Corporate Crypto Treasury: Holding Digital Assets

Yes. A U.S. corporation can hold digital assets on its balance sheet, and no provision of corporate law prevents it. Most boards treat this as a wallet question. My view is that the deciding constraint is evidentiary: a company answers to directors, auditors, lenders, and counterparties, and it has to prove to every one of them that it controls what it reports.

The short version

  • Get the authorization in writing before the first purchase. A board minute with no policy behind it reads, a year later, as approval of nothing in particular.
  • Access to the assets is an accounting control. 15 U.S.C. § 78m(b)(2)(B) tells reporting companies to permit access only under management’s authorization and to reconcile recorded balances to the assets at reasonable intervals.
  • Proving control of a self-custodied asset to a third party is harder than proving a bank balance. Settle the method before the company holds anything.
  • The accounting standard covers less ground than people assume. ASC 350-60 reaches only assets meeting six criteria, so ask the auditors whether a given token qualifies.
  • The credit agreement predates the asset class. Read the covenant definitions ahead of the purchase.

Who authorized this, and under what policy

A family holding entity answers to the family. A corporation has directors who owe duties, and the decision gets reviewed later against whatever record exists. Delaware puts a corporation’s “business and affairs” under “the direction of a board of directors” (8 Del. C. § 141(a)), and Wyoming lists what a public corporation’s board attends to, including major risks and “the effectiveness of the corporation’s internal controls” (W.S. 17-16-801(c)).

None of that says whether a company should hold anything. It says who decides. A treasury policy that survives an audit settles the purpose of the position, a size limit written to stay meaningful when the price moves, permitted assets and venues, who may instruct a transfer and at what quorum, what triggers a sale, and who escalates a breach. The entity question resolves differently here, because an operating company already has a governance apparatus the position must fit inside.

Access to the assets is the control that matters

The federal books-and-records provision states what a control over a bearer asset must accomplish:

“(iii) access to assets is permitted only in accordance with management’s general or specific authorization; and (iv) the recorded accountability for assets is compared with the existing assets at reasonable intervals and appropriate action is taken with respect to any differences”

15 U.S.C. § 78m(b)(2)(B)

That binds issuers registered under section 12 or reporting under section 15(d). A private company sits outside it, though the language is what an auditor, a lender, or an acquirer will use.

Clause (iii) is where a crypto treasury departs from every other asset the company owns. A wire needs a bank to act on the instruction, so a rogue signer can be stopped downstream. A signed transaction needs nobody, so (iii) has to live in the signing quorum itself: a threshold where no individual can sign alone, a documented recovery path, roles assigned by office, and key rotation on any departure.

Clause (iv) is the monthly reconciliation, the cheapest item here: the ledger against on-chain balances at addresses the company has identified as its own, signed by somebody who cannot move the assets. In my experience it is the first artifact anyone outside asks for.

How anyone outside the company verifies the position

Boards underestimate this. The PCAOB confirmation standard, effective for audits of fiscal years ending on or after June 15, 2025, states the hierarchy directly:

“Audit evidence obtained from a knowledgeable external source is generally more reliable than evidence obtained only from internal company sources.”

PCAOB AS 2310.06

For a bank balance the auditor picks the confirming party and controls the channel. A self-custodied position has no external party to ask, so the company becomes the only source, which the hierarchy ranks lowest.

The substitutes carry limits. A message signed with the private key demonstrates control of that key at the moment of signing. It does not establish that nobody else holds a copy, and it says nothing about the balance sheet date unless performed then. On-chain balances are public, so existence is easy to check; ownership is the assertion that takes work. A regulated third party restores an external source, which is most of what “qualified” is doing.

Valuation, reporting, and the covenants nobody reread

Measurement here moved recently, so treat the current standard as a question for the company’s auditors rather than settled ground. ASU 2023-08, effective for fiscal years beginning after December 15, 2024, added Subtopic 350-60, whose scope reaches only assets meeting all six of its criteria. Three of them rule out a lot: the asset must convey no enforceable rights to underlying goods, services, or other assets, it must be fungible, and it must not have been created or issued by the reporting entity or a related party. Confirm the scope conclusion for your asset, and the tax side separately, since the IRS treats digital assets as property (IRS, Digital assets).

