Coal — A Complete Market Guide (2026)

Data as of 6 July 2026. Prices are quoted as multi-year and full-year averages, not a single day’s snapshot, so this report stays useful over time. Reserves, production splits, balances and historical series are estimates from agency data, rounded for clarity. This report is for information only and was prepared with AI assistance — see the disclaimer at the end.

Coal is the fuel the modern world was built on — and the one it keeps declaring dead while burning more of it than ever. In 2025 the coal market set another all-time record: 8.85 billion tonnes consumed, more than every other mined commodity in this series combined, generating a third of the world’s electricity and the coke behind most of its steel. It is also the most polarised market in commodities: written off by Western capital, priced for extinction, and yet — as 2022’s energy crisis proved, when its price quadrupled — still the system’s fuel of last resort. This report is the free, big-picture primer on how the coal market actually works — thermal versus coking, the seaborne trade and its chokepoints, who mines it and who quietly finances it, and what actually drives the cycle. For the company-level data behind the charts — every producer screened by production, reserves and cost — go to Metal Pilot.

TL;DR & Key Takeaways

• What it is: combustible sedimentary rock, sold as two nearly separate commodities — thermal coal (power generation, ~⅔ of demand) and metallurgical (coking) coal (steelmaking’s carbon source) — graded by energy content and coking quality.
• Market structure: production is enormous but home-bound — China mines and burns roughly half the world’s coal, India another eighth — while the price is set in the seaborne trade (~1.47 Gt): Indonesia and Australia export most of it, China and India buy most of it, and there is no cartel — Beijing’s production and import policy is the closest thing coal has to supply management.
• Demand story: a plateau, not a cliff. The IEA’s base case has global demand flat-to-slightly-declining through 2030 — advanced economies’ collapse offset by India (+200 Mt by 2030) and Southeast Asia — while the demand quality erodes as renewables cut coal’s power-sector share.
• Price regime: violently cyclical with an energy-crisis kicker: Newcastle thermal averaged ~$26/t in 2000, $138 in 2008, $63 at the 2015 trough, an unprecedented $384 in 2022 (spot ~$457) and ~$112 in 2025 — while the sector’s equities swung from mass bankruptcy (2016) to record dividends (2022–23).
• Biggest swing factor: Chinese import appetite (a half-billion-tonne swing variable) against seaborne supply discipline — with gas prices setting the substitution ceiling and transition policy the long-run floor.

Numbers to remember (coal at a glance)

Figure 1. Coal at a glance

Figure data: IEA Coal 2025 , IMF/FRED and Energy Institute Statistical Review ; see Sections 1.3–2.7.

Why it matters now: coal sits at the centre of the energy transition’s great contradiction — the fastest-growing electricity systems on Earth still run on it, Western divestment has starved the sector of capital exactly as its cash flows peaked, and every energy crisis reprices it overnight. For investors, that combination — structural decline priced in, cyclical scarcity not — has made coal the most contrarian and, episodically, the most profitable corner of the commodity complex. The big-picture case is below.

How to read this guide: new to coal? Read straight through from Section 1. Here for the investment case? Jump to Section 4 — how to own it, how to screen the miners, and what drives the cycle. Every technical term is defined in the Vocabulary (Section 7).

1. Coal & the market basics

1.1 What coal is — physical basics & quality

Coal is fossilised plant matter, compressed over hundreds of millions of years into a combustible rock whose rank — the degree of that transformation — defines its uses and value. The rank ladder runs from lignite (brown coal: young, wet, low-energy, burned at the mine mouth because it isn’t worth shipping) through sub-bituminous and bituminous coal (the workhorses of power generation and the seaborne trade) up to anthracite (the hardest, highest-carbon rank, a specialty product). Cutting across rank is the market’s fundamental split — two commodities sharing one name:

• Thermal (steam) coal — burned to raise steam for electricity and heat: roughly two-thirds of world demand. Priced on energy content (kilocalories per kilogram), with the seaborne benchmarks at 6,000 kcal/kg (Newcastle), 5,500 (the China-India workhorse grade) and 4,200 (Indonesian low-rank). Sulfur and ash content set discounts.
• Metallurgical (coking) coal — baked in coke ovens into coke, the carbon backbone and structural support of the blast furnace: ~15% of demand but a distinct, premium market. Its quality axis is coking behaviour (measured by coke strength after reaction, fluidity, vitrinite content), splitting the market into hard coking coal (HCC) — premium low-volatile Australian material the world’s steel mills are built around — semi-soft, and PCI (pulverised injection coal). Met coal demand is steel demand, one step from the iron ore guide’s blast furnaces.

Coal’s economics are logistics economics. It is cheap per tonne and heavy, so most coal never travels: China and India burn their own production near the mine, and only ~16% of world output reaches the sea. That seaborne minority, though, sets the global price — and its infrastructure (dedicated rail corridors, export terminals, capesize and panamax fleets) is the industry’s real capital stock. The value chain is short and blunt: mine (open-cut or underground) → washing/preparation → rail → port → power plant or coke oven. Nothing about it is technologically scarce; everything about it is politically and financially contested — coal combustion produces ~15 Gt of CO₂ a year, about 40% of energy-sector emissions, making it decarbonisation’s first target and the divestment movement’s founding cause.

Figure 2. The coal value chain — two commodities, one logistics chain

Source: industry value-chain primers. The single logistics chain splits at conversion into the thermal route (power and cement, ~⅔ of demand) and the metallurgical route (coke ovens for blast-furnace steel), described in the text above.

