Comments close on 27 July 2026 on one of the most consequential payment proposals the Federal Reserve has made in years. On 20 May the Board proposed a "payment account", an account that legally eligible financial institutions could use only for clearing and settling their payments[1]. Many of the institutions that could use it are not federally insured.
What the account is, and is not
The design is deliberately narrow. According to Morrison Foerster's analysis, a payment account would give access to the Fedwire Funds Service, FedNow, the National Settlement Service and Fedwire Securities for free transfers, but not to FedACH[2]. It would carry no interest, no intraday credit and no discount window access. Each Reserve Bank would set a closing balance limit based on the institution's expected payment flows, with a hard cap of $1 billion.
Two other signals matter. The Fed has asked Reserve Banks to pause decisions on access requests from Tier 3 institutions, those neither federally insured nor federally supervised, with the pause expected to end on or before 31 December 2026. And a 19 May executive order directed the Fed to review wider account access for uninsured institutions and non-bank financial companies, with a report due by mid-September.
Separating payments from banking
The important idea is the separation. For decades, direct access to the Fed's payment rails came bundled with the full privileges of a bank: interest on reserves, intraday credit and emergency lending. That bundle is why access was restricted, and why fintechs have relied on sponsor and correspondent banks to move money for them.
A payment account unbundles it. An eligible institution could settle payments directly, in real time, without receiving the privileges that create risk for the Fed and the deposit insurance fund.
Access to the rails is being separated from access to the safety net. That removes the main reason many fintechs need a bank in the middle.
Who gains and who loses
The likely beneficiaries are fintechs with novel banking charters, the category Morrison Foerster identifies as the primary users. For them, direct access means faster settlement, lower cost and less dependence on a partner bank that can change terms or exit the relationship.
The banks that lose are the ones earning revenue from settling payments on behalf of fintechs and payment companies. Sponsor banking will not disappear, because many fintechs will still need deposit accounts, card issuing and credit that a payment account does not provide. But settlement, the most commoditised part of the service, becomes something a growing number of clients can do for themselves.
What I would do
For banks in the sponsor and correspondent business: segment fintech clients by whether they are likely to become eligible for a payment account, and model the revenue at risk. Then move up the value chain, into services a payment account does not provide, such as credit, liquidity, fraud and compliance tooling, and multi-rail orchestration.
For fintechs: a payment account is not a bank charter. Eligibility depends on legal status, and the balance cap and lack of credit mean treasury and liquidity planning still matter. Treat it as a way to reduce dependency, not to remove the need for banking partners entirely.
The final rule may change after the comment period. The direction, payments access that is open to more institutions but deliberately limited, now looks settled.