Every lending market rests on a simple promise: collateral sits behind a loan, and if the borrower fails, the collateral covers the loss. The promise holds only if the collateral is actually there, and only once.

Rehypothecation breaks the 'only once' part. When the same asset is pledged, reused, and pledged again, the market ends up carrying more exposure than the underlying capital could ever support. In calm conditions this looks like efficiency. In stressed conditions it looks like a chain that was never as strong as its first link.

Key Takeaways

  • Rehypothecation lets one pool of collateral back several loans, so total exposure grows without new capital
  • The leverage multiplier rises sharply as the reuse rate rises, from 2x at 50% reuse to 10x at 90%
  • Reused collateral turns a price risk into a counterparty risk, because recovery depends on who holds the asset next
  • Stress shows up everywhere at once, since every link in the chain relies on the same underlying asset

The Common Misunderstanding

Most traders think about collateral as a static object. You deposit an asset, you borrow against it, and the asset waits in a vault until you repay. Risk, in this picture, is a question of price: will the collateral fall far enough to trigger a liquidation?

That view is the basis of most introductory lending advice. Look at the loan-to-value ratio, keep a buffer, and the position is safe. It is a useful frame, and collateral ratios are central to how any lending protocol stays solvent.

But the frame quietly assumes that the collateral is sitting still. It assumes the asset you posted is the same asset that would be sold if things went wrong.

In a rehypothecated system, that assumption is false. The collateral may have been lent onward, pledged to a third party, or wrapped into a new token that is itself being used as collateral somewhere else. The ratio on your screen describes your loan. It says nothing about how many other loans lean on the same asset.

One observation a week on liquidity, flow, and structure. 4 minutes. No price calls.

Subscribe →

What Actually Happens

Rehypothecation is the practice of a lender or custodian reusing the collateral it holds. The term comes from traditional finance, where prime brokers have long re-pledged client assets to fund their own borrowing. The mechanics are straightforward, and the consequences are not.

Step 1: The first pledge

A borrower posts 100 units of an asset to a lender and receives a loan. In a non-rehypothecated system, those 100 units stay locked until the loan is repaid.

Step 2: The reuse

In a rehypothecated system, the lender is allowed to take the 100 units and pledge them to a second party, usually to raise funding for itself. The second party may do the same with a third.

Nobody has created a new asset. There is still only one pool of 100 units. But there are now several claims against it, each of which believes it is secured.

Step 3: The multiplier

The arithmetic of reuse is a geometric series. If a fraction r of the collateral is reused at each step, the total exposure supported by the original capital is roughly 1 divided by (1 minus r).

  • At 50% reuse, 100 units of capital supports about 200 units of exposure
  • At 75% reuse, about 400 units
  • At 90% reuse, about 1,000 units

The key point is the shape of the curve. Moving from 50% to 75% doubles the multiplier. Moving from 75% to 90% more than doubles it again. Small changes in how freely collateral is reused produce large changes in system-wide leverage.

This is also why the leverage is hidden. Each participant sees a modest, well-collateralized position. No single balance sheet shows the multiplied total, because it only exists in the sum across the chain. That is the same pattern described in how collateral chains create hidden leverage across DeFi, where each layer looks reasonable and the stack does not.

Step 4: The risk changes type

Once collateral is reused, the nature of the risk changes. A borrower holding collateral in a segregated vault faces price risk. A borrower whose collateral has been re-pledged faces price risk plus counterparty risk.

If the lender fails, the borrower is no longer an owner with a claim on a specific asset. They are an unsecured creditor with a claim on a failed institution, standing in a queue behind everyone else.

That is the structural cost of rehypothecation. It does not make losses larger on the way up. It makes recovery uncertain on the way down.

Step 5: Why the chain breaks together

In a chain of reused collateral, every link depends on the same underlying asset holding its value and remaining accessible. When that asset drops, margin calls arrive at every layer simultaneously.

Each participant tries to meet the call the same way: by selling the asset, or by recalling the collateral they lent onward. Those two responses compete for the same limited supply. The result is the kind of synchronized forced selling described in how DeFi leverage cascades across protocols.

The mechanism is not a panic. It is arithmetic. A system with a high multiplier has little room between 'everything is fine' and 'everyone needs the same asset back at once'.

Example from Crypto Markets

Crypto has produced both versions of this story: the centralized one and the on-chain one.

