Mentioned in This Article
1. LlamaRisk proposes an Aave freeze on three liquid staking tokens
LlamaRisk, a risk adviser partly funded by the Aave DAO, proposed on November 10 an Aave freeze covering three tokens on Aave V3 Core: StakeWise's osETH, Lombard's LBTC and Rocket Pool's rETH. Using figures from November 9, the three assets account for $538.6 million of collateral backing $364.1 million of loans. Six smaller reserves are also included.
The stated purpose is to reserve V3 Core, one of Aave's lending markets on Ethereum, for established and highly liquid collateral, while shifting riskier assets toward Aave V4, the protocol's next major version.
The proposal has drawn a quick response. On November 11, a forum participant signing as the founder of StakeWise asked that the osETH portion be split off and decided in its own vote. No vote on the plan had appeared in the Aave DAO's Snapshot space as of that date.
Risk advisers play a central role in how Aave operates. The protocol is governed by holders of its AAVE token, but most of the day-to-day technical work of setting collateral rules, borrowing limits and liquidation parameters is carried out by specialist firms that study each asset and recommend changes for the DAO to approve. LlamaRisk is one of those firms, and its recommendations carry significant weight in the governance process.
2. Five assets named, two of them already frozen
The proposal identifies five assets in total. Two of them are already frozen on Core. Kelp DAO's restaked ether token, rsETH, has been frozen since April. Threshold Network's bitcoin token, tBTC, was frozen on October 31 after Threshold paused minting and redemptions.
That leaves osETH, LBTC and rETH as the three significant new freezes. osETH and rETH are liquid staking tokens: each represents ether (ETH) staked to help secure the Ethereum network while giving holders a tradable token they can use elsewhere in decentralized finance. LBTC is Lombard's yield-bearing bitcoin (BTC) token, designed to let bitcoin holders earn a return while keeping exposure to the asset.
Tokens like these are popular collateral on lending markets because they continue to earn staking or yield income while they sit in a borrower's position. That same feature adds layers of risk, since their value depends not only on the underlying asset but also on the smart contracts, operators and redemption mechanisms behind each token.
3. How much money is in scope
Across everything the proposal covers, LlamaRisk counted $1.33 billion of collateral backing $1.03 billion of debt. The three largest new freezes alone represent $538.6 million of collateral and $364.1 million of debt, based on November 9 figures.
LlamaRisk also put a price on what the DAO currently earns from this business. It estimated that the debt in scope generates about $3.7 million a year in revenue for Aave's treasury.
That figure frames the trade-off at the heart of the proposal. Aave's governance would be choosing to stop growth in a revenue-producing segment of its largest market in exchange for a cleaner, lower-risk collateral base on V3 Core.
Seen that way, the decision is less about the revenue itself and more about what kind of collateral the DAO wants V3 Core to rely on as the protocol prepares for its next version. Giving up a few million dollars a year in income is the cost LlamaRisk is asking token holders to accept in exchange for a market built on established, liquid assets.
4. What does an Aave freeze do?
A freeze on Aave stops users from making new deposits of an asset and from opening new borrows against it, and it sets the asset's loan-to-value ratio to zero. The loan-to-value ratio is the share of a deposit's value that a user is allowed to borrow, so a zero setting means the asset no longer supports fresh borrowing power.
Crucially, a freeze does not force anyone out. LlamaRisk CEO Adam Sadowski has described the mechanism as "a one-way valve where users can always exit, but no new exposure can come in." The proposal states that existing positions would remain untouched and unaffected.
In practical terms, borrowers already using osETH, LBTC or rETH as collateral could keep their loans open, repay them or withdraw. What they could not do is add more of the frozen asset or borrow more against it. Over time, as users close positions, exposure to the frozen assets would be expected to shrink naturally.
Aave has used this tool before. The protocol carried out an emergency freeze on October 29, and freezes have become a standard part of how its risk advisers manage assets they no longer want to grow on a given market.
