Initiated 2026-10-07 · Price $25.43 (close 2026-10-06, stockanalysis) · Mkt cap $3.10bn · Energy / coal (US thermal, seaborne thermal and metallurgical) · Model: verified

Rating: Sell — Conviction: Medium

Peabody runs the biggest US coal mine (North Antelope Rochelle, 65.0mm tons in 2025), and the stock is priced for a recovery that hasn't shown up in the numbers yet. Adjusted EBITDA fell from $1.36bn in FY2023 to $455mm in FY2025 and was $106.5mm in H1 2026. The price assumes Centurion, the new Queensland met mine, ramps cleanly and that met prices stay firm. Even if met prices hold at today's levels, the Base case still sits well below the price once ~$390mm of legacy and minority claims are counted. An unresolved arbitration claim from Anglo American adds downside the cases don't include. Sell.

Conviction tests (3a-v-c): T1 pass (Base −46% at +1pp, −38% at −1pp) · T2 pass (Gordon −52%, exit −32%) · T3 pass (all FY0 facts and historicals from EDGAR and 8-K releases; no est inputs) · T4 fail — Anglo American ICC arbitration over the terminated $3.8bn deal, reported ~$755mm claim (~$6/sh)

Model value range vs price
Bear $0.00Base $14.77Bull $49.75Price $25.43

Business overview

Peabody sells thermal coal to US utilities and thermal and steelmaking (met) coal into Asia. It reports four segments. FY2025 revenue was $3.86bn, split Powder River Basin (PRB) $1.15bn, Seaborne Metallurgical $1.04bn, Seaborne Thermal $0.91bn and Other U.S. Thermal $0.71bn. PRB is ~70% of tonnage but earns about a dollar a ton. Seaborne operations are ~20% of tons and 51% of revenue (10-K). Earnings turn on three variables: the seaborne met price, met unit cost (which now depends on Centurion's longwall ramp, after Q2 met costs of $155/st exceeded the $148 realized price), and US power burn, which sets PRB volume against a largely fixed cost base.

Competition

In met coal, Peabody competes for Asian and Indian steel mills with BHP, Anglo American, Glencore, Whitehaven and the US producers Warrior (HCC), Core (CNR) and Alpha (AMR). The pressure that matters for Centurion's ramp is new premium supply. Warrior raised 2026 sales guidance to 13.0–14.0mm st in August, with its new Blue Creek mine at ~5mm st and 90% contracted (Q2 release, 2026-08-05). That tonnage competes for the same buyers Centurion has to win. In the PRB, Peabody's rivals (Navajo Transitional, Eagle Summit and others; 10-K) all depend on the same utility load. The pressure shows up first in volume: Q2 PRB shipments were 16.4mm st against 20.0mm a year earlier, on mild weather and plant outages. The closest listed comparable, CNR, trades at 8.1x EV/EBITDA with a 5.8% FCF yield. BTU trades at 10.8x with negative FCF.

Bull case

  1. Centurion ramps on plan — longwall output of 1.5–2.0mm st in H2 2026, rising toward 4.7mm st by 2028 at ~$105/t (2024$), roughly doubles met volume at a better cost position. Plays out if the rock fault zone clears and Q3's guided ~$20/t met cost improvement shows up. Model: volume_growth, unit_cash_cost
  2. Met price recovery holds — PHCC was ~$247.5/t FOB in mid-May on Chinese supply tightness. Q2 realized $148.04/st, up from $114.79 a year earlier. Plays out if Chinese and Indian steel output holds and Queensland supply stays disrupted. Model: realized_price
  3. US power demand supports PRB — data-center load keeps coal plants running, holding PRB at ~85mm st. The market then values Peabody on a mid-cycle multiple, not as a melting ice cube. Plays out if utilities defer retirements. Model: volume_growth, exit_ev_ebitda

Bear case

  1. Centurion geology — Q2 roof-control problems in a rock fault zone cut 2026 Centurion guidance to 2.0–2.5mm st, from 3.5mm in February. If that persists, met costs stay above $125/st and the volume step-up slips. Plays out if Q3 met costs miss the guided improvement. Model: unit_cash_cost, volume_growth
  2. Thermal secular decline — PRB tons fell 18% y/y in Q2 while costs rose above price ($14.06 vs $13.63/st). Plant retirements turn a high-volume, thin-margin business into a cash drain. Plays out if retirements resume on schedule. Model: volume_growth, terminal_growth, exit_ev_ebitda
  3. Seaborne prices revert — met falls back to ~$135/st and export thermal to ~$52 as new supply (Blue Creek, Queensland restarts) arrives. Plays out if Chinese steel output falls. Model: realized_price
  4. Anglo arbitration — Anglo claims Peabody wrongfully invoked a MAC to walk away from the $3.8bn deal. A reported ~$755mm claim is ~24% of market cap. Model: not in the value range — a binary legal outcome; see T4.

