Perpetual futures are the most-traded instrument in crypto, and they carry a mechanism with no equivalent in traditional futures: the funding rate. For traders who understand it, funding creates a recurring, relatively market-neutral source of yield. For traders who do not, it quietly drains an otherwise sound position. This explainer walks through how the perp-spot basis pays, how a delta-neutral structure captures it, and the specific ways the trade fails.
Why perpetuals need funding
A traditional futures contract has an expiry date. On that date, its price converges to spot, because the contract settles against the underlying. A perpetual future never expires, so there is no settlement event to force convergence. Left alone, a perp could drift arbitrarily far from the spot price of the asset it tracks.
Funding is the tether. At fixed intervals (commonly every eight hours, though some venues use one-hour or continuous accrual), longs and shorts exchange a payment. The size and direction of that payment are set by the funding rate, which is derived primarily from the gap between the perp's price and an index price built from spot markets.
When the perp trades above spot, the basis is positive, funding is typically positive, and longs pay shorts. When the perp trades below spot, funding goes negative and shorts pay longs. The payment is an economic incentive: it pushes the crowded side to close and rewards the side that absorbs the imbalance, nudging the perp back toward spot. Crucially, the exchange does not pay funding. It is a transfer between traders.
The basis trade: capturing funding delta-neutral
If positive funding means longs pay shorts, the obvious idea is to be short the perp and collect. The problem is directional risk: a short perp loses money when price rises. The solution is to neutralize price exposure entirely.
The classic structure is cash-and-carry, ported from traditional finance:
- Buy spot (or hold the asset): you own 1 unit of BTC.
- Short the perp for the same notional: you are short 1 unit of BTC.
Your net delta is roughly zero. If BTC rises, the spot leg gains what the perp leg loses, and vice versa. Price direction no longer matters much. What you are left holding is the funding stream. As long as funding is positive, the short perp leg collects a payment every interval while the spot leg just sits there. That is the yield.
The annualized return is simply the funding rate scaled to a year. An eight-hour rate of 0.01 percent, the rough long-run baseline on many venues, compounds to roughly 11 percent annualized; in heated bull markets, funding can spike several multiples higher for sustained stretches. The trade is sometimes described as "delta-neutral basis capture," and at scale it is a staple of crypto market-neutral funds.
When it does not pay
The structure looks clean on paper. The risks are real and concentrated in a few places.
Funding flips. Funding is not fixed. It floats with positioning, and it can turn negative. A negative funding rate means your short perp leg now pays instead of collects, inverting the trade. You can re-flip the position (long perp, short spot via a borrow) to keep harvesting, but each flip costs spread and fees, and borrow to short spot is not always cheap or available. A trade sized for steady positive funding becomes a slow bleed the moment regime shifts.
Liquidation on the perp leg. The hedge is only neutral if both legs survive. The perp leg uses margin and can be liquidated if price moves hard against it before the spot gain is realized or rebalanced, especially on lower-margin configurations. A sharp rally liquidates an underfunded short perp, leaving you suddenly long-only spot into a position you intended to be neutral. Margin buffer and active rebalancing are not optional; they are the trade.
Execution and basis risk. You must enter both legs at prices that lock in the basis you modeled. Slippage on entry, a moving spread between spot and perp, and the cost of rebalancing as delta drifts all erode the funding you collect. On thin pairs, getting filled on both sides at the intended ratio is harder than the spreadsheet suggests.
Venue and counterparty risk. The spot and the perp often sit on the same exchange or within the same custody surface. Funding yield in the single digits to low double digits annually is poor compensation for exchange insolvency, a withdrawal freeze, or a depeg of the settlement stablecoin. Splitting legs across venues reduces single-point failure but adds transfer latency and capital inefficiency.
Reading funding before you trade it
The edge in this trade is not the structure, which is widely known. It is in seeing where funding is actually being paid, how persistent it is, and how it compares across venues, because the same asset can carry very different funding on different exchanges at the same moment. Cross-venue dispersion is both a risk and, occasionally, the opportunity itself.
This is where consolidated data matters. Athenum provides funding rate analytics across 14 exchanges in one view, so you can compare rates, watch for flips, and gauge persistence before committing capital. See the live funding view.
The honest summary
Funding rate arbitrage is not free money. It is a paid service: you are providing liquidity to the crowded side of a leveraged market, and the funding rate is your fee. When positioning is one-sided and persistent, that fee is attractive and largely uncorrelated with price direction. When funding flips, when the perp leg gets liquidated in a violent move, or when a venue fails, the trade stops paying and can cost more than it ever earned. Size the perp margin for survival, model the funding net of fees and rebalancing, and never let a "neutral" trade quietly become directional.