August 25, 2026

Four stablecoin developments bankers should know about

The Stablecoin Banker is a periodic newsletter keeping bankers on top of the stablecoin industry. I highlight top stories that are relevant to banks, with my insights and commentary to draw out the most important conclusions.

If you find this content useful, please feel free to forward it to your friends and colleagues. They can also subscribe directly or read back issues at thestablecoinbanker.com. If you're interested in stablecoin services for your bank, feel free to reach out to us at Omnia.

"My job is not about incumbent protection."
— Jonathan Gould, Comptroller of the Currency (Aug 19, 2026)

In This Issue

• Rain completes the consumer stablecoin payment loop
• Treasury's GENIUS proposal starts closing the U.S. market to foreign stablecoins
• FASB proposes moving stablecoins into cash & equivalents
• Swift made correspondent banking run 24/7. It's still correspondent banking.

Plus, tidbits you may have missed in our Coupon Clippings section.

Rain Assembles the Stablecoin Payments Loop

Rain, the stablecoin card processor and principal member of both Visa and Mastercard, acquired Ansa, a stored-value platform that runs branded prepaid and wallet balances for merchants. The acquisition follows Rain's November purchase of onchain-rewards firm Uptop. The company also launched the Agentic Payments Alliance, a standards coalition with 26 founding members including Visa, Mastercard, Shift4, Circle, Fiserv, Lithic, and Remitly, focused on authorization, agent identity, fraud detection, and loyalty for agent-driven commerce. Sources [Rain, American Banker]

My Take:

I want to share a hypothesis about where Rain may be going. They now hold every piece of a consumer payment loop that can settle in stablecoin without ever touching network interchange. It starts with loyalty. The merchant enrolls the customer in a loyalty program at checkout, which provisions a stored value balance with an incentive to load. That loyalty credential is a stored-value account on Ansa's rails, able to transact with any Ansa-powered merchant. The stored value balance can be held in stablecoin, with an attached Rain-issued debit card. The user can pay with their stored value at the terminal. For unsupported merchants, the user can pay with their Rain card using the same balance.

Starbucks has one of the most successful stored value loyalty programs out there, with $1.6 billion+ in unredeemed balances and 35 million 90-day actives. By contrast, the average SMB doesn't have the repeat spend at sufficient scale to attract meaningful stored value behavior. In Rain's model the stored value is the customer's own portable stablecoin balance, spendable across every participating merchant. It generates income to fund attractive rewards in three ways: first, the balance generates back-end yield; second, the stored value transaction at the merchant avoids the entire card processing stack, saving the merchant ~3-5% on the transaction; third, the attached Rain Visa card generates interchange at non-participating merchants that flows back towards participating merchants.

It's worth pointing out why stablecoins uniquely enable this model. Rain's direct-to-visa settlement model eliminates prefunding capital inefficiency and cuts out an inefficient traditional depository bank. Loyalty program transactions, which look a lot like account-to-account payments, use stablecoins to avoid settlement delay and NSF or ACH return risk. Stablecoins also make this work globally.

Meanwhile, Rain is getting closer to Visa and Mastercard via their Agentic Payments Alliance. How might the networks react to this play? I wonder: could an apparent interchange threat today turn into an opportunity. They understand that agentic commerce favors stablecoins, because card economics don't work for the majority of emergent transactions, and they lost payments in some foreign pay-by-app markets. Thus, Rain may be their ticket to the stablecoin based payments version of cards: an experimental parallel stablecoin network that becomes big enough to prove out functionality across geographies and transaction types, which the networks can then absorb and scale further. They can bring standards for reversibility, fraud and dispute management, and complex pricing which have evaded the stablecoin ecosystem thus far. Visa and Mastercard have said that they don't yet know how they'll monetize agentic payments, yet they engage to learn. Will they take the same posture if Rain tries to shift card spend onto stablecoin rails?

Supposing this comes to pass, there will be a dual impact on community banks: shifting of deposit balances that back card spend today to stored value balances, and loss of interchange revenue. Banks that want a seat in this loop should be evaluating stablecoin conversion, settlement, and issuing roles now.

