“Self-custody” is the most striking phrase in the Securities and Exchange Commission’s new custody proposal. It is also easy to misunderstand. The proposal does not give investment advisers a general license to hold any client token themselves. Instead, it would make the availability of an eligible custodian a key condition for an adviser’s proposed self-custody pathway, with a fresh determination required every quarter, as described by SEC Commissioner Hester Peirce.
The SEC issued the proposal on October 1, 2026, and included a route for state trust companies to serve as custodians, subject to conditions (SEC proposal). For funds and advisers, the change is less a broad shift away from third-party custody than a proposed framework for documenting when one custody route is available, appropriate, and supportable. The measure remains a proposal, not a final rule.
Methodology note
This analysis is based on the SEC’s proposed rule page for File No. S7-2026-35, the Commission’s October 1 press release, and statements from Commissioners Hester Peirce and Mark Uyeda. The SEC page labels the measure a proposed rule and says comments are due 60 days after publication in the Federal Register. The page does not list a calendar deadline. This article distinguishes the proposed text from the commissioners’ explanations and does not treat either as an adopted requirement.
The proposal is narrower than the headline suggests
The SEC proposal would amend custody rules under the Investment Advisers Act and Investment Company Act. It concerns registered investment advisers and regulated funds, including registered investment companies and business development companies. Its stated goals include modernizing existing custody rules, addressing crypto-related custody practices, and updating related recordkeeping and reporting (SEC press release).
That scope matters. Commissioner Peirce’s statement explains that the proposed Advisers Act amendments would apply to crypto assets that are funds or securities, and that proposed Investment Company Act custody rules would apply to crypto assets that are securities or similar investments. The proposal is therefore not a single custody rule for every token or every person who owns crypto (Peirce’s statement).
Nor does “self-custody” here mean an individual investor holding a private key without an intermediary. Peirce uses the term for an adviser holding client assets itself. She notes that the distinction is important and suggests “shelf-custody” would have better described the arrangement. The proposal is about the duties of regulated advisers and funds, not a restriction on an individual’s choice to hold their own assets (Peirce’s statement).
Availability becomes a recurring control
Under Peirce’s account of the proposal, an adviser could use the proposed self-custody option only in limited circumstances and subject to conditions. Before taking custody, the adviser would need to determine that no permitted custodian is available for the relevant crypto asset. It would need to repeat that determination quarterly (Peirce’s statement).
That turns a vendor search into a continuing control. An adviser could not treat a one-time review as permanent justification. If a permitted custodian later becomes available, the basis for using the proposed exception could change. The adviser would need a process to notice that change and assess what it means for the custody arrangement.
The rule would not make the risk disappear by moving key control in-house. Uyeda’s statement says the proposal contemplates safeguards such as safeguarding expertise, cybersecurity protections, annual reviews, internal reporting, account statements, and client disclosures. He also emphasizes that an adviser’s fiduciary duty would continue to apply and that self-custody creates an inherent conflict of interest (Uyeda’s statement).
The alternative pathway is a state trust company. Peirce says an adviser or regulated fund would need a reasonable basis, after due inquiry, to believe that the company is authorized by the relevant state banking authority to provide crypto custody and has written policies designed to safeguard crypto assets and related cash from theft, loss, misuse, and misappropriation. The proposed review would occur before engagement and annually thereafter (Peirce’s statement).
The two paths are not interchangeable checkboxes. One depends on whether a permitted custodian is available for a specific asset; the other depends on whether a state trust company meets the proposed conditions and can provide a suitable service. Both would require governance, evidence, and ongoing review.
A worked example for an adviser evaluating custody
Consider a registered adviser evaluating custody arrangements for a crypto asset held in client accounts. The first step is to determine whether the asset falls within the proposed rule’s coverage. Under the scope described by Commissioner Peirce, the answer depends on whether the asset is a fund or security for the relevant Advisers Act purpose, or a security or similar investment for the relevant regulated-fund account. A token label alone does not settle that legal question.
