Aave’s main lending market would stop taking new deposits of rsETH, osETH, LBTC, tBTC, rETH and a handful of smaller reserves under a proposal from LlamaRisk, one of the protocol’s risk managers. Those assets currently hold about $1.33 billion of collateral on Aave V3 Core. The plan, posted to Aave’s governance forum on Tuesday, would leave the loans those assets back in place but cut off new borrowing against them, with the aim of moving restaking tokens, wrapped bitcoin and other riskier collateral toward Aave V4 over the coming months.
What LlamaRisk is proposing
The proposal, filed as an ARFC on the Aave governance forum, would turn Aave V3 Core into a market for what LlamaRisk calls “liquid, bluechip collateral.” As of Nov. 9, the assets in scope held $1.33 billion of collateral backing $1.03 billion of debt.
The first step is a freeze. A frozen reserve takes no new deposits and no new borrowing, and its loan-to-value ratio drops to zero, including inside Aave’s E-Mode categories. Existing borrowers can still repay, withdraw or move their positions. For frozen reserves that can still be borrowed and carry more than $20,000 of debt, the reserve factor would rise to 99%. Borrowers would pay the same rate, but nearly all of the interest would go to the DAO instead of suppliers.
What stays on Core is a short list: WETH, the staking tokens wstETH, weETH and cbETH, WBTC and cbBTC, the major stablecoins including USDC, USDT, GHO and DAI, plus XAUt, AAVE and LINK. USDe, sUSDe and syrupUSDT also stay.
Who holds the exposure
Most of the money sits in two restaking tokens. rsETH accounts for $661.8 million of collateral against $605.7 million of WETH debt, and osETH for $279.4 million against $256.6 million. Together they back $862 million of WETH borrowing, almost entirely through E-Mode loops at 12 to 14 times leverage, LlamaRisk says.
The rsETH book is also concentrated. The token has already been frozen on Core since April, and its supply there has fallen 45%, from 525,800 to 290,700 rsETH. The remaining debt sits in eleven positions larger than $200,000, and the three biggest hold 69% of it. osETH is the opposite case: its supply has climbed to 122,900 tokens, close to its 123,000 cap.
The bitcoin and rETH side is smaller and more spread out. LBTC holds $177.4 million of collateral against $72.1 million of debt, tBTC $117.4 million against $52.9 million across 309 borrowing positions, and rETH $81.8 million against $35.4 million across 317 positions. tBTC has been frozen on Core since Oct. 31, after Threshold paused minting and redemptions, which AltcoinReporter covered at the time. LBTC is expected to continue on Aave through a dedicated spoke on the V4 Ethereum Core Hub.
Why the loops get different treatment
The usual way to shrink a risky reserve is to lower its liquidation threshold in steps. LlamaRisk says LBTC, tBTC and rETH positions sit far enough from liquidation for that to be an option: a 10-point cut would leave less than $100,000 of LBTC collateral liquidatable. It does not work for the leveraged loops. For rsETH and osETH, a cut of just 2 to 5 points would make most of the collateral liquidatable at once.
So the proposal keeps a different lever in reserve. If the loops do not unwind, Aave could raise the base WETH borrow rate by up to 100 basis points, in 50-point steps no more than once a month, until the trade stops paying. A Merkl campaign would rebate the increase to WETH borrowers who hold neither asset, and a matching reserve factor change would keep the WETH supply rate flat. LlamaRisk describes this as “the final step of the process,” to be used only if talks with issuers and the largest borrowers fail. “We will not make abrupt parameter changes that would liquidate significant positions,” the proposal says.
LlamaRisk puts the revenue generated by the debt in scope at about $3.7 million a year, before any of it migrates to other collateral or to V4.
Two different numbers
Aave founder Stani Kulechov described the plan minutes after it was posted as “potentially moving $1.52B in collateral backing $1.17B in debt out of V3 Core.” That is about $190 million more collateral than the forum post lists. Neither post explains the gap. The forum table, which breaks the total down by asset, adds up to $1.33 billion and $1.03 billion, the figures used here.
What happens next
The proposal is at the request-for-final-comment stage, with no vote scheduled yet. If it is approved, the freeze and reserve factor changes would go to an on-chain AIP, and LlamaRisk says it would then report on the unwinding every month. It would also extend the rsETH E-Mode so loopers can swap into wstETH, weETH or cbETH without repaying their WETH debt first.
The plan continues a cleanup that started this summer, when the DAO moved to wind down low-adoption reserves and six whole markets. That round deliberately left the large Core positions alone. This one targets them and leans on V4, where each borrower pays a premium based on the risk of their collateral. LlamaRisk notes that it is funded in part by the Aave DAO.
FAQ
Will rsETH or osETH borrowers on Aave be liquidated?
Not by the freeze itself. Existing positions stay open and can be repaid, withdrawn or migrated. LlamaRisk says liquidation threshold cuts would not be used on the rsETH and osETH loops; a gradual WETH rate increase is the last-resort tool.
Which assets would leave Aave V3 Core?
rsETH, osETH, LBTC, tBTC and rETH, plus small reserves including eBTC, BTC.b, tETH, LDO, mUSD and a Pendle PT token. rsETH and tBTC are already frozen.
Is the change live?
No. It is a request for final comment and needs DAO approval and an on-chain AIP before any parameter changes.
Disclaimer: This article is for informational purposes only and does not constitute financial advice.