The centralized version

In 2022, several large crypto lenders took customer deposits and deployed them elsewhere. Customers believed they were holding assets at an interest-bearing account. In practice, those assets had been lent to trading firms, committed to yield strategies, or otherwise made unavailable.

When prices fell and counterparties could not repay, withdrawals froze. Users discovered that their claim was a claim on the company, not on the coins. The loss was not simply the price decline. It was the gap between the assets people thought were held in reserve and the assets actually available to return.

The on-chain version

On-chain, the reuse is more transparent but follows the same logic. A trader deposits ETH into a lending protocol and receives a receipt token. That token can be posted as collateral in a second protocol, which issues another token, which can be posted in a third.

The liquid staking market made this especially common. A staked ETH token represents a claim on locked ETH. Traders could deposit it as collateral, borrow ETH against it, buy more of the staked token, and repeat the loop. The result is a larger position on the same capital, with each turn adding to the multiplier. Leverage staking works exactly this way, and it relies on one assumption above all: that the staked token keeps trading close to the asset it represents.

In mid-2022, that assumption was tested. The staked ETH token traded at a discount to ETH while large holders needed liquidity. Anyone looping the position faced a squeeze from two directions: the collateral was worth less, and the ability to exit was limited. The positions did not fail because the underlying asset collapsed. They failed because the relationship between the layers shifted.

The lesson from both versions is the same. The visible balance on a platform describes a claim. Whether the claim is good depends on what sits behind it and how many other claims point to the same place.

What Traders Can Learn

This is not an argument that collateral reuse is inherently bad. It makes credit cheaper, deepens liquidity, and lets capital work harder. The point is that it carries a cost that does not show up in a standard risk checklist. A few observations follow from the mechanics.

Leverage can exist without you taking any. A trader using no borrowing at all can still be exposed to a system whose total leverage is high. The effects reach everyone through prices, liquidity, and the failure of counterparties. This is why crypto liquidations cascade in ways that individual position sizing cannot explain.

The question changes from 'how much?' to 'who holds it?' For any yield-bearing or lent-out asset, it is worth asking whether the asset is segregated, whether it can be re-pledged, and what the terms say about reuse. These are structural facts, not predictions.

Layers are leverage. Every wrapper, receipt token, or derivative placed on top of an asset adds a dependency. Fewer layers means fewer places for the multiplier to hide.

Stress reveals the real structure. In normal markets, a reused asset and a segregated asset behave identically. The difference appears only when many parties want their collateral back at once. By then, the structure is fixed.

Understanding this does not require predicting when stress will arrive. It only requires recognizing that a system's resilience depends on how often its collateral has been counted. Simpler exposure, such as holding spot instead of leveraged positions, removes one layer of this dependency by design.

FAQ

What is rehypothecation in crypto?

Rehypothecation is when a lender or platform reuses the collateral it holds, pledging it to someone else to raise funding or earn yield. In crypto this happens on centralized lenders that lend out customer deposits, and on-chain when receipt tokens are reused as collateral in other protocols.

Is rehypothecation the same as recursive borrowing?

They are related but not identical. Rehypothecation means the holder of collateral reuses it. Recursive borrowing, or looping, means a borrower repeatedly deposits, borrows, and re-deposits to build a larger position. Both multiply exposure on the same capital, and both create dependencies that surface under stress.

How can I tell if my collateral is being reused?

Check the terms of service or protocol documentation for language about re-pledging, lending out, or commingling assets. On-chain, look at whether the receipt token you hold is accepted as collateral elsewhere. If a platform pays yield on a deposit, the asset is usually being put to work somewhere.

Why does rehypothecation increase counterparty risk?

Because the collateral is no longer sitting in a place the borrower controls or can verify. If the party that reused it fails, the original owner becomes an unsecured creditor and recovery depends on the failed party's remaining assets and the order of claims.

Related Concepts

Conclusion

Rehypothecation does not create new capital. It creates new claims on the same capital, and the multiplier on those claims grows faster than intuition suggests as reuse increases. In calm markets the extra leverage is invisible, and in stressed markets it arrives all at once, through counterparties and prices that depend on the same underlying asset.

The practical shift is to look past the balance on the screen and ask how many hands the collateral has passed through. Collateral is only safe when it is only used once.