We appreciate the opportunity to submit this comment letter on the proposal implementing the Guiding and Establishing National Innovation for U.S. Stablecoins (GENIUS) Act.
Our comment focuses on one issue: How to ensure that stablecoins maintain the “singleness of money”—always redeemable at par value both in fact and in expectation. To achieve these goals, stablecoins need to be fully backed only by assets with minimal risks or to hold capital to absorb losses of any reserve assets, such as uninsured demand deposits, that have credit or liquidity risks.
The proposal states that the capital requirements for permitted payment stablecoin issuers (PPSIs) would focus primarily on the operational risk of stablecoin issuers. It states that credit risk, market risk, and interest rate risk are minimal for stablecoin issuers and can be addressed through other means, such as reserve asset liquidity and diversification requirements.
We do not agree that these risks for uninsured demand deposits are minimal and are similar to the risks of the other reserve assets. We also do not agree that liquidity or diversification requirements would be sufficient to address the potential for loss if the value of uninsured deposits fell because of rising risks of bank failures. Were such losses to occur when a bank failed, the value of a stablecoin’s reserves could fall to below 1-to-1. The risk of such an outcome could prompt a run on a stablecoin because holders were no longer confident of convertibility at par value.
Minimizing run risk is critical because a run on a stablecoin could have systemic consequences for other stablecoins that hold similar assets. It also could lead to stresses in broader markets as PPSIs were forced to sell reserve assets to meet redemption requests.
For stablecoin issuers, the OCC should require higher capital for uninsured demand deposits because they have greater credit and liquidity risk than other eligible reserve assets, such as coin, currency, and Treasury bills, and could lead to reserves backing the stablecoin to fall to less than 1-to-1. The capital requirements should be set at a level that provides strong incentives for PPSIs to hold all of their reserves in the safest assets (e.g., central bank reserves, Treasury bills, and insured bank deposits).
Uninsured bank deposits are not risk-free and not always highly liquid. The bank failures in March 2023 demonstrated this (Liang and Dudley, 2026). That episode saw the near-instantaneous loss of access to tens of billions of dollars in deposits at Silicon Valley Bank (SVB) and Signature Bank. Uninsured depositors faced the prospect of receiving less than 100 cents on the dollar and experiencing prolonged delays in accessing their funds. Although that outcome was avoided when the U.S. government issued a systemic risk exception to insure all the deposits of these two institutions and the Federal Reserve established a special lending facility, that response occurred only after the depositor run. Moreover, uninsured depositors cannot be assured that such systemic responses would necessarily occur in the future. They are made only on a case-by-case basis and only after a bank has failed.
The risk that the value of the reserve assets could fall below par value could prompt a run on the stablecoin backed by bank uninsured deposits. The run on the Primary Reserve Fund, a prime money market mutual fund (MMMF), in 2008 is an example of how the promise of par-value redemption when backed by short-term uninsured debt of financial firms can lead to runs and cause systemic damage to the broader financial system (Liang, 2025).
Before 2008, prime MMMFs promised stable value and were allowed to invest in risky deposits and commercial paper of financial companies without any capital to absorb losses. When Lehman Brothers filed for bankruptcy on September 15, 2008, the Reserve Primary Fund held about 1.2 percent of the fund’s $62.6 billion in assets in Lehman unsecured commercial paper and medium-term notes. That single credit loss, amounting to little more than one cent on the dollar of the fund’s total portfolio, was sufficient to cause the fund to “break the buck”: its net asset value fell below the $0.995 threshold at which a $1.00 per share price could no longer be maintained.
The announcement that the fund had broken the buck triggered a run of extraordinary speed and scale. Within two days, redemption requests exceeded $40 billion—nearly two-thirds of the fund’s total assets. The fund was ultimately suspended and wound down under SEC order, with investors waiting months to recover half of their funds, and years for further distributions. The broader market consequences were severe. In the four weeks following the Lehman bankruptcy, prime money market funds as a group lost approximately $450 billion in assets—roughly 21 percent of their total—as institutional investors fled to government only MMMFs. Commercial paper markets, which depended heavily on prime money market fund demand, seized up for even the highest-rated issuers.
