Key Highlights
- A BIS working paper examined weekly U.S. data from January 2019 to June 2025 to assess the link between deposit rates and stablecoin growth.
- The baseline model found that a 1-basis-point increase in bank deposit rates was associated with roughly 3 percentage points less stablecoin growth over 10 weeks.
- An instrumental-variable model estimated a larger effect of about 9 percentage points, but its instrument-strength statistic warrants caution.
A new working paper from the Bank for International Settlements (BIS) finds that higher bank deposit rates are associated with slower stablecoin market-cap growth, suggesting that returns available through traditional banking may influence stablecoin demand.
The paper, Are Stablecoins and Bank Deposits Substitutes?, was published on October 13 by researchers Rashad Ahmed and Iñaki Aldasoro. It examines weekly U.S. data from January 2019 to June 2025, comparing changes in bank deposit rates with stablecoin market capitalization.
The findings suggest that stablecoin demand may respond to changes in returns available through traditional financial institutions. However, the study does not establish how much money moves directly from bank deposits into stablecoins or whether changes in stablecoin demand affect bank lending.
Higher deposit rates are linked to slower stablecoin growth
Bank deposits and stablecoins can both be used to hold funds and make payments, but they offer different features. Bank accounts may pay interest, while fiat-backed stablecoins generally do not automatically provide holders with a return simply for holding the tokens.
The BIS researchers used an ordinary least squares (OLS) model to examine the relationship between changes in bank deposit rates and stablecoin market capitalization. The baseline estimates indicate that a 1-basis-point increase in bank deposit rates was associated with approximately 3 percentage points less cumulative stablecoin growth over the following 10 weeks.
Stablecoin market capitalization grew by about 15% on average over a 10-week period in the sample. Under the model’s estimate, a comparable period following a 1-basis-point rate increase would therefore see growth of roughly 12%, all else equal.
The estimate indicates slower growth, not necessarily a contraction in stablecoin supply. It also does not establish that users moved funds directly from stablecoins into bank accounts.
Instrumental-variable model estimates a larger effect
The researchers also used an instrumental-variable (IV) approach to address the possibility that bank deposit rates and stablecoin demand influence each other. The paper notes that bank deposit rates tend to respond more strongly when the federal funds rate rises above approximately 3%, while short-term Treasury bill rates track changes in the federal funds rate more consistently.
Using this approach, the researchers estimated that a 1-basis-point increase in bank deposit rates was associated with about 9 percentage points less stablecoin growth over 10 weeks.
However, the model’s first-stage F-statistic was 9.78, slightly below the conventional benchmark of 10 often used to assess instrument strength. This warrants caution when interpreting the IV estimate.
The researchers also conducted a placebo test using the market capitalization of other crypto assets. They found no statistically significant response, supporting the possibility that the observed relationship is specific to stablecoins rather than a broader reaction across the crypto market.
USDC responds more strongly than USDT
The study found different responses for the two largest stablecoins in its analysis. The researchers estimated that USDC’s market-cap growth fell by about 4.8% over 10 weeks following a 1-basis-point increase in bank deposit rates, compared with approximately 1.9% for USDT.
The authors suggest that the difference may reflect how the stablecoins are used. USDC has stronger connections to U.S. institutions and decentralized finance (DeFi) activity, while USDT has a larger presence in offshore markets and some developing economies.
The study does not identify the locations of individual stablecoin holders, so these differences remain a possible explanation rather than a confirmed account of user behaviour.
DeFi stablecoin balances also show a negative relationship
The researchers examined whether stablecoin balances held in DeFi lending protocols responded to changes in bank deposit rates. The analysis controlled for DeFi yields, which can influence whether users choose to hold stablecoins in lending protocols or elsewhere.
For USDC, the researchers estimated that balances in DeFi lending protocols declined by about 3.4% over 10 weeks following a 1-basis-point increase in bank deposit rates. USDC balances outside lending protocols showed a larger estimated decline of 5.5%.
For USDT, balances in DeFi lending protocols fell by an estimated 6.1%, while the response for balances outside lending protocols was small and statistically insignificant.
The authors caution that these estimates are less precise and rely on a shorter sample than the main analysis.
Findings remain subject to data and methodological limits
The study uses 339 weekly observations covering January 2019 to June 2025. Its findings are based on aggregate market data and do not identify individual holders, the banks they use, or where stablecoin redemptions ultimately send funds.
A decline in stablecoin market capitalization does not necessarily translate into an equivalent increase in bank deposits. Funds could move into other crypto assets, remain in cash, or be used for other purposes.
The researchers tested whether the results changed after accounting for major market events, including the TerraUSD collapse, the FTX failure, the wind-down of Binance USD (BUSD) and the disruption to USDC associated with Silicon Valley Bank in 2023. They also tested alternative deposit-rate measures and controls for crypto-market activity and DeFi yields.
The negative relationship persisted across these robustness checks.
Earlier BIS research examined stablecoins and traditional finance
The new paper follows other BIS research and commentary on the relationship between crypto assets and the traditional financial system.
On September 16, a BIS study found that estimates of Bitcoin transfer values could differ by as much as sixfold depending on the measurement method.
On August 29, BIS General Manager Pablo Hernández de Cos said stablecoins were not yet a credible means of payment at scale. He pointed to tokenized bank deposits as a more direct connection between blockchain-based finance and the traditional banking system, while raising concerns about bank funding costs, interoperability and monetary sovereignty.
A separate BIS study published in July examined dollar-backed stablecoin flows across more than 130 economies. It found that those flows were largely unaffected by capital-flow restrictions, raising questions about monetary control in emerging and developing economies.
The earlier report covered Bitcoin transfer measurement, stablecoin payments and monetary policy, while the latest study examines the relationship between bank deposit rates and stablecoin demand.
BIS paper raises questions for stablecoin regulation
The latest findings raise questions about whether stablecoin issuers should be required to hold part of their reserves as bank deposits and whether they should be allowed to pay interest to holders.
The authors discuss how a reserve requirement could support liquidity management. However, concentrated uninsured deposits at a small number of banks could also create financial-stability risks.
Whether stablecoin issuers can pay interest may influence how users choose between stablecoins and traditional bank accounts.
The study does not establish that stablecoins are replacing bank deposits on a one-for-one basis or that slower stablecoin growth leads to reduced bank lending. Whether changes in stablecoin demand materially affect bank funding or credit provision remains unresolved.
Also Read: FSB Flags Stablecoin and AI Risks Ahead of G20 Meeting
