Aave Utilization Rate: Interest-Rate Kinks and Liquidity

Quick answer

Aave borrow utilization compares a reserve’s debt with its debt plus available liquidity. In the V3 strategy examined here, borrow utilization is debt ÷ (available liquidity + debt). It influences the variable borrow rate, but a utilization percentage alone cannot tell you the rate, supplier return or amount you can withdraw.

To interpret utilization, compare it with the reserve’s kink and rate slopes, then check the liquidity available for your intended transaction. The worked examples below show why these inputs matter.

The kink changes the sensitivity of borrowing costs

The official V3 interest-rate documentation describes two segments separated by optimal utilization. In the versioned strategy source, the calculation is:

Use ratios consistently: 80% is 0.80. The “slope” parameters are rate increments allocated across each segment, not the rate increase for every one-percentage-point rise in utilization. At the kink the rate is base + first slope; at 100% it is base + first slope + second slope. The formula is continuous at the kink, although its sensitivity changes.

Hypothetical example · Borrow-rate kink

For the hypothetical curve above, an increase from 70% to 80% utilization raises the borrow rate from 4.5% to 5%. From 80% to 90%, the rate instead rises from 5% to 25%. Below the kink, each additional utilization percentage point adds 0.05 percentage points to the rate; above it, it adds 2 percentage points. This difference follows from distributing a 40-point increment across the final 20 utilization points.

A borrower near the kink can therefore face a substantial cost change after a relatively small change in reserve usage. Repayments, new supply and withdrawals change reserve usage; parameter updates can also change the curve itself.

Why the same utilization can produce different rates

“80% utilized” is not a comparable price without the strategy parameters. Hold the base rate at 1%, the first slope at 4 percentage points and the second at 40, then change only the kink:

Strategy B’s value is 1% + 4% × 0.80 ÷ 0.90, rounded to two decimals. These are constructed examples, not two observed reserves. Changing the base or slope parameters would change the comparison again.

For a real comparison, match the chain, market, underlying asset, deployed strategy and observation block. The reserve documentation defines the reserve-level context; a token name by itself is insufficient. A governance proposal is also not proof that its parameters are already active.

Borrow utilization and supply utilization have different denominators

The inspected implementation calculates available liquidity from the virtual underlying balance, adjusted for liquidity added or taken in the operation. It then uses two ratios. The unbacked amount represents minted aTokens whose corresponding underlying assets have not yet been supplied:

When the unbacked amount is zero, the two ratios coincide. When it is positive, supply utilization is lower for the same debt and available liquidity. If there is no debt, this implementation returns a zero liquidity rate and the base variable borrow rate.

This is why “borrowed divided by supplied” is a useful introduction but an incomplete reconciliation procedure. Two dashboards can show different utilization values if they use different supply denominators, debt definitions or observation blocks. Compare those inputs before deciding whether the figures conflict. Virtual liquidity used in the formula should not be substituted for a transaction-level withdrawal simulation.

Why suppliers do not receive the borrow rate

For the examined strategy, the supplier liquidity rate is variable borrow rate × supply utilization × (1 − reserve factor). The reserve factor allocates a share of interest away from suppliers. It is not a haircut applied to the principal you supply.

Here is the same hypothetical strategy at two utilization levels. Assume available liquidity and debt total 100 units, no unbacked amount, and a 10% reserve factor:

The higher supply rate comes with less available liquidity in this example. Comparing the opportunity requires both the rate and liquidity conditions, as well as how long those conditions may persist.

These are instantaneous annualized strategy rates, not a realized annual return or a complete displayed APY. Borrowing accrual, supplier accrual, UI annualization and separately advertised incentives need their own conventions. The Pool documentation also describes flash-loan premium allocation to liquidity providers; the strategy multiplication should not be presented as an exhaustive accounting of every supplier-income source.

High utilization is a liquidity question as well as a rate question

A percentage hides scale. With no unbacked amount, 90% utilization could describe 90 units of debt and 10 available units, or 90 million and 10 million. The ratios are identical; the capacity to accommodate a particular withdrawal is not.

If a simplified reserve has 10 units available and a supplier wants to withdraw 15, a higher advertised rate does not supply the missing 5 units. For an actual withdrawal, verify the reserve’s transaction-level liquidity and the account’s eligibility at the relevant block. Do not infer a waiting time or guaranteed failure from utilization alone.

The official withdrawal rules also constrain withdrawals when the supplied asset backs debt: the operation must preserve the required health factor. A position can therefore face an account-level constraint even when the reserve has available liquidity. Conversely, additional collateral capacity does not create cash in the reserve.

Utilization and E-Mode answer different questions

Our Aave E-Mode explanation covers category permissions and collateral terms. E-Mode can affect the collateral side of a position; it does not fix the variable interest rate or guarantee exit liquidity.

How to verify a real reserve

1. Identify the chain, Pool address and underlying asset address. Record the observation time and block.
2. Resolve the reserve’s interest-rate strategy address and deployed implementation. Match that implementation to the source version rather than assuming every V3 deployment uses this file.
3. Read the base rate, optimal usage ratio and both slope parameters for that reserve. In the inspected strategy, getInterestRateDataBps returns basis-point parameters; getInterestRateData returns ray-scaled values. Do not mix those units.
4. Read debt, the relevant liquidity inputs, unbacked amount and reserve factor at the same block. Account for operation inputs when reproducing calculateInterestRates.
5. Recalculate both ratios and rates, then compare like-for-like with the display. Check rounding, APR/APY treatment and incentives before interpreting a mismatch as an error.
6. For a withdrawal decision, separately check account health and simulate the operation. A rate calculation is not a transaction guarantee.

Sources and limitations

Official documentation and the pinned strategy revision were checked Oct 11, 2026. The curve, comparison and worked calculation are hypothetical examples derived from that formula; none represents a live reserve configuration or historical observation. Values are rounded for explanation and do not reproduce every fixed-point rounding operation.

The analysis covers the pinned V3 strategy implementation. Check the deployed contract before applying its formulas to a specific market; other V3 implementations and Aave V4 may differ. This page explains the mechanism rather than tracking live rates. Rules and parameter changes require a fresh source review.