How commodities trade: Hedging in practice: trading price risk for basis risk

3.1 A farmer and a miller hedge opposite sides of the same price

Geman's basic case is a farmer who in January sells futures maturing at harvest. A short hedge sells futures for the quantity and date of the planned sale and buys them back when the crop is sold. If prices fall, the loss on the crop is offset by a gain on the futures; if they rise, the reverse.

An illustrative example, built on October 2026 WASDE numbers, shows the arithmetic. US wheat yielded about 48.1 bushels an acre in 2026/27, so a 1,000-acre farm grows about 48,000 bushels, roughly ten of the 5,000-bushel contracts Morgan describes. Suppose, hypothetically, that at planting the farmer hedges at a futures price that implies a local price equal to the 2025/26 average farm price of $5.06. US production then fell from 1,989 to about 1,534 million bushels and the projected 2026/27 farm price rose to $6.30. The farmer sells his wheat locally at $6.30 but loses $1.24 a bushel on the futures. His net price stays at $5.06, about $60,000 less than if he had not hedged. That is the cost of certainty, not a mistake. The national season-average prices stand in for one farm's local price.

A flour miller faces the mirror image. Having agreed to supply bakers at fixed prices months ahead, he fears a rise in wheat, so he buys futures: a long hedge. When wheat rises, his futures gain covers the dearer grain. The miller's purchases and the farmer's sales rarely match in size or timing. Radetzki and Wårell note that the clearing house's book must balance in every month, so speculators must at minimum fill the gap between short and long hedges.

• Geman (2005), Commodities and Commodity Derivatives · pdf p. 16
• Radetzki & Wårell (2021), A Handbook of Primary Commodities in the Global Economy · pdf pp. 158-162
• USDA (2026), World Agricultural Supply and Demand Estimates, October 2026 web ↗ · U.S. wheat supply and use
• Morgan (1979), Merchants of Grain · pdf p. 453

3.2 A copper fabricator hedges a fixed-price order

Consider a hypothetical wire and cable maker that in September 2025 signs a contract and agrees to supply a utility with cable for delivery over the following year at a fixed price. The quote is built on copper at the September 2025 world price of $9,984 a tonne. The fabricator does not yet own the copper, so it is exposed to a rise. It buys LME copper futures for the tonnage it will need: a long hedge, the case Tilton and Guzmán describe for a firm that will need refined copper in three months.

By September 2026 copper had reached $14,474 a tonne, 45% higher. Unhedged, the fabricator would pay about $4,490 a tonne more than its price assumed: about $4.5 million on an illustrative 1,000-tonne order. Hedged, it sells its futures at a gain of about $4,490 a tonne as it buys physical cathode, and the two cancel.

Two things remain. First, cash. Here the futures gains flowed in daily, but had copper fallen, the fabricator would have paid variation margin every day while its offsetting benefit, cheaper metal, arrived only as it bought cathode. Pirrong notes that even when the hedge and the physical position move in lockstep, their cash flows do not, so a hedger needs a credit line to bridge them. Second, premiums, which can move against it: the fabricator buys cathode of a particular brand at a particular place, paying the exchange price plus a physical premium that the futures do not cover.

• World Bank (2026), Commodity Price Data (Pink Sheet), monthly prices to September 2026 web ↗ · copper, Sept 2025 and Sept 2026
• Tilton & Guzmán (2016), Mineral Economics and Policy · pdf p. 96
• Pirrong (2014), The Economics of Commodity Trading Firms · pdf pp. 16, 23
• Geman (2005), Commodities and Commodity Derivatives · pdf p. 182

3.3 Basis is the gap a hedge cannot close

Geman defines the basis as the spot price minus the futures price, usually quoted as a premium or discount: a local cash price a dollar over futures has a basis of plus one. It reflects the cost of cleaning a local grade to deliverable quality, transport to or from the delivery point, and storage until the futures month. Pirrong's summary is that hedging exchanges flat price risk for basis risk. Since basis usually moves less than the flat price, the trade is worth making, but the basis can still jump.

Basis risk has several sources. The hedge may use a related product: an airline hedging jet fuel with heating oil futures. Location bottlenecks widen the basis: Pirrong cites a pipeline explosion in late 2000 that cut gas supply into California and blew out the gap to Henry Hub, and the surge in US oil output that made the gap between WTI and internationally traded crudes wider and far more variable. And timing mismatches create calendar spread risk, as when corn stored in October is hedged with a December contract.

Geman shows that basis risk depends mainly on the correlation between spot and futures. Pirrong gives the metals case: copper cathode in an LME warehouse hedges well with LME futures, but concentrate does not, because only its copper content is priced off the LME, while its treatment and refining charges cannot be hedged.

For professional traders, basis is the business. Pirrong reports that Trafigura hedges all flat price risk it can, and that to the extent it speculates, it speculates on basis relationships. Distress usually comes from basis and spread positions: Cook Industries had to shrink drastically after losses on soybean calendar spreads in 1977. Pirrong reports that Glencore lost about $300 million in cotton in May-July 2011 through basis moves likely caused by a corner of ICE cotton futures; Blas and Farchy, citing Glencore's results, put the loss at more than $330 million.

• Geman (2005), Commodities and Commodity Derivatives · pdf pp. 19, 26-27
• Pirrong (2014), The Economics of Commodity Trading Firms · pdf pp. 12-14, 23-24
• Morgan (1979), Merchants of Grain · pdf pp. 300-302
• Blas & Farchy (2021), The World for Sale · pdf p. 265

3.4 Options buy insurance rather than a lock

A call option gives the holder the right, not the obligation, to buy a futures contract at a set strike price until expiry; a put gives the right to sell. The holder pays a premium up front. Radetzki and Wårell explain that the premium has two parts: intrinsic value, the gap between the strike and the futures price, and time value, which shrinks to zero at expiry. The holder can lose no more than the premium paid.

An option hedge suits someone who wants protection but also wants to keep the upside. Geman's airline that buys call options on fuel pays the premium in January and in September pays either the strike or the lower market price, whichever is less. A futures hedge costs nothing up front but forgoes favourable moves; the option costs money but works like insurance.

Professional traders often prefer futures. Pirrong notes that Trafigura uses options little, because its aim is to remove price exposure, and an option leaves a one-sided exposure that has to be paid for.

Producers sometimes hedge years ahead. Radetzki and Wårell describe metal producers selling forward their output for several years and gold mines using forward sales as loan collateral. Such hedging can feed back into prices. Humphreys argues that systematic forward selling by gold miners in the late 1990s mobilised dormant gold and added to the slump. As prices recovered in 2003-04, producers began abandoning forward sales and some bought their hedges back.

• Radetzki & Wårell (2021), A Handbook of Primary Commodities in the Global Economy · pdf pp. 152-153, 159-161
• Geman (2005), Commodities and Commodity Derivatives · pdf p. 19
• Pirrong (2014), The Economics of Commodity Trading Firms · pdf p. 23
• Humphreys (2015), The Remaking of the Mining Industry · pdf pp. 40, 103

See it in the table

Check your understanding

A wheat farmer hedged at an implied $5.06 a bushel and harvest prices rose to $6.30. What was his net price, ignoring basis changes and costs?

Why can a fully hedged copper fabricator face a cash squeeze if copper falls?

Which hedge carries the most basis risk?