at better economics or the U.S. Steel granulated pig iron project (10-K 2025, Item 7) is sanctioned; EBITDA $280 million at 6.5x, about $18.60. Probability weighted: roughly $12.90 against $10.33, near 10% a year over two and a quarter years. A fair return, not a deep value return.
7. Catalysts and timeline
Granite City renewal, most likely between December 2026 and February 2027 since prior extensions were signed in January. Q3 results in late October test the raised guide. Algoma enforcement recovery is unquantified upside, and sub-2x leverage could open a buyback.
8. Kill criteria (pre-registered)
- No announced Granite City extension or replacement tonnage by the FY2026 results release, or an extension that pushes FY2027 Domestic Coke Adjusted EBITDA per ton guidance below $45 (2026 guide is $51-52).
- First FY2027 consolidated Adjusted EBITDA guidance below $230 million.
- Net debt above $530 million at 12/31/2026, against a company guide implying $458-524 million.
- A dividend cut, or insider net open-market selling above $1 million in any trailing six months.
9. Verdict and summary
WATCH, conviction 3. SunCoke is a real operating recovery that the market has already found: Phoenix Global added roughly $60 million of annual EBITDA for $296 million, coal-to-coke yields turned favorable, the Middletown turbine is back and guidance went up twice, so the stock trades 2% off its 52-week high after a 79% six-month run and the February insider buys at $5.53 to $5.84 are long since in the money. At 5.8x guided EBITDA with a 17.7% headline FCF yield it still looks cheap, but that yield is flattered by a cash tax refund and working capital, depreciation of $171 million runs nearly double the $90-100 million of capex on batteries that get closed rather than rebuilt when a contract dies (Haverhill I, Q1 2026), and 650 thousand tons of Granite City capacity comes off contract with U.S. Steel in December 2026 while already running in turn-down mode. Probability weighting the renewal gets to about $12.90 against $10.33: a fair price for a decent business with a binary event inside ninety days, not a margin of safety. A multi-year Granite City renewal at similar economics, or a materially lower price, would make this an IDEA.
Research for discussion, not investment advice. Positions and sizing are the reader's decision.