What is spot–futures arbitrage?

If someone offers you an "arbitrage" strategy and cannot explain it in two sentences, walk away. Here is ours in two sentences: we buy gold at today's price and, at the same instant, sell a contract to deliver gold at a future date for a slightly higher price. When those two prices meet, the difference we locked in is the profit.

The basis: why two prices exist for the same thing

Gold has a price today — the spot price. It also has a price for delivery in three months' time — the futures price. These are almost never the same number, and the gap between them has a name: the basis.

The futures price is usually higher, for reasons that have nothing to do with anybody's opinion about gold. Somebody holding physical gold for three months has to store it, insure it, and tie up capital that could have earned interest elsewhere. Those costs get priced into the future delivery. The market is not predicting gold will rise — it is charging for time.

That is the key insight, and it is what separates arbitrage from speculation. The gap exists because of arithmetic, not opinion.

Why the gap always closes

A futures contract has an expiry date. On that date, the contract settles at the spot price — it has to, because delivery happens then. So as expiry approaches, the futures price and the spot price converge. They must. It is the one thing in financial markets you can rely on.

So if you hold both sides — long spot, short futures — the gap between them shrinks to zero, and the amount it shrinks by is yours.

A worked example

Now watch what happens if gold falls 5% before expiry. The spot position loses roughly $120.60 an ounce. But the futures position — which was sold — gains roughly $122.40. The two cancel out. What remains is the $35.20 that was locked in at the start.

Now run it the other way: gold rises 5%. The spot position gains, the futures position loses, they cancel out, and the $35.20 is still there. This is what "market neutral" means in practice. The direction of gold is not a variable in the outcome.

Why this needs an algorithm

Three reasons, and all of them are about speed rather than cleverness.

• Both legs must fill together. If you buy spot and the futures price moves before your sell order lands, you are no longer hedged — you are simply long gold, which is an entirely different risk. Machines place both orders in the same instant. Humans cannot.
• The gaps are small and frequent. A 1.4% basis captured repeatedly across a year is meaningful. Catching it once by hand is not a business.
• The opportunity moves. The widest spread is not always in the same contract month. Software watches every combination continuously; a person watches one screen.

What can go wrong — honestly

Arbitrage is lower risk than directional trading. It is not no risk, and anyone who tells you otherwise is selling something.

• The basis can compress. When markets are calm or financing costs fall, the gap narrows. Narrower gap, smaller return. Some months are simply thinner than others.
• Execution slippage. In violent markets, the two legs can fill at slightly different moments and prices, eating into the spread you thought you had locked.
• Counterparty risk. The whole structure depends on the broker honouring both sides. This is why the choice of trading platform matters as much as the strategy.
• Margin. The short futures leg requires margin. If it is not managed with room to spare, a sharp move can force a position to close early — at exactly the wrong moment.

How to check whether it is real

Anyone can describe arbitrage. The question is whether the account actually does it. Ask for read-only access to a live account and look at the trade history: you should see positions opened in pairs, at nearly the same timestamp, on opposite sides. If you see a stream of one-directional trades, whatever you are looking at is not arbitrage, regardless of what it is called.

We publish read-only logins to our own live accounts so you can run exactly that check before speaking to us.