Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
What Is a Yield Loop?
A yield loop uses recursive borrowing on yield-bearing assets to amplify returns — until a borrow rate spike or price drop unwinds the whole stack at once.
A yield loop is a DeFi strategy where you deposit a yield-bearing asset as collateral, borrow against it, and use the borrowed funds to buy more of the same asset. Then you repeat the cycle — amplifying both your yield and your liquidation risk with each pass.
Most DeFi strategies ask you to deploy capital once and collect a return. A yield loop asks the same capital to do the job several times over. Each cycle adds more collateral and more debt, growing your effective position well beyond your original deposit. That amplification is the point — and it’s also the hazard.
The strategy has moved from CT threads into mainstream protocol interfaces. Summer.fi Multiply, Morpho vaults, and Contango all offer looping with automated execution. But automation doesn’t change the underlying math, and it doesn’t protect you from a bad outcome. Understanding how a yield loop works at the level of capital exposure — not just clicks — is what separates informed use from an expensive mistake.
Key Takeaways
- A yield loop compounds a yield-bearing asset’s return by recycling borrowed capital through repeated deposit-borrow cycles.
- The profit comes from the spread between the asset’s staking or native yield and the cost to borrow against it — when that spread flips negative, the loop bleeds slowly until you exit or get liquidated.
- Cascade liquidations can happen in minutes during sharp price drops, as every looped position unwinds simultaneously and amplifies the sell pressure.
—
What a Yield Loop Actually Does to Your Capital
The simplest yield loop looks like this: you deposit 1 ETH worth of wstETH (Lido’s wrapped staked ETH) into Aave. Aave lets you borrow up to 90% of that value in ETH — the loan-to-value, or LTV, ceiling. You borrow 0.85 ETH, swap it for more wstETH, and deposit that too. Then borrow again. Then deposit again.
After five cycles, your deposited collateral might be worth 3.5 ETH — but your original capital was 1 ETH. You now have 3.5 ETH of staking yield working for you, offset by 2.5 ETH of variable-rate debt. If wstETH yields 3.8% and your borrow rate is 2.5%, you’re capturing a 1.3% spread on 3.5 ETH instead of 1 ETH. That’s yield amplification through recursive borrowing.
What beginners miss is that the 2.5 ETH of debt is real. Every loop deepens both your position and your obligation. The nominal size of the trade is 3.5 ETH. The collateral at risk if prices move is the full 3.5 ETH, minus whatever the liquidation engine can recover. Your original 1 ETH is the floor of what you might get back — in a bad unwind, you could recover much less after liquidation penalties and gas.
DeFi looping works because the borrowed asset (ETH) and the collateral (wstETH) are tightly correlated. wstETH is just staked ETH with yield accruing. The price ratio between them stays close to 1:1 plus accumulated staking yield. That tight relationship keeps the health factor stable during normal market conditions. It’s only when that correlation breaks — or when borrow rates spike past the yield — that the loop becomes a problem.
—
The Maths Behind a Yield Loop (And Why Spreads Matter)
The yield loop earns money through the yield spread: what the collateral asset generates minus what the borrowed asset costs. If that spread is positive, each loop cycle adds a small return. Multiply across five or six cycles and the effective APY climbs meaningfully above what holding the asset outright would produce.
Here is a concrete example using realistic 2025–2026 figures:
| Variable | Example Value |
|---|---|
| wstETH staking yield (gross) | 3.8% APY |
| ETH borrow rate on Aave (variable) | 2.5% APY |
| Yield spread per loop | 1.3% |
| Number of loops | 5 |
| Effective APY on original capital | ~5.9% |
The effective APY rises because each loop adds another layer of the spread without adding new capital. But the table above hides a critical variable: the borrow rate is not fixed. Aave’s ETH borrow rate can spike during high utilisation periods — sometimes to 5–8% or higher. When the borrow rate crosses the asset yield, the spread turns negative. The loop then costs money every day you hold it.
Aave’s E-Mode (efficiency mode) is designed for this use case. It raises the LTV ceiling for correlated asset pairs — for example, allowing up to 93% LTV on wstETH/ETH — which lets loopers extract more yield from each cycle. But higher LTV also means a tighter health factor buffer. More efficiency, less cushion. Two benefits that always trade against each other.
