BIS Study Finds Rising Bank Deposit Rates May Slow Stablecoin Growth
BIS Working Paper 1382 uses weekly U.S. data from 2019 to 2025 to estimate the semi-elasticity of stablecoin market capitalization with respect to deposit rate changes, employing local projections and instrumental variable methods. The results show that the two are partial substitutes: deposit rate increases predict slower stablecoin growth, with USDC responding more strongly than USDT. The authors also emphasize that the analysis concerns the allocation of funds between assets and does not establish that stablecoin growth necessarily reduces aggregate bank deposits or lending.
ImportanceMaterialEvidenceE2 unreplicatedWrite-upQuick
Stablecoins are digital tokens designed to maintain a stable value, usually against the US dollar. Like bank deposits, they can provide payment and liquidity services, but they generally pay little or no interest. Bank deposits likewise provide liquidity and can pay interest. Do the two compete for the same funds? In BIS Working Paper 1382, published in October 2026, economists Rashad Ahmed and Iñaki Aldasoro provide an empirical answer using U.S. weekly data.
Their core finding is that stablecoins and bank deposits are partial substitutes. When banks raise deposit rates, the growth of stablecoin market capitalization slows. Specifically, against average ten-week stablecoin growth of 15%, ordinary least squares estimates imply a reduction of roughly 3 percentage points per basis point of deposit repricing; instrumental variable estimates are approximately three times larger and pass a placebo test using non-stablecoin cryptocurrencies.
The key identification strategy exploits the nonlinear way in which banks pass changes in policy rates through to deposit rates. Whereas other money market rates track the federal funds rate roughly linearly, banks reprice deposits more aggressively once it exceeds approximately 3%. This nonlinearity provides an exogenous source of variation for the instrumental variable, helping to address the endogeneity arising from possible two-way causality.