Initiated 2026-10-06 · Price $24.67 (close 2026-10-06, stockanalysis) · Mkt cap $3.17bn · Energy / coal (Illinois Basin and Appalachian thermal; oil & gas mineral royalties) · Model: verified

Rating: Sell — Conviction: High

ARLP is the best-run, lowest-cost thermal coal producer in the Illinois Basin. It pays a 9.7% distribution covered 1.2–1.4x. But the units are priced as if that cash stream were flat. The coal underneath is a utility-contract business whose prices reset lower as legacy contracts roll off, and its roughly $320mm of mine-closure and black-lung liabilities sit outside most headline multiples. Valued as a slowly declining annuity, every base-case path sits well below the price, so the call is Sell.

Conviction tests (3a-v-c): T1 pass (base −36% at +1pp, −26% at −1pp) · T2 pass (Gordon −34%, exit −28%) · T3 pass (one est input, AllDale borrowings, moves base 9.4%) · T4 pass

The rule gives High. I'd call it Medium: the call rests on a −2% terminal decline and on deducting legacy liabilities. Without the legacy liabilities, Base is −21%.

Model value range vs price
Bear $3.33Base $16.95Bull $35.07Price $24.67

Business overview

ARLP sold 32.97mm tons of coal in FY2025: 26.49mm from the Illinois Basin at $50.97/t and 6.48mm from Appalachia at $82.53/t. Coal sales were $1,932.5mm of $2,194.8mm total revenue. Nearly all of it is thermal coal sold to US utilities under term contracts, and only ~3.0mm of the 34.3mm tons committed for 2026 go to export. The second leg is mineral royalties: oil & gas royalties of $137.8mm in FY2025, plus coal royalties. Q2 2026 oil & gas segment EBITDA was a record $38.0mm of $185.7mm consolidated Adjusted EBITDA, and the $206.2mm AllDale III & IV acquisition (closed 2026-07-01) roughly lifted 2026 oil volume guidance from 1.58mm to 2.0mm bbl. Three variables drive earnings: re-contracted coal prices as legacy contracts roll off, segment cost per ton (Q2 $38.68, FY2025 $40.27), and oil and gas prices on the royalty book.

Competition

In the Illinois Basin ARLP produced ~26mm tons in 2025, ahead of Foresight (14.9mm), Prairie State (6.3mm), Peabody (5.6mm), Knight Hawk (4.2mm) and Sunrise (4.0mm). The basin's major mines made 61mm tons of the Interior region's 82.5mm, out of 527.5mm tons of US production (Coal Age, 2025 data). The real competitor for a utility's dispatch is natural gas. Pressure shows up first in price at re-contracting, not in market share: Illinois Basin realization fell 7.2% in FY2025 as higher-priced legacy contracts expired, and Q2 2026 was only +0.5% year on year. Per ARLP's Q1 2026 release, PJM's 2027/28 capacity auction cleared at the cap and selected every megawatt of coal capacity. The closest listed Illinois Basin comparable, Hallador (HNRG), trades at 12.6x trailing EV/EBITDA, but that multiple sits on depressed earnings and on its power plant.

Bull case

  1. Power demand keeps Illinois Basin plants running. Data-center and industrial load in PJM and MISO defers coal retirements. Volume holds at or above 35mm t, and 2027–28 contracts re-price flat to up. Plays out if utilities extend unit lives and reserve margins stay tight. Model: volume_growth, realized_price, exit_ev_ebitda
  2. The royalty leg grows. With a full year of AllDale and Brent near $100 (macro log, 2026-10-05), oil & gas royalties add ~$1.5/t of blended revenue in 2027. Plays out if Hormuz disruption keeps oil elevated. Model: realized_price
  3. Costs improve. Tunnel Ridge's new district and Hamilton's longwall return keep segment cost at or below $37.50/t. Plays out if no longwall setbacks come with the 2027 moves. Model: unit_cash_cost

Bear case

  1. Thermal decline resumes. Once the load-growth premium fades, utilities restart retirements, and volume falls ~4% a year from 2027. Model: volume_growth, exit_ev_ebitda
  2. Contracts re-price lower and oil normalizes. Only 29.4mm t is priced for 2027, versus 34.3mm committed for 2026. Legacy contracts keep rolling off, as in FY2025's −7.2% Illinois Basin price, and royalty income falls with oil. Model: realized_price
  3. Cost and legacy burden. Wage inflation and longwall moves push cost toward $42/t. Maintenance capex runs ~$240mm a year, and $318.8mm of ARO, black-lung and workers' comp obligations absorb cash. Model: unit_cash_cost, fixed_costs, capex_pct_rev; liabilities via other_claims

Valuation & balance sheet

Metric (definition) Current Own history (range or 5y avg) Peers Source, as-of
EV/EBITDA, model basis (mkt cap − cash + funded debt incl. est. AllDale borrowings + legacy liabilities) 5.98x FY2025 · 5.56x base 2026 n/a (unverified) — model; 8-K; Q2 10-Q
EV/EBITDA trailing (stockanalysis) 5.42x n/a (unverified) CNR 7.98x · BTU 10.47x · HNRG 12.62x (BTU, HNRG on depressed EBITDA) stockanalysis, 2026-10-06
Forward P/E (stockanalysis consensus) 8.10x n/a (unverified) CNR 18.79x · BTU 23.21x · HNRG n/m stockanalysis, 2026-10-06
FCF yield / distribution yield (stockanalysis, TTM) 10.72% / 9.73% n/a (unverified) CNR 5.87% / 0.44% · BTU −5.71% / 1.18% · HNRG −7.15% / — stockanalysis, 2026-10-06

Model-implied value range (from model-summary.json; energy module, DCF (Gordon) and DCF (exit multiple), midpoints): Bear $3.33 · Base $16.95 · Bull $35.07 per unit, i.e. implied returns of −87% / −31% / +42% vs $24.67. These ranges show how the bull and bear drivers translate into value; they are not price targets.