Treasury Starts Building a Border Wall for Stablecoins

US Treasury issued a proposed rule implementing Section 3 of the GENIUS Act, defining what it means to "issue" a payment stablecoin in the United States and to "offer or sell" one to a U.S. person. The issuer definition is drawn around whoever bears the redemption obligation — the operative question for white-label and bank-branded arrangements — and issuance by anyone other than a permitted issuer becomes unlawful, with penalties up to $1 million and five years' imprisonment. Comments are due October 19, 2026.

The GENIUS Act takes effect January 18, 2027, and from that point marketing a non-permitted stablecoin as a payment stablecoin to U.S. persons is prohibited. From July 18, 2028, digital asset service providers including exchanges, custodians, and transfer services may not offer or sell any payment stablecoin to U.S. persons unless it comes from a permitted issuer, and must verify a foreign issuer's compliance before offering its coins. The proposal carves out self-custody, direct peer-to-peer transfers, and same-owner transactions, and offers a safe harbor for inadvertent issuance conditioned on policies against U.S.-targeted marketing. Treasury poses dozens of questions, including how far providers must go to ensure offers don't reach U.S. persons. Sources [Treasury, CoinDesk]

My Take:

The NPRM confirms what the statute implied: GENIUS is on track to close the U.S. market to foreign stablecoins that don't meet its requirements, and the closure arrives sooner than the 2028 date suggests. The NPRM makes the impact of this prohibition clearer. Consider Tether, whose USDT is the largest stablecoin in the world, with most of its adoption outside the U.S. It is not GENIUS-compliant. For example, its reserves include gold and bitcoin, which are disallowed by GENIUS. Tether launched USAT, a compliant U.S. coin, for exactly this reason. But after July 2028, a U.S. provider cannot offer a swap from USAT into USDT to a U.S. person, because that swap is an offer of USDT. (The reverse direction appears to survive — swapping USDT into USAT doesn't offer USDT for sale.) A U.S. person holding USAT who wants to pay a counterparty abroad in USDT will have no legal onshore venue for the trade. And the squeeze starts in January 2027: offering the swap remains legal, but marketing it to U.S. persons does not.

Resourceful users will route through self-custody wallets and exempted protocols because the carve-outs allow it. But most wallet providers monetize the transactions they facilitate, which makes them digital asset service providers, and therefore they won't be able to explicitly market stablecoin swaps via exempted protocols to their users. The marketing prohibition effectively balkanizes the global stablecoin network, which may be the point. Isolating the U.S. market pressures foreign issuers everywhere toward GENIUS-style reserves in U.S. deposits and Treasuries, extending American stablecoin policy globally and entrenching dollar dominance. It also recreates the offshore structures the industry built in 2022–2024 to stay clear of Washington. We'll see very awkward arrangements for U.S. providers to offer non-GENIUS coin access.

Community and regional banks entering this business should choose their supported stablecoins with the 2027 and 2028 dates in view. The exposure is sharpest for banks with meaningful international business, whose customers may expect access to foreign coins to complete cross-border transactions that will soon have no compliant onshore path.

Stablecoins Reach the Top of the Balance Sheet

FASB issued a proposed Accounting Standards Update on cash equivalents with two components. First, all entities presenting cash equivalents would disclose their significant components and amounts, whether or not any are digital assets. Second, new illustrative examples would clarify how the existing cash-equivalents definition applies to digital assets such as stablecoins. Under the examples, a stablecoin qualifies only if the holder has a direct, on-demand contractual redemption right against the issuer at a fixed amount, redemption carries no significant fees or restrictions, and the issuer maintains segregated reserves of cash and Treasury bills with original maturities of three months or less on at least a 1:1 basis; secondary-market saleability alone is not sufficient. FASB is not changing the definition itself, and cash-equivalent presentation remains an election. Comments are due November 19, 2026. Sources [FASB, CoinDesk, JofA]

My Take:

FASB didn't change the definition of cash equivalents because it didn't need to. The GENIUS Act codified redemption rights and reserve quality, so certain compliant coins already fit. The board is simply saying so with examples, and requiring everyone to disclose what sits inside the cash-equivalents line. When FASB took this project up last November, I wrote that reclassification would broadly unblock corporates to hold and use stablecoins like cash deposits and money market instruments, and in March, when the SEC gave payment stablecoins a 2% net capital haircut, that stablecoins were marching up the balance sheet toward being formally treated as money. This proposal is the final step in that march: qualifying holdings move out of the intangibles footnote and into cash.

