Scott Shields 祝维沙
The U.S. Digital Asset Market Clarity Act (CLARITY Act) has recently produced a phenomenon worth studying.
On July 17, 2025, the House of Representatives passed the bill by 294 votes to 134, with 78 Democrats voting in favor. After more than a year of negotiations in the Senate, Senators Cynthia Lummis, John Boozman, and Tim Scott released a final Senate draft on September 14, 2026, stating that it reflected 126 substantive changes requested by Democrats. The next day, the Senate voted on cloture on the motion to proceed to H.R. 3633. The motion failed, 49-50, short of the three-fifths threshold. Strictly speaking, this was not a final vote on the bill itself; the bill failed to clear the procedural threshold required to move to consideration.[5][6][7]
There were, of course, direct political reasons for the procedural failure: the 60-vote threshold, disputes over public officials’ crypto conflicts of interest, stablecoins and bank deposits, and securities-law boundaries were all part of the negotiations.[10][11] But a closer look shows that the problem was not simply whether the parties could compromise.
Many of the questions raised by U.S. legislators, regulators, banks, the crypto industry, consumer groups, and law-enforcement agencies were technically serious. Different sides kept identifying loopholes, proposing counterexamples, and asking about institutional consequences. That is a strength of the U.S. legislative process: unresolved problems are forced into the open.
The question this article asks is not, “Why did this particular vote fall eleven votes short?” It is: after prolonged negotiations and 126 substantive revisions, why did technical and institutional boundary problems—apart from the ethics dispute—continue to reappear?
I. Six Problems, Five of Them Sharing a Common Root
The major disputes surrounding the CLARITY Act can be organized into six broad categories.
The first concerns conflicts of interest involving public officials and their families in crypto assets. This is fundamentally a political-ethics problem, and the principal tools remain disclosure, recusal, trading restrictions, enforcement, and other public-ethics rules. Whether sufficiently strict rules could be agreed upon was also a real political constraint on the September procedural vote.
The remaining five are the ones most directly relevant to the regulatory framework examined here.
Second: should a digital asset be treated as a security, a commodity, or some other category?
Third: which regulator should have jurisdiction—the SEC or the CFTC?
Fourth: when is a blockchain or protocol genuinely decentralized, and who actually controls it?
Fifth: where should responsibility fall among DeFi protocols, software developers, protocol operators, and anti-money-laundering obligations?
Sixth: how should stablecoins be treated in relation to yield, bank deposits, payments, and credit?
These questions look different, but the five regulatory questions share a common feature:
They begin by asking, “What is this thing?”
Traditional financial regulation usually begins with classification. Is it a security or a commodity? A bank or a payments company? An exchange or a broker? A deposit or an investment product?
Once the category is chosen, the system then determines the regulator, the applicable law, the licensing regime, and the allocation of responsibility. That approach has a historical logic in traditional finance.
The difficulty is that crypto finance has changed the underlying conditions.
The same token may be part of an investment contract when capital is raised and later circulate simply as a transferable digital asset. A protocol may have no conventional company, yet someone may hold an administrator key, upgrade authority, control of the front end, or governance power. A software developer may merely write code—or may in practice control user assets and transaction rules. A stablecoin may function as a payment instrument while also competing with bank deposits; if passive yield is added, it may take on deposit-like economic characteristics.
Each new definition therefore creates a boundary. Each exception requires an anti-evasion rule. Each safe harbor requires another definition of who qualifies for it. The statute grows longer, and the boundary problems multiply.
This suggests that the problem may not be only a shortage of rules. The regulatory starting point itself may also need adjustment.
II. Do Not Begin with “What Is It?” Begin with “What Happened?”
A different starting point changes the structure of the problem. Instead of first asking, “What exactly is this token?” ask six more basic questions:
Who is the actor? Who authorized the action? What transaction occurred? Was delivery completed? Who bears responsibility? Do the accounts match the actual assets?
These are the six basic facts developed in Verifiable Thought and The Revolution from Double-Entry Bookkeeping to Verifiable Finance: actor, authorization, transaction, delivery, responsibility, and accounting.[1][2]
The core principles of Verifiable Thought are equally simple:
Key facts should be verifiable;
key processes should be replayable;
key judgments should disclose their grounds so they can be reviewed.
Verifiable Thought does not seek to abolish legal classification, nor can it replace political compromise, legislative choice, or judicial adjudication. It changes the sequence: legal judgments should rest first on a clearer, more continuous, and reviewable factual structure.
Establish the facts before assigning the legal relationship; identify control and responsibility before deciding classification, regulatory jurisdiction, and liability rules.
Take DeFi. There is no need to begin with a philosophical debate over whether a system is “decentralized enough.”
