U.S. Coal Mining (Industry Group) — An Investor's Primer

Industry group: NAICS 2022 code 2121 — Coal Mining (United States) NAICS = North American Industry Classification System, the federal code that defines this industry. This four-digit group contains exactly one child industry: 21211 Coal Mining.

Read this first, then go one level down. This four-digit group is a pass-through: it holds a single child industry, 21211 Coal Mining, so every dollar of revenue, every firm, and every worker in "Coal Mining (2121)" is also in 21211. The full analysis — the surface-vs-underground split, the company-by-company investable universe, the cost economics, and the complete source list — lives in the 21211 primer. This page gives you the group-level identity, our federal ground-truth figures for this exact level, and a map of what you'll find below. For anything beyond the summary, read the 21211 primer.

---

1. Overview

Coal mining is the business of extracting coal from the ground — either by stripping the soil and rock off shallow seams (surface mining) or tunneling into deeper ones (underground mining) — and cleaning it for sale. It is a shrinking, deeply cyclical commodity industry: U.S. output fell from a 2008 peak near 1,172 million short tons (MMst) to about 512.5 MMst in 2024 — the lowest since 1964 — before a weather-and-gas-driven rebound to roughly 533 MMst in 2025 [3][5]. A short ton is 2,000 lb; MMst = million short tons.

For an investor, coal producers are price-takers: they cannot set the price of coal, so a modest move in price produces an outsized swing in profit against their largely fixed costs. Layered under that normal cycle is a structural decline — coal's share of U.S. electricity fell from about 51% in 2001 to ~15% in 2024 [6]. The result is an unusual profile: cash-generating, out-of-favor assets that throw off large dividends and buybacks in good years, against the real risk that demand fades before the coal in the ground is ever mined.

---

2. What's inside — why the group equals its one child

At the four-digit level, NAICS classifies all U.S. coal mining into a single industry, 21211. There is no second sibling to roll up, so 2121 and 21211 are the same industry measured at two labels — the group figures below are the 21211 figures. This is why this page is short.

The economic diversity all sits one level further down, inside 21211, which itself splits by method into two national industries:

• 212114 Surface Coal Mining — moves the most tons (~60%) but earns less than 40% of revenue, because its dominant product is cheap, low-energy Powder River Basin (PRB) thermal (steam) coal burned for electricity.
• 212115 Underground Coal Mining — moves fewer tons (~40%) but earns the majority of revenue, because it produces high-value metallurgical ("met," or steelmaking) coal for export, plus high-heat thermal.

That surface-vs-underground contrast — one half a domestic-electricity story in structural decline, the other a global-steel-and-export story that is durable but not growing — is the heart of the coal investment case. It is developed in full in the 21211 primer; this page does not repeat it.

---

3. How big it is — our federal ground-truth for 2121

Two yardsticks measure this industry: business size (Census dollars, firms, jobs) and physical size (tons). Keep them separate.

Business size — authoritative federal figures for NAICS 2121 (U.S. Census Bureau; 2022 Economic Census and 2023 County Business Patterns, or CBP) [2]. Because 2121 = 21211, these match the child figures exactly:

Implied pay is roughly $106,000 per worker — well above the private-sector average, reflecting skilled heavy-equipment operators, longwall crews, electricians, and engineers (a calculation from the figures above, not a published wage) [2].

Concentration is only moderate at this level — the four largest firms earn 39.7% of revenue (the "CR4," or four-firm concentration ratio), the top eight 54.5%, the top 20 76.3%, and the top 50 92.7%, with a Herfindahl-Hirschman Index (HHI, where under 1,500 counts as "unconcentrated") of just 523.1 [2]. But tonnage is far more concentrated than the dollar figures suggest: EIA controller data show the top 20 companies moving ~84.5% of national output — a few giant low-cost mines produce most of the coal while a long tail of small operators fills out the firm count (detailed in the 21211 primer) [4].

Physical size — EIA tonnage (the U.S. Energy Information Administration, or EIA, is the authoritative federal source for coal's physical facts — not the U.S. Geological Survey, or USGS, which covers metals and industrial minerals):

Reserves vs. resources — don't confuse them. The U.S. has a demonstrated reserve base near 468 billion tons, but only about 10.6 billion tons sit at currently producing mines [3]. The U.S. is not short of coal; it is short of demand and of premium met quality. Most coal in the ground is not an investable reserve.

