How Futures Prices Are Actually Determined
Investors watching a futures price move without understanding what anchors it to today's spot price can badly misread what the market is actually signaling about the future. The cost-of-carry relationship explains, with real arithmetic, exactly why futures prices sit where they do, and why a rising futures price is not automatically a bullish forecast.
The core mechanism: cost of carry and no-arbitrage pricing
For assets that can be stored and traded freely, financial assets like stock indices and currencies, and physical commodities like gold or oil, a futures contract's price is tied tightly to today's spot price through the cost of carry: the combined cost of financing the purchase of the asset today, any physical storage or insurance cost involved in holding it, minus any income the asset generates while being held, dividends on a stock index, interest on a currency, a convenience yield on a commodity in short supply. The relationship is captured by futures price = spot price x (1 + financing cost + storage cost minus income yield), adjusted proportionally for the length of time until the contract's expiration.
The relationship holds so tightly in liquid markets specifically because of arbitrage: if the futures price drifts meaningfully away from what the cost-of-carry formula implies, a trader can lock in a riskless profit by simultaneously buying the underpriced side and selling the overpriced side, and the act of doing so, at scale, across many market participants, pushes the price back toward the formula's implied level almost immediately in liquid markets. This mechanism is precisely why the cost-of-carry relationship functions as a reliable anchor rather than a loose approximation for financial futures on liquid underlying assets, even though the same logic applies somewhat more loosely, because physical storage and transport are not instantaneous or costless, for bulky physical commodities.
The math: two worked examples across different carry costs
Worked example 1: gold, a non-yielding asset. Gold trades at $2,000 per ounce, with financing costs of 5% annually and no income yield at all, since gold pays no dividend or interest to whoever holds it. A one-year futures contract should price at roughly $2,000 x (1 + 0.05) = $2,100. If the market instead quotes the one-year futures contract at $2,150, an arbitrageur can borrow money to buy gold at $2,000, simultaneously sell the futures contract at $2,150, and lock in a riskless profit: at expiration, the arbitrageur delivers the gold against the futures contract for $2,150, repays the loan plus 5% interest, or $2,000 x 1.05 = $2,100, and pockets the difference of $2,150 minus $2,100 = $50 per ounce, entirely risk-free. This exact kind of trading activity is what pulls an overpriced futures contract back down toward the cost-of-carry level.
Worked example 2: a dividend-paying stock index. A stock index trades at 4,000 points, with financing costs of 5% annually and a dividend yield on the underlying stocks of 2% annually, income the futures buyer forgoes by holding the futures contract instead of the actual stocks, since a futures position collects no dividends. The one-year futures price should be roughly 4,000 x (1 + 0.05 minus 0.02) = 4,000 x 1.03 = 4,120. Notice the dividend yield partially offsets the financing cost here, producing a smaller futures premium over spot, roughly 3%, than the non-yielding gold example's full 5% premium. If the dividend yield were instead higher than the financing rate, say 6% against 5% financing, the formula would actually produce a futures price slightly below the spot price, a state called backwardation, illustrating that a futures price trading below spot is not inherently a bearish signal, it can simply reflect a high income yield on the underlying asset that the futures holder is giving up.
What the evidence shows about carry and futures pricing
The cost-of-carry relationship has been tested extensively against actual market data across financial futures markets, stock index futures, currency futures, and interest rate futures in particular, and the empirical fit is consistently tight, with observed deviations from the formula's implied price typically small and short-lived, closing within minutes to hours in the most liquid contracts as arbitrage capital responds, consistent with the theoretical no-arbitrage logic holding up robustly in practice for financial assets that are easy to borrow, store, and transact in at low cost.
The fit is measurably looser, though still broadly directionally consistent, for physical commodity futures, where storage is neither instantaneous nor free, and where an additional factor absent from pure financial assets, the convenience yield, plays a documented role. Convenience yield reflects the practical value of holding physical inventory on hand, an oil refiner benefits from having crude oil in storage rather than only a futures contract promising delivery later, insuring against the operational risk of a supply disruption, and this value is genuinely difficult to observe or estimate directly, unlike financing costs or dividend yields, which are readily quoted. Studies of commodity futures curves consistently find that convenience yield rises sharply during periods of tight physical supply and falls, sometimes toward zero, when inventories are abundant, and this shifting convenience yield is a primary documented driver of commodity futures curves moving between contango and backwardation over time, a pattern with a much cleaner, better-behaved explanation in financial futures markets than in physical commodity markets.
