Two men are looking at the same absence of a crisis and drawing opposite conclusions.
Rob Nichols, president of the American Bankers Association, says the fact that no deposit flight has followed the GENIUS Act is "irrelevant" — because the rules aren't final yet. Sen. Kevin Cramer, watching the same banks ask Congress to reopen a settled provision, says the tell is elsewhere: "If we're waiting for a moment when the banks are going to say, 'OK, now it's good enough,' that moment's not happening. No is their default."
That is the whole fight in miniature. One side dismisses the only data that exists; the other, per the Blockchain Association's framing, suspects the objection was never about data at all.
One side dismisses the only data that exists; the other suspects the objection was never about data at all.
Last week the thread here was the SEC and charter track — who gets to issue, and under what supervisory roof. This week the picture sharpened on two fronts at once: the OCC put a date on its rulebook, and the bank-versus-issuer dispute stopped being about framing and started being about specific statutory verbs.
Start with the two clocks, because they are colliding.
Comptroller of the Currency Jonathan Gould told the Wyoming Blockchain Symposium the OCC "will have a final rule out by November," racing to beat a Jan. 18 statutory deadline before the law takes effect in January 2027. The agency already blew past an earlier July target.
The document in question runs 376 pages, released in February, open for comment through May. It covers the full life cycle of a payment stablecoin — reserves, redemption at par, liquidity, audits, custody, wind-downs. Gould expects to begin processing issuer applications in 2027, and reported an eightfold jump in digital-asset chartering activity versus the prior administration.
So the regulatory machinery is committing to hard dates. And roughly four weeks before a September 15 vote, the bank lobby chose that moment to reopen a yield provision its own lobbyists helped negotiate months ago.
The reopening is dressed as a technical tidy-up. Nichols says banks want to "strengthen" the bill, not kill it. The specific asks: swap in the phrase "substantially similar to interest," and strike the word "solely."
Neither edit is modest, and the framing is the sleight of hand.
"Substantially similar to interest" is an elastic legal standard. It is broad enough to catch conduct nobody legislated against — including, by its own logic, banks' cash-back and loyalty-tier programs, which per the Blockchain Association could get caught in the standard.
The second edit is quieter and more consequential. "Solely" is the word that confines GENIUS's yield ban to issuers. Strike it, and the prohibition stops being about who mints the token and starts reaching everyone downstream — exchanges, distributors, platforms — sweeping in conduct Congress deliberately left alone.
That is the move: not a cap on issuers paying yield, which is already in the bill, but an expansion of the ban's blast radius to the entire distribution chain. "Strengthen" is doing a lot of work for two words that widen a prohibition.
Now the evidence, because this is where the argument fails its own test.
Nichols's core claim is deposit flight — that yield-paying stablecoins will drain money out of banks. Asked why the post-GENIUS deposit data shows no such thing, he answers that the data is irrelevant until the rules are final.
But it is the only real data anyone has, and it points the other way. In every quarter since GENIUS became law for which figures exist, domestic deposits have grown — more than $800 billion across three quarters.
Scale compounds the point. U.S. bank deposits sit near $19 trillion. The global stablecoin market is a fraction of that even after years of growth, and a CRA International analysis of 2019–2025 data found no statistically significant relationship between stablecoin growth and community bank deposit outflows — per the Blockchain Association, worst case under 7%, realistically under 1%.
Then there is the number that quietly detonates the prudential framing.
Money market funds hold $7.93 trillion as of this month. They pay real, market-leading returns. They compete directly with bank deposits for exactly the yield-seeking dollars banks claim to be defending — and no one has asked Congress to cap them.
That silence is the argument. If the concern were genuinely about deposits fleeing to higher-yielding cash alternatives, the $7.93 trillion MMF complex is the obvious, decades-old threat. It draws zero objection.
A yield cap demanded against a fraction-of-a-percent competitor while a multi-trillion-dollar one operates unbothered is not a stability safeguard. It is a moat. The distinction between prudential and competitive isn't rhetorical here; it's the difference between what the banks say the edit does and what the numbers say it does.
The guardrails Nichols says he wants, meanwhile, are already in the statute: SEC and CFTC jurisdiction, mandatory reserves, custody rules, disclosure requirements. The proposed edits don't add protection. They subtract something that, per the Blockchain Association, roughly 67 million Americans can already do.
But statutory verbs are only half the battlefield. The other half runs through accounting, and it is where the real competitive question gets decided.
On Aug. 19, the Financial Accounting Standards Board proposed conditions for when a stablecoin can be classified as a cash equivalent under U.S. GAAP. The definition itself stays unchanged; FASB is adding illustrative examples to resolve inconsistent treatment.
The test has three parts: an on-demand contractual redemption right, a direct redemption right with the issuer for a known cash amount, and at least one-to-one segregated reserves in short-term, highly liquid assets.
The crucial line: secondary-market liquidity alone will not qualify. A holder who can only sell a token on an exchange, without a direct issuer redemption right, fails the test. Reserves mixed with crypto or gold fail it too, on valuation risk.
This is where the interest fight actually lands for institutions. A stablecoin that clears the FASB bar becomes a cash equivalent on a corporate balance sheet — a treasury instrument that sits alongside Treasury bills and money market funds, which the proposal would require companies to disclose as distinct components of cash equivalents.
A stablecoin that fails it is just a crypto asset a company happens to hold. The difference determines whether stablecoins compete as treasury instruments at all — regardless of what the yield statute says.
So the question of whether dollars on-chain can behave like interest-bearing cash is being answered in two places simultaneously. One is the word "solely" in a bill facing a September 15 deadline. The other is a redemption-rights test with a comment period open until Nov. 19 — the same month Gould promised his final rule.
The deadlines are stacking into a single quarter, and the substance is splitting across two rulebooks that never reference each other.
Watch the verbs and watch the accounting. The bank lobby is fighting over statutory language because it can be reopened four weeks before a vote. But whether interest-bearing stablecoins ever function as competitive treasury dollars will be settled as much by GAAP and redemption rights as by which words survive markup.
The evidentiary asymmetry is the thing to hold onto. The side invoking financial stability is dismissing the only stability data that exists, while ignoring the $7.93 trillion competitor that would matter most if the concern were real. That is not how a prudential case behaves. It is how a moat gets built.