Blog/Policy

The CLARITY Act vote: what happened, and what the bill actually does

The vote failed 49–50. The agencies have spent a year building an alternative — but a rule is not a law, and the difference decides where capital commits, where builders incorporate, and whether the right to hold your own keys is durable.

STSwop TeamSep 16, 2026Updated Sep 16, 202611 min read

The CLARITY Act is the bill meant to give crypto exchanges and token issuers a federal rulebook, splitting oversight between the Securities and Exchange Commission and the Commodity Futures Trading Commission instead of leaving classification questions to case-by-case enforcement lawsuits. On September 15, 2026, a Senate cloture vote to take it up failed 49–50 — eleven votes short of the 60 needed, and one short of even a simple majority — over three unresolved disputes: ethics restrictions on public officials' crypto income, developer liability for non-custodial software, and whether exchanges can pay yield on stablecoin balances. The bill is not dead, but no new vote is scheduled.

The more useful question is what happens without a statute — because the answer is not nothing. The SEC and the CFTC have spent the past year building a substitute out of rulemaking and interpretive guidance, and much of it is already operating. That substitute is real and it is useful. It is also structurally weaker than a law in precisely the way that decides whether capital commits and whether builders incorporate here.

What the bill would actually do

The Senate has been negotiating its own companion to H.R. 3633, the Digital Asset Market Clarity Act, which the House passed 294-134 in July 2025. The core idea is a jurisdictional line that does not currently exist in statute: the bill defines a "digital commodity" category for tokens on networks decentralized enough that no single party controls them, hands the CFTC primary authority over spot trading in that category, and leaves the SEC in charge of securities offerings and primary-market fundraising — with a new, limited exemption from full SEC registration for issuers that meet specific conditions.

Right now, whether a given token is a security or a commodity gets fought out case by case, mostly through SEC enforcement actions rather than a rule anyone can read in advance. Exchanges, issuers, and app developers have all cited that uncertainty as the reason they build outside the U.S. or avoid listing certain assets. A statute that draws the SEC/CFTC line up front is the thing the bill is actually trying to sell — regardless of which side of the political fights below you're on.

The September 15 vote

Cloture requires 60 votes to end debate and move a bill toward a floor vote. The count came in at 49-50, so the motion failed and the bill cannot currently proceed. Notably, several Democrats who had spent months working on the bill's text — including Sens. Kirsten Gillibrand, Mark Warner, Cory Booker, Raphael Warnock, Ruben Gallego, Angela Alsobrooks, and Catherine Cortez Masto — voted no on this particular motion. That is a signal about the state of the text, not about the concept: Republican sponsors had released a revised draft the Sunday before the vote, adding ethics restrictions meant to address Democratic concerns, but the changes did not go far enough to close the gap.

The three fights holding it up

Reporting on the negotiations converges on the same three sticking points, and none of them is really about whether crypto should have a market-structure law — they're about specific text inside this one.

  • Ethics rules on officials' crypto income. Democrats want stronger restrictions on sitting officials' personal financial ties to crypto ventures, a concern sharpened by public reporting on the President's own crypto-related income. Republicans added some ethics language in the Sunday revision; Democratic holdouts said it wasn't enough. Reporting put the coalition's collapse squarely here rather than on the market-structure framework itself: no voting Democrat supported the motion, and four Republicans opposed it.
  • Section 604 — developer liability. This section incorporates language from the Blockchain Regulatory Certainty Act: a developer who writes non-custodial software, and who cannot control or execute a user's transaction, would not be classified as a money-transmitting business solely for building or publishing that software. A companion amendment preserves criminal liability for anyone who "knowingly" facilitates illicit transactions. The fight is over exactly where "publishing code" ends and "operating a service" begins — and how much comfort that line actually gives a developer before the fact rather than after an indictment.
  • Stablecoin yield. As drafted, the bill would let exchanges pay yield on customers' stablecoin balances. Opponents — banking-industry lobbyists and some Democrats — argue that functions like an unlicensed bank deposit product, since it pays a return on a balance without the capital and disclosure rules banks operate under. It's also commercially loaded: stablecoin reward programs are a real revenue line for large exchanges today, so this isn't an abstract argument for the companies on either side of it.

What happens next

A failed cloture vote is a procedural stop, not a repeal. Senate leadership can schedule another cloture vote whenever it believes the votes are there, and that usually means the underlying text has to move first. As of publication, no new vote is on the calendar, and each of the three disputes above is still open. Because the House and Senate would need to agree on identical text before anything reaches the President, a Senate breakthrough would still require reconciling with the House-passed H.R. 3633 rather than simply adopting whatever the Senate eventually passes.

What the agencies are building instead

While the bill stalled, both agencies moved. Treating the failed vote as “no rules” misreads the situation badly — the more accurate description is that rules are arriving, just from regulators rather than from Congress.

