Stablecoin Yield Ban: 10% Odds, and a Definition That Won't Survive Contact With Code

A stablecoin yield ban won't pass. TD Cowen's Jaret Seiberg put 10% odds on Congress enacting one over the next two years in a Monday policy note, and that number is about right. Here's the part everyone's skipping: the ban doesn't need to pass to do its job.

Banks have already won the thing they actually wanted, which is ambiguity. Every month the OCC handles third-party rewards on a case-by-case basis is a month a compliance officer at a mid-size exchange has to justify a product to a supervisor holding no rulebook. That's not a court fight. It's a slow leak, and leaks are cheaper than lobbying.

The numbers behind the 10%

Start with the Senate, because that's where this dies. The Clarity Act failed a cloture vote 49-50 on Sept. 15, and Seiberg has treated it as dead for this Congress since the Democratic sweep. Key crypto advocates retired or lost their seats. Skeptics got elevated instead. Elizabeth Warren is likely to chair Banking. Sherrod Brown took Ohio back after roughly $64 million in crypto-backed spending failed to move him. You don't buy a Senate seat for $64 million and then watch the guy win anyway without learning something about your own political tap into.

Now the statute everyone keeps misreading. The GENIUS Act, signed July 18, 2025 as Public Law 119-27, bars permitted and foreign payment stablecoin issuers from paying "any form of interest or yield" solely for holding, using, or retaining a payment stablecoin. Read that again. It covers issuers. It says nothing about the exchange sitting between the issuer and the customer. That gap is what the American Bankers Association calls a loophole and what I'd call the entire business model of consumer crypto.

Seiberg's argument for why a narrow ban could still clear 60 votes is procedural, and it's sharper than it looks. It may be easier to find 60 votes on yield alone than on a 400-page market structure bill that Democrats want loaded with ethics and investor-protection language the President would veto. That's the narrow path. Ten percent is what a narrow path looks like when you price it honestly.

The definition nobody can enforce

Here's where I part company with the smart money. Everyone's arguing about vote counts. The real fight is over a distinction that can't be enforced.

Sen. Angela Alsobrooks, a lead Democrat on crypto policy, backed paying rewards forusingstablecoins rather than for passively holding them. That distinction shaped the Tillis-Alsobrooks compromise text during Clarity talks. It sounds reasonable in a committee room. It's a legal fiction at the contract level.

If you hold 50,000 USDC and the platform pays you 4% a year because you route one swap a month through it, that's yield with a chore attached. Any competent Solidity dev can wrap the chore. You don't need a novel primitive. You need a calldata check and a multiplier keyed to average balance. The economics are identical. Only the labeling changes.

So ask the question the bill text can't answer. If a platform can pay you for swiping a card but not for leaving a dollar on deposit, what exactly is Congress protecting? The consumer, or the deposit franchise?

The steelman for the banks

I'll grant the ICBA its argument, because it's better than most people in this industry admit. Community banks fund local lending out of deposits. If yield-bearing payment stablecoins pull even a few percent of a $19 trillion deposit base into tokenized money market exposure, the lending compression lands hardest on small banks in counties with three branches and no alternative. That's a real transmission channel, not a talking point. And small banks have a genuine track record of getting Congress to move. They get carve-outs in nearly every financial bill that passes.

There's also a consumer-protection case that doesn't come from bad faith. Yield has to come from somewhere. When a platform pays 5% on a dollar peg, that return is funded by lending, basis trades, or token emissions, and at least two of those three can blow up. Banks will argue, correctly, that nobody is stress-testing the middleman.

But here's the thing. Seiberg flagged the optics problem himself, and it's fatal. Congress would be voting to stop voters from earning a return on cash at a moment when they can get one from a savings account or a money market fund. Democratic leadership has no appetite for a floor vote that gets framed as taking money out of retail accounts. That's why the yield-only bill doesn't unify the caucus, and why 10% stays 10%.

Where this actually lands

My verdict: the ban fails, and the ambiguity wins anyway. Watch the OCC, not the Senate.

Final GENIUS rules take effect in January 2027 and leave third-party rewards to case-by-case review after banks had pushed the agency to close the question outright. That's the pressure point. A supervisory posture isn't a statute, so it can't be lobbied once and forgotten. It gets relitigated in every exam cycle, in every state, in front of every charter application. House Democrats are already sketching a market structure approach with a bigger SEC role, which means the next Congress gets a fresh set of definitions to argue about.

What does that mean for the people actually building? Stop waiting for a safe harbor that isn't coming. The platforms that survive the next two years will be the ones whose yield products don't depend on a statutory carve-out to exist, and whose risk models can survive a supervisor reading them out loud. If the AI can hold a wallet, who writes the risk model? Same question for a rewards program paying 4% on a dollar.

Seiberg's one-in-ten is the correct call. But it's the wrong scoreboard. The bank lobby lost the vote and won the weather.