Start with a number that frames the opportunity, provided you read it carefully. In a 2025 YouGov survey commissioned by BVNK of 4,658 crypto-active or crypto-intending adults across fifteen countries, 77% said they would open a stablecoin wallet if their bank or fintech app offered one. That is not a representative sample of American consumers, still less of credit union members, and it proves nothing about where your members move money today. What it suggests is that among people already inclined toward this, availability binds harder than appetite.

Hold that against the sector's demographics. The average North American credit union member is often cited at 53, roughly fourteen years above the US population median. There is a clock on top: the GENIUS Act takes effect on the earlier of 18 January 2027 or 120 days after the primary federal stablecoin regulators finalise implementing rules.

Credit unions have machinery for this that banks are currently having to assemble. That is the argument, and it is comparative rather than a claim of exclusivity.

Under the GENIUS Act, a payment stablecoin like USDC, is a digital asset designed for payment or settlement that its issuer must redeem for a fixed monetary amount. A permitted issuer must maintain at least one-to-one backing in eligible liquid reserves and disclose its redemption policy and reserve composition. The token itself is not federally insured. Depending on network and product design, it can support around-the-clock transfer and programmable settlement, though neither instant settlement nor an unbroken market peg comes guaranteed with the label. Much of the anxiety in this sector comes from collapsing the instrument into the wider asset class, and pulling those apart is usually where a productive board conversation begins.

The defensive case deserves a fair hearing, and the estimates now come from serious places. They measure different things, which is where most of the confusion in this debate starts.

That Federal Reserve note makes a point which rarely survives the summary: deposits can be reduced, recycled back into banks, or restructured from retail into wholesale funding. Which of those happens matters enormously and is not settled.

Yield is the live political fight. America's Credit Unions co-signed a joint trades letter to the Senate Banking Committee pressing to close the inducement loophole, and has argued that a ban on inducements belongs in the CLARITY Act. The White House Council of Economic Advisers came out the other way with modelling on yield prohibition and bank lending, and the ABA said it had studied the wrong question. Everyone in that exchange has a book to talk.

One distinction gets flattened in almost all of this coverage, and it should shape your strategy. Retaining a member relationship and retaining insured share funding are different objectives, and a stablecoin product mostly serves the first. A credit union-branded, core-integrated wallet keeps the interface, the transaction data, and the service relationship inside your orbit. It does not automatically keep the funds as insured shares on your balance sheet, because a stablecoin is the issuer's obligation rather than a share at the distributing institution. NCUA has been explicit that digital assets held through third-party vendors, or in custody where a state charter permits it, are not shares and not NCUSIF-covered. Tokenised shares are the product that addresses funding retention, and they sit inside NCUA's May supplemental proposal.

The strategic question is therefore not offer it or lose the deposit. It is which combination of funding retention, relationship retention, and payment utility you are trying to preserve, and those three need different products.

Banks are organising, and anyone telling you otherwise has not been reading. The Clearing House has announced a bank-led on-chain money initiative, owned by 25 large financial institutions and explicitly intended to be accessible to banks of all sizes. FIS is building Project Keystone with six institutions, bank-owned and bank-administered. The community bank-owned IBAT DTX Consortium has passed 40 participants. Citi's analysts think bank tokens could eventually rival stablecoins on volume.

So exclusivity is not the credit union advantage. Speed might be, because every one of those initiatives had to be convened, capitalised and governed from a standing start. Credit unions already run the equivalent institutions and have done for decades:

A rail intended to keep liquidity inside the network sits naturally with all four charters. The advantage, if it is real, is compressed coordination and procurement time rather than a structure nobody else can build.

Work is underway on several tracks, at varying stages of maturity, and the distinctions matter:

None of that is production at scale. All of it is further along than most boards realise.

Filene put the strategic question about as well as it can be put in its read on the June consortium announcements: the argument has moved on from whether tokenised money reshapes payments to who ends up setting the standards and keeping the economics. Participant or customer, and the gap between those two compounds annually.

Most rooms assume this is a big institution game. Having had the conversation in a few of them, I think the opposite is closer to the truth.

Twenty years of technology shifts have punished small credit unions with dull consistency. Digital banking, real-time payments, data warehousing, compliance automation: each arrived carrying a fixed cost that a $3 billion institution could spread and a $150 million one could not. The sector now counts 4,250 federally insured credit unions, down from 4,411 a year earlier. My own view, and it is a view rather than something the data establishes, is that most of that consolidation is a technology cost story wearing a strategy costume.

A shared issuer changes the arithmetic, though not as completely as vendors will tell you. Licensing, reserve management, redemption obligations and issuer-level compliance can be centralised, which genuinely lowers the fixed cost of participation. Distribution is not plug and play: a participating credit union still handles member due diligence, BSA and sanctions controls, consumer disclosures, cybersecurity, private key and custody risk, third-party oversight, reconciliation and the limits of its own charter authority.

With that caveat stated plainly, the hypothesis is worth testing. A $60 million credit union could put comparable capability in front of members as a $6 billion one, because the expensive layer is shared. Small institutions ought to be the loudest voices in these consortium discussions rather than the last ones invited.