Valuation & balance sheet

Metric (definition) Current Own history (range or 5y avg) Peers Source, as-of
EV/EBITDA, TTM (aggregator EV) 10.82x n/a (unverified) CNR 8.07x · HCC 10.78x · AMR 14.04x stockanalysis, 2026-10-06
EV/EBITDA, model basis (mkt cap − cash + debt + other claims ÷ FY2025 Adj. EBITDA) 7.33x — — model inputs, 2026-10-07
FCF yield, TTM −5.53% — CNR 5.80% · HCC 0.32% · AMR −0.06% stockanalysis, 2026-10-06
Forward P/E 23.2x — CNR 18.8x · HCC 14.6x · AMR 51.1x stockanalysis, 2026-10-06

Peer multiples are all on trough-ish earnings; AMR's is distorted by near-zero EBITDA.

Model-implied value range (from model-summary.json; energy module, DCF (Gordon) and DCF (exit multiple), midpoints): Bear $0.00 · Base $14.77 · Bull $49.75 per share, i.e. implied returns of −100% / −42% / +96% vs $25.43. These ranges show how the bull and bear drivers translate into value; they are not price targets. The price sits about 30% of the way from Base to Bull, so the market is paying for a good part of the Centurion-and-met-price recovery already. Base's 5.0x exit multiple is a 32% de-rating from today's 7.33x on the model's basis. The Gordon method ($12.31, at −2% terminal growth) embeds a lower terminal multiple than the exit method ($17.23), so the 40% gap between them is a further de-rating in the Gordon method, not noise. Bear prints $0.00 because the floor binds: in that case segment margins shrink to ~$75–180mm a year, below ~$110mm of corporate costs. FCF runs −$190mm to −$330mm a year, and $390mm of other claims sit ahead of equity. That is a cash-burning business, not a merely weaker version of Base. No single product is ≥50% of revenue, so no tail sensitivity was run.

Balance sheet: net leverage −0.61x (net cash), coverage not meaningful, liquidity $959.1mm (Q2 release), nearest material maturity is the $78.8mm 3.25% converts due 2028 (in the money at ~$19.09; modelled as converted), then $250mm 0.50% converts due 2031. Ratings: n/a (unverified). Off the debt line: ARO ~$692mm and retiree medical $108mm against $460mm of restricted reclamation collateral (Q2 10-Q).

Model note: built and LibreOffice-verified (3,511 cells match). No unverified inputs; no assumptions without basis; no scenario CHECKs. Other claims ($389.9mm) = NCI + ARO + OPEB − restricted collateral, a judgment netting. Peabody gives no EPS guidance, so there is no guidance line to reconcile (Base Y1 EPS −$0.64). The Anglo claim and the Centurion securities suit are not modelled.

Scenario stress test

Reasoned from the bull/bear drivers above. The model column comes from the scenario overlays (Base case + shock).

Scenario Effect Mechanism Magnitude Model Δ value vs Base ($/sh)
S1 Fast equity crash − Risk-off de-rating; contracts and coal demand unchanged Low −$1.05
S2 Slow bear / recession − Steel and power demand fall; met and export thermal prices drop, PRB burn falls. Milder than Base − Bear ($14.77) because it is a two-year hit; the bear case is permanent High −$8.65
S3 Rapid rate shock − Higher discount rate on a short, declining cash flow stream; debt is fixed-rate Low −$0.90
S4 Slow rate grind − Sticky diesel, explosives and wage inflation on thin margins, plus higher rates High −$6.14
S5 Soft-landing cuts + Cheaper money, steady steel and power demand Low +$1.96
S6 Recession-driven cuts − Demand and price loss outweigh lower rates High −$7.99
S7a Credit liquidity shock − Sentiment de-rating; net cash, no funding stress Low −$0.84
S7b Slow default cycle − Weaker steel and utility counterparties, softer pricing Med −$3.43
S8 Stagflation + Energy prices outrun mining cost inflation; US thermal contracts lag Med +$2.36
S9a Dollar spike − ~51% of revenue seaborne; netbacks fall vs Australian supply Med −$2.43
S9b Dollar slide + Seaborne netbacks rise; AUD cost base partly offsets Med +$2.43
S10 Melt-up + Low-beta name participates little Low +$0.52
S11 Energy supply shock + Oil/LNG disruption lifts Newcastle thermal through gas-to-coal switching; diesel costs rise Med +$3.98
S12 Mega-cap/AI derating − Data-center power-demand premium on US thermal unwinds Low −$0.52

Currently active/on watch per the playbook: S3 partially active; S8, S10, S11 on watch.

Model value change vs Base, by scenario
S2 Slow bear / recession−$8.65S6 Recession-driven cuts−$7.99S4 Slow rate grind−$6.14S7b Slow default cycle−$3.43S9a Dollar spike−$2.43S1 Fast equity crash−$1.05S3 Rapid rate shock−$0.90S7a Credit liquidity shock−$0.84S12 Mega-cap/AI derating−$0.52S10 Melt-up+$0.52S5 Soft-landing cuts+$1.96S8 Stagflation+$2.36S9b Dollar slide+$2.43S11 Energy supply shock+$3.98

What would change the call

Upgrades if: Centurion sells ≥1.5mm st in H2 2026 and met segment costs fall below $125/st for two quarters. Or the arbitration resolves for ≤$100mm. Or PRB volume recovers to a ≥85mm st run-rate with costs below price. Downgrades if: n/a; already Sell. The Sell would be confirmed by Q3 met costs above $140/st, a 2026 Centurion guide below 2.0mm st, or an adverse arbitration award.

Watch items

Sources