tokenized deposits drain US bank lending — Research published this week has put a hard number on a risk that bankers have discussed in private for two years: widespread adoption of tokenized deposits could reduce U.S. bank lending capacity by $580 billion, or roughly 5% of total bank lending. A paper released on August 25 modeled what happens when bank deposits gain the speed of blockchain transactions, and the conclusion is that the deposit base becomes structurally less stable. That matters because banks lend against stable deposits, and any technology that makes balances move in seconds instead of days forces lenders to hold more reserves and extend less credit.
The mechanics are rooted in fractional reserve banking. When a customer deposits $1,000, the bank keeps a fraction, typically 3% to 10%, and lends the remainder. That lent money circulates through the economy and is redeposited, generating the money multiplier that converts roughly $22 trillion in U.S. bank deposits into about $12 trillion in lending. The system depends on deposits being sticky. A customer who deposits on Monday is not expected to withdraw on Tuesday, and banks lend against that statistical floor with confidence.
Why tokenized deposits drain US bank lending capacity
Basel III codifies the stickiness assumption through the Liquidity Coverage Ratio, which requires banks to hold enough high-quality liquid assets to cover 30 days of net cash outflows under stress. Retail deposits are assigned the lowest outflow assumption, typically 3% to 10%, because individuals rarely move entire balances in a single day. Corporate deposits face outflow assumptions of 20% to 40%, and interbank deposits are treated as the least stable category. Tokenized deposits threaten to reclassify every balance into the highest outflow category, since the technology makes any deposit as mobile as an interbank transfer.
That reclassification is no longer hypothetical. In July 2026, LayerZero and Keeta launched tokenized bank deposits across Ethereum, Solana, Base, and Keeta, covering nine fiat currencies. The infrastructure now exists for deposits to move between institutions at blockchain speed. JPMorgan’s Kinexys platform already processes tokenized deposit transfers between institutional counterparties, and in Japan, MUFG, SMBC, and Mizuho are piloting tokenized government bonds settled through tokenized central bank reserves. USBC, Uphold, and Vast Bank launched the first retail tokenized dollar deposits in late 2025, marking the point at which the theoretical risk became operational.
The paper’s three scenarios produce sharply different outcomes. In the low adoption case, covering 5% to 10% of deposits, lending capacity contracts by roughly $120 billion, an amount banks could absorb through minor adjustments to reserve ratios and overnight funding markets. The moderate scenario, covering 15% to 25% of deposits, yields the $580 billion headline figure. That contraction equals roughly the total outstanding balance of U.S. auto loans, or about one third of all commercial and industrial loans outstanding, and would tighten credit availability for borrowers at the margin.
The high adoption scenario covers 35% to 50% of deposits and projects a $1.2 trillion reduction in lending capacity. At that scale, banks would need to restructure their funding models entirely, shifting toward wholesale markets, securitization, or Federal Home Loan Bank advances. Each alternative is more expensive than deposits, and the paper estimates average mortgage rates could increase by 15 to 30 basis points while small business loan rates rise by 25 to 50 basis points. The borrowers most affected would be first-time homebuyers and small businesses, the rate-sensitive segments of the credit market.
Regulators outside the United States are already treating tokenized deposits as legitimate financial infrastructure rather than experimental technology. The Bank of England endorsed tokenized deposits as belonging in UK payments rails, with deputy governor for financial stability Sarah Breeden stating they should coexist with stablecoins and a potential digital pound. South Korea has begun trialing tokenized deposits for government operational spending, signaling that adoption pressure is emerging from policymakers, not only from crypto startups. The direction of travel points toward broader integration rather than restriction.
The critical variable in the paper’s modeling is not how much deposits move but how fast they move. Traditional ACH transfers settle in one to three business days, and wire transfers cost $25 to $50 and take hours. Both impose friction that limits movement velocity. FedNow, the Federal Reserve’s instant payment system launched in 2023, settles in seconds but still operates within banking hours and per-transaction limits. Tokenized deposits eliminate those frictions entirely, allowing balances to be transferred across institutions and networks at any hour, in any size, without intermediary delays.
Industry response has been split along predictable lines. Crypto firms have spent two years building the infrastructure to put bank deposits on chain, while incumbent banks have spent two years modeling the consequences. The research paper now quantifies those consequences, and the $580 billion figure under moderate adoption is large enough to warrant attention from both camps. Whether the moderate scenario is realistic depends on how quickly consumers and corporations adopt wallets capable of holding and transferring tokenized balances, and on whether regulators allow deposit tokens to circulate outside closed institutional networks. The tokenized deposits drain US bank lending situation remains a defining test case for the cycle ahead.
For now, the moderate scenario remains a forecast rather than a measurement. But the infrastructure LayerZero, Keeta, JPMorgan, and the Japanese megabanks have deployed in the past twelve months suggests the forecast period is shorter than many bank treasurers assume. If even the low adoption case materializes, the adjustment will be visible in reserve ratios and overnight funding volumes. If the moderate case arrives, the contraction in credit supply will be felt by borrowers who never touched a blockchain, and the cost of that adjustment will land in the interest rates they pay. The research framing tokenized deposits as a drain on US bank lending capacity is now backed by quantitative modeling, and the number is large enough that both the crypto industry and the banking sector will need to engage with it rather than dismiss it.
Source: Crypto.news.