Nothing here is final. Both stablecoin rulemakings remain proposals with closed comment periods, and the agency has not published final implementing rules. The sequence:

Two proposed constraints deserve attention. First, the GENIUS framework contemplates issuance by a credit union subsidiary rather than by the insured credit union directly. Second, NCUA's February proposal would limit federally insured credit union investment to NCUA-licensed permitted payment stablecoin issuers.

That second point is not yet binding and could change before a final rule. If it survives substantially as drafted, it could encourage capital formation around credit union owned issuers, which some in the sector will read as a favourable position for a cooperative consortium. Treat that as a contingent possibility rather than a fact about the market, because the proposal does not establish it.

The operational conclusion is less exciting than the regulatory debate. Almost no credit union should be standing up a licensed issuer subsidiary. Nearly every one should be working out what participation on somebody else's would require.

This is the ladder I put on the whiteboard, ordered by what a mid-sized institution can plausibly get past a board this year. Note that the first two rows answer different questions.

Tier 1 is where I would spend the effort, and it stays under-discussed because most commentary on this subject is written for banks whose customers do not send money home. Inclusiv's Juntos Avanzamos network now covers more than 167 credit unions across 34 states, serving Hispanic and immigrant members with bilingual staff and alternative ID acceptance. The World Bank puts the global average cost of sending $200 at 6.36% as of its most recent quarterly reporting, and the Federal Reserve has published its own assessment of what payment stablecoins do to cross-border cost and settlement time.

Be careful with the savings arithmetic, because this is where the sector is being sold a story. Stablecoin rails may lower cost and settlement time, but the all-in price a member pays still absorbs FX spreads, on- and off-ramp charges, last-mile delivery, compliance, and whatever is peculiar to the corridor. Anyone quoting a headline sub-1% figure is quoting a network fee, not a member price. The opportunity is real, and it is corridor-specific, which makes the diligence corridor-specific too.

Beneath all the tiers sits the vendor question: what is the cheapest work available to you this quarter? Three things to ask your core and your digital banking provider, in writing:

Fiserv's FIUSD uses Finxact as its underlying ledger, and the Bank of North Dakota's Roughrider Coin beta shows a state-owned bank testing the model. BND is clear that the public cannot access it and that development viability, total investment and return are still to be determined, so treat it as a proof of seriousness rather than a proof of production. A core that cannot answer question one has answered question one.

The numbers are strong enough that overstating them would be a waste, and each needs its label:

None of that proves stablecoin demand among your members, and I would not present it as though it does.

Then the honest part, because overselling this in a boardroom costs you the room. A stablecoin wallet is not a growth strategy, and nobody has ever joined a credit union because it supports USDC. What it buys is a place in the consideration set, which happens to be where credit unions are losing younger members. The problem is rarely that a 26 year old weighs you up and picks a neobank. It is that you never entered the comparison. Pitch this as a credibility product with an acquisition side effect, and the business case holds in front of a board.

As of 11 August 2026, the CLARITY Act has passed the House 294 to 134, cleared Senate Banking and sits on the Senate calendar without a floor vote before the August recess, so passage may slide into 2027.

The provision to follow is narrower than the coverage suggests. GENIUS already prohibits permitted issuers from paying interest or yield directly. The unresolved question is whether exchanges, affiliates or other third parties may pay rewards tied to stablecoin balances. That choice could materially affect adoption and deposit competition, though it is one variable among several, alongside where demand actually comes from, how reserves are allocated and recycled, foreign demand, and how banks and credit unions respond. America's Credit Unions and its league system have filed detailed feedback on exactly this point.

A short note on AI, since the two keep getting bundled. They are separate compliance regimes and should be run separately. The EU's AI Omnibus entered into force on 27 July 2026, pushing the high-risk obligations for Annex III systems, creditworthiness assessment among them, out to 2 December 2027. In the US, CFPB Circulars 2022-03 and 2023-03 were withdrawn in May 2025, though ECOA and Regulation B still require specific and accurate reasons for adverse action regardless of model complexity. The OCC has issued updated, tailored model risk guidance, most relevant above $30 billion, with generative and agentic AI left outside its scope pending further work, and it supervises banks rather than credit unions. Worth tracking. Not a stablecoin compliance issue.

The comfortable version of this argument says community banks are stuck and credit unions have a clear run. It is not true, and the IBAT consortium is the proof: 40 community banks organising collectively are exactly the institutions competing with credit unions for the same members.

So the advantage is not exclusivity, and any pitch built on that will not survive its first informed question. It is that corporates, CUSOs, leagues and pooled investment networks already exist, already have governance, and already carry the trust required to move money between competing institutions. Those cut coordination and procurement time, which is the scarce resource here. Ron Shevlin's advice to community banks applies equally on this side of the fence: find out what is already happening inside your member base, lean on your core, and choose the use case that pays for itself.

The head start, such as it is, is measured in quarters and it is shrinking.

Credit unions have spent a decade being told they lacked the scale to compete on technology, and for most of that decade the warning was fair. This time the movement starts with the institutions it needs already built and already trusted. That is a real head start, and it is not a permanent one, because the banks have now worked out that they need the same thing and have started building it.

The window is open. It closes when somebody else sets the default.