If you hold dollars in crypto and want them to earn something without taking a view on whether bitcoin goes up, there are broadly two routes. You can lend them — to a protocol, an exchange, or a borrower — and collect interest. Or you can put them behind a market-neutral trading strategy and collect a market-structure spread.
Both are described as non-directional. Both are quoted as a percentage. That is roughly where the similarity ends, and the differences are the whole decision.
This piece compares the two on four dimensions that actually separate them: where the return comes from, what each one risks, how liquid each is, and — the one that matters most and gets discussed least — what each does in a drawdown. If you want the mechanics of the trading side first, the funding rate arbitrage guide covers construction and failure modes in full, and the glossary defines the terms used here.
No verdict at the end. The two have different shapes, and which shape suits you depends on what you are being paid to bear.
Where each return actually comes from
This is the question that determines everything downstream, and "yield" as a word deliberately obscures it.
Stablecoin lending pays you for credit and liquidity. When you deposit USDC into a lending market, a borrower pays interest because they want leverage or working capital. Your return is that borrower's cost, minus the platform's cut. You are a lender. The risk you are paid to bear is that the borrower, the protocol, or the stablecoin itself fails.
Funding rate arbitrage pays you for taking the other side of leverage demand. A position holding spot and shorting the perpetual collects the funding payment that leveraged longs pay to maintain their exposure. Your return is their cost of leverage. You are a liquidity provider to the derivatives market. The risk you are paid to bear is that the payment inverts, the spread compresses, or the venue holding your collateral fails.
Both are, at bottom, selling something to leveraged traders. Lending sells them capital; funding arbitrage sells them the other side of their position. That similarity is worth holding onto, because it explains why both returns compress in the same conditions — when leverage demand falls, both get worse at once.
What each one actually risks
Published comparisons tend to list risks generically. Here is the specific version.
Stablecoin lending carries:
- Depeg risk. The asset you are denominated in can stop being worth a dollar. USDC broke its peg during the 2023 banking stress; TerraUSD went to zero.
- Smart-contract risk. On-chain lending markets have been exploited. Euler Finance lost $197M in 2023.
- Counterparty and custody risk. On a centralised platform, your deposit is a claim on that platform.
- Opacity about the source. A quoted rate does not tell you what is generating it. Some products advertising stablecoin yield are themselves running funding arbitrage underneath — in which case you hold both risk sets and see only one.
Funding rate arbitrage carries:
- Funding inversion. The primary return source can go negative and stay there. Bitcoin funding was negative for 46 consecutive days through mid-April 2026.
- Spread compression. The more capital runs the trade, the less it pays. Carry fell to around 4% by mid-2025 as exchanges industrialised the short side.
- Execution risk. The two legs must fill together; slippage between them is unhedged exposure against a return measured in basis points.
- Margin and forced closure. A rally can liquidate the short leg of a hedged position. Worse, exchanges can force-close profitable hedges through auto-deleveraging, which ranks its targets by unrealised profit — meaning a hedge that is working is first in the queue.
- Exchange counterparty risk. Collateral must sit on trading venues. This is the largest tail exposure and holding more assets does not reduce it.
The honest comparison: lending concentrates risk in one or two places — the asset's peg and the platform. Funding arbitrage spreads it across more places, each individually smaller, several of which are operational rather than credit. Neither list is shorter than the other. They are differently shaped.
Liquidity and lock-up
A real difference, and one that is easy to check before committing.
Most stablecoin lending is open-ended: deposit and withdraw on demand, subject to the market having liquidity. That flexibility is genuine, and it is worth something — though it is also the first thing to disappear under stress, when withdrawal queues form precisely as you want out.
Funding arbitrage as a managed product typically runs on lock periods, because positions need to be held through funding cycles that a short mandate would have to exit early. A lock is a real cost: you cannot act on your own change of view. It also buys something, which is that the strategy is not forced to unwind into a bad market because of redemptions.
The useful question is not which is more liquid on paper. It is which one is still liquid on the day you want your money, and the answer for both is: less than advertised.
What each does in a drawdown
This is the dimension that separates them, and it is the one most comparisons skip.
Think about what a sharp market-wide selloff does to each.
Lending: borrowers get liquidated, leverage demand collapses, and rates fall — often sharply, right when you would most like to be earning. If the stress is severe enough to pressure the stablecoin itself, or a protocol it depends on, the principal is exposed rather than the yield. The failure mode is correlated with everything else going wrong.
Funding arbitrage: perpetuals swing to a discount, funding inverts, and the position starts paying out. Simultaneously, margin is stressed, liquidity thins, and the probability of forced closure rises. In October 2025, auto-deleveraging closed the short legs of professional delta-neutral books during a liquidation cascade, leaving them holding unhedged spot into a falling market — without any exchange failing.
So both deteriorate in the same conditions. Neither is a diversifier against crypto beta in the moment that matters, whatever the delta-neutral or non-directional framing implies. If you are holding one of these as the stable part of a crypto portfolio, the honest expectation is that it will have its worst month in the same month everything else does.
That is not a reason to avoid either. It is a reason not to size either as though it were uncorrelated.
A decision framework, not a verdict
Four questions, in the order they should be asked.
1. What risk are you actually willing to be paid for? If you would rather hold credit and peg risk than operational and venue risk, lending suits you. If the reverse, funding arbitrage does. This is a preference, not a right answer.
2. Can you see what is generating the return? A funding-arbitrage strategy can tell you which venues hold collateral, how the return splits between funding and basis, and what happens when funding inverts. A lending rate often cannot tell you where it comes from. Opacity is itself a risk, and it is the one most consistently underpriced.
3. What period is the quoted number from? Any return figure for either route calibrated on 2024 conditions is describing a market that no longer exists. Ask what the rate was over the last twelve months, not the last twelve days.
4. Does the number need a denominator check? Funding carry is routinely quoted on notional rather than on capital committed, which makes it look materially larger than what you would receive. Lending rates mix APR and APY quoting. Neither is dishonest, and both need normalising before they can be compared.
If you want to see how we answer the second and third questions for our own strategy, how we run it sets out the venue policy and risk limits, the fund plans show how lock periods map to mandates, and about us covers the structure behind them.
By VectorTraders Editorial Team · Published SEP 22, 2026.
This article is educational and is not investment advice. We operate a funding rate arbitrage strategy and are not a neutral party to this comparison. Rates and conditions described change continuously. Capital deployed in either route is at risk, including total loss.
This article is for information only and is not investment advice. Past performance does not guarantee future results.