The basis trade is the closest cousin in DeFi strategy — it also earns a spread between two correlated positions, and the same rate-spike risk applies.
—
What Assets Work in a Yield Loop
Yield loops are not ETH-only. Any asset that earns yield while also serving as accepted collateral can theoretically loop — though the borrowed asset must be closely correlated to the collateral. If it isn’t, price divergence creates severe liquidation risk.
Three asset categories dominate looping today.
The most common loop collateral is liquid staking tokens (LSTs). wstETH (Lido’s wrapped staked ETH) and weETH (EtherFi’s liquid restaking token) both accrue ETH staking and restaking rewards while being accepted as collateral on major money markets. A wstETH loop on Aave or Morpho is the canonical yield loop — tightest correlation, highest LTV ceiling, and deepest liquidity. The liquid staking guide covers how LSTs and LRTs are structured if you want to understand the asset before using it as loop collateral.
Yield-bearing synthetic stablecoins form the second category. sUSDe, Ethena’s staked synthetic dollar, earns yield from Ethena’s delta-neutral position. A sUSDe loop borrows USDC against the sUSDe collateral — lower liquidation risk because both sides are pegged to the dollar, but exposed to depeg events (more on that in the risks section).
Tokenized real-world assets (RWAs) are the newest entrant. Hamilton Lane’s SCOPE and Apollo’s ACRED both offer yield-bearing tokenized fund positions that protocols are beginning to accept as collateral. The yield spreads on RWA loops can be attractive, but liquidity is thinner and exit options are narrower.
The rule for all three categories is the same: if the borrowed asset drifts away from the collateral in price, the health factor drops and the loop can get liquidated before you can react.
| Asset | Why It Works in a Yield Loop |
|---|---|
| wstETH (Lido) | Accrues ETH staking yield; tightly correlated to ETH borrow asset |
| weETH (EtherFi) | Earns restaking rewards; accepted on Aave, Morpho with high LTV |
| sUSDe (Ethena) | Earns delta-neutral yield; both sides dollar-pegged, so lower liquidation risk |
| SCOPE (Hamilton Lane) | Tokenized fund yield; accepted on select money markets as RWA collateral |
| ACRED (Apollo) | Tokenized credit yield; newer entrant, thinner liquidity than LSTs |
Keep in mind that the list above reflects protocol support as of mid-2026. Collateral eligibility changes — always check the protocol’s current collateral parameters before building a loop.
—
Yield Loop vs. Yield Farming — The Actual Difference
The confusion between yield loops and yield farming is common. Both involve putting crypto to work for more crypto. But the mechanics, risks, and reward structures are entirely different.
Yield farming typically means depositing capital into liquidity pools or protocol vaults in exchange for reward tokens — usually a protocol’s native governance token. The return depends on how long those tokens hold value and whether the liquidity pool fee income is high enough to compensate for impermanent loss. The risk is mostly token depreciation and impermanent loss on the LP side.
A yield loop doesn’t seek reward tokens. It uses borrowed capital to amplify the native yield of an existing asset. No governance token is involved. The return is the spread between what the collateral earns and what the debt costs. The risk is liquidation, not impermanent loss.
| Yield Farming | Yield Loop |
|---|---|
| Deploys capital into pools/vaults for reward tokens | Uses recursive borrow-and-redeposit cycles |
| Earns protocol reward tokens and LP fees | Earns the spread between asset yield and borrow rate |
| Primary risk: token depreciation, impermanent loss | Primary risk: liquidation, borrow rate spike, cascade |
| Single capital deployment | Capital amplified through multiple loops |
Both strategies can run simultaneously — a user can deposit a looped wstETH position into a farming vault that also distributes incentive tokens. But stacking them without understanding each layer adds compounding risk, not just compounding yield. The yield farming guide covers the farming mechanics in depth for anyone who wants the full picture.
—
Where Yield Loops Actually Run
A yield loop can be built manually on any money market that accepts yield-bearing collateral — but doing it by hand means repeating the borrow-swap-deposit cycle multiple times, paying gas on every step. In practice, most loopers use one of four venue types.
Aave V3
The foundational money market for ETH-based yield loops, Aave V3 raises LTV ceilings for correlated asset pairs through its E-Mode feature. Liquidity depth keeps borrow rates relatively stable during normal market conditions. Aave does not automate the loop — users build and manage it themselves or use an automation layer on top.