1.2 Units & measurement conventions

This report uses the trade’s conventions, stated here once. Quantities are million and billion tonnes (Mt, Gt) of product coal; note that global statistics mix physical tonnes of very different energy content (a tonne of lignite carries roughly half the energy of Newcastle-spec coal), which is why the IEA also tracks coal in energy terms. Prices are US dollars per tonne (USD/t), always tied to a quality basis: the thermal benchmarks quote kcal/kg NAR (net as received — the energy actually delivered; GAR, gross as received, reads higher for the same coal, a classic trap in Indonesian grades). This report’s reference series is Newcastle 6,000 kcal/kg NAR FOB Australia for thermal and premium hard coking coal (PLV) FOB Australia for met. Freight (panamax and capesize routes from Kalimantan, Newcastle and Richards Bay) links FOB to delivered prices. Flow vs. stock: production and demand are flows; reserves (~1,070 Gt — over a century of output) and stockpiles (Chinese state reserves, mandated power-plant inventories) are levels.

Table 1. Coal units and conventions

Source: IEA Coal 2025 conventions; index specifications per globalCOAL/Argus/Platts methodologies.

Numbers intuition: a 1 GW coal power plant burns roughly 3 Mt a year; one tonne of steel needs ~770 kg of coking coal; a capesize cargo is ~170,000 t. The seaborne trade (~1.47 Gt) was worth roughly $180–200 billion in 2025 across thermal and met — and at 2022’s crisis prices the same tonnage grossed over half a trillion dollars, the windfall that rebuilt every surviving coal balance sheet.

1.3 Pricing & benchmarks — two markets, many indices

Coal pricing is regional, quality-specific and index-based — there is no single “coal price.” The thermal complex hangs on three physical marker prices: Newcastle 6,000 NAR FOB (the Pacific and global reference, assessed by globalCOAL/Argus/Platts and traded as liquid ICE futures), API2 (CIF Rotterdam — the Atlantic marker, now pricing a shrinking European market) and API4 (FOB Richards Bay), plus the Indonesian ICI ladder for the low-rank grades that dominate real tonnage and China’s domestic 5,500 kcal Qinhuangdao price — which, because China is half the market, quietly disciplines them all: when Chinese domestic prices sag, import demand evaporates and the seaborne complex follows. The met complex prices off PLV HCC FOB Australia (Platts/Argus daily assessments, futures on SGX/CME), with CFR China its shadow. The two complexes usually move together in crises and separately otherwise — met follows steel, thermal follows power and gas. And gas is the hidden hand of thermal pricing: coal-to-gas switching in Europe and Asia ties Newcastle and API2 to TTF and JKM gas prices (the mechanism that transmitted 2022’s gas panic into coal’s greatest bull market — see the natural gas guide ).

Table 2. Key coal benchmarks & price markers

Source: index methodologies (globalCOAL, Argus , Platts) and exchange specifications, 2025.

The long price story is three shocks on a rising floor. Through the early 2000s Newcastle sat near $26–33/t; China’s flip to net importer drove the 2008 spike ($138 average) and the post-crisis supercycle ($130 in 2011), before the long bust ground the market to $63 by 2015 — the trough that bankrupted half the US industry. The 2016–18 recovery (Chinese supply reform) gave way to the pandemic low ($62 in 2020), and then the extraordinary part: Russia’s invasion of Ukraine, a global gas panic and a post-COVID power surge produced 2022’s $384 average — with spot near $457 — the highest price, in real terms, coal has ever traded. The normalisation since ($188 in 2023, $144 in 2024, $112 in 2025) still leaves prices well above every pre-crisis norm, cushioning exporters even as volumes plateau. Met coal ran the same arc harder: PLV averaged ~$365 in 2022 and ~$190 by 2025 as Chinese steel softened.

Table 3. Average annual Newcastle thermal coal price, 2000–2025 (USD/t, 6,000 kcal NAR FOB)

Source: IMF Primary Commodity Prices / Global price of Coal, Australia , 2000–2025. Calendar-year averages, rounded; Newcastle-basis Australian thermal coal.

Figure 3. Newcastle thermal coal price, milestone years 2000–2025 (USD/t)

Figure data: Table 3 (full annual series). Milestone years shown; the 2022 average of $384 came with a ~$457 spot record.

2. Supply, demand & the market balance

2.1 Where coal is mined — basins & geology

Coal is the most widely distributed fossil fuel — every inhabited continent mines it — but production concentrates where thick, shallow seams meet demand or ports. China dominates absolutely: ~4.8 Gt a year, roughly half the world’s output, led by the great northern provinces (Shanxi, Shaanxi, Inner Mongolia) whose mega-mines feed the world’s largest rail corridors. India’s state-run belt (Jharkhand, Odisha, Chhattisgarh) produces over 1 Gt with the world’s biggest open-cuts. The export basins are different animals, built around quality and water access: Indonesia’s Kalimantan (low-rank thermal, barge-fed, the biggest export flow on Earth), Australia’s Hunter Valley (premium thermal through Newcastle) and Bowen Basin (the world’s coking-coal heartland), Russia’s Kuzbass (rail-constrained), South Africa’s Mpumalanga, Colombia’s Cesar-Guajira, the US Powder River Basin (vast, cheap, landlocked) and Appalachia (met). Mongolia’s Tavan Tolgoi ships premium coking coal to China by truck and rail — the market’s newest big corridor.