This episode illustrates several risks relevant for setting capital requirements for PPSIs: First, even though the loss that triggered doubts about the ability to redeem at par was very small, without any capital to absorb losses, the share price fell to below $1.00. Stablecoin issuers currently hold significant shares of their reserves in demand deposits, of which a large share is very likely above the insured amount of $250,000. For example, Circle USDC had almost 14% of its portfolio in the deposits of regulated financial institutions as of June 2025.1
Second, the first-mover advantage creates incentives to redeem immediately, which creates the incentive to run during times of stress. This dynamic could be even faster for stablecoins where holders can submit redemption requests any time of any day (24/7/365) globally on multiple blockchains.
Third, the costs of the systemic spillovers from a single break-the-buck event were far larger than the event itself. For payment stablecoins, spillovers could also be very large, with flights from stablecoins with similar risky assets but without capital to absorb possible losses. Banks would also be at risk as PPSIs withdrew their uninsured deposit reserves to forestall runs or meet redemption requests.
In addition, the connection between risky bank deposits and a PPSI’s ability to maintain par value creates a two-way risk. Stress at the bank that holds the PPSI’s uninsured deposits could trigger a stablecoin run and a stablecoin run would accelerate the deposit outflow from the already-stressed bank.
The Reserve Primary Fund episode underscores both the magnitude of the risk that deposit-based reserves create for stablecoin issuers and the inadequacy of a capital framework that does not directly address that risk. We strongly urge the OCC to require PPSIs to hold capital specifically calibrated to the credit risk of the deposits held as reserves. The capital requirement should be high enough to ensure that stablecoin holders will never view the par value of their stablecoin holdings as potentially being at risk.
Stablecoins that cannot ensure that they will always trade at par value (and always be perceived that way) cannot function effectively as money. Money serves three functions in an efficient system: unit of account, medium of exchange, and store of value. Establishing appropriate capital requirements for risky assets held as reserves would ensure store of value and unit of account properties, which would allow stablecoins to function effectively as a medium of exchange.
In the wildcat banking era of the 19th century, banks issued their own bank notes to be used as currency, but holders could not be confident that the value of the notes represented the same unit of account or were stable; instead, they were information-sensitive assets (see Gorton and Zhang, 2023). Transactions were inhibited by the ability of holders of these bank notes to assess their fluctuating values, and many banks failed.
Liquidity requirements and diversification requirements are not sufficient substitutes for capital requirements. Liquidity requirements can help to reduce fire sales of less liquid assets in the face of runs and redemption requests, but liquidity cannot fill the initial capital hole when the value of uninsured demand deposits falls. Diversification requirements can limit concentration but could only be a suitable substitute for capital if they were to require that PPSIs diversify enough such that all demand deposits would be insured, no more than $250,000 per institution. This requirement would be burdensome and unworkable for sizable PPSIs with assets of tens of billions.
Capital charges that do not reflect risks create bad incentives. Because PPSIs are not permitted to take deposits and make loans, as banks do, their capital requirements can be lower and simpler than for banks while still being sufficient to support the singleness of money. At the same time, ignoring differences in credit risk and liquidity risk of uninsured deposits from Treasury bills, which can be material, distorts incentives and will lead to riskier outcomes, as an issuer is not required to internalize the costs.
The OCC’s proposed rule represents a carefully considered effort to implement the GENIUS Act’s framework for payment stablecoins. We urge the OCC to adopt risk-sensitive capital requirements for deposits held as reserve assets and to consider supplementary capital charges for very large PPSIs because of greater disruptions to the system if they were to experience runs and were to fail. We appreciate the OCC’s careful attention to these issues and welcome further dialogue on this issue.