Morpho
A curated lending layer with vault-based architecture, Morpho lets curators set risk parameters, and users deposit into vaults rather than individual asset markets. Gas efficiency is better than direct Aave interaction, and the vault structure gives clearer collateral and LTV rules at a glance.
Summer.fi Multiply
The most accessible entry point for a new looper, Summer.fi’s Multiply product (formerly DeFi Saver) handles the full deposit-borrow-swap-redeposit cycle in one transaction. Users set a target exposure multiplier and can configure automated deleveraging if the health factor drops below a threshold. It does not, however, protect against a market move that outruns the deleveraging trigger.
Contango
Approaching looping as a structured product built on top of money markets, Contango gives users a fixed-term yield position rather than a perpetual loop to manage daily. This is closer to a crypto derivatives use case than a raw lending position, which changes both the risk and the fee structure.
Gas cost matters at low loop counts on Ethereum mainnet. A three-loop position built manually on Aave during moderate congestion can cost $40–80 in gas, which erodes the spread advantage for smaller positions. Morpho and automation layers reduce that friction, while Solana and Base deployments cut gas costs further — though liquidity and collateral options are narrower on those chains.
—
Yield Loop Risks — What Can Go Wrong and How Fast
Most guides list DeFi looping risks in a flat bullet list. That misses the point. Each risk type moves at a different speed and demands a different response.
Liquidation Risk
Every yield loop has a health factor — a ratio that compares collateral value to debt value, adjusted by the liquidation threshold. When the health factor drops to 1.0, the protocol auto-liquidates enough collateral to bring the position back to safety. In a correlated pair like wstETH/ETH, the health factor is usually stable because the two assets move together. But a short-lived spike in the price ratio between them, or a temporary borrow rate surge, can compress the buffer on a high-LTV loop faster than you can react.
Borrow Rate Spikes
Variable borrow rates can flip a profitable yield loop into a slow loss without any price movement at all. Money market borrow rates respond to utilisation: when most of the supplied ETH is already borrowed, the rate rises sharply. A position earning 1.3% net spread at a 2.5% borrow rate starts losing money the moment the rate climbs to 4%.
Cascade Liquidations
When a sharp price drop triggers liquidations across many looped positions at once, liquidators sell collateral to cover debt, pushing prices down further and triggering more liquidations. DL News reported on the Q4 2025 USDe depeg as a case study in looping at systemic scale: as USDe lost its dollar peg, sUSDe-collateralised loops across multiple protocols began unwinding simultaneously, creating selling pressure that amplified the depeg. The 2022 stETH/Celsius event followed the same logic — a correlated pair temporarily depegged, health factors dropped across overleveraged positions, and the cascade consumed capital far beyond any single user’s exposure.
Smart Contract Risk
Every protocol in the loop stack is a failure point. A Morpho vault exploit, an Aave parameter change, or a bug in a looping automation contract can drain a position before the user can exit. The more protocols a yield loop touches, the wider the attack surface.
The table below maps each risk to a concrete check you can do before and during a loop:
| Risk | Practical Check |
|---|---|
| Liquidation risk | Set a health factor alert at 1.3 or higher; know your liquidation threshold before entering |
| Borrow rate spike | Monitor the variable borrow APY daily on DefiLlama or the protocol dashboard; use a rate-cap alert |
| Cascade liquidation | Reduce loop count and target an effective LTV below 75% during volatile market conditions |
| Smart contract risk | Use protocols with multiple audits; avoid new vaults without a track record |
Understanding the health factor and liquidation threshold is essential before entering any yield loop. The guide to LTV and liquidation buffers covers the math in depth so you can model your exact cushion before committing capital.
—
Does Yield Looping Inflate DeFi TVL?
Short answer: yes, and significantly. The number on a protocol’s DefiLlama page does not represent net new capital in the way most users assume.
When one user runs a yield loop with 10 ETH through five deposit-borrow cycles on Aave, each cycle adds to the protocol’s recorded total value locked. The user’s 1 ETH of original capital might show up as 5 ETH in TVL — the 1 ETH deposited in cycle one, plus the 0.85 ETH in cycle two, plus each subsequent loop. The protocol has not attracted five times more capital. It has the same 1 ETH, recycled through its own machinery.