Two qualifications matter. First, the examples require a direct redemption right against the issuer. That burdens holders, who would need redemption accounts with every issuer they touch, and it undercuts the intermediaries who make multi-stablecoin convenience possible, since their customers must contractually disintermediate them to qualify. Issuers should be drafting flow-through provisions that extend the legal redemption right through intermediaries to end holders. Second, the successful examples look through to reserves with original maturities of three months or less, while GENIUS permits investments with remaining maturity under 93 days. Read literally, that divergence could disqualify even USDC, which routinely holds near-maturity paper with longer original maturities. I suspect a drafting oversight here considering companies regularly mark 2a-7 money market funds as cash, and those hold the same assets.

Bankers that want to block stablecoins (speaking to you, BPI) should submit their comment and keep stablecoins stuck in intangibles. Once a CFO can show stablecoins inside the cash line, corporate treasury use cases open up: operating funds in a 24/7 liquid instrument, real-time rebalancing across entities at no cost, paying suppliers and contractors in stables. When this change lands, corporate stablecoin demand will arrive through banks' commercial customers, and banks that can hold, convert, and move stablecoins from the deposit account will be the ones who keep the relationship.

Swift Made Correspondent Banking Run 24/7. It's Still Correspondent Banking.

HSBC and Standard Chartered completed the first bank-to-bank tokenized deposit settlement across Swift's blockchain ledger, which went live in July with 17 "pioneer" banks. The ledger acted as an orchestration layer between HSBC's Tokenised Deposit Service and Standard Chartered's tokenized deposit rails, matching and netting obligations between the two banks; final settlement ran through existing banking systems, and no common deposit token moved between them. Swift says the design "leverages existing compliance processes" and that banks retain full authority over keys, assets, funding, and settlement through RTGS systems or correspondent relationships. Sources [Payment Expert, Swift]

My Take:

Swift is using tokenization as a faster accounting database for the largest banks. The tokens exist to produce net settlement instructions, letting member banks process Swift payments around the clock instead of waiting on gross settlement through central banks and their operating hours. When this project hit "design complete" in April, I called it a database upgrade for the existing messaging network, and the first live settlement confirms the architecture: same participants, same roles, same market structure, now open on weekends. The benefit accrues mainly to correspondent pairs lacking bespoke bilateral nostro/vostro arrangements that already achieve 24/7 clearing with deferred net settlement.

Credit where due: 24/7 cross-border availability is real, and it is one of the pain points stablecoins have been attacking. But this fixes only one step in a correspondent transaction: the interbank financial transaction between two correspondents. It does not touch how each bank processes the payment: compliance holds, screening, AML, and posting all remain per-bank, by design. And it does nothing for the thousands of banks that depend on correspondents. Even if every bank could become a user of this new system, they remain one node in a multi-bank hop with gates and hurdles at each step. Meanwhile, a stablecoin transaction moves point-to-point in real time. Swift says that 75% of cross-border payments reach beneficiary banks within 10 minutes. Stablecoins reach the beneficiary wallet or institution in seconds 100% of the time.

Community banks should not assume Swift's roadmap will close the gaps stablecoins currently fill. All 17 pioneers are global correspondents, the model requires a bank to issue tokenized deposits on its own ledger and hold its own keys, and Swift has published nothing about community or regional participation. If this rail ever reaches smaller banks, it will be through a correspondent — the same dependency, on a newer database.