---

4. The investable universe (summary)

The listed group is short, spans both surface and underground, and every name is a leveraged bet on the coal price — none is a large-cap. The public routes are a handful of producers — Peabody Energy (BTU, mostly PRB surface), Core Natural Resources (CNR, both — PRB surface + Appalachian met/thermal, formed by the Jan 2025 Arch–CONSOL merger), Warrior Met Coal (HCC, pure premium met), Alpha Metallurgical Resources (AMR, met), Alliance Resource Partners (ARLP, thermal income partnership), and Ramaco Resources (METC, met) — plus one royalty owner (Natural Resource Partners, NRP), a foreign-listed name (Coronado, ASX: CRN), and a single niche global exchange-traded fund (COAL). Major private / tribal / foreign owners not on public markets include Navajo Transitional Energy Company (NTEC), American Consolidated Natural Resources (ACNR), and Foresight Energy.

There is no large pure-play coal streaming or royalty company analogous to the precious-metals streamers, so "royalty-style" coal exposure has far fewer public choices than in metals or oil and gas. The full company-by-company table — which child each name sits in, what you're buying, and the private owners — is in the 21211 primer.

---

5. How the money works

Both halves make money on volume × a per-ton spread, and both sit largely outside management's control — but they diverge sharply on price. Surface/PRB thermal is cheap to mine but low-energy, so it sells for very little (2024 subbituminous coal averaged ~$18/ton) — high volume, razor-thin margin; a $1/ton price move can be roughly half a PRB producer's unit margin. Underground met trades on volatile seaborne benchmarks (above $300/tonne in 2022, range-bound near $190–235 more recently), so a met producer's earnings behave like a geared call option on the coking-coal price [4].

Coal has no single standard cost yardstick — unlike metals miners' all-in sustaining cost (AISC) or oil-and-gas lifting and finding-and-development (F&D) costs — because reported "cash cost per ton" excludes different items. Freight is often decisive for PRB, where rail can cost as much as the coal itself. Producers now harvest existing low-cost leases rather than expand, and pay production royalties (mostly to the federal government for PRB coal) plus severance taxes and reclamation fees. The cost mechanics, freight economics, and where the two children diverge are worked through in detail in the 21211 primer.

---

6. What drives demand

The two halves serve almost separate end-markets. Surface/thermal is an electricity story in structural decline: power generation is ~90% of coal consumption, and coal's electricity share fell from ~51% (2001) to ~15% (2024), recovering to ~17% in the colder, higher-gas 2025 [5][6]. Cheap shale natural gas is the swing competitor and the main reason coal lost the power market. Electricity demand is growing again (data centers, electrification), but EIA expects most new supply from gas, solar, and wind — load growth does not translate one-for-one into coal demand. The near-term help is slower plant retirements (only 2.6 gigawatts retired in 2025, the least since 2010) [7].

Underground/met is a steel-and-export story — durable but not growing. Met coal is coked for blast-furnace steelmaking with no drop-in substitute at scale today, but U.S. steel is now made mostly in scrap-melting electric-arc furnaces that need no coke, so U.S. met producers survive by exporting into a roughly flat global steel market. Full demand detail is in the 21211 primer.

---

7. Regulation

Coal is a leasable mineral (Mineral Leasing Act of 1920), not hardrock — the General Mining Law of 1872 does not apply. The federal government keeps title and leases development rights through the Bureau of Land Management (BLM), and owns most PRB coal in the ground. A 2025 law cut the federal coal royalty to no more than 7% through Sept. 30, 2034 and directed new leasing — a real but reversible producer tailwind [8]. Mine reclamation runs through the Surface Mining Control and Reclamation Act (SMCRA); worker safety through the Mine Safety and Health Administration (MSHA); mine drainage and downstream power-plant rules through the Environmental Protection Agency (EPA). Crucially, downstream power-plant rules often matter more to a thermal miner than mine-site rules, because they decide whether the customer keeps burning coal. The full regulatory map — BLM leasing, SMCRA bonding, MSHA's 2024 silica rule, EPA power-plant standards, and ESG capital-access risk — is in the 21211 primer.

---

8. Consolidation

Modern coal structure was forged in bankruptcy: debt-financed acquisitions at the 2011 met peak collided with the shale-gas collapse, producing a 2015–2016 bankruptcy wave (Patriot, Walter, Alpha, Arch, Peabody). Survivors emerged deleveraged and export-focused — most consequentially Core Natural Resources, the Jan 2025 Arch–CONSOL "merger of equals." But antitrust caps consolidation within the PRB: when Peabody and Arch tried to combine their PRB mines in 2019, the Federal Trade Commission won an injunction and the deal collapsed. With no reason to expand into a declining market, producers now compete on cost and return cash rather than grow. Detail is in the 21211 primer.

---

9. Risks

1. Commodity-price (and volume) cyclicality — the central risk. Thin, variable per-ton margins mean small price moves swing earnings, dividends, and share values hard; met is one of the most volatile bulk commodities and PRB thermal whipsaws with natural-gas prices [3][4].
2. Secular demand decline / stranded-asset risk — what sets coal apart. Coal's power share fell from ~51% to ~15% in two decades; reserves, equipment, and rail spurs can be made worthless if the customer plant retires first [6].
3. Customer / end-market and logistics concentration — who burns (or cokes) the next ton, freight cost, and terminal access can matter more than spot price.
4. Regulatory reversal, reclamation/bonding liabilities, and ESG capital-access drag — two-directional policy risk plus obligations that can outlast production.