5. osETH carries most of the new exposure
Of the three tokens, osETH accounts for the largest share of the new exposure. It backs $279.4 million of collateral supporting $256.6 million of loans in WETH, the wrapped form of ether used inside smart contracts.
Those positions run at roughly 14 times leverage. Such high leverage typically comes from looping, a strategy in which a user deposits staked ether, borrows ether against it, converts the borrowed ether into more staked ether and repeats the process. Looping amplifies the gap between staking yield and borrowing cost, but it also leaves positions highly sensitive to any change in the relative price of the staking token or in borrowing rates.
Because osETH dominates the numbers, it is also the part of the proposal most likely to be contested, which helps explain why StakeWise has asked for it to be handled separately.
The risk in such structures is concentrated in a single relationship: the price of the staking token relative to ether itself. Under normal conditions the two move closely together, since each staking token is a claim on staked ether. If that link weakens, even briefly, positions at 14 times leverage have very little room before they approach liquidation, and large liquidations can in turn place pressure on the lending market and its lenders.
6. Reserve factor raised to 99% on frozen reserves
The plan goes beyond the freezes themselves. It would raise the reserve factor to 99% on seven borrowable frozen reserves that still carry more than $20,000 of debt.
The reserve factor is the share of the interest paid by borrowers that goes to the DAO's treasury rather than to the lenders supplying the asset. Setting it at 99% means almost all interest from those reserves would flow to the protocol instead of to suppliers.
LlamaRisk says the change would not alter borrow rates. Borrowers would pay the same interest as before; the difference lies in who receives it. For suppliers of those reserves, however, the change sharply reduces the incentive to keep funds in place, which is consistent with the broader goal of winding down the frozen markets.
7. Contingency steps if exposure does not fall
LlamaRisk has also set out what could happen if the freezes alone fail to reduce exposure. The proposal permits stepwise reductions to liquidation thresholds, the collateral level at which a loan becomes eligible for liquidation.
Lowering a liquidation threshold means a position must hold more collateral relative to its debt to stay safe, which pushes borrowers to repay or add collateral. Doing so gradually gives users time to adjust rather than triggering sudden liquidations.
Only as a final step would the WETH borrow rate be increased, aimed at the leveraged loops built on rsETH and osETH. A higher borrow rate would make those loops less profitable or unprofitable, encouraging users to unwind them.
LlamaRisk has committed to moving carefully. "We will not make abrupt parameter changes that would liquidate significant positions," the proposal states.
8. StakeWise asks for a separate osETH vote
The most direct pushback so far has come from StakeWise. On November 11, a poster signing as the project's founder asked LlamaRisk and the DAO to "move the proposed osETH freeze to a separate vote."
Splitting the decision would allow Aave token holders to approve freezes on the other assets while considering osETH on its own merits. Given that osETH represents the bulk of the new exposure and carries the most leveraged positions, a standalone vote would put the most consequential part of the plan under its own spotlight.
For StakeWise, the stakes are straightforward. A freeze on the largest Aave market would remove one of the most significant sources of demand for osETH as collateral, limiting how much new borrowing the token can support. Projects issuing staking tokens generally depend on deep integration with lending markets to make their tokens attractive to holders, so a freeze can affect a token's usefulness well beyond the positions directly involved.
9. Aave news: what's next for the proposal
The plan is an Aave Request for Final Comments, or ARFC, a late stage in Aave's governance process in which a proposal is refined on the forum before going to a vote. It remains under discussion on Aave's governance forum.
As of November 11, no vote had been posted to the Aave DAO's Snapshot space. Snapshot is an offchain voting platform that DAOs use to gauge token holder sentiment without paying onchain transaction fees. A Snapshot poll would be the next visible step, and the outcome of the StakeWise request will determine whether osETH is voted on together with the other assets or separately.
Whatever the result, the proposal signals a clear direction for Aave's markets: a tighter, more conservative V3 Core, with newer and more complex forms of collateral expected to find their home on V4.