Start by verifying who can modify the protocol, who holds upgrade authority, who can freeze assets, who controls the front end, who receives the economic benefits, and who bears responsibility when losses occur.
If there truly is no central actor, regulation should not invent one merely for convenience. If actual control exists, the system should not be treated as uncontrolled simply because it calls itself DeFi. Many abstract classification disputes can thereby be converted into factual questions that can be tested.
III. Stablecoins Show Most Clearly Why the Starting Point Matters
Section 404 of the CLARITY Act would prohibit covered digital-asset service providers and their affiliates from paying U.S. customers passive, deposit-like interest or yield merely for holding payment-stablecoin balances, while allowing bona fide activity- or transaction-based rewards.[8]
Why is this issue so difficult? Because it reaches the core contradiction in the present stablecoin structure. If a stablecoin is a payment instrument, it does not need to become a new form of bank deposit. But if a stablecoin earns yield simply by “sitting there,” it begins to compete directly with commercial-bank deposits. If deposits migrate from banks into stablecoins at scale, the issue is no longer only the development of the crypto industry; it can affect bank lending capacity and the structure of the monetary system.
That is why the issuance structure itself deserves reconsideration. The stablecoin model proposed here is the author’s alternative design for this structural conflict; it is not a logical consequence of the September 15 procedural vote.
In How Stablecoin Issuance Can Rationally Adapt to Section 404 of the CLARITY Act, we identified at least three structural features of the current mainstream model that deserve redesign.[3]
First, stablecoin names and issuers are excessively fragmented.
There is only one U.S. dollar, yet the market can contain USDT, USDC, and a growing number of institution-branded “dollar stablecoins.” A monetary unit that should have a unified identity is turned back into a set of institutional brands.
Second, the ultimate guarantee and responsibility structure is not sufficiently clear.
Reserve assets ultimately remain inside banks, Treasury securities, and the traditional financial system, while the circulating on-chain instrument is a stablecoin created by a private issuer. In a severe stress event, the ultimate chain of responsibility is not self-evident.
Third, this structure can create new infrastructure concentration.
Large stablecoin issuers, exchanges, and custodians may simultaneously control issuance, redemption, and liquidity gateways.
The question, therefore, is not merely how to add more rules to the existing structure. It is whether the stablecoin structure itself can be redesigned.
One simpler structure we have proposed is this: keep the money inside the banking system and let the stablecoin serve only as the payment layer.
Commercial banks would continue to perform deposit, reserve, clearing, and core credit functions. The stablecoin would become a digital payment representation of dollars that actually remain within the banking system. Qualified banks could issue the same stablecoin under common rules rather than each creating a separate branded coin. The user would see a unified dollar stablecoin, while the back end would show which bank issued each unit, which deposit and reserve assets support it, and which institution bears the redemption obligation.
This is the model developed in Transparent Stablecoins: How an Open, Verifiable Issuance Network Can Be Built:[4]
One monetary identity, multiple bank issuers;
one network, distributed responsibility;
one verification framework, no single point of credit.
Yield can then be separated from payments. A payment stablecoin need not pay interest merely for being held. Users who want a return can enter a genuine lending market.
Aave provides a useful functional example: users supply assets to liquidity pools, borrowers use those assets, and suppliers earn yield generated by borrowing activity; the rate changes with utilization and borrowing demand. The economic source of the return is lending, not a stablecoin issuer paying passive yield simply to encourage holding.[9]
Aave is not presented here as a ready-made regulatory template for Section 404. It illustrates a narrower point: payment functionality and lending yield can be separated institutionally.
Payments and lending can therefore be separated. Banks manage the money; stablecoins provide the payment layer; lending protocols conduct lending; and a verification layer proves the critical facts.
This is not a compromise halfway between banks and crypto. It is a redesign of their respective functions.
IV. How Regulation Itself Can Change
The stablecoin example is not only about stablecoins. More generally, when critical financial facts can be recorded continuously, read by machines, and replayed after the event, the regulatory toolkit itself can change.
Traditional financial regulation broadly follows:
Classification → registration → disclosure → inspection → enforcement.
That architecture will not disappear. But digital signatures, real-time ledgers, open verification structures, and artificial intelligence now make an additional layer possible:
Fact definition → evidence generation → continuous verification → anomaly detection → responsibility tracing.
Traditional regulation relies heavily on periodic reporting, inspections, and ex post enforcement. New technical conditions make it possible for key facts to remain verifiable at much higher frequency and, in some cases, continuously.
The existence of reserves need not be assessed only at month-end. Transaction authorization can leave digital evidence. System changes can be replayed. The path of funds and the allocation of responsibility can form a machine-readable chain that can be reviewed.