The valuation policy needs an owner: which market, observed at what time, applied the same way every period. Where fair value applies, the movement lands in reported results, and any covenant defined on earnings or net worth inherits it. Working capital is the collision that actually happens: an asset the company cannot sell on a payroll timeline fails the test that makes it treasury. Companies answer with a stablecoin working balance or by borrowing against the position, trading market exposure for counterparty exposure.

What I actually see

The first pattern is the approval with nothing behind it. The minutes record that the board approved an allocation. No policy states the size limit, the venue, the transaction authority, or the reporting cadence, and two years on the CFO who ran it has left.

The second is a segregation-of-duties failure in the one place it was load-bearing. Two officers are required to release a wire; one officer holds the seed phrase and also prepares the reconciliation that would catch a problem. Every control built for cash was abandoned where the asset became a bearer instrument.

The third is the auditor conversation that happens last. The company holds for a year, then meets its auditor in fieldwork and finds the evidence it can produce falls short of the assertion already in draft financials. The worst version is a scope limitation on a material balance.

The check I would run: hand your controller a blank page and ask for four things, without you in the room. The board resolution and the policy it rests on. A current list of every address the company controls, with the quorum to move each. Last month’s reconciliation of the ledger to on-chain balances, signed by someone who cannot move the assets. A memo stating how existence and ownership get evidenced at year end. If any takes longer than a day, the position is not under control yet.

Where this goes wrong

The position stops being a treasury matter the moment somebody outside the company needs an answer about it.

The specific failures: a covenant whose defined terms predate the asset class, since current assets, tangible net worth, and permitted investments were drafted around cash and receivables, so a fair value swing breaches a ratio nobody was watching. A key holder who leaves without a rotation, leaving live access outside the org chart. Treasury activity that drifts into moving value for other people, which can pull an operating company inside the money services business definition and its registration regime on FinCEN Form 107. Staking or other network participation that changes what business the company is in. And the case that concentrates all of it: a position that is down, a shareholder asking who approved it, and a record that answers badly.

The decision rule

  1. Decide what the position is for before sizing it, and write the purpose down.
  2. Adopt a treasury policy by board resolution covering size limit, permitted assets and venues, authority, and reporting cadence.
  3. Separate approval authority from execution ability, and keep both away from whoever reconciles.
  4. Set the signing quorum above any single individual, assign roles by office, and document recovery and rotation.
  5. Agree the audit evidence with your auditor in writing before the first purchase.
  6. Reconcile the ledger to on-chain balances monthly, signed by someone who cannot move the assets.
  7. Read the credit agreement’s definitions and negative covenants first, and take ambiguity to the lender.
  8. Confirm accounting scope with the auditors and tax treatment with the CPA each period.

Where this sits

These questions sit on the same ground the rest of this library covers. Custody settles who can move the assets, and for a company it also settles what evidence exists that it controls them. Wyoming entity questions settle where the position is held.

The join is what fails. Corporate counsel drafts the resolution, the CPA closes the books, the auditor tests the assertion, and custody gets set up by whoever was closest to the operational problem. Those four rarely read each other’s work, and the gap surfaces as a control that exists on paper with no counterpart in the signing quorum. In a company, the CFO owns that seam.

Sources

Related

Last updated: 3 August 2026.

This article is general education, not legal, tax, accounting, audit, or investment advice, and nothing here recommends that any company hold digital assets. Accounting and audit requirements here have changed recently, and treatment depends on facts this page cannot know. Talk to a qualified attorney, CPA, and your company’s auditors about your own situation.

Sources

    Jake Claver

    Written by

    Jake Claver

    Family office professional working on how substantial holdings are held, structured and passed on. Qualified Family Office Professional. Finance degree, University of North Texas. Board member, Arkansas Blockchain Council. Author of Wealth in Numbers and Infinite Banking for Crypto Investors.