Aave's two stablecoin protection pools on Ethereum V3 fell by a combined $12.43 million between September 17 and October 17, 2026, even as the additional reward rates for USDC and USDT increased. The dollar value of the positions declined from $128.52 million to $116.09 million, a 9.67% drop. The change reduced the capital available to absorb losses in the related lending reserves, while increasing the risk compensation earned per dollar of capital. The balance decline should not, however, be treated as proof of equivalent withdrawals or realized losses.
USDC and USDT balances fall by $5.3 million and $7.13 million
The positions are part of Aave's Umbrella mechanism. Participants stake the relevant assets and continue to receive base lending income, while earning additional rewards for taking the risk that their capital could be used, or potentially slashed, if a reserve develops a shortfall.
At 23:01 UTC on September 17, the USDC protection pool held $58.7396 million on a dollar basis. By 15:01 UTC on October 17, that figure had fallen to $53.4380 million, a decline of about $5.3 million. The USDT pool dropped from $69.7839 million to $62.6557 million over the same period, down about $7.13 million. The combined decrease was $12.4299 million.
The comparison points are just under 30 days apart. Because the balances are measured in dollars, they are affected both by changes in the quantity deposited or withdrawn and by changes in the value of the underlying assets. A lower dollar balance therefore does not establish that all of the reduction came from withdrawals, nor does it indicate that a loss occurred.
| Ethereum V3 Umbrella position | Capital on Sept. 17 | Capital on Oct. 17 | Additional reward on Sept. 17 | Additional reward on Oct. 17 |
|---|---|---|---|---|
| USDC | $58.7396 million | $53.4380 million | 1.66% | 1.82% |
| USDT | $69.7839 million | $62.6557 million | 1.46% | 1.63% |
Higher reward rates leave annualized incentives almost unchanged
The additional reward rate for USDC rose by 0.16 percentage points, while the USDT rate increased by 0.17 percentage points. Base lending yields also moved during the period, rising from about 3.60% to 3.77% for USDC and from about 3.02% to 3.12% for USDT. Base yield reflects the return from lending activity; the additional reward compensates participants for staking and potential slashing risk. The two components should be calculated separately.
Multiplying each position's dollar balance by its corresponding additional reward rate produces an indicative annualized amount of about $1.9939 million on September 17. The same calculation using the October 17 balance and reward rate gives about $1.9939 million. The precise figures are $1.99392312 million and $1.99385928 million, respectively, which both round to roughly $1.994 million.
This is an annualized run rate derived from snapshots at two points in time, not a measure of rewards already paid or expenses actually incurred. Actual distributions depend on the path of rate changes, staked balances, reward allocations and participant activity during the period.
The reward curve can increase returns per dollar staked
The Umbrella reward controller indicates that the additional reward rate does not necessarily rise because the overall incentive budget has increased. Its emission curve links the total amount emitted to actual staking relative to the target staking level.
When staking reaches the target, total emissions reach their maximum. As staking moves from the target to 120% of the target, total emissions decline progressively, reaching a level close to 80% of the maximum. Once staking reaches or exceeds 120% of the target, total emissions remain broadly stable. Under those conditions, a fixed or nearly fixed reward flow distributed across a smaller capital base can produce a higher percentage return per dollar staked.
If staking falls below the target, the mechanism can also raise the reward per dollar to attract participation, although total emissions remain below their peak. The two historical balance and reward-rate snapshots alone do not establish whether the same section of the curve applied to both pools at both observation times. That would require a review of the relevant contract parameters and balance states.
A 20-day cooldown keeps capital exposed during withdrawals
Aave's asset and network isolation design means that USDC staking is primarily intended to cover a shortfall in the USDC reserve, while USDT staking corresponds to the USDT reserve. The combined balance shows the overall scale of the protection capital, but each reserve's coverage depends on the capital assigned to it. The total cannot be used as a substitute for the protection available to either individual reserve.
Umbrella staked assets may be slashed only after realized reserve shortfalls exceed the applicable loss-absorption allowance. The loss-absorption layer is funded first by DAO resources, with the relevant staking pools involved only afterward. The withdrawal process also affects when capital becomes available to participants: Aave imposes a 20-day cooldown followed by a two-day withdrawal window. Rewards continue to accrue during the cooldown, but the staked capital remains exposed to potential slashing throughout that period.
The positions are wrapped-aToken Umbrella stakes on Ethereum V3, identified as stkwaEthUSDC.v1 and stkwaEthUSDT.v1. Because these tokens have an exchange rate, the dollar balances also reflect changes in asset valuation and in that exchange rate.
On September 25, 2026, Aave's governance proposed renewing the funding required for Umbrella reward payments. The proposal covers accumulated but unclaimed rewards, amounts owed to former stakers and remaining emissions through the scheduled end of distributions on December 2, 2026. It does not change the emission rate, target liquidity, cooldown parameters, loss-absorption arrangements or distribution end date. The requested amount includes both historical obligations and future emissions, so it should not be treated as an entirely new annual incentive budget.
As of October 17, the additional reward rates for both pools were higher than on September 17, while the underlying protection capital had declined. The data therefore show higher incentives per unit alongside a smaller staking base. Determining whether the balance reduction was driven mainly by asset prices, deposit levels or withdrawals requires further comparison with on-chain balances and the applicable contract parameters.