Science1 publisher2 min readPublished
A Penn State analysis links Aave's lending rates to U.S. Treasury yields
Siddharth Bhambhwani's comparison of 39 months of Aave rates against U.S. Treasury yields finds the two connected, even though the protocol sets its rates from pool utilization and never looks at government debt.
The Scientist · Science desk
What happened
- Siddharth Bhambhwani compared Aave's borrowing and deposit rates with U.S. Treasury yields from January 2023 to March 2026 and found the yields significantly influence crypto lending rates.
- Deposits based on stablecoins have often paid substantially more than traditional bank savings accounts, and DeFi users generally lack deposit insurance.
- A DeFi protocol generally ignores a borrower's credit score, income and employment history, and the markets run around the clock without requiring a conventional bank account.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- decision Someone weighing a stablecoin deposit against a Treasury bill is choosing on the spread and the protocol risk, since the level underneath both appears to move with the same driver.
- constraint For the pools in this sample, holding a DeFi position because on-chain credit is priced independently of the risk-free curve loses its stated justification.
- exposure Depositors taking the on-chain premium accept smart-contract failure risk while their yield stays tied to the same inflation and policy expectations that price government debt.
Aave prices credit with a formula. The rate on a pool comes from utilization, the ratio of assets borrowed out of it to assets deposited into it, so a pool with little borrowing pays little [5]. The formula does not look at the Treasury market. If the government curve moves Aave's rates, it can only do that by changing how much depositors leave in the pools and how much borrowers take out [15].
"While this system is not directly linked to traditional markets by design, my study found that Treasury yields and DeFi rates are nevertheless connected," Bhambhwani said in a Q&A published by Penn State [6]. He is an assistant clinical professor of accounting at the university's Smeal College of Business [1].
The sample runs 39 months and covers one platform [14]. A design like that can establish that the two series move together. It cannot separate Treasury yields acting on pool rates from both series reacting to the same conditions, and Treasury yields themselves respond to expectations about inflation, economic growth and monetary policy [12].
What was compared is the price of on-chain credit [2]. An investor who holds crypto for diversification is making a claim about asset returns. This test does not reach it.
Stablecoin deposits have often paid substantially more than traditional bank savings accounts [7], and that gap can persist while both rates rise and fall in step. The extra yield comes with a different risk. Smart contracts can contain vulnerabilities and have been exploited, and recovering funds after a hack or a failure can be difficult [9].
Borrowing demand depends on collateral. DeFi borrowing generally requires collateral worth more than the loan: lock up $100 to borrow $80 [10]. Utilization therefore moves with crypto prices as well as with what cash pays elsewhere. If the collateral falls close to the borrowed amount plus accrued interest, the software sells it automatically and repays the loan [10].
What to watch
- The coefficients in the Finance Research Letters paper: how much of the variation in Aave rates the Treasury series explains, and whether stablecoin pools and volatile-asset pools behave alike.
- Whether the relationship holds if Treasury yields fall sharply after March 2026, where the sample ends.
- Replication on lending protocols other than Aave, which would show whether this is one platform's rate curve or the on-chain credit market generally.