Rising Aave borrow rates threaten to flip Ethena’s USDe yield loops into negative carry

LlamaRisk has proposed increasing borrowing costs for Ethena’s USDe across five Aave V3 markets, tightening the economics for positions that borrow the stablecoin and recycle it into yield-bearing sUSDe.

The Sept. 9 recommendation would increase USDe’s base variable borrow rate from 5% to 6% on Core, Plasma, Monad, Mantle, and Avalanche. It would also reduce Slope1 by one percentage point on every deployment.

At the utilization levels captured in the proposal, modeled borrower APRs would rise by 13 to 89 basis points across markets holding about $323.8 million of USDe debt against $1.18 billion supplied.

Infographic compares projected Aave USDe borrowing rates across Core, Plasma, Monad, Mantle and Avalanche after proposed base-rate and Slope1 changes.

LlamaRisk framed the changes as recommendations and said they would be implemented through the Risk Steward process.

The increase is not uniform because the proposal changes two parts of the rate curve at once. The higher base pushes borrowing costs up, while the lower Slope1 offsets part of that move at each reserve’s utilization.

Avalanche, the most heavily utilized market in the snapshot, gets the largest offset and a 13-basis-point increase. Mantle, with the lowest utilization, absorbs nearly the full base-rate rise.

The change targets leveraged sUSDe positions. TokenLogic’s staged repricing program described borrowers recycling USDe into sUSDe to capture the spread between staking yield and Aave’s borrowing cost.

The program argued that a higher borrowing floor should reduce those loops and could lift the yield available to remaining sUSDe holders toward 5.3%, but only as loop-funded supply unwinds.

By Sept. 10, Aavescan’s sUSDe snapshot showed a 4.72% supply APY. Its Core, Plasma, and Monad market pages displayed USDe borrow APRs above that level. The Mantle and Avalanche pages did as well, leaving a simple borrow-and-stake loop with negative carry before incentives, transaction costs, and other frictions.

Borrower rates can change as utilization moves, so the proposal’s market-level APRs describe the modeled effect at the captured Sept. 9 conditions rather than fixed costs.

Editorial credits