Aave's Stable Vault

By: Thejaswini M A

Translated by: Block unicorn

Nothing is more expensive than a decision placed before people. In comparison, various fees are negligible. People pay merely for convenience and ease, nothing more.

This is how platforms extract value from users; they deprive users of choice. "Couch locking"—this is a term I love from Tim Wu's book "The Age of Extraction".

Can't pick stocks? Of course you can, index funds, S&P 500 index funds...

Not sure if you should lend? Let's just call it a savings account, and you'll be proud.

They only charge removal fees. Removing not just the right to decide, but sometimes the best benefits too. People simply don't care.

The role of DeFi is to add. But which chain, which pool, what rates, when to migrate, where to bridge? Is this really Aave? Or something Claude Fable put together before July 12? Aave has 2.5 million users and has operated for six years. Revolut has 65 million users. Is it fair to say Aave must be smarter?

From January to July, Aave's USDC pool rates fluctuated between 2% and 9%.

In the DeFi field, price volatility is the norm. You watch prices rise and fall, and move funds at the right time.

But this structure simply doesn't work elsewhere. New banks cannot explain to users that rates are determined by borrowing demand, and that rates might plummet to 2% despite marketing it as a savings account. People won't put money into chaos. This is why people never use crypto apps daily, let alone crypto savings apps.

On July 9, Aave Labs uncovered a workaround called Stable Vault. Today, I want to show you how it works, who profits from it, and why I think ordinary users will end up using it anyway.

Stable Vaults allow any business to start offering savings account services with just one integration. It could be a neobank, e-wallet, or payroll processing platform. Deposits go directly into Aave's lending market. Users can view rates in the apps they use daily; if the rate is right, they open an account.

The rates are fixed. I know, in the crypto field, this is hard to say out loud.

Aave's market pays borrowers' repayments for the week. Stable Vaults sit on top of this and provide operators with a dial. The app inputs a number, say 4%. From then on, regardless of what Aave does underneath, the vault pays 4% yield per second. This is the app's problem, not yours. Any yield generated by the strategy exceeding that number belongs to the operator.

Perspective: The Person Putting in Money

They all enjoy rated insurance. This spring, Aave's USDC pool paid 2%, while a vault promising to pay 4% still paid 4%, with the operator covering the difference.

Elsewhere, transferring risk comes at a cost, and it's no different here. Take fixed-rate mortgages, for example; they cost 50 to 100 basis points more than adjustable-rate mortgages, and this premium is essentially the fee borrowers pay for fixed rates.

Users don't need to create wallets, save seed phrases, set up bridges, or choose blockchains. They get support hotlines, account recovery, Face ID, and addresses provided by the company (in case things go wrong). Aave's app is SOC 2 certified and supports Two-Factor Authentication (2FA), and promotes both features simultaneously, because this is exactly the product or service customers are buying.

Users lose the most amount. When the pool pays 9%, they only get 4%; when the pool pays 6%, they still only get 4%. This is a fixed ratio set by the app based on user tiers. Dynamic ratios refer to real-time data you can actively view; fixed ratios completely hide the intermediary's cut.

Users also choose a second counterparty. With this setup, two new risk models are added to their balance sheet: one is the financial health of the fintech company operating the app, and the other is the code quality of the private scripts hidden in the backend that handle fund allocation. In pure Decentralized Finance (DeFi), your only risk lies in the core protocol code. But here, if the intermediary company goes bankrupt, or their private backend scripts crash causing loss of funds, your money will be lost just the same, even if Aave itself runs perfectly.

In a true swap market, since both parties can shop around, fixed rates are driven down close to fair value. But here, operators set rates unilaterally, and customers have no reference. Users won't compare a 4% rate with Aave's 6%, but rather with their own bank's rate. Aave's app page displays its rates alongside the national average savings rate of 0.40% set by the Federal Deposit Insurance Corporation (FDIC); in comparison, any rate looks very attractive.

Perspective: The Operator

Suppose a new bank has $200 million in idle user stablecoins. It already has funds and users, and these users were acquired at a cost. It only needs to complete one integration, promote a 4% yield, and if the strategy returns 6%, it only needs to recognize $4 million in revenue on its balance sheet annually, where this revenue was originally a cost. Low input, yet yield as high as 2%, quite good.

Rise is a payroll service company that pays contractors across 190 countries, having processed over $1.5 billion in payroll cumulatively. In the past, if companies prepaid payroll a week in advance, these USDC would sit idle, so Rise developed Rise Earn, depositing it into Aave's USDC pool on the Arbitrum platform until payday.

Rise charges 1% of the interest, no other fees. Assuming a yield of 6%, Rise charges 6 basis points. Employees actually receive 5.94% yield, and they see only Aave's real-time rate.

If using the same funds, a Stable Vault operator would charge 200 basis points. The intermediary's cut increases by 33 times.