Table 4. Leading coal-mining countries, 2024–25 (estimated)

Source: IEA Coal 2025 and Energy Institute Statistical Review 2025 ; 2024 actuals with 2025 records noted. Figures rounded, physical tonnes.

Figure 4. Leading coal-mining countries, 2024–25 (Mt)

Figure data: Table 4.

At the asset level coal has true giants — single mines producing more than entire exporting countries. The table lists the flagship operations across the major systems; note the split personality: the biggest thermal mines are volume machines feeding domestic power, while the Bowen Basin’s met mines earn multiples of their tonnage in value.

Table 5. Major coal mines & complexes (approximate recent annual output)

Source: company reports and national statistics, 2024–25; output approximate and rounded (ROM vs saleable bases differ). Screen producers and assets on Metal Pilot.

2.2 Demand & consumption

Coal demand hit 8.85 Gt in 2025 — an all-time record — and its composition explains both its resilience and its fate. Power generation takes about two-thirds: coal still produces ~34% of the world’s electricity, and in the fastest-growing power systems (India, Indonesia, Vietnam) it remains the base of the stack even as renewables take the growth. Steel takes ~15% via coking coal — demand that tracks the blast-furnace economics of the iron ore guide and shares its long-run EAF threat. Cement and industry absorb most of the rest. The geography is the starkest in this series: China burns ~56% of the world’s coal — more than the rest of the world combined — with India (~14%) second and rising; the entire OECD, which built the industry, now accounts for well under a fifth and shrinks every year. The IEA’s base case sees the total flat through the late 2020s, easing ~3% by 2030: China plateauing (its 2025 use was unchanged), India adding 200+ Mt by 2030 (+3%/yr), Southeast Asia growing, and advanced economies continuing their collapse. Coal demand no longer grows — but it is not yet falling either, and that stubborn plateau is the market’s defining fact.

Table 6. Global coal demand by end use, 2024–25 (share of consumption)

Source: IEA Coal 2025 and IEA World Energy Outlook sector splits; shares approximate, rounded.

Figure 5. Global coal demand by end use, 2024–25

Figure data: Table 6.

Table 7. Coal consumption, world and leading consumers, selected years (Mt)

Source: IEA Coal 2025 and Energy Institute Statistical Review historical series; physical tonnes, approximate and rounded — energy-content trends differ slightly.

Figure 6. The great rotation of coal demand, 2000–2025 (Mt)

Figure data: Table 7. Bars are scaled to each year’s composition (the world total rose from ~4.7 Gt in 2000 to ~8.85 Gt in 2025); segment values are absolute Mt.

2.3 Supply: producing countries & the export layer

Coal’s supply pyramid inverts iron ore’s: the giants barely export. China’s ~4.8 Gt and India’s ~1.1 Gt are almost entirely domestic — state-directed systems run for energy security, whose production policy (Beijing’s post-2021 output surge added the equivalent of Australia’s entire industry in two years) sets the context for everyone else. The seaborne layer (~1.47 Gt in 2025, shrinking ~5% from 2024’s record) is where investors live: Indonesia (~555 Mt) dominates thermal volume, Australia (~363 Mt) dominates value (premium thermal plus two-thirds of seaborne coking coal), with Russia (~198 Mt, sanctioned westward, rail-capped eastward), the US (a swing exporter), Colombia, South Africa and Mongolia’s overland met flow completing the map. Reserves are geologically irrelevant to any investment horizon — ~1,070 Gt, over a century of production — led by the US, Russia, Australia, China and India; what is scarce is permitted, financed, port-connected capacity, because Western banks, insurers and governments stopped funding new thermal mines a decade ago. That capital starvation is the supply side’s defining feature: the seaborne industry is running off its existing asset base, which is precisely why crisis prices spike so hard.

Table 8. World coal production, selected years (Mt)

Source: IEA and Energy Institute Statistical Review series; physical tonnes, rounded.

Table 9. Coal reserves by country (Gt)

Source: Energy Institute Statistical Review (proved reserves); over a century of current production — reserves are not the constraint, capital and permits are.

Figure 7. The seaborne export layer, 2024 (Mt)

Source: IEA Coal 2025 trade data and national statistics, 2024; approximate.

2.4 The supply–demand balance

Coal’s balance is set in the seaborne market, and its swing variable wears one flag: China’s import decision. Chinese imports are not structural need — they are an arbitrage between domestic mine prices, safety-campaign-driven supply swings and the seaborne price, which makes them the market’s shock absorber and its shock generator at once: ~543 Mt in 2024 (an all-time record for any country), then a ~53 Mt cut in 2025 as record domestic output crowded imports out — a single-year swing bigger than Colombia’s entire industry. Around that pivot: India’s imports grind higher with its power build-out (~240–250 Mt), Japan-Korea-Taiwan shrink slowly, and Vietnam and Southeast Asia are the growth buyers. The 2025 result was the first genuine seaborne surplus since the crisis — trade down ~5% to 1,468 Mt, Newcastle back near $100 — resolving, as always, through the cost curve: Russian exporters (squeezed by rail costs and sanction discounts) and high-cost Indonesian and Australian tonnes idle first, and Glencore-style discipline (up to 10 Mt of announced Colombian cuts) does the rest. The buffer stock sits in plain sight: Chinese port and plant inventories (typically 300+ Mt across the system) and mandated utility stockpiles — the balance’s thermometer, watched the way copper traders watch LME sheds.