A further and separate empirical finding concerns using the shape of a futures curve, the pattern of prices across different expiration months, as a forecasting tool for the future spot price itself, a question distinct from whether the cost-of-carry formula correctly prices a single contract relative to today's spot. Research testing whether current futures prices are unbiased predictors of the eventual future spot price has produced genuinely mixed results across different asset classes and time periods studied, with some evidence of a modest, historically documented risk premium embedded in certain futures markets, meaning the futures price is not always a pure, unbiased average of possible future spot prices, but instead can include compensation demanded by one side of the market, hedgers or speculators, for bearing the risk the other side wants to shed.
This risk-premium finding connects directly back to the cost-of-carry framework rather than contradicting it. The cost-of-carry formula explains why a futures price sits where it does relative to today's spot price, a largely mechanical relationship driven by financing, storage, and yield. The separate risk-premium question asks something different: whether today's futures price, correctly calculated under cost of carry, also happens to be a good forecast of tomorrow's spot price, which depends on factors the carry formula does not address at all, chiefly how much compensation hedgers are willing to pay speculators to take the other side of a risk they want to shed. Keeping these two, related but distinct, questions separate is what allows an investor to correctly interpret both what a futures price is today and what it may or may not be telling you about tomorrow.
Applying this in a real portfolio
For an investor evaluating a futures-based exchange-traded product, common in commodity exposure funds that use futures rather than holding the physical commodity directly, understanding the cost-of-carry relationship explains a return dynamic that catches many investors off guard: a fund that must regularly sell an expiring futures contract and buy the next month's contract, a process called rolling, experiences a return drag when the futures curve is in contango, since it is systematically selling low and buying high across the roll, a real, structural cost distinct from and in addition to the fund's stated expense ratio, and one that has meaningfully eroded long-term returns in several widely held commodity futures funds during extended contango periods.
For an investor trying to interpret a rising futures price as a market signal, the practical lesson from worked example 1 is to first ask how much of that rise is simply explained by the passage of time and the financing cost embedded in the cost-of-carry relationship, rather than immediately reading it as a bullish forecast about the underlying asset's future direction, a distinction that matters considerably when a futures price is being cited in financial media as evidence of where a market expects prices to go.
For an investor considering stock index futures specifically as an alternative to buying the underlying stocks, worked example 2's dividend-yield adjustment is directly relevant: because a futures position collects no dividend income, an investor comparing a leveraged futures-based equity exposure to a direct stock holding needs to account explicitly for the forgone dividend yield when judging whether the futures route or the direct-ownership route is more cost-effective for their specific holding period and objectives.
A further practical checkpoint worth building into any evaluation of a futures-linked investment product is comparing the fund's actual realized return over a multi-year period against the return of simply holding the underlying spot asset directly over the identical period, rather than trusting the fund's stated benchmark index alone. Because the cost-of-carry relationship, and the roll dynamics it drives, can create a persistent gap between a futures-tracking fund's return and the underlying spot asset's own return, this direct comparison, easy to run using a fund's own published historical data, often reveals a meaningfully different real-world outcome than an investor would expect from the fund's name or marketing materials alone.
An investor comparing several commodity-linked funds side by side should specifically check each fund's disclosed methodology for which futures contracts it holds and how frequently it rolls them, since two funds tracking the same underlying commodity can produce meaningfully different long-run returns purely from differences in roll timing and contract selection, a detail buried in fund prospectuses that matters considerably more to long-term results than the headline commodity exposure the fund is marketed around.
Actionable breakdown
- Reading a futures quote correctly
- Separate carry-driven price levels from directional forecasts.
- Check whether the underlying yields income you would forgo.
- Note whether the curve sits in contango or backwardation.
- Evaluating futures-based funds
- Ask how the fund handles rolling expiring contracts.
- Check historical roll drag during past contango periods.
- Compare total costs against a direct-ownership alternative.
- Using the cost-of-carry framework
- Estimate the expected premium from financing and income yield.
- Treat large, sustained deviations as informative, not noise.
- Remember commodity convenience yield resists clean estimation.
Common pitfalls
Reading a rising futures price as a pure bullish forecast: it may simply reflect financing costs compounding over time, not a changed market view.
Ignoring storage costs for physical commodities: they can be substantial for bulky goods and materially change the cost-of-carry math.
Overlooking roll drag in futures-based funds: a persistent contango curve quietly erodes long-term returns beyond the fund's stated expense ratio.
Assuming backwardation always signals bearishness: it can simply reflect a high dividend or convenience yield being given up by the futures holder.
The bottom line
Futures prices are anchored to spot prices through financing, storage, and income costs, and reading a futures quote through that lens, rather than as a raw price forecast, is the difference between correctly and incorrectly interpreting what the market is signaling.
All articles · What a futures contract obligates you to do · Futures trading mechanics · Futures market strategies · Options and derivatives guide