The trade
A cash-and-carry trade (also called the basis trade) has two legs of the same size: buy the coin on spot and sell a dated futures contract on it, then hold both until the future expires. It only pays when the future trades above spot, which is the usual state for BTC dated futures (contango). The gap between the two prices is the basis, and it is what the trade earns.
Why the return is locked at entry
A dated future settles against the spot index on its expiry date, so its premium has to shrink to zero by then. Whatever BTC does in between, the spot leg gains what the short loses, or the other way round. What is left at expiry is the premium you sold at: known on the day you open the trade, before costs. That is the difference from almost every other crypto yield, whose rate is reset as you go.
Annualising: days to expiry matter
A 1.5% premium is a very different return on a contract that expires in a month and one that expires in a year. To compare them, scale the premium to a year: annualised basis = (futures / spot − 1) × 365 / days to expiry. This site uses that simple, not compounded, rate on the futures term structure page and in the box above. It describes holding to expiry; it is not a yield you earn by the day if you close early.
Worked example
An illustrative trade, with round numbers rather than live prices: BTC spot at $100,000, a quarterly future at $101,500 with 91 days to expiry. Put $10,000 into 0.1 BTC on spot and sell 0.1 BTC of the future.
- Premium: $101,500 / $100,000 − 1 = 1.5%, or $150 on $10,000.
- Annualised: 1.5% × 365 / 91 ≈ 6.0% a year, simple.
- After fees of 0.05% on each of four fills (open and close both legs, $20): $130, about 5.2% annualised.
| BTC at expiry | Spot leg | Short future | Total |
|---|---|---|---|
| $80,000 | −$2,000 | +$2,150 | +$150 |
| $100,000 | $0 | +$150 | +$150 |
| $120,000 | +$2,000 | −$1,850 | +$150 |
The total is the same at every expiry price: the $150 premium, before fees. For comparison, the nearest BTC quarterly future (expiring 25 Dec 2026) currently shows +5.64% annualised on Binance and Bybit.
Costs that eat the return
- Trading fees and spread. Four fills if you close both legs, two if the future settles and you sell spot at the same time. On a premium of a percent or two, they are a real share of the return.
- Margin on the short. The future needs collateral of its own. Money posted as margin is money not in the spot leg, so the return on the capital you really commit is lower than the premium on the spot leg alone.
- Funding the spot leg. If the spot purchase is borrowed, or the cash could earn a rate elsewhere, that cost comes off the basis. A premium below what the cash earns risk-free is a loss in real terms.
- Early exit. The return is locked only at expiry. If the basis widens after you enter and you have to close, you buy the future back at a bigger premium than you sold and realise a loss.
The perp version: short the perp, collect funding
The same hedge works with a perpetual future instead of a dated one: hold spot, short the perp, and collect funding while it is positive. There is no expiry, so no premium has to converge, and the return is whatever funding pays from interval to interval. It can be higher than a quarterly's basis in a hot market, but it is not locked: funding is reset every interval and can turn negative, so the short then pays. BTC perp funding is 4.6% as an APR right now, weighted by open interest. Funding arbitrage covers that version and its cross-exchange variant; settled funding history shows what it actually paid and the calculator runs the numbers for a position.
| Dated future | Perpetual | |
|---|---|---|
| Return comes from | Premium at entry | Funding payments |
| Known in advance | Yes, if held to expiry | No, reset each interval |
| Ends | At expiry | When you close |
| Can turn negative | Only by exiting early | Yes, when funding flips |
Risks
- Exchange risk. Both legs often sit on one exchange, or the spot on one and the future on another. If an exchange fails, halts withdrawals or changes its rules, the hedge can break and funds can be lost.
- Liquidation of the short leg. When BTC rallies, the short future loses before the spot gain can be used as margin, especially across exchanges. Without enough collateral the short is liquidated and the trade is left unhedged, long spot only.
- Basis widening before expiry. A wider premium is a paper loss on the open trade and a realised one if you must close early.
- Contract details. Coin-margined (inverse) futures pay in BTC, so the hedge ratio drifts with price; expiry times, settlement indices and delivery rules are set per exchange.
- This page explains how the trade works. The basis shown is a gross rate, not a risk-free yield, not a recommendation and not financial advice. CME Bitcoin futures positioning, where much of the institutional basis trade sits, is on the CME COT page.
Checked against the exchanges' own documentation on 19 October 2026: Binance: USD-M market API reference · Bybit: instruments and delivery times · Bybit: delivery prices · Binance: introduction to funding rates. Not financial advice.