Hedging Basics with Futures

The short answer

A hedge takes a risk you are exposed to and swaps it for a smaller, different one. It does not remove loss. It converts an uncertain outcome into a narrower range of outcomes, and it charges a fee — in costs, in carry, and in the upside you give up — whether or not the thing you were worried about ever happens.

Hedging is the original purpose of futures markets, and it is genuinely useful to producers and holders of real exposure. It is also more mechanically demanding than a single directional position, and it is the wrong place to begin with leverage.

What a hedge is

A hedge is a second position, opposite in direction to an exposure you already hold, taken to reduce the effect of price moves on the total.

The classic case has nothing to do with speculation. Someone holds an asset, or is contractually committed to buying or delivering one, and wants the price risk between now and a future date to matter less. They take an offsetting futures position. If the price falls, the loss on the holding is partly recovered on the short contract. If it rises, the gain on the holding is partly given back.

Note "partly" in both sentences. That word is the whole subject.

What a hedge is not

Four things a hedge is routinely described as, and is not:

• It is not loss avoidance. It reduces sensitivity to one specific price. The residual, the costs and the foregone upside all remain, and each is a real number.
• It is not one-sided. A hedge that protects you from a fall protects you from a rise with equal enthusiasm. If you wanted only the downside removed, you were thinking of a different instrument with a different price.
• It is not a way to hold a losing position longer. Opening an opposing leveraged contract against something you should have reduced is not risk management; it is two positions, two sets of costs and two liquidation prices, arranged so that the loss becomes harder to see.
• It is not a simplification. It adds a position with its own margin requirement, its own funding or roll cost and its own mechanics. There is more to monitor after hedging than before.

Every hedge costs something

Assume the risk you hedged never materialises. You still pay:

1. Transaction costs on both legs, entry and exit — four transactions where you previously had none on the hedge side. Run them through the fee calculator before deciding the hedge is cheap.
2. Carry on the futures leg — funding on a perpetual, roll spread on a dated contract, for the entire life of the hedge. See funding and carrying costs.
3. Margin capital — collateral posted against the hedge leg is capital doing nothing else, and it must stay there.
4. The foregone move — if the price goes the way you were hoping, you do not receive it.

That list is the price of the swap, and it is paid up front in the sense that it does not depend on the outcome. A hedge is only worth its cost when the risk being removed is one you genuinely cannot carry.

Which points at an uncomfortable alternative. If an exposure is large enough to need hedging, selling part of it removes exactly the same risk, costs one transaction instead of four, posts no margin and creates no second liquidation price. Hedging earns its complexity when the holding cannot be sold — a committed delivery, a locked position, a tax or timing constraint. When it can be sold, it usually should be.

Basis risk

Basis risk is the gap between what you hold and what you hedged it with.

A perfect hedge would use an instrument identical to the exposure, in the identical quantity, over the identical period. In practice at least one of those fails. No contract exists on the exact thing you hold, so you use a related one. The contract size does not divide evenly into your exposure, so the quantity is approximate. Your horizon does not match the expiry, so the timing is approximate.

Each mismatch leaves a residual. That residual is basis risk, and it is not a defect in how you set the hedge up — it is a structural feature of hedging with standardised contracts. The best you can do is know how large it is and stop describing the position as protected.

Two instruments that have moved together historically can stop doing so precisely when it matters. Correlations tend to be least stable in exactly the conditions that make people want a hedge, which is the same lesson correlation and portfolio risk makes about diversification. A hedge is diversification with a deadline and a margin requirement attached.

Losing on both legs

The numbers below are invented for teaching. They describe no real instrument and are not a trade idea.

You hold $10,000 of an asset. No contract exists on it, so you hedge with one contract on a related index, notional $10,000, held short. Costs over the period — fees on both legs plus carry on the short — come to $60.

Case A — the hedge works imperfectly. The asset falls 8%, so the holding loses $800. The index falls 5%, so the short gains $500. Net: −$800 + $500 − $60 = −$360. The hedge worked. It did not prevent a loss; it reduced one.

Case B — the correlation fails. The asset falls 8% again, losing $800. This time the index rises 3%, so the short loses $300. Net: −$800 − $300 − $60 = −$1,160.

Unhedged, Case B would have cost $800. Hedged, it cost $1,160. The hedge did not offset the loss, it added to it, and the costs were paid on top of both.

Case B is not exotic. It is what happens whenever the thing you hold underperforms the thing you hedged with — a sector-specific problem, an asset-specific problem, a liquidity event that hits one and not the other. Any hedge built on a proxy carries this outcome permanently.

There is a third case worth naming. If the hedge leg is held on isolated margin and the index moves far enough against it, the hedge leg can be liquidated on its own — leaving the original exposure unhedged, at the worst possible moment, plus the realised loss and the liquidation fee. Margin arrangements across the two legs are part of the hedge design, not an afterthought.

Where this belongs in a learning path

Hedging is not a beginner's first use of leverage, and it should not be the reason anyone opens a futures account.

The reasoning is mechanical rather than moral. A hedge requires you to manage two positions, two margin balances, two cost streams and two liquidation levels, and to form a judgement about how closely two instruments will track each other under stress. That is strictly more than a single directional position demands, and it is demanded of you while you are also carrying the exposure you were worried about in the first place.

A trader who cannot yet size one leveraged position correctly cannot size two offsetting ones, and the failure will be harder to notice because the position looks protected. "Hedged" is a comfortable word. It has talked a lot of people into carrying more total exposure than they would have accepted in a single line.

What to do with this

If you are reading this to work out whether to hedge something, run three checks first.

Can the exposure simply be reduced? If yes, do that instead — it removes the same risk more cheaply and adds nothing to monitor. If no, quantify the cost: fees on both legs, carry for the expected duration, and the margin you will tie up. Then estimate the basis risk honestly by asking what would have to be true for both legs to lose at once, and whether that scenario is plausible.

If the cost and the residual together are not clearly smaller than the risk you are removing, the hedge is not doing what you want it to do. That is a complete answer, and it is the usual one.