Rehypothecation in DeFi

What Is Rehypothecation in DeFi?

Understanding what is rehypothecation in DeFi starts with a simple observation: DeFi protocols don't lock collateral in a vault and walk away. They put it to work. When you deposit ETH into Aave, that ETH doesn't sit idle — it's lent to other borrowers. When you receive aETH (your receipt token) and deposit that into another protocol as collateral, you've just rehypothecated your original position.

Same asset. Multiple claims. Compounding risk.

The Traditional Finance Origin

This isn't a crypto invention. Prime brokers on Wall Street have rehypothecated client assets for decades. A hedge fund posts $10 million in Treasuries as margin — the prime broker pledges those same Treasuries as collateral for its own borrowing. The 2008 financial crisis exposed exactly how dangerous this gets: Lehman Brothers had rehypothecated client assets so aggressively that when it collapsed, untangling who owned what took years.

DeFi replicated this behavior at protocol level, but made it transparent and permissionless. Whether that's better or worse depends on who's asking.

How Rehypothecation Works On-Chain

Here's a concrete chain of events:

1. You deposit 10 ETH into Aave and receive 10 aETH
2. You deposit that aETH into Compound or a yield vault as collateral
3. You borrow USDC against it
4. You deploy that USDC into a liquidity pool on Curve
5. You receive Curve LP tokens — and deposit those somewhere else

At each step, the same underlying value gets encumbered again. The original 10 ETH now supports three or four distinct debt positions across separate protocols. Capital efficiency is extraordinary. So is the fragility.

Warning: In a sharp market downturn, liquidation at any layer can trigger cascading liquidations across every layer above it. I've seen a $50M position unwind across four protocols in under 15 minutes during high-volatility episodes.

Liquid Staking: The Most Widespread Form

The most pervasive example of DeFi rehypothecation today is liquid staking. When you stake ETH through Lido, you receive stETH — a token representing your staked position that earns staking rewards. That stETH can then be:

• Deposited on Aave as collateral to borrow stablecoins
• Provided as liquidity on Curve's stETH/ETH pool
• Wrapped into wstETH and used in Pendle or other yield-splitting protocols

By May 2026, Lido holds over 9 million staked ETH. A meaningful portion of that circulates through secondary and tertiary DeFi positions. The underlying ETH is locked in the Ethereum beacon chain — completely illiquid — while multiple derivative claims on it trade freely. That gap between the liquid derivative and the illiquid underlying is where rehypothecation risk lives.

See Staking Yield Comparison: Liquid vs Traditional Staking Returns in 2026 for data on how these layered positions affect net yield.

The Capital Efficiency Argument

Proponents aren't wrong. Rehypothecation allows the same capital to generate yield at multiple layers simultaneously. A user depositing 10 ETH might earn:

In aggregate, that 10 ETH could generate returns that would be impossible from a single-layer deployment. Traditional finance calls this "financial engineering." DeFi calls it composability.

Myth vs Reality

Myth: On-chain transparency eliminates rehypothecation risk.

Reality: Transparency tells you that the risk exists, not when it triggers. You can see every aToken deposited as collateral on-chain — but predicting the exact price level at which a cascade of cross-protocol liquidations begins requires modeling interactions between protocols that weren't designed to coordinate. Most risk dashboards significantly underestimate this correlation.

Myth: High TVL means a protocol is safe.

Reality: Total Value Locked counts the same asset multiple times when it's been rehypothecated across protocols. A system showing $5B TVL might have $1.5B in unique underlying assets — the rest is layered derivative claims on those assets.

The Systemic Risk Problem

This is where rehypothecation gets genuinely dangerous. When collateral is reused across protocols, correlations between supposedly independent systems spike to nearly 1.0 during stress events. The protocols aren't independent anymore — they're coupled.

The June 2022 stETH depeg is the clearest case study. stETH briefly traded at a ~5-8% discount to ETH. That discount triggered margin calls on protocols that had accepted stETH as collateral at parity. Those margin calls forced selling of stETH, deepening the discount, triggering more margin calls. The loop took weeks to fully resolve and contributed directly to the collapse of several leveraged entities that had built positions across multiple rehypothecation layers.

For a deeper look at how smart contract interactions amplify these risks, Smart Contract Security Vulnerabilities in DeFi Protocols covers the technical attack surfaces that make cascades harder to stop once they start.

What Differentiates Sustainable vs. Dangerous Rehypothecation

Not all rehypothecation is equally risky. The key variables:

• Liquidation buffer depth — Is there enough collateral cushion at each layer to absorb price volatility before liquidations trigger?
• Asset correlation — Rehypothecating ETH into an ETH-denominated position doubles your directional exposure. Rehypothecating into a stablecoin position partially hedges it.
• Protocol liquidity depth — Can the underlying collateral actually be liquidated at the assumed price, or does the liquidation itself move the market?
• Oracle reliability — Cross-protocol rehypothecation chains often depend on multiple price oracles. One stale or manipulated feed can corrupt the entire stack. See Oracle Network Reliability: Comparing Chainlink vs Band vs Pyth for a breakdown of how different oracle designs handle these failure scenarios.

The difference between capital efficiency and systemic fragility often comes down to whether those buffers were sized for normal conditions or stress conditions. Most protocols, historically, size for normal.

Rehypothecation isn't inherently reckless. But treating it as free yield without accounting for the layered liquidation risk it creates is one of the more reliable ways to get wrecked in a downturn.