The SEC. Chairman Paul Atkins’ Project Crypto initiative has been working toward a token taxonomy that sorts assets into categories rather than litigating them one at a time — digital commodities, digital collectibles, digital tools, and tokenized securities. On August 18, 2026 the Commission went further and proposed Regulation Crypto Assets, a tailored offering regime for investment contracts involving crypto assets, drawing on Commissioner Hester Peirce’s earlier token safe-harbor work. It contemplates a time-limited startup exemption — on the order of a capped raise while a network works toward maturity. The SEC said plainly that part of the point is to reduce the incentive for issuers to organise offshore.

The CFTC. Its Crypto Sprint has been more concrete still. Listed spot crypto products began trading on federally regulated U.S. markets for the first time in December 2025, with designated contract markets permitted to list spot products — including leveraged contracts for retail — under the same oversight applied to futures. And in a joint release in March 2026, the agencies published a taxonomy naming sixteen assets as digital commodities under primary CFTC oversight, Bitcoin, Ether and Solana among them.

That is a meaningful amount of clarity, and it arrived faster than the legislative process did. Anyone claiming the United States has no crypto policy is describing 2022.

Why a rule is not a law

Here is the part that the vote count actually decides. A rule and a statute can say identical things and still be worth different amounts to the person deciding where to put money or where to incorporate, because they differ in how hard they are to undo.

An agency rule is the product of the agency that wrote it. A future commission can propose a new rule that supersedes it. A court can vacate it. Congress can pass a resolution of disapproval. A proposal — and Regulation Crypto Assets is still a proposal — can simply never be finalised. None of that requires anyone to act in bad faith; it is the ordinary machinery of administrative law working as designed. A statute is not immune to change either, but changing one requires both chambers and a signature, which is a far higher bar than a change of chair.

That difference shows up directly in two places.

  • Capital underwrites duration. A fund committing to a ten-year vehicle, or a bank deciding whether to custody, is pricing a regulatory regime across a horizon longer than any single administration. Guidance that could be withdrawn by a successor gets discounted — not to zero, but discounted. Meanwhile the funding environment has tightened on its own: new venture fund formation hit its lowest quarterly total since 2020 in Q1 2026, with roughly $1.1 billion committed across eight new funds, and the top thirty firms have been absorbing the large majority of what is raised. When capital is scarce and concentrated, the marginal allocator has every reason to wait for the version that survives an election.
  • Builders incorporate where the rules keep. Choosing a jurisdiction is a multi-year, expensive, hard-to-reverse decision, and founders make it against the rules they expect to exist at exit, not the rules today. That is why offshore structuring persisted through a friendlier turn in Washington, and why the SEC named onshore issuance as an explicit goal of its own proposal. Compliance cost compounds the same way: licensing and compliance can run into the millions annually at scale, which is survivable against a stable rulebook and brutal against one that might be rewritten mid-build.

So the honest reading of September 15 is not that crypto lost its rules. It is that crypto kept its rules on loan. Agency action can deliver most of the substance of a market-structure regime; what it cannot deliver is the commitment that the regime will still be there in four years — and duration is the specific thing both capital and founders are buying.

To be clear about the disagreementThe three fights that sank this vote are real, and reasonable people land in different places on all of them — particularly the ethics provisions, which is where the coalition actually broke. The argument here is narrower and, we think, separable: whatever the right substantive answers are, settling them in statute is worth more than settling them in guidance. That case does not depend on any particular bill passing.

The thing worth writing into law: self-custody

Strip the jurisdictional argument back and one property underneath all of it deserves protecting in statute rather than in guidance, because it is the property that makes the system open at all: the ability to hold your own keys.

A self-custodial wallet is an account nobody grants you. No branch, no credit file, no minimum balance, no proof of address, no correspondent bank deciding your country is not worth the compliance overhead. You install software and you have an account. That reads like a technical detail until you look at who is currently outside the system.

The World Bank’s Global Findex 2025 counts 1.3 billion adults with no account at a bank, financial institution, or mobile-money provider — more than half of them, roughly 650 million, concentrated in eight countries: Bangladesh, China, Egypt, India, Indonesia, Mexico, Nigeria and Pakistan. The same survey finds 86% of adults worldwide own a mobile phone. The gap between those two numbers is not a technology problem. It is a permission problem: the device is already in the hand, and what is missing is an institution willing to open the account.

The cost of that gap is measurable. Remittance corridors into Sub-Saharan Africa still average over 6% in fees, on flows that are the entire household budget for many families. And the alternative is being adopted rather than merely proposed — roughly $33 trillion in stablecoin transfers settled on public blockchains in 2025, up 72% year over year, while 71% of Latin American firms now report using stablecoins for cross-border settlement, the highest rate anywhere. People are not waiting for permission to join a better rail.

Self-custody is what keeps that participation genuinely open rather than handing it to a second gatekeeper. A custodial app can onboard someone in Lagos or Dhaka, and can also freeze them, reprice them, or exit their market on a quarter’s notice — which reproduces the original problem behind a nicer interface. Holding the key is the difference between being admitted to a financial system and being a participant in one.

