Options on Futures

Walk onto a commodity desk and ask to trade "a call on crude oil," and you'll almost never be buying the right to take delivery of actual barrels. You'll be buying a call on the futures contract for crude oil — a completely standardized, exchange-traded, cash-settleable instrument that happens to track the same underlying price. Options on futures (rather than on the spot asset itself) dominate the commodity world, and much of the interest-rate world too, for a reason that sounds like a footnote but changes the pricing machinery in a real way: entering a futures contract costs nothing upfront. That single fact — no premium to pay just to take the position — is the whole story of this lesson.

Why "no upfront cost" changes everything

Every bound and every parity relation you've built so far — for stocks, for indices, for currencies — started from the cost of holding the underlying today: pay S_0 and you own the asset. A futures contract is different in kind. Entering a long futures position costs \$0 to initiate (margin is a refundable performance bond, not a purchase price), and its value is marked to market and settled in cash every day. The "asset" underlying a futures option is therefore not a thing you pay F_0 to own — it's a zero-cost bet whose current fair (forward) level is F_0.

Re-running the same no-arbitrage arguments from the bounds lesson with a zero-cost underlying in place of a stock gives the futures analogue directly — just swap S_0 for the discounted futures price F_0 e^{-rT} everywhere a plain-vanilla formula uses spot:

Lower bound: c \ge \max\!\big((F_0 - K)e^{-rT},\ 0\big).
Upper bound: c \le F_0 e^{-rT}.
Put–call parity: C - P = (F_0 - K)\,e^{-rT}.

This is exactly the pattern from index and currency options, where a continuous yield q replaced part of spot's cost of carry — except here the substitution is total. Since the futures price already embeds the full cost of carry (storage, financing, any convenience yield), pricing off F_0 means you never have to separately estimate storage costs or a commodity's convenience yield at all: the futures market has already done that work and packaged it into one quoted number. This is the essence of what practitioners call Black's model — the same lognormal machinery as Black–Scholes, just with the forward/futures price standing in for spot everywhere.

Watch the two bounds pivot around a discounted futures price, the same shape as the stock-option bounds from earlier — only now driven by F_0, not spot:

Why commodity desks live here

A barrel of crude oil is a genuinely awkward thing to hold: it needs a physical tank, insurance, transport, and a buyer who actually wants delivery at a specific port on a specific day. The spot market for many commodities is thin, fragmented, and expensive to access directly. The futures market for the same commodity, by contrast, is deep, liquid, exchange-cleared, and — crucially — can be traded (and unwound) by anyone with a brokerage account, with zero interest in ever touching a barrel of oil.

Options on futures inherit all of that liquidity. A desk that wants to hedge a client's exposure to natural gas prices, or write a call to collect premium on a wheat position, trades a contract that settles against the same exchange-traded futures price they already use to hedge everything else — instead of having to separately estimate a spot price that may barely trade, plus storage costs, plus a convenience yield nobody agrees on. Pricing and hedging both collapse onto one clean, observable number: F_0.

Three-month WTI crude futures trade at F_0 = \$78.00 per barrel. A call struck at K = \$75.00 is priced with r = 5\%. The lower bound is

(F_0 - K)e^{-rT} = (78.00 - 75.00)\,e^{-0.05 \times 0.25} = 3.00 \times 0.9876 \approx \$2.96.

Notice what never entered this calculation: crude's spot price, storage costs at Cushing, tanker rates, or any convenience yield. The futures price already reflects all of that — which is exactly why practitioners reach for a futures option instead of trying to price one off a messy physical spot market.

In April 2020, as global demand collapsed and storage tanks at the Cushing, Oklahoma delivery point neared physical capacity, the front-month WTI futures contract briefly traded at negative prices — sellers paid buyers roughly \$37 a barrel to take oil off their hands, because nobody had anywhere left to actually put it. It's a vivid, extreme illustration of exactly why options on futures are so valuable to a desk: the buyer of a futures option never has to worry about finding a tank. They can let the option lapse, or close it in cash, entirely insulated from the physical delivery chaos that made spot crude briefly worth less than nothing. Options on the physical commodity itself would have forced someone, somewhere, to solve a very real storage problem; options on the futures contract let the whole market simply mark a price and move on.

The single most common slip: plugging the spot price into a futures-option formula out of habit. Once you're pricing an option on a futures contract, spot doesn't appear anywhere — every formula uses F_0, the futures price itself, discounted at the risk-free rate.
Exercising an American option on a futures contract does not hand you a barrel of oil. It hands you a futures position struck at K, immediately marked to market against the current futures price — the physical delivery (if any) only happens later, when that futures contract itself expires, and most traders close out long before that point.