The price sits about 40% of the way from Base to Bull. The market is capitalizing ~$410mm of annual free cash flow as if it would last, while Base has it fading to ~$300mm by year 5 and declining 2% a year after that.

Re-rating: the Base exit of 4.75x is a 21% de-rating from today's 5.98x on FY2025 EBITDA and 15% below the 5.56x on base 2026 EBITDA (~$751mm). The two methods agree within 10% (exit $17.73 vs Gordon $16.16), so both embed the de-rating.

Tail sensitivity (Illinois Basin is ~70% of revenue, and the bear case is a plateau, not a break): blended price $63→$55/t, cost $39→$45/t, volume −6% then −8% a year, NWC 9% of revenue, exit 3.0x and WACC 12.25% put the Bear value at $0.00.

Balance sheet: net leverage 0.86x and coverage 12.6x (FY2025 Adjusted EBITDA $698.7mm, against $590.0mm of debt at 2026-06-30 plus $206.2mm est. AllDale borrowings, less $111.2mm cash). Company-reported total leverage was 0.82x at Q2, before AllDale. Liquidity $424.0mm at Q2, of which $305.5mm was revolver availability. Nearest material maturity: the $400mm 8.625% notes due 2029-06-15; the $56mm securitization facility renews sooner. Ratings: n/a (unverified); none retrieved.

Model note: verified (LibreOffice matched all 3,511 formula cells). Unverified inputs: the AllDale borrowing tranche ($206.2mm, the full purchase price, since the cash/debt split isn't disclosed; ~$1.60/unit). Assumptions without basis: none. No scenario consistency CHECKs. Historical EBITDA is company Adjusted EBITDA, which adds back impairments and digital-asset marks. Projected revenue is blended per coal ton (coal plus royalty and other revenue) and excludes pass-through transportation. Fixed costs ($170mm) are a judgment calibrated to H1 2026's run-rate. Bitcoin ($37.9mm) and equity-method investments ($71.3mm), together ~$0.85/unit, are not credited; legacy liabilities ($318.8mm, ~$2.48/unit) are deducted. NCI isn't deducted. There's no EPS guidance. Base Y1 EPS of $2.78 compares with $2.42 of FY2025 GAAP net income per unit and $0.68 in H1 2026, which carried Q1's Hamilton longwall move. Non-note maturities are placeholders. WACC 10.70% is CAPM at a 5.31% 10Y and a judgment beta of 1.2 (reported 0.31), weighted 84.3/15.7 with 7.44% after-tax debt.

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 − Sentiment de-rating; contracted utility volumes unchanged Low −$0.99
S2 Slow bear / recession − Load falls, utilities defer shipments, oil royalties drop. Milder than Base − Bear ($13.62) because the contract book cushions a two-year hit; the bear case is a permanent decline High −$5.22
S3 Rapid rate shock − Higher discount rate on a fading stream; little floating debt Low −$1.22
S4 Slow rate grind − Sticky wage and supply inflation; fixed-price contracts lag Med −$2.79
S5 Soft-landing cuts + Cheaper money, steady power demand Low +$1.16
S6 Recession-driven cuts − Volume and royalty loss outweigh lower rates High −$4.13
S7a Credit liquidity shock − Sentiment-only; 2029 notes far off, leverage <1x Low −$0.86
S7b Slow default cycle − Utility counterparties mostly investment grade; mild price and funding hit Low −$1.04
S8 Stagflation + Oil royalties and coal re-pricing lift revenue; mining costs rise in step Low +$0.66
S9a Dollar spike − Only ~9% of 2026 tons exported; small netback hit Low −$0.24
S9b Dollar slide + Small export netback gain Low +$0.24
S10 Melt-up + Low-beta income name participates little Low +$0.49
S11 Energy supply shock + Royalties reprice at once; gas-to-coal switching supports coal; diesel costs rise Low +$1.21
S12 Mega-cap/AI derating − Data-center power-demand premium unwinds Low −$0.49

Currently active/on watch per the playbook: S3 active (partially, per state.md); S8, S10 and S11 on watch. S11 is the name's largest positive row, but it's small.

Model value change vs Base, by scenario
S2 Slow bear / recession−$5.22S6 Recession-driven cuts−$4.13S4 Slow rate grind−$2.79S3 Rapid rate shock−$1.22S7b Slow default cycle−$1.04S1 Fast equity crash−$0.99S7a Credit liquidity shock−$0.86S12 Mega-cap/AI derating−$0.49S9a Dollar spike−$0.24S9b Dollar slide+$0.24S10 Melt-up+$0.49S8 Stagflation+$0.66S5 Soft-landing cuts+$1.16S11 Energy supply shock+$1.21

What would change the call

Upgrades if: 2027 priced tons pass ~32mm t with the Illinois Basin price at or above $51/t, and segment cost holds at or below $38/t, or the price falls into the Base range. Downgrades if: this is already Sell. The call strengthens if Q3 shows 2027 contracts re-pricing below $50/t in the Illinois Basin or distribution coverage dropping below 1.1x.

Watch items

Sources