(The full nine-item risk register, including the classic valuation mistake, is in the 21211 primer.)

---

10. How to invest, and the outlook

Match the name to the product. Pure met (Warrior/HCC, Alpha/AMR, Ramaco/METC) has healthier, export-driven pricing; diversified (Core/CNR) blends PRB surface, high-heat thermal, and met with export terminals; thermal-heavy income (Alliance/ARLP); largest surface controller (Peabody/BTU); royalty/fee models (NRP, NACCO) carry less direct commodity exposure. Expect a boom-bust dividend and buyback pattern — clean post-restructuring balance sheets mean up-cycle cash is returned, not reinvested. This is a capital-return story, not a growth story, and the market prices these as "melting ice cubes" (low earnings multiples, high free-cash-flow yields). Private routes — direct/private-equity mine ownership, contract mining, mineral-and-royalty interests, and distressed debt — must be underwritten mine-by-mine, because bonding, black-lung, water-treatment, and plant-life liabilities are the real risk.

Outlook. The near term is more resilient than a straight-line decline suggests — the 2025 production rebound, the slowest retirement pace since 2010, the royalty cut and new leasing, and data-center power demand all support existing low-cost operators [5][7][8]. But the structural verdict is unchanged: policy can extend the runway and fatten interim cash returns; it does not restore thermal coal's competitiveness against cheap gas and renewables. The realistic frame for both public and private investors is a high-cash-return, high-volatility, finite-life exposure — best sized as a tactical, cash-harvesting position, not a long-duration compounder.

→ For the full analysis, go to the [21211 Coal Mining primer]. This group-level page is a summary; the child primer carries the surface-vs-underground economics, the complete investable universe, the cost-curve mechanics, and the full source list.

---

Sources

Figures for NAICS 2121 are identical to its sole child, 21211; the complete, numbered source list (company filings, EIA reports, statutes) lives in the 21211 primer. Key sources cited on this page:

1. U.S. Census Bureau, 2022 NAICS Definitions — 2121 / 21211 Coal Mining (scope; single-child structure; leasable-mineral note), 2022. https://www.census.gov/naics/?input=2121&year=2022
2. U.S. Census Bureau, 2022 Economic Census (2121/21211 receipts $27.43 bn, 218 firms, CR4/CR8/CR20/CR50, HHI 523.1) and 2023 County Business Patterns (491 establishments, 42,347 employees, $4.50 bn payroll). Histometrics ingested federal ground-truth for NAICS 2121. https://data.census.gov/
3. U.S. Energy Information Administration (EIA), Annual Coal Report 2024 (production 512.5 MMst; surface ~306 / underground 206.2; 524 mines; consumption 410.9, electric power 373.3; exports; reserves), 2025. https://www.eia.gov/coal/annual/
4. U.S. EIA, Annual Coal Report 2024 — underground/met detail (met price ~$180/ton; ~79% of met exported; controllers ~84.5% of output), 2025. https://www.eia.gov/coal/annual/pdf/acr.pdf
5. U.S. EIA, U.S. coal production and coal-fired generation increased in 2025 (~533 MMst, +~4%), 2026. https://www.eia.gov/todayinenergy/detail.php?id=67684
6. U.S. EIA, Coal's share of U.S. electricity generation (~51% 2001 → ~15% 2024 → ~17% 2025), 2026. https://www.eia.gov/todayinenergy/detail.php?id=67005
7. U.S. EIA, U.S. coal-fired power plant retirements slowed in 2025 (2.6 GW retired, least since 2010; ~6.4 GW planned 2026), 2026. https://www.eia.gov/todayinenergy/detail.php?id=67427
8. 119th Congress, Public Law 119-21 (§§50201–50203) — federal coal royalty ≤7% through Sept. 30, 2034; expanded leasing, 2025. https://www.congress.gov/119/plaws/publ21/PLAW-119publ21.htm

---

Reference years: business statistics = 2022 Economic Census and 2023 County Business Patterns (U.S. Census Bureau — Histometrics' ingested ground-truth for NAICS 2121: $27.43 bn receipts, 218 firms, 491 establishments, 42,347 employees, $4.50 bn payroll, HHI 523.1); physical production/consumption/exports = calendar-year 2024 EIA with the 2025 rebound noted; policy = 2025. NAICS 2121 contains exactly one child industry (21211), so group and child figures are identical by construction. For the full company-level and source detail, see the 21211 primer. Forward-looking statements reflect cited EIA outlooks, not guarantees of investor returns.