The core argument of The Revolution from Double-Entry Bookkeeping to Verifiable Finance is that finance is moving from internal accounting constraints toward an additional layer of externally verifiable constraints—from primarily trusting institutions toward verifying facts.[2]
Verifiable Thought extends the same method to judgments about finance, policy, investment, and AI: opinions should not be presented as facts, but the factual basis, reasoning process, invalidation conditions, and responsibility boundaries of a judgment can be made visible so that the judgment can be reviewed and revised.[1]
This is why the CLARITY Act is worth studying. Its difficulties do not simply mean that U.S. institutions have failed.
On the contrary, public argument has forced unresolved issues into the open. The questions raised by legislators are serious: conflicts of interest, bank deposits, securities-law boundaries, DeFi control, developer liability, and AML vulnerabilities all deserve scrutiny.
A good institutional mechanism must first be able to discover problems. But a mechanism that discovers problems well does not guarantee that old methods can solve new ones.
The September procedural vote’s failure to reach 60 votes provide is a valuable real-world test. When five technical disputes repeatedly return to classification boundaries, control, actor identity, and jurisdiction, it is reasonable to ask whether a more basic factual-verification layer should come before classification.
If such a layer can remove part of the boundary problem, what is needed is not merely amendment 127 or 128, but a different way of framing the problem.
Conclusion: From “What Is It?” to “What Happened?”
The traditional financial system begins by asking: What is it? A security or a commodity? A bank or a payment institution? An exchange or software?
Verifiable finance can add another set of questions: What happened? Who authorized it? Who is responsible? Where is the evidence? Can it be verified?
These are two different starting points. We have repeatedly argued:
From trusting institutions to verifying facts.
This does not mean eliminating institutional, legal, or governmental credit. It means adding a new source of credit for a machine-based society:
Verification of verifiable facts can itself produce trust and credit.
The logic is simple:
Facts → verification → trust → credit → consensus.
The same applies to stablecoins.
The important task is not to create yet another new “dollar.”
It is to make reserves, issuance, transactions, redemption, yield, and responsibility continuously verifiable financial facts.
The same applies to regulation.
Real transparency is not merely sending more documents to regulators.
Real transparency means that critical facts can be independently verified.
Real regulation is also not merely the accumulation of more rules.
Key facts should be verifiable, key processes replayable, and key responsibilities traceable.
And no matter how far AI and machine systems develop, one principle should remain:
Final verification authority remains with human institutions.
This does not mean that human beings should manually recheck every machine verification. It means that rule-setting, exception handling, dispute resolution, and final attribution of responsibility remain human institutional functions.
The deeper question raised by the CLARITY Act may therefore be not only, “How should the United States regulate crypto more clearly?” but:
In an era when financial facts can increasingly be verified, should the new financial order still rely primarily on repeatedly classifying new things?
The difficulty of amending the CLARITY Act may be one signal of a broader transition. Not every new problem is best solved by adding another patch to an old framework. Sometimes the harder task is not finding the answer, but moving to a different vantage point from which the problem itself can be redefined.
References
[1] Zhu Weisha Scott Shields 《可验证思想》 [Verifiable Thought]. 2026.
[2] Zhu Weisha Scott Shields 《从复式记账到可验证金融学的革命——从信任机构到验证事实》 [The Revolution from Double-Entry Bookkeeping to Verifiable Finance: From Trusting Institutions to Verifying Facts]. May 2026.
[3] Zhu Weisha 《稳定币发行如何合理适应CLARITY法案第404条——从单一机构信用到开放式可验证稳定币网络》 [How Stablecoin Issuance Can Rationally Adapt to Section 404 of the CLARITY Act: From Single-Institution Credit to an Open Verifiable Stablecoin Network]. 2026.
[4] Zhu Weisha 《透明稳定币：无链如何实现开放式可验证发行网络》 [Transparent Stablecoins: How an Open, Verifiable Issuance Network Can Be Built]. 2026.
[5] U.S. House of Representatives, Office of the Clerk. “Roll Call 199: H.R. 3633, CLARITY Act.” July 17, 2025.
[6] U.S. Senate. “Roll Call Vote No. 234, 119th Congress, 2nd Session: Cloture on the Motion to Proceed to H.R. 3633.” September 15, 2026.
[7] Cynthia Lummis, John Boozman, and Tim Scott. “Lummis, Boozman, Scott Release Final Clarity Act Text.” September 14, 2026.
[8] U.S. Senate Committee on Banking, Housing, and Urban Affairs. “CLARITY Act Section-by-Section Analysis,” Sec. 404. May 12, 2026.
[9] Aave. “Supplying Tokens.” Aave Help. Accessed September 2026.
[10] Reuters. “US Senate Republicans release new crypto bill text ahead of critical vote.” September 14, 2026.
[11] Reuters. “US Senate fails to advance sweeping cryptocurrency bill in blow for industry.” September 15, 2026.