If the asset is within scope, the adviser would then assess whether a permitted custodian can maintain it. The proposal responds to a market gap: the release, as quoted in Peirce’s statement, says few traditional permitted custodians have offered robust custody services for a substantial range of crypto assets (Peirce’s statement and its cited proposing release). That general observation is not proof that a particular asset lacks an available custodian. The adviser would still need asset-specific evidence, such as documented inquiries, service capabilities, and the terms on which a provider is willing to hold the asset.
If no permitted custodian is available, the proposed exception could become relevant, but it would not become automatic permission to hold the asset. The adviser would need to meet the final rule’s conditions if adopted, demonstrate the required expertise and safeguards, and revisit availability each quarter. If an eligible state trust company can provide custody instead, the adviser would assess that company’s state authorization and written safeguards, then repeat the proposed review annually.
This example separates three questions that are often collapsed: whether an asset is covered, whether a permitted custodian is available, and whether the adviser can satisfy the conditions for a specific custody route. The SEC’s proposal addresses these questions as part of a regulatory design. It does not establish that any individual adviser or trust company already qualifies.
The implementation work is mostly evidence and controls
Firms can prepare for the proposal without assuming it will be adopted as written. A useful readiness file would identify which client assets may be within scope, which custody providers have been evaluated, what each provider can actually support, and when each assessment was performed. It would also identify who owns quarterly availability reviews and who is responsible for escalating a material change.
For a state trust company, diligence would need to go beyond a marketing description. The adviser would want to verify the company’s authorization, understand its written controls, and assess how those controls address key management, access, reconciliations, incident response, and the protection of related cash. Those operational checks are practical analysis points; the precise obligations will depend on the final rule, if one is adopted.
Portfolio operations will also need clear separation between the adviser’s investment decisions and its custody controls. Where the same organization can select an asset and hold it, conflict management, independent review, client reporting, and access controls become central. A custody policy should explain not only who can move assets, but how the firm verifies balances, records exceptions, and responds when a custody provider’s capabilities change.
The comment process is another part of the control environment. The SEC’s proposal page identifies File No. S7-2026-35 and says comments are due 60 days after Federal Register publication. It does not show a specific calendar deadline. Firms that intend to comment should use the SEC’s docket page to track the publication date and submission instructions rather than infer a deadline from the date the Commission issued its proposal.
What the data does and does not tell us
The proposal and commissioners’ statements show the direction of the SEC’s proposed approach: more defined custody options, a conditional path for adviser-held client assets, and a route for eligible state trust companies. They also identify the recurring availability test as a key limit on the proposed self-custody option (SEC proposal; Peirce’s statement).
They do not tell us which specific tokens will qualify, which providers will be accepted, or how the Commission will revise the proposal after reviewing public comments. The SEC has not adopted the measure as a final rule. Its eventual scope, wording, effective date, and transition arrangements could change.
The proposal also does not establish that self-custody is safer or more dangerous than third-party custody in every case. The relevant risks depend on the asset, the firm’s systems, governance, key controls, and operational resilience. A provider’s regulatory status alone does not reveal the quality of its controls or how well they fit a particular fund.
Finally, the statements by individual commissioners are useful explanations, not substitutes for the proposed rule text. Firms should read the SEC’s proposal and seek qualified legal advice before making custody decisions. The distinction between a proposed pathway and a binding requirement is essential.
Why this matters
For funds and advisers, custody is a market-structure question as well as an operational one. The availability of reliable custodians can affect which assets a regulated product can support. A conditional self-custody option may help address gaps, but it also puts more weight on internal control design and recurring evidence.
For custody providers, the proposal highlights the importance of asset coverage, state authorization, written safeguards, and the ability to support institutional due diligence. For analysts and investors, it is a reason to distinguish a policy proposal from a live compliance regime and to ask which assets and types of account are actually covered.
The practical test is not whether a firm says it can custody crypto. It is whether the asset is within scope, whether an eligible provider is available, and whether the firm can demonstrate controls that meet the rule ultimately adopted. Those facts will matter more than the phrase “self-custody” on its own.
For more research and tools, visit Presolt.com.
Compliance disclaimer
This material is for informational and educational purposes only. It is not investment, legal, tax, or accounting advice, and it is not a recommendation to buy, sell, or hold any digital asset. Crypto assets are volatile and may lose some or all of their value. Consult qualified professionals about your circumstances.