Coupon Clippings

Erebor Raises Again, Its Valuation Comes Back to Earth

Erebor — the Palmer Luckey–backed national bank chartered in February for crypto, defense, and frontier-tech clients — is in talks to raise $1.5 billion at a $9.5 billion post-money valuation, with deposits up from $1.1 billion in late March to $4.1 billion at the end of Q2. By my math, this round prices the bank at 4.5x post-money book, down from roughly 7.6x at the December Series B. I've flagged VC return expectations as an open question on VC-funded banks before because the regulated capital that defines a bank is at odds with the return goals of most venture investors: every successive round comes at a lower book multiple while diluting the investors before it. JPMorgan trades at 2.6x and the fintech-sponsor banks that supposedly carry a tech premium sit at 1–2x. Meanwhile, the bank runs with a minimum 12% Tier 1 leverage ratio and, according to Luckey, "we'll have the most conservative loan-to-deposit ratios of any bank in history." This implies that the bank will need to make up the difference with substantial non-interest income, not yet demonstrated, unusually high NIM, or impressive gains in operating efficiency. Having worked with many banks, I believe the latter is by far the most accessible approach for de novos that focus their service set and built their own software (both of which Erebor is doing). Barring these strategies playing out, investors will need to adjust their return expectations. [Full Story]

The OCC Chartered a President's Stablecoin and Rejected a European Neobank

The OCC granted preliminary conditional approval for a national trust charter to World Liberty Trust Company, allowing the Trump-family-backed World Liberty Financial to issue and redeem its $4 billion USD1 stablecoin. Days earlier, the OCC rejected Dutch neobank bunq's third U.S. charter attempt, citing inadequate capitalization, inexperienced U.S. management, and an "unrealistic" path to profitability — the second published denial in three weeks, after Wise, which I covered last edition. Some speculate that the concurrent timing of the denials alongside the World Liberty approval was a political move to demonstrate that the OCC is applying an objective risk standard to applications, and not simply following the interests of the Trump administration. Senator Elizabeth Warren was undeterred and introduced a bill the following day that would bar charter approvals tied to a president or members of Congress. In any case, the useful artifact here is the denial letters themselves, which the OCC now publishes; they convert the gaps — compliance depth, U.S. management, capitalization — into a preparation checklist for everyone still in the queue. Bank strategy teams watching the charter pipeline should read them the way applicants will: as the exam answers, published in advance. Sources [CoinDesk, Banking Dive]

Is the Bank Lobby Warming to Stablecoins - Or Just Protecting its Own Hide?

The ABA and Bank Policy Institute jointly asked the OCC to delay finalizing its new weekly and quarterly reporting forms for permitted stablecoin issuers until the substantive GENIUS Act rules are final, to define the reporting terms consistently with Call Report instructions, and to drop the line item collecting how much of an issuer's reserves are held in tokenized form. The rest of the letter is considerably less accommodating: it wants issuers filing the full forms on top of whatever they already file, and objects to the FDIC's lighter weekly form for issuers under $1 billion as an invitation to regulatory arbitrage. So this is not the banking lobby warming to stablecoins. The tokenization line item is the only ask in the letter that isn't really about issuers at all, and I'd argue that's the point — it would be the first federal reporting field in which "tokenized" is a reportable attribute of a balance sheet position. There is still no Call Report line for tokenized deposits, and none proposed. Thus, this change sets a precedent for requiring banks to report on how much of their deposit base is tokenized, which is a question no bank wants on a Call Report because it triggers supervisory questions about whether an instantly redeemable, always-on liability deserves its own outflow assumption under LCR. Maximum visibility into the competition, none into ourselves. [Full Story]

Zaria Files for a Trust Charter Built Around Tokenized Collateral

Zaria Systems applied to the OCC for a special-purpose national trust charter consolidating corporate trustee, agency, loan servicing, collateral management, and backup servicing for structured finance. The application demonstrates how trust charter applicants are expanding beyond pure crypto activity: a single federal license covering work that would otherwise require state-by-state money transmitter licensing or a full depository charter. That is the point bankers should absorb — trust charters are not just for crypto companies, they work for anyone holding assets, moving money, or filling statutory roles like trustee, and plenty of those activities have historically belonged to depository banks only because no other nationally recognized charter existed. I wrote in January that federal regulators are unbundling banking, creating the specialized middle tier between money transmitter and full bank charter that Europe has had for years; Zaria's filing extends that thesis from payments and custody into the back office of tokenized credit. Expect more applicants who want the charter without the balance sheet. [Full Story]

Thanks for reading. If you don't hear from us for a bit, don't worry—we'll be back when there's something important to share.

Omnia is a provider of stablecoin infrastructure for banks that want to capture growing demand for stablecoins. If you're interested in learning more about us, please get in touch.

Getting value? Forward to a colleague and subscribe.