The Stablecoin Banker, November 17
November 17, 2026
Three 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.
"I'll be damned if crypto billionaires decide this race."
- Senator-elect Sherrod Brown (Sept. 2026)
In This Issue
- USDC In, Deposit Out: Pay-In Is Becoming Table Stakes
- Congress Flips. GENIUS Doesn't.
- The Float Is Shrinking. The Flow Isn't.
Plus, tidbits you may have missed in the Coupon Clippings section.
USDC In, Deposit Out: Pay-In Is Becoming Table Stakes
On November 9, PNC opened a limited pilot that lets eligible corporate and institutional clients receive USDC on Ethereum. The service runs on Coinbase Payments products, and incoming USDC is converted 1:1 into dollars and deposited into the client's PNC account. On October 30, Walmart-backed OnePay began letting select customers fund OnePay Checking with USDC sent on any of 12 blockchains. Stripe's Bridge takes the coins and converts them to dollars, which land at OnePay's sponsor banks, Coastal Community Bank or Lead Bank, with no conversion fee at launch. Neither product lets customers send stablecoins out. On November 5, Samsung's Stablecoin Account went live for eligible U.S. Galaxy users, who can hold, send and receive USDC on Solana inside Samsung Wallet, with Bastion running the accounts and Coinbase holding the USDC. Users can send to other Samsung users, to external wallets, or to bank accounts in more than 60 countries in local currency, but the account does not offer a USD deposit balance. Sources [PNC, OnePay, The Block]
My Take:
PNC and OnePay are adopting the stablecoin-as-money strategy, treating inbound stablecoin as one more way to fund a deposit balance. Samsung offers the contrasting approach: a native USDC balance in the wallet.
Stablecoin-as-money says that stablecoins are just another form of bearer funds, similar to cash. For a bank, this makes sense: in the same way that a depositor can fund an account with cash, PNC and OnePay allow funding with a stablecoin. Rather than hold the stablecoin in a bank-owned wallet booked as a balance sheet asset, their partners, Coinbase and Bridge, deliver dollars to the bank. The customer gets the simplest possible experience. The implication is that liquidity becomes more important, because providers offering this service need to convert every stablecoin they receive into fiat. Circle charges up to 5 bps for instant redemption, and other liquidity providers like Bridge have their own fees, so there's a real cost to offering stablecoin-as-money. OnePay's terms even let the sponsor bank credit the customer before Bridge settles, which means the bank is fronting some of that liquidity itself. Since the service only works to increase deposit balances and associated NIM, PNC and OnePay can offer the conversion at negative margin.
Samsung is not following this approach, principally because it's not able to offer a USD balance. It doesn't have a sponsor bank and doesn't embed a deposit account in Samsung Wallet. With the ability to send funds out of the wallet to local fiat rails in more than 60 countries (fees undisclosed), it doesn't really need one. Its economics don't depend on deposit yield either. Samsung advertises a 3.75% rewards rate, which suggests that the rewards Circle pays through Coinbase, the sub-custodian, are flowing straight through to the wallet holder. Samsung's incentive is to drive wallet adoption.
The other notable difference is outbound payments. Neither PNC nor OnePay supports sending stablecoins out today. Whether for risk, liquidity, or simply the desire to hold on to deposits, these products offer only inbound stablecoin payments. Samsung offers full outbound capability, to other Samsung accounts or to any on-chain wallet, which positions it for peer-to-peer payments that PNC and OnePay don't offer. Being inbound-only is a choice. Cross River already debits outgoing stablecoin transfers from a customer's dollar balance, and OnePay calls this launch the first of several stablecoin features. Its partnership with Tempo, announced in April, is explicitly about payouts.
For bank deposit products, stablecoin-as-money is becoming the accepted way to treat stablecoins. Disclosed stablecoin balances separate from deposits haven't disappeared; they've moved to wallets like Samsung's, to fintechs, and to bank-issued coins like SoFiUSD. Banks adding stablecoin support should start with stablecoin-as-money on the way in, and decide now how they'll send stablecoins out, because outbound payments are where Samsung and the other wallets compete.