Table 10. Seaborne coal trade & the Chinese pivot, 2015–2025 (Mt)

Source: IEA Coal 2025 and Coal Mid-Year Update 2025 ; China customs. Rounded; trade includes overland flows in IEA totals — treated here as the traded market.

Figure 8. China’s import pivot, 2015–2025 (Mt)

Figure data: Table 10.

At the country level, the net positions split the world into three camps: the home giants (China, India — massive producers and the biggest importers, because coastal plants buy seaborne coal even as inland mines overproduce), the export specialists (Indonesia, Australia, Russia, Colombia, Mongolia — mining multiples of their own consumption), and the pure importers (Japan, Korea, Taiwan, Vietnam, and what remains of European demand).

Table 11. Coal net positions, major players, 2024–25 (Mt)

Source: IEA Coal 2025 and EI Statistical Review , 2024–25; approximate, rounded. China and India also import heavily at the coast while their inland mines overproduce, so their net-position figure shows gross seaborne imports — the market-watched swing — rather than the residual of these rounded production and consumption estimates.

Figure 9. Coal net positions, 2024–25 (Mt)

Figure data: Table 11.

2.5 Supply structure: thermal vs met, state vs seaborne

Coal’s supply structure is best read as a two-by-two: thermal vs metallurgical on one axis, state-domestic vs seaborne-commercial on the other. The state-domestic quadrant (China’s provincial giants, Coal India’s monopoly) is ~80% of tonnage but operates outside market economics — production targets, price caps and energy-security mandates rule it, and its spillovers (China’s 2021–25 output surge) dominate everything else. The seaborne thermal quadrant is the investable core: Indonesian low-rank volume (price-elastic, cheaply expanded — the closest thing coal has to swing supply), Australian premium tonnes, and the Atlantic rump. The seaborne met quadrant is the premium franchise: Australia’s Bowen Basin supplies about two-thirds of traded coking coal, with the US, Canada (Glencore’s Elk Valley, acquired from Teck in 2024) and Mongolia the balance — an oligopoly-quality market whose prices routinely triple thermal’s. Unlike every metal in this series, there is no recycling and no secondary supply — coal burns once — so the only supply responses are mine economics and policy. And uniquely, new supply is capital-rationed by choice: Western financing for greenfield thermal mines has effectively ceased, making existing export assets quasi-annuities — depleting, un-replicated, and cash-generative far beyond their book values.

Table 12. The coal supply matrix, 2024–25

Source: IEA Coal 2025 trade splits (thermal 1,111 Mt, met 357 Mt in 2025); structure per industry reporting. Approximate.

2.6 Trade flows & chokepoints

The seaborne map is a Pacific story with an Atlantic remnant. The great flows: Kalimantan → China and India (barges to anchorage transshipment — the highest-volume coal corridor on Earth), Newcastle/Hunter → Japan, Korea, Taiwan, China (premium thermal), Bowen Basin → every blast furnace economy (met), Russia → China (eastbound rail through the strained BAM/Trans-Siberian corridors — the sanctions-era reroute), Mongolia → China (the Gashuunsukhait truck-and-rail met corridor), and the Atlantic’s survivors (Colombia and South Africa serving residual European and Mediterranean demand, plus US swing cargoes). The chokepoints are real and mostly inland: coal moves through a handful of mega-ports and rail arteries whose capacity and politics bind the trade far more than any strait — though the Malacca route carries most North Asian imports and the Panama Canal’s 2023–24 drought restrictions (a concluded episode) showed how quickly logistics can reprice US-Asia arbitrage.

Table 13. Coal trade chokepoints & corridors

Source: port and rail operator disclosures, IEA Coal 2025 and trade reporting, 2024–25. Volumes approximate.

Figure 10. Major coal trade corridors by volume (Mt, approx.)

Source: corridor volumes approximated from Table 13, Figure 7 and IEA Coal 2025 trade data . “Indonesia → Asia” is chiefly Kalimantan to China and India — the largest coal flow on Earth; “Atlantic → Europe” combines Colombian and South African tonnes.

2.7 Market organisations & supply coordination

Coal has no cartel and never sustained one — but it has the most consequential single-government supply management in commodities. Beijing runs coal policy as energy policy: the NDRC sets production targets (the post-2021 directive added ~600 Mt of annual output in two years — an OPEC-scale intervention executed in reverse), caps domestic prices in crises, gates imports informally (the 2020–23 unofficial ban on Australian coal — a concluded episode that redrew trade flows for two years), and maintains the mandated stockpile system described below. India runs the second state system through Coal India’s near-monopoly and import-substitution targets. The seaborne side coordinates only through capital discipline: after the 2015–16 bankruptcies and a decade of divestment pressure, no listed Western producer builds greenfield thermal capacity, an involuntary restraint more binding than any quota. The formal institutions are thin — the IEA (statistics and forecasts), the World Coal Association (advocacy), globalCOAL (trading standards) — and the loudest coordination is negative: the financing and insurance exit (hundreds of institutions with coal-exclusion policies) that functions as a distributed, permanent supply constraint on the Western industry.

Table 14. Who shapes the coal market

Source: IEA Coal 2025 , policy reporting 2020–26.