Where the legislation actually bitesThis is why Section 604 is not a side issue. Whether publishing non-custodial software makes you a money transmitter decides whether the tools that let people hold their own keys can be built in the United States at all. Settled in guidance, it can be resettled. Settled in statute, builders can commit. Leaving it unresolved protects no one — it exports the work, and opts the country out of the rail everyone else is already using.

None of which is a claim that self-custody is easy or finished. Key loss is a real failure mode, and recovery design is where most wallets are still weak — we are writing about that side of it separately later this week. The argument here is narrower: an open financial system is one where the account requires nobody’s approval, and that property only survives if the right to build for it is written somewhere harder to change than a rule.

What this means for Swop users and builders

None of the three fights above change the thing that actually determines whether your assets are safe today: who holds the signing key. Swop is fully self-custodial — keys are generated and held on your device; Swop never holds them. That doesn't move regardless of how the SEC/CFTC jurisdictional question gets settled, because the bill is about exchange and issuer market structure, not wallet custody.

The part of this fight that matters most directly to builders is Section 604. Until something like it becomes law, whether writing and publishing a smart contract, a DEX front end, or a non-custodial wallet interface exposes a developer to money-transmission liability keeps getting decided case by case rather than by a bright-line federal rule — which is exactly the ambiguity the digital-commodity classification fight is also trying to resolve for exchanges and issuers. Swop runs on Solana, Ethereum, Base, and Polygon. Transactions on Swop are gas-sponsored — you don't need to hold SOL or ETH to transact — and none of that changes with this bill either way. On the access question above, two verified details matter more than they sound: that gas sponsorship, and the fact that losing your phone doesn't mean losing your funds, since you can log in with your email on any phone and your Swop wallet comes back with it. Needing to acquire a native token before your first transaction, and needing to protect a seed phrase perfectly forever, are the two places self-custody most often fails the people it is meant to include.

FAQ

If the bill failed, does the U.S. have no crypto rules?

No — that describes 2022, not now. The SEC has proposed Regulation Crypto Assets, a tailored offering regime for crypto investment contracts, and has been working toward a token taxonomy under its Project Crypto initiative. The CFTC's Crypto Sprint has gone further in practice: listed spot crypto products began trading on federally regulated U.S. markets in December 2025, and a joint March 2026 release named sixteen assets as digital commodities under primary CFTC oversight.

If the agencies are already doing it, why does legislation matter?

Because a rule and a statute differ in how easily they are undone. A future commission can supersede a rule, a court can vacate it, Congress can disapprove it, and a proposal can simply never be finalised — all ordinary administrative process, no bad faith required. Changing a statute takes both chambers and a signature. Capital committing to ten-year horizons and founders choosing a jurisdiction are both buying duration, and guidance that a successor could withdraw gets discounted accordingly.

Why does self-custody matter to the legislation debate?

Because self-custody is the property that makes the system open: a wallet is an account nobody has to grant you, with no branch, credit file, or proof of address in the way. The World Bank's Global Findex 2025 counts 1.3 billion adults with no financial account while 86% of adults own a mobile phone — a permission gap, not a technology gap. What legislation decides is whether the software that lets people hold their own keys can be built domestically: Section 604 governs whether publishing non-custodial code makes a developer a money transmitter. In guidance that can be resettled; in statute, builders can commit.

What is the CLARITY Act?

Federal legislation that would split crypto oversight between the SEC and CFTC: a new "digital commodity" category for sufficiently decentralized tokens goes to the CFTC for spot trading, while the SEC keeps authority over securities offerings, with a limited registration exemption for qualifying issuers.

Did the CLARITY Act pass the Senate?

No. A cloture vote to advance it failed 49–50 on September 15, 2026 — eleven short of the 60 required, and one short of a simple majority. The House passed its own version, H.R. 3633, 294-134 in July 2025. The bill isn't dead, but no new Senate vote is scheduled.

What are the three issues blocking the CLARITY Act?

Ethics restrictions on officials' personal crypto holdings and income; developer liability for non-custodial software under Section 604; and a stablecoin-yield provision that critics say functions like an unlicensed bank deposit product.

Does the CLARITY Act affect self-custody wallets like Swop?

It targets exchange and issuer market structure, not wallet custody. Swop is fully self-custodial — keys are generated and held on your device; Swop never holds them — regardless of the outcome. Section 604's developer-liability question is the part most relevant to wallet and smart-contract builders.

When will the Senate vote on the CLARITY Act again?

No new vote is scheduled. Leadership can call another cloture vote once the ethics, developer-liability, and stablecoin-yield disputes narrow enough to find 60 votes.

ST

Written by the Swop product team. Editorial rules: a direct answer up front, no invented statistics, dates on everything, and links to primary sources.

One email, every other Tuesday.

Releases, market notes, and what we got wrong. No growth-hacking, no referral loops.

Subscribed — see you Tuesday.
Swop App

Reading about it is the slow way.