Futures: contracts, margin, and mark to market

The textbook definition of a future — a standardised agreement to buy or sell an asset at a fixed price on a fixed date — is accurate and almost useless. It describes the economics and omits the mechanism that actually determines whether you survive the trade.

The defining feature of a future is not the price agreed. It is that the exchange's clearing house re-prices your position every single day and moves cash between accounts to settle the difference. A forward contract lets a loss accumulate quietly until maturity. A future does not let anything accumulate. Every evening, your profit is paid to you in cash and your loss is taken from you in cash, and if you cannot pay, your position is closed by someone else at a price you did not choose. Understanding that one operational fact explains most of the ways people lose money in futures, including the ways in which they lose money on positions that were fundamentally right.

What is actually written in the contract

A futures contract is a specification document, and reading it is not optional. Six fields matter.

The underlying — which index, which grade of crude, which government bond basket. The contract size or multiplier, which converts index points into money. The tick size, the minimum price increment, and the tick value it implies. The delivery month and last trading day. The settlement method: physical delivery of the asset, or cash settlement against a final reference price. And the final settlement price definition, which is usually not the last trade but an average or auction price computed over a defined window.

Concretely: the E-mini S&P 500 future traded on CME has a multiplier of USD 50 per index point and a minimum tick of 0.25 index points, so one tick is worth USD 12.50 and one full index point is worth USD 50. Eurex lists futures on the SMI and on European benchmarks with their own multipliers denominated in francs and euros. Do not assume; open the specification page for the contract you intend to trade and read the six fields above before you place an order. More retail damage is done by traders who misread a multiplier than by traders who misread the market.

Cost of carry: why the future is not the spot price

A futures price is not a forecast. For an asset you can hold, it is arithmetic.

If you could buy the index today and sell a future above the no-arbitrage level, you would borrow the purchase price, collect the dividends, deliver into the future and pocket a risk-free difference. Arbitrage removes that, which pins the future to spot adjusted for the cost of financing minus the income foregone:

F ≈ S × (1 + (r − q) × T)

where r is the financing rate, q the dividend yield, and T the time to expiry in years.

Take an index at 12,000, a financing rate of 1.0%, a dividend yield of 3.0%, and three months to expiry. Then F ≈ 12,000 × (1 + (0.01 − 0.03) × 0.25) = 12,000 × 0.995 = 11,940. The future trades 60 points below the index.

This is not bearishness. It is the fact that a holder of the index collects three months of dividends the futures holder does not. Swiss equities have historically carried relatively generous dividend yields, and Swiss franc financing rates spent years at very low and at times negative levels — the arithmetic above is precisely why SMI futures routinely traded below the cash index in that period, and why anyone reading that discount as a bearish signal was reading a financing calculation as sentiment.

When interest rates are themselves uncertain and correlated with the underlying, futures and forward prices separate slightly, because the daily cash settlement of a future is reinvested or financed at rates that move with the market. Cox, Ingersoll and Ross set out the relationship formally in 1981. For equity index futures over a few months the difference is small; for interest rate futures it is not, and that is the origin of the convexity adjustment that fixed income desks apply.

Margin is collateral, not a down payment

Three terms get confused constantly, and the confusion is expensive.

Initial margin is the good-faith deposit the clearing house requires to open a position. It is sized to cover a plausible adverse move over the time it would take to liquidate you — one or two days for a liquid contract. It is not a partial payment toward ownership, and it does not cap your loss. Clearing houses compute it with portfolio risk models: CME uses SPAN, Eurex Clearing uses its Prisma methodology, and both raise requirements when volatility rises — usually at the exact moment you would least like to find more cash.

Maintenance margin is a lower threshold. If your account equity falls below it, you receive a margin call to restore the balance to the initial level.

Variation margin is the daily settlement itself. In futures this is genuinely a payment, not a pledge: yesterday's loss leaves your account and lands in the account of whoever was on the other side. This is why a futures position has no unrealised P&L in any meaningful sense. Everything is realised, every day.

The daily cycle, with real numbers

Buy one E-mini S&P 500 future at 5,000.00. Notional exposure is 5,000 × USD 50 = USD 250,000. Suppose the clearing house requires USD 14,000 initial margin and USD 12,700 maintenance.

Day 1. The contract settles at 4,970. You are down 30 points × USD 50 = USD 1,500. Your balance falls to USD 12,500, below the USD 12,700 maintenance level. You receive a call to restore the account to USD 14,000, so you wire USD 1,500.

Day 2. Settlement at 5,010. You gain 40 points, USD 2,000. Balance is USD 16,000. The USD 2,000 above the initial requirement is genuinely yours and can be withdrawn.

Day 3. Settlement at 4,930. You lose 80 points, USD 4,000. Balance falls to USD 12,000, again below maintenance, and you post USD 2,000 to return to USD 14,000.

Cumulatively the index has moved from 5,000 to 4,930 — 70 points, or USD 3,500 of loss. The cash flows confirm it: −1,500 + 2,000 − 4,000 = −USD 3,500. You have deposited USD 17,500 in total and hold a balance of USD 14,000.

Notice what a 1.4% move in the underlying did to a 5.6% deposit. Notice also that the index finished the three days near where a patient long-term holder would shrug at, while the futures trader was called for cash twice.