No one is committing fraud, and no protocol is gaming the data. Recursive borrowing just works this way: every borrow-and-redeposit creates a new deposit record without adding fresh external capital. DefiLlama’s methodology surfaces this issue when you look at “circulating supply” versus “TVL” for looping-heavy protocols — the gap between the two is the fingerprint of yield loop activity.
The practical takeaway for anyone evaluating a looping protocol: look past raw TVL. Net capital inflows, utilisation rates, and the ratio of borrowed to supplied assets are more useful signals of genuine protocol health. A protocol with $1 billion in TVL and an 85% utilisation rate has far less real liquidity buffer than a $400 million TVL protocol at 40% utilisation. The TVL number tells you how much looping is happening — not how much capital is actually at risk or available.
—
Frequently Asked Questions
Is a yield loop the same as yield farming?
No. Yield farming deploys capital into liquidity pools or protocol vaults in exchange for reward tokens. A yield loop uses recursive borrowing to amplify the native yield of a single asset — the mechanism, risk profile, and reward type are completely different. A looper earns a yield spread while a yield farmer earns governance or incentive tokens. The comparison section above has the full breakdown.
How many loops is too many in a yield loop?
There is no universal cap, but more loops means a higher effective LTV and a thinner health factor buffer. Most experienced loopers target an effective LTV below 80% across the full stack. Above that threshold, a modest price move can trigger liquidation before you can react. Automated platforms like Summer.fi allow you to set a deleveraging trigger so the position partially exits if the health factor drops below a chosen threshold.
Can a yield loop get liquidated?
Yes. When the health factor on a yield loop drops to 1.0, the lending protocol auto-liquidates enough of your collateral to restore the required collateral-to-debt ratio. You cannot lose more than your deposited collateral — DeFi protocols liquidate before the position goes insolvent — but you can lose most of it in a sharp downturn, especially if the liquidation happens during high gas periods where the liquidation penalty and gas costs eat into the remaining collateral.
What is the difference between a yield loop and leverage trading?
Both use borrowed capital to amplify exposure. The difference is the goal. Leverage trading targets price appreciation — you borrow to buy more of an asset you expect to rise. A yield loop targets the spread between a yield-bearing asset and a borrow rate, usually while staying correlated within a tight pair like wstETH/ETH. A loop gone wrong can still produce losses if that correlation breaks or if the spread inverts.
Does a yield loop work with stablecoins?
Yes. Stablecoin loops borrow one stablecoin against another yield-bearing stablecoin — for example, using sUSDe as collateral to borrow USDC. The liquidation risk is lower because both sides track the dollar. But a depeg event can collapse the trade rapidly: the Q4 2025 USDe depeg triggered cascades across every sUSDe-backed looped position in the protocols that supported it, showing that “lower risk” in a stablecoin loop is not the same as no risk.
—
Where To Start
If the yield loop strategy makes sense to you after reading this, here are five practical checks before you commit capital.
The most common mistake is skipping the math and starting with the APY number. That’s backwards. The APY is the outcome — your health factor, your effective LTV, and your exit cost are the inputs. Check those first. Then decide whether the spread is worth the exposure at your chosen loop count.
- Calculate your effective LTV across all cycles before entering. Confirm the health factor at your entry price stays above 1.3. Most protocol interfaces show this in real time — check it, not just the APY figure.
- Start with one or two loops, not five. The first two cycles capture most of the spread amplification. Loops three through six add meaningful liquidation risk for marginal yield gain. Add loops only after you’ve held the position through a volatile period.
- Pick the asset with the tightest correlation. A wstETH/ETH loop on Aave E-Mode is the most stress-tested structure. Stablecoin loops introduce depeg risk. RWA loops introduce liquidity risk.
- Set health factor alerts before closing the interface. Aave, Morpho, and Summer.fi all support notifications. Set a warning at 1.3 and an emergency level at 1.1 — below 1.15 during volatility, there is very little time to deleverage manually.
- Know your exit path before entering. Understand which steps are required to close the position and how much that costs in gas. A manual multi-loop unwind on Ethereum mainnet during congestion can cost more than one full day of earned yield. Automation layers reduce that cost but add smart contract exposure.