Congress Flips. GENIUS Doesn't.
Democrats won control of both the House and the Senate on November 3, and the new Congress convenes January 3. Sen. Elizabeth Warren is expected to chair the Senate Banking Committee, and Rep. Maxine Waters is expected to return as chair of House Financial Services, though Rep. Sean Casten has challenged her for the post. Crypto in America reported that House Financial Services Democratic staff are drafting a crypto market structure bill for the next Congress that would give the SEC broader authority than the CFTC, add consumer protections such as monthly proof-of-reserves attestations and SIPC-like coverage, and set ethics rules covering crypto activity by the president and other federal officials. The CLARITY Act will need to be reintroduced in the new Congress. Meanwhile, the SEC and CFTC have sent crypto rules to the White House for review, following through on their chairs' September pledges to act without legislation. On November 12, President Trump nominated four commissioners, one Republican and one Democrat for each agency. Industry groups expect little progress on CLARITY before the new Congress convenes and see narrower items, such as taxation, stablecoin yield and illicit-finance provisions, as possible additions to must-pass legislation, though no deal has been announced. Sources [Crypto in America, The Block, Cointelegraph, crypto.news]
My Take:
Gridlock has switched sides. In 2024, inaction in Congress kept a stifling regulatory environment that froze crypto in the U.S. Today, inaction does the opposite, and that makes a deal more likely than a standoff.
Start with what the new Congress can't easily undo. GENIUS is law and in implementation: the OCC's final rule takes effect January 18, and the deadline for exchanges and wallets to drop non-compliant stablecoins arrives in July 2028, before the next presidential election. Federal regulators have approved about 15 digital-asset trust banks, four of them already open, and four insured banks with digital-asset plans, including Nubank and Revolut. And with CLARITY dead, the SEC and CFTC have said in no uncertain terms that they will fill the gap with rulemaking. Every month without legislation, more of the rulebook gets written without Congress.
That gives both sides a reason to come to the table. The industry and its backers want legislative clarity heading into 2028, since agency rules are easier to unmake than statute. Democrats want anti-corruption and consumer protection provisions, and limits on regulators legislating through rules. While it's still anyone's guess whether the president would veto a market structure bill that carries ethics provisions, Democrats have a better shot at their financial regulation goals now than they did in 2024.
For bankers, two things are back on the table. The first is yield. The banks overplayed their hand on CLARITY, which left rewards in a gray area that works well for stablecoin issuers, and yield is like water: it will find a path to the consumer. The second, and less contested, is how deposit tokens are treated in market structure legislation, which matters to every bank that has invested in them. Banks following real-world asset tokenization also stand to gain from clarity that opens products like tokenized equities trading and intraday fixed income liquidity. Bankers should push their lobbyists toward a constructive tone: arrive with asks, not objections, and use the bill to get the rules banks want while opening product doors that let them compete with larger banks.
The Float Is Shrinking. The Flow Isn't.
Circle reported on November 4 that USDC in circulation ended the third quarter at $73.9 billion, flat from a year earlier, while reserve income fell 8% to $658 million as its reserve return rate dropped to 3.54%. Activity rose over the same period: USDC on-chain volume reached $13.9 trillion, up 49%, and minting and redemptions hit a combined record of $208 billion. On the earnings call, CEO Jeremy Allaire said each USDC changed hands about 190 times in the quarter, describing circulation as the stock and transaction volume as the flow. Total stablecoin supply then fell $6.66 billion in the week to November 7, the steepest weekly drop of 2026, according to DefiLlama data. Meanwhile, Ramp's Andrew Chapello said more than 2,500 businesses on Ramp now pay vendors in stablecoins, with over 70% of that volume moving outside banking hours and most of it without holding a stablecoin balance, and Stripe opened a public preview of stablecoin payouts to wallets in more than 160 markets for U.S. businesses. Sources [Circle, Blockchain Reporter, Ramp, Stripe]
My Take:
Do you remember the stream of increasingly bullish stablecoin market cap forecasts back in 2025? In April, Standard Chartered predicted $2 trillion by the end of 2028. In June, Treasury Secretary Scott Bessent said the market could reach $3.7 trillion before the end of the decade. In September, Citi raised its bull case to $4 trillion by 2030. The results this year look a lot different: according to rwa.xyz, stablecoin market cap stands at $277 billion today, down from a peak above $300 billion in May. This is bad news for stablecoin issuers and good news for banks, and I think the optimistic growth rates are still years away. The news here shows the industry moving strongly toward stablecoins for payments, which fundamentally doesn't need a lot of stablecoin value to function. A year ago I wrote that "the market cap for a payments-focused stablecoin will be low due to high turnover," and that's what the numbers now show.