Solvency is not the same as liquidity

This is the lesson that generalises beyond futures. Daily settlement converts credit risk into liquidity risk. The clearing house has protected itself and everyone else from your default — but it has done so by demanding that you be able to fund adverse moves in cash, immediately, regardless of whether your view eventually proves correct.

A hedger is especially exposed to this. If you are long physical inventory and short futures against it, the hedge is sound in aggregate, but the two legs settle differently: the futures leg bleeds cash daily when prices rise, while the gain on the physical inventory is unrealised and finances nothing. The most-cited case is Metallgesellschaft's US refining subsidiary in 1993, which held long-dated supply commitments hedged with a rolling stack of short-dated futures. When oil prices fell, the futures leg generated margin calls of a size the parent was unwilling to keep funding, the positions were unwound, and the reported losses ran well over a billion US dollars. Whether the underlying strategy was sound is still argued. That the funding requirement was the binding constraint is not.

The practical rule: a futures position needs two budgets. One for the loss you are willing to take, and one for the cash you must be able to produce along the way. They are different numbers, and the second is usually larger than people plan for.

Convergence, expiry and the roll

As expiry approaches, the carry term shrinks toward zero and the futures price converges on the spot price. On the last trading day the two are effectively the same, which is what makes the contract a usable hedge.

Two consequences follow. First, if you do not want to make or take delivery, you must act before the last trading day. Physically settled contracts — commodities, government bonds — will deliver against you if you hold them through expiry, and the operational and financing consequences are unpleasant for anyone not set up for it. Cash-settled contracts simply pay out against the final settlement price, which is defined in the specification and is typically an average or auction print rather than the last trade.

Second, maintaining a long-term position means rolling: closing the expiring contract and opening the next one. Most benchmark contracts follow a quarterly cycle expiring in March, June, September and December. Each roll pays a spread and crosses a bid-offer, and if the curve is unfavourable the roll has a persistent cost that compounds. This is the main reason a futures-based position on a commodity can drift far from the commodity's spot return over years, and the sibling lesson on hedging and basis risk develops what that gap means for a hedger.

Swiss practicalities

Two points matter locally.

Currency. A Swiss-resident trader holding a USD-denominated future carries two exposures: the contract's exposure to the index, and the franc value of the margin balance and every daily settlement. Those settlements happen in the contract currency. A profitable position can shrink in franc terms; the hedge on the second exposure, if you want one, is a separate FX position.

The coverage rule Swiss institutions live under. The clearest statement of the discipline this lesson is really about comes from pension law. The ordinance governing occupational pension investments (BVV2/OPP2) permits a Swiss pension fund to use derivatives, but attaches two conditions: their use must produce no leverage effect on the fund's total assets, and every obligation arising from a derivative position must be covered — cash or equivalent liquidity behind a position that can require a payment, the underlying itself behind a written call. A margin deposit is explicitly not treated as coverage.

That rule exists because of exactly the mechanic described above. A futures position lets an institution change its market exposure without moving the capital that exposure represents, and daily settlement then demands cash on a schedule the position's economics do not control. The regulator's answer was not to ban the instrument but to require that the money be identifiable somewhere before the position is opened. A private investor is under no such obligation, which is precisely why the discipline is worth importing voluntarily: decide which assets stand behind the contract, and confirm they can be turned into francs on the day the clearing house asks.

What to take away

• The contract specification — multiplier, tick value, last trading day, settlement method — determines your real exposure, and misreading a multiplier has ruined more retail accounts than misreading a market.
• The futures price is spot adjusted for financing minus income; an index yielding 3% against a 1% financing rate trades roughly 60 points below a 12,000 spot level at three months, and that discount is arithmetic, not a forecast.
• Variation margin is a cash settlement, not a pledge: in the worked example a 70-point adverse move produced two margin calls and USD 3,500 of realised loss on a USD 250,000 notional held against USD 14,000 of collateral.
• Daily settlement replaces counterparty risk with funding risk, which is why a fundamentally sound hedge can still be liquidated — plan the cash you must be able to produce separately from the loss you are willing to accept.
• In Switzerland, futures P&L is settled in the contract currency and creates a second, unhedged franc exposure — and the pension-fund rulebook's answer to leverage, requiring every derivative obligation to be covered by identifiable assets rather than by margin, is a standard a private investor can usefully apply to themselves.

Sources

• John C. Hull, Options, Futures, and Other Derivatives — chapters on futures mechanics, margining, hedging and the cost-of-carry relationship.
• John C. Cox, Jonathan E. Ingersoll and Stephen A. Ross (1981), "The Relation between Forward Prices and Futures Prices", Journal of Financial Economics.
• CME Group contract specifications and SPAN margin methodology documentation.
• Eurex Clearing, Prisma margin methodology documentation and Eurex contract specifications for Swiss and European index futures.
• Bank for International Settlements, Quarterly Review — recurring analysis of central clearing, margin procyclicality and liquidity demands in stressed markets.
• Swiss Financial Market Infrastructure Act (FinfraG / FMIA) and FINMA materials on clearing and reporting obligations.
• Ordinance on Occupational Retirement, Survivors' and Disability Pension Plans (BVV2 / OPP2), provisions on the use of derivative financial instruments and the coverage requirement.
• Christopher L. Culp and Merton H. Miller (1995), "Metallgesellschaft and the Economics of Synthetic Storage", Journal of Applied Corporate Finance — the funding-versus-solvency argument, from the side that defended the strategy.