Perspective: Aave and Stable Vault

Aave's selling point is that its vaults can set different rates based on user loyalty, activity, or tier. For example, premium members can get 5% yield, while other users get 3.5% from the same borrowing interest pool. Fintech companies issuing their own stablecoins can register them as deposit assets, achieving closed-loop transactions. Moreover, yielding balances do not churn, meaning the yield itself is a retention tool.

Do operators get these yields for free? Of course not. It shorts the bid-ask spread. This spring, when Aave paid a 2% yield, all vaults promising yields higher than this had to pay higher costs.

This leads to April 18, when the Kelp DAO bridge vulnerability triggered a massive run on Aave on April 18, pool utilization reached 100%, completely freezing all withdrawals, and trapping both operators' book profits and user funds in the same locked queue.

When capital utilization hits the cap, no one can profit from it, including vaults. Surplus funds pile up on the books, alongside users' principal.

If liquidity recovers, operators will clear the surplus funds accumulated during the period users couldn't exit. This surplus is the fee the market pays for insufficient liquidity. It is the users who provide the insufficient liquidity. If liquidity fails to recover, bad debt enters the pool, the vault will face a shortage, and Aave's documents show authorizers can recharge the system. Behind the word "can" here lies no reserve fund.

Aave will claim their contracts were never exploited, that it was Kelp's bridge that had issues not Aave's code, and rsETH was frozen within hours. All this is true. Previously, just before their risk manager resigned, they voted to accept extremely high-risk collateral, with Loan-to-Value ratios as high as 93%, forcing ordinary users to bear the brunt of this crashing app.

So now, does Stable Vault look like the final piece of the puzzle?

Rise runs payroll floating through Aave. Kraken white-labeled Aave v3 integration into its Tydro protocol on its L2 layer, then pointed its retail product Earn to that protocol, so when Kraken users use the Earn feature, they are also Aave users. Cap Finance stores stablecoin reserves in Tydro.

Horizon partners with institutions like Circle and Franklin Templeton to lend against tokenized treasuries. The Aave App targets consumers directly. Stable Vaults are open to all other institutions, calling it diversified investment.

Aave doesn't need more deposits. Kulechov told The Block in March that there is excess liquidity in the DeFi field, and the focus must shift to lending. His point is correct, and this is why USDC yields dropped from 8% to 2-3%. The problem Aave has always faced is that DeFi funds are extremely liquid; they churn if yields drop by 50 basis points. Through an app that controls fund flow like payroll, the most unstable funds in finance can be transformed into assets as stable as deposits.

Aavenomics 3.0 launched on June 27 and now automatically buys back AAVE from revenue. Revenue needs to keep flowing regardless of market conditions. In a bear market, stable deposits are key to keeping buybacks running. How to get these deposits? The answer is: Stable Vaults.

Coinbase offers about 4% yield on USDC balances. Robinhood launched Earn services on July 1, with yields around 7%, and currently has 28 million funded accounts. Both call it a savings feature.

Coinbase runs on Morpho and Ethena. Robinhood runs on Morpho and Maple, with its risk parameters set by a company called Steakhouse.

Both had to build this system themselves: custody protocols, custodians, risk teams, and months of legal work. Aave's contribution is making all this unnecessary. With just one integration, any app on earth can display a number on screen and track the gap between that number and the borrower's actual repayments in real time.

The reason banks can use this system is because of a century of legal support behind it. Reserve requirements, audits, deposit insurance, and regulators who can conduct surprise inspections at any time—all this is built on a consensus reached long ago: banks will lend out your money, so when loans go wrong, someone must step up to take responsibility.

All functions of Stable Vaults can be achieved in just 20-30 minutes. But you need to create a wallet first, bridge some USDC, and then supply it to Aave. This way, you need no KYC verification, no communication with operators, and no waiting for fund rebalancing. Moreover, you won't suffer any loss from the spread. You will get 6% yield, not 4%, and can view the pool throughout.

I understand the system considers problems from a perspective far more long-term than my logic. Moreover, when people don't do this, I don't think they are stupid.

Iyengar and Huberman's research on retirement plans found that as the number of fund choices increases, participation rates decrease. Faced with more choices, people ultimately choose not to participate in any plan. All consumer financial products since then have been built on this conclusion.

For fifteen years, self-custody has been the right choice, this has been well known for a long time. Even so, most on-chain credit card spending still goes through custodial platforms. This is a repeatedly mentioned and massive preference. Moreover, their security algorithms are superior to ours. For someone with $2,000 and no crypto background, the most likely way to lose funds is losing seed phrases or sending them to the wrong address. An app with face recognition and account recovery features can eliminate this failure mode that could cause them loss. They pay 200 basis points to buy insurance for their own risk, this is a reasonable expense.

So Aave's approach is correct. This is exactly what a business with liquidity and no user loyalty should do, and all consumer apps in the crypto field are competing to move in the same direction, because no matter where we go, the same logic exists.

In the end, this is an acceptance of human nature. People crave safety, predictability, and most importantly—convenience. Life is hard enough, why manage a private bank account? They just want to close the app and look at that static number.