2.8 Strategic reserves & stockpiles

Coal’s strategic reserves are Asian and administrative rather than OECD-style SPRs. China operates the world’s only formal coal-reserve programme: after the 2021 power crisis, the NDRC built a government coal reserve system (announced targets in the hundreds of millions of tonnes of dedicated reserve capacity) on top of mandated inventories — power plants are required to hold roughly 20–30 days of burn, and the port-plus-plant system typically carries 300+ Mt at any time. The visible slice (Bohai-rim port stocks) is the market’s highest-frequency balance indicator. India mandates plant stockpiles (chronically thin ahead of monsoon season — the recurring spring scramble is a seasonal trading pattern in itself) and holds pithead stocks at Coal India. The OECD holds no strategic coal — its residual plants keep commercial piles, and several governments instead maintain reserve power plants (Germany’s grid reserve) rather than fuel. The asymmetry matters: the two countries that could destabilise the seaborne market hold the buffers, and their stockpile policies — not any exporter’s — decide how import demand responds to shocks.

Table 15. Coal stockpile systems

Source: NDRC/CEC policy announcements as reported (IEA Coal 2025 ); Indian CEA stock data. Approximate, indicative levels.

3. The companies & the value chain

3.1 The largest coal companies

Coal’s corporate league table splits into three worlds. The state giants dwarf everyone: Coal India (the world’s largest coal producer, ~780 Mt, listed in Mumbai under government control) and China’s CHN Energy/Shenhua, Shaanxi Coal and China Coal (each a top-five global producer, Shanghai/Hong Kong-listed under state control). The seaborne commercial tier is where Western investors live: Glencore — the largest exporter-trader, now spanning premium thermal (Australia, Colombia) and coking coal (Elk Valley, acquired from Teck in 2024) after shareholders voted to keep the coal business it had offered to spin off; Peabody (the US giant that went through Chapter 11 in 2016 and, in 2025, walked away from its $3.8bn deal for Anglo American’s coking-coal mines after the Moranbah North fire — leaving Anglo’s exit in limbo and the sector’s biggest M&A story unresolved); Core Natural Resources (the 2025 Arch-CONSOL merger), Australia’s Whitehaven (transformed by buying BHP’s Daunia and Blackwater met mines — production up 60%) and Yancoal, and Indonesia’s export complex (Alamtri/Adaro, Bumi, Bayan, ITMG). The third world is the exiting majors: BHP (BMA met only, thermal sold/closing), Anglo (selling), Rio (out since 2018) — the great unwinding that handed the assets to specialists and traders. Companies are compared on production, reserves and cost — never market capitalisation.

Table 16. Leading coal companies, 2024–25 (approximate)

Source: company reports 2024–25 (Whitehaven FY2025 ; Peabody-Anglo termination ; Glencore production reports incl. Colombian cuts ); output approximate on differing bases (saleable vs ROM, attributable vs managed). Listing notes: the Chinese and Indian majors trade publicly under controlling state shareholders; BMA (BHP/Mitsubishi) met volumes sit inside BHP. No market-capitalisation figures are shown by design.

3.2 Company archetypes along the value chain

Coal’s business models are defined less by geology than by stance toward decline. The state production machines (Coal India, the Chinese groups) run for energy security — volume targets, capped prices, dividend yield for the state — and are only loosely coupled to world prices. The seaborne pure-plays (Whitehaven, Yancoal, Alamtri, Core) are the sector’s high-torque instruments: un-hedged price exposure, and — because growth capex is socially rationed — cash flows returned at yields the rest of the equity market rarely prints. The met-coal specialists (BMA, Elk Valley, Mongolian producers) own the premium franchise with steel-cycle risk. The trader-consolidators (Glencore above all) buy what others are shamed into selling — the “last owner” strategy: acquire depleting assets cheaply, run them for cash, capture the scarcity premium the financing exit creates. And the exiters (BHP, Anglo, Rio, South32’s spin-offs like Thungela) convert coal into corporate virtue at whatever price the shrunken buyer pool offers — the mechanism that keeps handing value to the archetypes above. There is no explorer/developer class to speak of: nobody finances coal exploration anymore, which is itself the sector’s most important structural fact.

Table 17. Coal company archetypes

Source: company strategies per filings and disposals record, 2018–25; the Metal Pilot project-type taxonomy.

Figure 11. Coal company archetypes by price sensitivity

Source: illustrative ranking of the archetypes in Table 17 by coal-price sensitivity — qualitative, not a measured index. The trader-consolidator (“last owner”) is the structural winner of the divestment era.

3.3 Infrastructure & balance-sheet assets

A coal company’s filings describe a logistics and liability portfolio around a simple pit. The mines (open-cut draglines/truck-shovel, or longwall underground — the productivity champion with the gas and safety risks Moranbah North demonstrated) are described by ROM capacity, strip ratio and wash-plant yield (saleable tonnes per ROM tonne — the quiet margin lever). The scarce assets are port allocations and rail contracts: Newcastle and Dalrymple Bay capacity entitlements, take-or-pay rail deals (a fixed-cost trap in downturns, an option in booms). Quality itself is a balance-sheet item — a reserve’s kcal/sulfur/CSR profile decides which index (and premium) it earns. And uniquely in this series, the liability side leads: rehabilitation and closure provisions, take-or-pay commitments, and the financing structure itself — with banks retreating, the sector runs on retained cash, prepayments and trader finance, so balance-sheet self-sufficiency is a screening criterion, not a nicety. Watch too the insurance and refinancing cliff dates; more coal assets have been repriced by a lapsed policy than by any geology.