Stablecoins are moving faster, reducing the overall stock required relative to payments volume. According to Visa and data provider Allium, monthly stablecoin volume for payments reached an all-time high of $63 billion in October, up over 100% from a year prior while market cap shrunk slightly. During this time, stablecoin use grew substantially in the "stablecoin sandwich" use case. Fiat in the sender's country is converted to stablecoin and delivered to an off-ramp in the destination country, which sends fiat to the recipient and offloads the stablecoin, ultimately causing a redemption at the issuer. Since the net stablecoin balance required for a sandwich transaction is zero, in a perfectly efficient world these transactions don't drive any increase in stablecoin market cap. Market cap creation is driven by frictional inefficiencies that slow down redemption operations and issuer redemption fees that incentivize holding and using the coins in payments instead of redeeming. As on- and off-ramp liquidity deepens alongside regulatory acceptance of stablecoins and wider use of fast payment systems, these forces moderate further dampening demand for stablecoins.
Stablecoins also make a weak savings product. Lending them out on the most established DeFi protocols has paid roughly the fed funds rate, or a little less, since early last year, and anyone with a money market fund or T-bills can earn the same without smart contract or hack risk. The real demand to hold digital dollars as savings comes from people in countries with unstable currencies or thin capital markets, and that pool is small in dollar terms: by my own rough estimate, about $1.5 trillion is held in local currency across 47 countries with chronic or acute currency instability. The $14 trillion Eurodollar market proves there's plenty of demand for dollars, but not for tokenized dollars. Its holders are sophisticated institutions already well served by fiat, and for them the switching costs still outweigh what a stablecoin adds.
What this implies is that for the foreseeable future, payment stablecoin issuer economics will shift away from reserve yield and towards fee income. Competition between issuers has already pushed many to share more yield; Open USD, announced in June, passes nearly all of its reserve yield to distributors. Payments volume in a world dominated by the stablecoin sandwich requires higher mint/burn volume (more often proxied by market making than by actual mint/burn), creating the opportunity for fee income. As evidence, look at Circle's October fee schedule update, which eliminated its free fast-redemption tier and instituted a 5 bps fee on instant burns, the exact operation needed to facilitate a stablecoin sandwich (for now, Circle exempts users of its payments products, like the Circle Payments Network, from the fee).
As issuers shift towards transaction fees, the opportunity for banks improves. Issuers charging for instant conversion creates room for banks to do the same. That doesn't contradict PNC and OnePay eating the cost: free inbound conversion that lands a deposit is a customer-acquisition cost, while instant two-way conversion for payment flows is where the fee is. Banks' advantage comes from a low cost of capital, the critical ingredient in maintaining liquidity on both sides of the trade, USD fiat and USD stablecoin. Banks that offer conversion between the two stand to increase their non-interest income.
