Most crypto yield is somebody else's risk repackaged. This one has a boring, checkable source, which is exactly why it is worth understanding.
Aave is a lending market. You deposit USDC, borrowers post other crypto as collateral and borrow it, and they pay interest. Your yield is that interest, minus defaults. The rate floats with demand. It is not a promise, a subsidy, or a token emission.
If someone offers you yield on a stablecoin, the only question worth asking is where the money comes from. Most crypto yield answers that question badly. Either the source is a token the protocol prints itself, which is a transfer from future holders rather than income, or the source is undisclosed, which usually means somebody is taking a risk on your behalf that you have not been told about.
Lending on Aave has an answer you can follow all the way down, and the answer is unglamorous, which is the highest compliment available here.
Aave is an over-collateralised lending market. Two sides use it.
Depositors put an asset in. In this case USDC, a dollar stablecoin. The deposit is pooled with everyone else's.
Borrowers want to borrow that USDC. To do it, they first post collateral in another asset, usually ETH or a large-cap token, worth considerably more than they intend to borrow. Then they draw USDC against it and pay interest for as long as they hold the loan.
That interest, minus a protocol cut, goes to the depositors. That is the whole engine. Your yield is other people paying to borrow the dollars you supplied.
This part often gets skipped, and it matters, because if you cannot explain borrower demand you cannot explain the yield.
People borrow stablecoins against crypto collateral for a few consistent reasons. They want cash without selling, often to avoid triggering a taxable disposal. They want leverage, borrowing dollars to buy more of the asset they already hold. Or they are running a trade that needs dollars on hand while keeping their existing position intact.
The important consequence: borrower demand rises when the market is excited and falls when it is not. Deposit yields follow. This is why stablecoin lending rates are high in a bull market and thin in a quiet one, and why nobody can promise you a fixed number.
The reason this is not simply unsecured lending to strangers on the internet is that borrowers post more than they take. A borrower might post $150 of ETH to borrow $100 of USDC.
If their collateral falls in value far enough that the loan is no longer comfortably covered, the position is liquidated: a third party repays part of the debt and takes some of the collateral at a discount as payment for doing so. Liquidators do this because it is profitable, which means the system does not rely on anyone's goodwill to stay solvent.
Well-run positions are liquidated long before they reach the point where the collateral is worth less than the debt. The buffer is the product.
When you supply USDC to Aave you receive an equal amount of a receipt token in return. On Base, supplying USDC gives you aBasUSDC. It represents your claim on the pool.
The neat property is that the balance grows on its own. Interest accrues continuously and shows up as your token balance increasing, rather than as a payout you have to claim. There is no staking period, no lock-up and no harvest step. Redeeming is a single call that burns the receipt token and returns USDC.
Because interest arrives as a rising balance rather than a separate payment, a naive accounting system that only tracks "USDC in and USDC out" will miss it entirely, and will also mis-classify a withdrawal, because the funds leave as a burn of the receipt token rather than a plain transfer. Anything reporting on this needs to understand the receipt token or the numbers quietly go wrong.
Aave sets rates algorithmically from utilisation, the share of deposited USDC currently borrowed.
Low utilisation means lots of idle dollars and little competition to borrow them, so rates are low. As utilisation climbs, rates rise, which attracts more depositors and discourages more borrowing. Above a threshold the curve gets steep on purpose, so that the pool never runs dry and depositors can always withdraw.
The practical result is that the rate moves continuously, sometimes by a lot within a week. A quiet market might pay a few percent on USDC. A frantic one has paid considerably more. Any specific number quoted to you is a snapshot, not a forecast, and should be presented that way.
This is the part that decides whether the yield is worth having, so here it is without softening.
Smart contract risk. Aave is code holding billions of dollars. It has been live for years across many chains, is among the most audited systems in the industry, and runs a large bug bounty. That is a strong track record, and a strong track record is not a proof. A flaw would be severe.
Bad debt in a violent crash. Liquidations assume someone can sell the collateral fast enough. In a sharp enough fall, with congested blocks and thin liquidity, a position can pass through the liquidation threshold and end up under water. The protocol has mechanisms to absorb shortfalls, but a large enough event could impair depositors.
Stablecoin risk. Your yield is denominated in USDC, so you are holding an issuer's promise that each token is redeemable for a dollar. USDC is fully reserved and regularly attested, which is about as good as this gets, but it is still a distinct risk from the lending itself.
Rate risk. Not a loss, but a disappointment worth pricing in. You can deposit at an attractive rate and watch it fall by half as borrower demand dries up. Nothing is broken when this happens; it is the mechanism working.
Chain risk. On an L2 like Base, you also depend on that network operating correctly. Base is a large, well-used chain with significant value on it, and it is still one more dependency than holding assets on Ethereum directly.
Stablecoin lending is not a way to get rich. It is a way to not be exposed to a falling market while still earning something. That is a much narrower and more useful job.
For a strategy that intends to hold crypto during one part of a market cycle and step aside during another, the question during the step-aside period is what to do with the cash. Sitting in idle dollars earns nothing. Chasing high yields reintroduces exactly the risk you just exited. Lending into a large, over-collateralised, transparent market earns a modest rate with an explainable source.
The right way to judge it is not against the best number you saw advertised somewhere. It is against holding idle cash, which is what the alternative genuinely is.