Table 18. Coal company asset types and metrics

Source: company disclosures (JORC reserves, annual reports), 2024–25.

4. Investing in coal

4.1 Ways to own it — exposure vehicles

Coal exposure is equities-first by structural necessity. Futures exist and are institutionally liquid — ICE Newcastle and API2 for thermal, SGX/CME coking coal — but they are professional hedging venues with no retail wrapper: the one dedicated coal ETF (VanEck’s KOL) closed in 2020, a divestment-era landmark, and nothing has replaced it — so the standard futures-product warnings (the roll-yield mechanics detailed in the oil guide ) apply only to those trading the curves directly. The equity menu is the real choice set: seaborne pure-plays (Whitehaven, Yancoal, Alamtri, Bumi, Core, Peabody) offering the cleanest torque and — their defining feature — outsized cash-return yields, because payout has replaced growth as the sector’s capital policy; Glencore as the liquid diversified proxy (coal inside a trading house); state majors (Coal India, Shenhua) as high-yield policy utilities with governance caveats; and met-coal names for the steel-linked premium segment. Two warnings define the asset class: valuations price decline (low multiples are the compensation for terminal risk, not a free lunch), and index/ESG exclusion means chronically thin ownership — amplifying both the cheapness and the violence of re-ratings when energy crises force generalists back in.

Table 19. Ways to own coal

Source: exchange specifications and issuer documentation; structural attributes only — no fees or prices, which change.

4.2 How to value & screen coal producers

Coal screening is cash-flow triage under a sunset clause. Start with cost position and quality: all-in cost against the relevant index (not “the coal price” — a 4,200 GAR Indonesian producer and a PLV met miner live in different markets), and the quality premium ladder (high-CV, low-sulfur thermal and true HCC earn structurally more per tonne and will hold demand longest — quality is longevity). Then the coal-specific balance-sheet tests: rehabilitation liabilities versus market value (some producers are worth less than their closure bill at conservative prices), take-or-pay exposure (fixed logistics costs turn downturns existential), and financing self-sufficiency (no bank will refinance trouble). Then the sector’s defining metric: cash-return yield and its durability — most coal equities are valued as run-off annuities, so the screen is reserves life × margin × payout ratio against the price you pay, stress-tested at trough prices ($60–80 Newcastle). For met names, add steel-cycle and EAF-substitution exposure (the same long-run force as in the iron ore guide ). These are the screens — production, reserves, quality, cost — you can run across every listed producer on Metal Pilot.

Table 20. Coal producer screening metrics

Source: company filings, 2024–25; thresholds indicative from the 2015–16 and 2020 trough experience.

Figure 12. Illustrative seaborne thermal cost curve

Chart source: illustrative; cost bands from company disclosures and industry commentary, 2024–25 — approximate, not company-level data. The dashed reference line is the 2025 average Newcastle price, ~$112/t (Table 3).

4.3 Macro regimes, rates & correlations

Coal’s regime map is energy-system stress overlaid on the Asian growth cycle. Demand-side, it behaves like a slower iron ore: Chinese and Indian power and construction set the pulse, so expansions and stimulus lift it, recessions dent it. But coal’s violent moves come from the energy-substitution channel: when gas is scarce or expensive (2021–22), power systems worldwide re-dispatch toward coal and its price rides the gas panic; when gas is abundant (the US shale decade, 2023–25’s LNG wave), coal is the fuel squeezed out. That makes coal the commodity complex’s crisis convexity trade — dormant through normal times, explosive when energy security fails — and 2022 the proof: while equities and bonds fell together, Newcastle quadrupled and coal equities tripled, the single best macro hedge of that year. (For the framework across all commodities, see the macro regime guide .) Rates and the dollar are second-order; weather (Northern winters, Indian monsoons, La Niña flooding Australian pits) is first-order and recurring.

Table 21. Coal across regimes

Source: price record (Table 3) with IEA demand data; regime characterisations are historical, not predictive.

On past performance, coal delivered both of the sector’s canonical lessons inside one decade. The equity amplifier, downside: the concluded 2011–2016 bust — Newcastle −52% from $130 to $63 and US gas eating coal’s home market — put Peabody, Arch, Alpha Natural and Walter Energy all through bankruptcy (Peabody’s equity, above $1,000 split-adjusted in 2011, was wiped out in its April 2016 Chapter 11 — a −99.9% outcome for holders), roughly half of US coal capacity restructuring within eighteen months. The upside: from pandemic lows to the 2022 crisis, Whitehaven rose more than tenfold (from under A$1 in 2020 to over A$10) while paying out billions — the same operating leverage, reversed, concentrated by the buyer-of-last-resort scarcity the divestment era created. Between those poles, long-run passive returns were poor: coal is a cycle trade and an income vehicle, not a compounder. Past performance is not indicative of future results.

On correlations (monthly, 2010–2025): Newcastle coal is strongly positive with natural gas (≈ +0.5–0.6 vs TTF/JKM in the switching era — the defining link), positive with oil (≈ +0.4, shared energy complex), moderately positive with iron ore (≈ +0.3 via China; met coal much higher), weakly negative with the dollar, and near zero with gold and equities through normal times — flipping sharply positive to energy-crisis conditions when everything energy moves together. The equities carry the additional ESG-flow factor: they re-rate on policy and index-exclusion news with no commodity move at all.

Table 22. Coal correlations (monthly, 2010–2025, approximate)

Source: author analysis of IMF/FRED and FRED series, monthly, 2010–2025. Correlations regime-dependent — they strengthen sharply in energy crises.