Coupon Clippings
The Bank Coin Gets a $270 Million Budget
Bloomberg reported that the 21 institutions behind Project Faro, among them Goldman Sachs, Citi, Bank of America, Wells Fargo and Deutsche Bank, have committed about $270 million to cover five years of operating costs for a joint U.S. dollar stablecoin. The issuer is expected to incorporate in the U.S., is targeting a launch in the first half of 2027, and plans tokens in other G7 currencies afterward, starting with the euro. On one hand, roughly $54 million a year is a generous budget for the narrow job of issuing a stablecoin, which can be run very efficiently, and Faro could be leaner still if member banks handle operationally intensive work like compliance. On the other hand, the banks can expect distribution to demand far more capital than $54 million a year. Think incentives for adoption, subsidies for liquidity providers and direct marketing. Fortunately, Faro's members collectively generate roughly $850 billion in annual revenue. [Full Story]
The OCC Writes the Agent Rulebook Before the Agents Arrive
On November 4, the OCC issued Bulletin 2026-54, a supervisory framework examiners can use to evaluate banks' use of what it now calls "Super Intelligence," including AI agents. The framework sets no new expectations for banks and says AI should not be treated as a separate risk category; instead, examiners score each use case on factors including consequence, reversibility and autonomy, and sort it into four illustrative tiers. The area where the framework overlaps with this newsletter is agentic payments, particularly those with stablecoins: autonomous transactions with low or no reversibility land in Tier 1, where examiners would look for the most intensive controls, and the OCC's own example is an autonomous payment agent. The OCC says such an agent sits in Tier 1 regardless of how well it performs in testing. One question is whether reversibility, whether on a stablecoin transaction or when an agent uses a credit card, would be enough to downgrade an agent to Tier 2. The framework suggests not: it counts an action as reversible only if an error can be undone before harm occurs and lists a payment sent as irreversible, so the realistic path to Tier 2 is a human overseeing the agent, not a chargeback. [Full Story]
Decentralized Providers Relying on Centralized Tech
Blockchain analytics firm Chainalysis disclosed on October 30 that an attacker using what it called "offensive agentic techniques" had reached its internal systems, and on November 4 it took its KYT transaction screening service offline after finding that names, password hashes and live tokens for all users had been exposed. KYT was fully restored on November 7, after customers allowlisted their IP addresses and rotated passwords and API keys. Over the same days, Nubank, Mercado Pago and Coincheck paused crypto transfers, each citing a problem at an outside screening or risk-monitoring provider, though none named Chainalysis. This outage highlights the vendor concentration risk in the crypto industry, reminding me of the cloud provider outages of the late 2010s, like Amazon's 2017 S3 outage, that showed how much of the internet had come to depend on a limited set of infrastructure. Services that support post-transaction monitoring can go offline periodically, but when it comes to real-time payments, outages mean payments don't move. In the age of AI hacking, banks will need to review all of their real-time vendor dependencies, decide in advance whether each screening check fails open or closed, and update their HA/DR plans assuming one of those vendors will go down. Sources [The Hacker News, PYMNTS]
Banks Link Two Countries. Stablecoins Link Them All.
On November 16, DBS and JPMorgan's Kinexys launched a service that lets institutional clients move USD tokenized deposits between the U.S. and Singapore 24x7 (except a three-hour Saturday downtime). Ant International is the first client, using it to manage cash across its entities in both markets. The link connects DBS Token Services to the Kinexys network, both permissioned blockchains, with Partior, in which both banks are founding shareholders, orchestrating settlement. Ant also announced live Hong Kong dollar transfers across multiple banks in September and October, settled in tokenized central bank money on the HKMA's EnsembleTX platform. Notice: these announcements connect exactly two countries (Ant's HKD settlement is within Hong Kong). Every additional corridor needs its own bilateral link between two banks' systems, so a company operating in many markets needs many of them. A dollar stablecoin on a public chain reaches every wallet on that chain at once, and the work shifts to the on-ramps and off-ramps at each end, which is the stablecoin sandwich described earlier in this issue. Notice also what is missing from this project: SWIFT, the interbank standard that has connected banks for decades, isn't part of it. DBS did use SWIFT's ledger for a weekend payment with Citi in September, but a major dollar corridor going live without it suggests SWIFT's position as the default connection between banks is weaker than it was. Banks building tokenized deposit networks should expect to repeat this work for every corridor, and decide now which corridors justify it and which they will leave to stablecoins. Sources [Cointelegraph, MarketScreener, Business Wire]