Skip to main content

The Commodity Cycle

Updated Jun 24, 2026 at 8:22pm

Research Draft High 1,278 words

The commodity cycle is the recurring boom-and-bust pattern in raw-material prices and, downstream of it, in mining-company profits and share prices. Its engine is a structural asymmetry of timing: demand for metals can move quickly, but supply cannot. Bringing a new mine from discovery to production routinely takes a decade or more, so when demand surprises to the upside, prices spike long before new tonnes arrive. High prices eventually trigger a wave of capital spending; by the time that capacity floods the market, demand has usually cooled, and the resulting glut crushes prices below the cost of production. The core tension for an investor is that the cycle's most powerful signal — sustained high prices — is precisely what sows the seeds of the eventual bust, and the worst returns are typically planted at the moment of greatest optimism.

How it's formed

The cycle is best understood as a capital cycle, the framework articulated by Marathon Asset Management and editor Edward Chancellor in Capital Returns (2015). The supply of capital into an industry, not demand alone, governs future returns: high prices and profits attract investment, investment expands capacity, excess capacity depresses returns, depressed returns starve the industry of capital, and the resulting scarcity eventually lifts prices again. Returns mean-revert because capital chases them.

The classic four phases:

1. Trough / under-investment. Years of low prices have shuttered mines and frozen exploration. Supply growth stalls. 2. Recovery. Demand recovers against a thin supply pipeline; prices rise faster than producers expect. Margins expand sharply because mining is a high-fixed-cost business — most of a price increase above operating cost drops straight to profit (operating leverage). 3. Boom / over-investment. Convinced the price is durable, companies sanction new projects, bid up acquisitions, and chase the lowest-grade deposits that are now economic. Capex peaks. This is the phase of maximum euphoria and minimum future return. 4. Bust. The long-lead projects finally come online into softening demand. Oversupply collapses prices; high-cost producers run at a loss; capex is slashed (after oil's 2014 price break, global upstream capex fell sharply from its peak — IEA data shows spending roughly 40–45% below the 2014 peak by the mid-2010s trough, and capital discipline kept it depressed for years). The cut sets up the next trough.

The supercycle is the multi-decade envelope around these shorter cycles. Buyuksahin, Mo & Zmitrowicz (Bank of Canada, 2016) used an asymmetric band-pass filter on a real commodity-price index back to 1899 and identified four supercycles — peaking around 1904, 1947, 1978, and 2011 — each tied to a great industrialization wave (US industrialization, 1930s rearmament, post-war Europe/Japan reindustrialization, and China). The Bank of Canada study reports a full trough-to-trough supercycle averaging about 32 years; Erten & Ocampo (2013), using a longer 1865–2010 sample, find four super-cycles lasting roughly 30–40 years each with amplitudes 20–40% above or below long-run trend.

How it's used in practice

Position against the capex cycle, not the price. The Marathon insight is to deploy capital where investment is depressed (favorable competitive conditions, cheap assets) and avoid sectors where capex is booming. Surging industry capex, frenzied M&A, IPOs of marginal producers, and "this time supply is constrained forever" narratives are classic late-cycle tells.

Read the cost curve. Each commodity has an industry cost curve (often expressed for gold as All-In Sustaining Cost, AISC — cash costs plus sustaining capex and overhead). Price tends to find a floor near the marginal producer's cost, because below it, supply shuts in. Knowing where a given miner sits on that curve tells you who survives the bust and who is squeezed first.

Exploit operating leverage — carefully. Because costs are largely fixed, miner earnings amplify metal-price moves. As an illustration: in 2025, industry gold All-In Sustaining Cost sat broadly in the $1,400–1,800/oz range (the median across miners was reported near $1,600/oz by mid-2025, with large caps such as Agnico Eagle nearer $1,400 and others above $1,800), while the gold price averaged roughly $3,400/oz for the year and traded above $4,000/oz late in 2025. With price well above cost, producers earned record margins — but the same leverage works brutally in reverse when price falls toward cost. (These are point-in-time figures, not durable constants.)

Distinguish the metals. Copper is the archetypal cyclical — demand tracks construction, electrification, and global GDP, so it tops and bottoms with the economy. Gold is largely counter-cyclical / monetary — driven by real rates, the dollar, and risk aversion rather than industrial demand, so the gold-vs-copper ratio is itself a read on the macro regime. A "commodity cycle" playbook must specify which commodity; lumping them together is a common error.

Adoption, debate & evidence

The cycle's existence at the price level is well documented; the supercycle dating by Bank of Canada and similar World Bank / academic work (e.g. Erten & Ocampo, 2013) shows long, statistically identifiable deviations from trend. The capital-cycle explanation — supply lags plus over-investment — is widely accepted among mining analysts and is the consensus mechanism in the academic literature for the upswing.

What is genuinely contested:

  • Identifying turns in real time. Supercycles are classified retrospectively; you only know the peak years later. "We are entering a new supercycle" is a perennial sell-side claim (made for the "green transition" / critical minerals through the 2020s) that is unfalsifiable in the moment and frequently wrong on timing.
  • Whether this cycle is structurally different. Each boom produces a scarcity narrative ("peak supply," "decade of underinvestment"); historically, high prices have always eventually called forth supply or demand destruction. Healthy skepticism is warranted.
  • Edge from the framework itself. The capital cycle is a sound analytical lens, not a timing system. It tells you the direction of mean reversion, not when. Cycles can run far longer than a value-minded investor expects, and miners can stay cheap or expensive for years.

Strengths & limitations

When it works: as a contrarian discipline. Its greatest value is keeping you from buying miners at peak margins (when they look cheapest on trailing earnings) and pointing you toward beaten-down, under-capitalized sectors. The framework correctly anticipated the post-2011 mining bust that followed the China-driven capex boom.

When it fails: as a precise market-timing tool. The lags are real but variable, demand shocks (recessions, stimulus, technology shifts) are unpredictable, and a low-cost producer can compound for years through a cycle. The framework also says little about individual company quality — balance sheet, jurisdiction risk, grade, management capital discipline.

The #1 misuse: extrapolating the current price. Buying miners because metal prices are high and margins are fat — the moment the cycle is most likely to turn. Trailing valuation multiples are most flattering exactly when forward returns are worst.

Sources

Disputes flagged: real-time supercycle identification and any "new supercycle" claim are contested and often promotional; AISC/margin figures are recent trade-press point-in-time numbers, not durable constants.