Figure 13. Coal correlations, monthly 2010–2025

Figure data: Table 22. Shaded on a fixed −1 to +1 scale (strong = 0.75); China and India power demand — the overriding driver — is not a price correlation and is discussed in the text.

4.4 Price drivers & cycles

The durable drivers, read through concluded episodes. Demand side: Asian power generation (the base load of the market), gas prices (the substitution ceiling and floor), weather (heat waves, cold snaps, monsoons — coal is the weather trade), and steel for the met complex. Supply side: Chinese production policy (the single biggest lever in the market), Indonesian volume elasticity, logistics (rail, ports, La Niña), and the capital starvation that mutes every supply response. The settled case studies: the 2008 and 2011 supercycle spikes (Chinese import flip meeting inelastic supply); the 2012–2016 bust (shale gas + Chinese slowdown + boom-built supply → the bankruptcy wave — proof that coal’s cost curve resolves gluts brutally); China’s 2016 supply-side reform (mandated mine closures doubled prices in six months — administrative supply management, demonstrated); the 2020–23 Australian import ban (politics redrew flows: Australian coal went to India and Europe, Indonesian and Russian coal filled China — and prices barely noticed, proof of the trade’s fungibility); and the 2021–22 energy crisis — gas panic, Russian sanctions, post-COVID demand and La Niña supply failures stacking into the greatest energy-price event since 1979, with coal the marginal fuel and marginal fortune. The IEA-consensus scenario frame below is illustrative, not forecast.

Table 23. Coal price drivers

Source: IEA Coal 2025 and the price record (Table 3). Case studies are concluded episodes.

Table 24. Illustrative scenario frame (conditions, not forecasts)

Source: illustrative scenario conditions from the 2012–2025 record; explicitly not predictions.

4.5 Risks, controversies & ESG

Coal is the asset class where ESG is the investment case’s counterparty. Transition risk is the terminal risk: every serious climate pathway phases unabated coal power out, OECD demand is already in free fall, and China’s extraordinary renewables build-out — the largest energy project in history — will eventually do to Chinese coal burn what it did to everyone else’s; the bear case is not a cycle but a schedule. Policy risk operates on both sides (carbon pricing and plant closures against coal; energy-security politics repeatedly for it — 2022 forced even Europe to reopen plants). The financing exit cuts both ways too: it starves supply (bullish) while shrinking the buyer pool for the equities (structural discount). Operationally, coal remains the deadliest major mining sector (methane and underground fires — Moranbah North’s 2025 fire killed a $3.8bn transaction, and Chinese mine-safety campaigns move national output), and its local externalities (air quality above all — coal combustion is among the largest single sources of premature deaths from air pollution) drive the social license collapse. Set against all of it: coal remains the cheapest firm power in the developing world, the only current route to primary steel at scale outside DRI, and — its defenders note — the fuel whose retirement the West preaches while financing none of the alternatives at the required scale. These are contested questions, and reasonable analysts weigh them differently — nowhere more sharply than here.

Figure 14. Coal risk map — likelihood vs. impact

Source: author’s qualitative assessment; see Section 4.5.

5. Future outlook & forecasts

Coal’s outlook is the most explicitly scenario-shaped in this series, because its demand curve is a policy choice as much as an economic one. Forecasts are scenarios, not measured facts — and coal’s forecasting record is uniquely humbling: “peak coal” has been declared repeatedly since 2013, and demand set fresh records in 2024 and 2025 regardless.

5.1 Demand

The IEA’s Coal 2025 base case: demand plateaus at ~8.8–8.85 Gt through the late 2020s and eases ~3% by 2030 (back to 2023 levels). The composition shifts beneath the flat total: China (56% of demand) holds a high plateau with a gentle decline (~−0.7%/yr to 2030) as its renewables juggernaut finally outruns power-demand growth; India adds 200+ Mt by 2030 (+3%/yr) and Southeast Asia keeps building; the OECD’s decline accelerates toward residual levels. Under climate-driven scenarios the 2030s bend steeply down; under energy-security-first politics the plateau extends. Met coal tracks the blast-furnace share of steel — eroding slowly with EAF growth, with premium HCC the most durable segment. The honest summary: the direction after 2030 is down in every scenario; the slope is the trillion-dollar argument.

5.2 Supply and the balance

Seaborne supply is entering managed decline with episodic scarcity. No Western greenfield thermal capacity is being built; Indonesian expansion is the marginal volume lever; Russian exports are rail- and sanctions-capped; Australian thermal mines are scheduling closures (Mount Arthur 2030). Against a plateauing demand base, that points to a structurally balanced-to-tight seaborne market punctuated by weather and policy shocks — the consensus view after 2025’s surplus year — with the cost floor near $60–90 and crisis prints unbounded (2022 proved the ceiling is demand destruction, not cost). The met market’s balance hangs on Bowen Basin discipline versus Mongolian growth and the steel transition. The permanent wildcards: Chinese policy (the half-billion-tonne import swing), and the capital-starvation premium — an industry forbidden to grow tends, on shocks, to be priced like one that can’t.

Table 25. Coal outlook indicators (scenario figures)

Source: IEA Coal 2025 , December 2025. Scenario projections, not measured data.

Figure 15. The coal plateau — demand by region to 2030 (Mt)

Source: IEA Coal 2025 ; regional paths approximate from IEA data and forecast. Bars are scaled to each year’s composition (the world total plateaus near 8.6–8.85 Gt); segment values are absolute Mt. Scenario projections, not measured data.

5.3 Catalysts to watch

The near-term list is weather-and-Beijing: Chinese import arbitrage (domestic vs seaborne prices — the swing variable), NDRC production and safety policy, the Indian monsoon and pre-summer stocking cycle, gas/LNG balances (the substitution channel), La Niña/El Niño (Australian-Indonesian supply weather), and Bohai port stockpiles. Corporate catalysts: the Anglo coking-coal sale’s eventual resolution (and the Peabody arbitration), further consolidation of exiting majors’ assets, and payout announcements across the pure-plays. Structural markers: Chinese renewables installation rates (the variable that ends the plateau), Indian and ASEAN plant commissioning, EAF-share milestones for met, and any crack in the financing exclusion wall (several institutions have quietly softened — a re-rating trigger if it spreads). What would confirm the bear case: two mild winters plus an LNG glut into disciplined Chinese output. What would light the bull case: any energy-security shock into today’s thin buffers — coal remains the system’s only fast-response reserve fuel.

Table 26. Coal catalyst calendar

Source: IEA , company calendars, market data providers.

6. Summary

Coal is the fuel the energy transition is supposed to retire and keeps setting records instead — 8.85 Gt consumed in 2025, a third of world electricity and the carbon behind most primary steel. It is two commodities in one: thermal coal priced off Newcastle ($26 in 2000, $138 in 2008, $63 at the 2015 trough, an unprecedented $384 in 2022, $112 in 2025) and premium coking coal priced off the Bowen Basin — both set in a seaborne trade (~1.47 Gt) that is a fraction of production, because China mines and burns roughly half the world’s coal at home and swings the traded market with an import decision (record ~543 Mt in 2024, cut by ~53 Mt in 2025). Supply has no cartel but plenty of policy: Beijing’s output directives, India’s state monopoly, Indonesia’s export machine — and the Western financing exit, which has frozen new thermal capacity and turned existing export mines into scarce, cash-gushing run-off annuities. The companies sort by stance toward decline — state machines, high-payout seaborne pure-plays, met specialists, and Glencore’s last-owner consolidation — screened on cost, quality slate, rehabilitation liabilities and payout durability, never market cap. The regime profile is crisis convexity: dormant in gas-abundant calm, explosive when energy security fails, with the concluded record spanning the 2016 bankruptcy wave (Peabody −99.9%) and the 2021–22 windfall (Whitehaven >10×). The outlook is a plateau with a schedule: flat to 2030 (India offsetting the OECD), declining after, at a slope set by Chinese renewables and Asian politics. The single most important variable to watch is Chinese import and production policy against the gas cycle, with weather the recurring wild card.

To go from this big-picture view to the actual companies — screening every coal producer by production, reserves, quality and cost — explore Metal Pilot.

7. Vocabulary

Every specialised term used in this report, defined once at first use in the body, collected here for reference.

Table 27. Coal vocabulary

Source: definitions follow IEA and index-methodology terminology, 2025–26.

8. Sources, methodology & disclaimer

8.1 Sources, methodology & data vintage

Agencies & official data: IEA Coal 2025 (December 2025) — executive summary , demand and trade chapters — and the Coal Mid-Year Update 2025 ; Energy Institute Statistical Review of World Energy (production, consumption, reserves history); IMF Primary Commodity Prices / Global price of Coal, Australia via FRED; FRED macro series.

Market & industry reporting: China customs and import coverage (Global Energy , IndexBox , The Coal Hub ); index methodologies (globalCOAL, Argus , Platts); worldstopexports trade compilation.

Company filings & statements: Whitehaven FY2025 annual report ; Peabody 10-K and the terminated Anglo transaction (Anglo statement ); Glencore production reports and Colombian curtailment ; Coal India, CHN Energy/Shenhua, Alamtri, Yancoal and Core Natural Resources disclosures, 2024–25.

Methodology: prices are calendar-year averages — Newcastle-basis thermal from the IMF series, met-coal levels from index reporting — never spot snapshots. Production, consumption and trade follow the IEA (physical tonnes; energy-content trends differ slightly and are flagged where material); reserves follow the Energy Institute. Country and company figures are rounded and mix reporting bases (ROM vs saleable; calendar vs fiscal years) as noted under each table. The balance is assessed through the seaborne trade, Chinese import behaviour and stockpile indicators — no audited global balance series exists for coal. Correlations use monthly data 2010–2025 and are approximate and regime-dependent. Scenario tables are illustrative conditions, not forecasts. Reserves, resources and projections are estimates, not measured facts.

Data as of: 6 July 2026. Intended update cadence: annually after the IEA’s Coal report (December) and the Energy Institute Statistical Review (June), plus ad hoc updates on major market events.

8.2 Disclaimer & disclosure

This report is for informational purposes only and is not investment advice, a recommendation, or an offer to buy or sell any security or commodity. Coal prices are volatile and exceptionally policy- and weather-sensitive; the figures here are estimates as of the stated date that will change, and the sector carries material transition and terminal-decline risks alongside its cyclical ones. Reserves, correlations and regime observations are estimates and historical patterns that may not persist. Do your own research and consult a licensed financial adviser before acting. This report was prepared with the assistance of AI; its figures were sourced from the references above and reviewed, but readers should verify any number before relying on it. The author holds no position disclosed as a conflict in respect of the companies named.