Where Should a Treasury Actually Live?

Where Treasury Should Sit in a Crypto-Native Stack

Founders searching for where to hold crypto treasury often start with the wrong mental model. They compare wallets, custody providers, multisig tools, and chains as if treasury placement were a question of technical storage. In practice, treasury location is not a wallet question at all. It is a control question. That distinction becomes painfully clear the first time a post-TGE project with eight figures in a multisig sits down with an accountant and hears a deceptively simple question: whose balance sheet does the treasury sit on? Silence follows, because no one explicitly decided.

Treasury Is Not a Wallet Question, It Is a Control Question

In most early-stage teams, treasury is treated as a technical primitive. A multisig is created, a handful of trusted people are added as signers, and the address is labeled “treasury.” Regulators, tax authorities, and courts do not care what the address is labeled. They look through the label and ask who effectively controls the assets. Control is inferred from who appoints and removes signers, who proposes and approves transactions, and which entity economically benefits from how the funds are deployed. If a company’s directors can change signers, instruct spending, and rely on treasury to fund payroll, that company will likely be treated as controlling the assets—regardless of whether a foundation exists.

This single determination quietly drives three downstream outcomes that shape a project’s entire risk profile. First, it influences which jurisdiction claims taxing rights over treasury activity and gains. Second, it determines which regulator claims enforcement authority if something goes wrong. Third, it influences who bears liability for treasury decisions—whether that liability lands on a corporate entity or identifiable individuals. Teams often experience these as separate issues, but they all collapse back to the same root: who controls treasury.

The arrival of the OECD’s Crypto-Asset Reporting Framework (CARF) as a live reporting regime from January 1, 2026 sharpens these questions further. CARF is not limited to consumer exchanges or hosted wallets. It is designed to capture entities that provide crypto-asset services, including facilitating transfers or maintaining control over assets for third parties. While managing one’s own treasury does not automatically trigger CARF, the framework intensifies scrutiny on the same control questions that already matter. Treasury placement is no longer a back-office design choice. It is a primary regulatory surface, alongside token issuance and user-facing services.

Most teams do not consciously choose to place treasury control inside their operating company. They drift into it. It feels convenient for the company that employs everyone to also “own” the funds that pay everyone. The multisig signers are employees or founders. The board approves budgets. Finance prepares treasury reports. Over time, a pattern of behavior forms that tells a clear story to outsiders: the operating company controls the assets. A foundation elsewhere in the structure does not change that story if it does not exercise control in practice. Convenience becomes exposure.

Why Foundations Became Treasury Holders (And When That Logic Breaks)

Foundations in jurisdictions such as Cayman, Panama, or Switzerland became the default treasury holder for token projects for understandable reasons. They offer a non-shareholder entity that can hold assets, articulate a public-purpose mandate, and fund multiple operating companies without collapsing token issuance, ecosystem development, and commercial operations into a single legal body. This separation allowed projects to argue that token networks were not simply corporate products, while still providing a wrapper capable of signing contracts, paying vendors, and holding large pools of capital.

At their best, foundations function as stewards. They hold treasury, define high-level policy about how funds may be used, and issue grants to operating companies or contributors for specific purposes. The operating companies execute those purposes without owning the underlying capital. This aligns reasonably well with how many decentralized projects behave.

The model breaks when the separation exists only on paper. A foundation does not meaningfully reduce exposure if it does not control treasury in practice. If the operating company appoints most multisig signers, proposes most transactions, and routinely receives large lump-sum transfers from treasury to cover general operations, authorities can treat the operating company as the true economic owner regardless of what the organizational chart says.

The boundary between foundation and operating company is therefore operational. One of the most important mechanisms in maintaining that boundary is how grants are structured. Foundations that issue purpose-specific, milestone-triggered grants preserve separation. Funds move in discrete tranches tied to defined activities. Foundations that periodically sweep large portions of treasury into an operating company to “run the business” erase that separation. At that point, the foundation may still exist legally, but economically the operating company looks like the owner.

Signer architecture compounds this effect. Multisig signers are not abstract roles. They are human beings located in specific jurisdictions. If most signers reside in a high-tax country or an aggressive enforcement jurisdiction, treasury risk tends to migrate there even if the foundation is incorporated elsewhere. A Cayman foundation whose treasury is effectively operated day-to-day by signers resident in Germany or the United States presents a very different risk profile than one whose signer set and governance machinery are aligned with the foundation’s jurisdiction and mandate.

This is why the question “should we use a foundation or a company?” is incomplete. The real question is whether governance design, signer architecture, and grant mechanics collectively support the story that the foundation controls treasury. Without that alignment, the foundation label does little work.

Designing Treasury Architecture So Control Lands Where You Intend

Among mature projects, a convergent pattern has emerged. Treasury sits inside an entity whose sole function is stewardship and grant-making. That entity does not run products, does not employ large operational teams, and does not directly conduct revenue-generating activities. Its purpose is to hold assets, set policy, and fund others. Operating companies never hold meaningful treasury. They receive funds through documented grants, service agreements, or milestone-based tranches. Transfers are limited, purposeful, and auditable.

Signer selection is treated as a first-order design choice rather than an afterthought. Signers are chosen with jurisdictional awareness, and rotation processes exist so that control does not silently concentrate in the hands of a small group of people in a single country. In some structures, licensed custodians or professional transaction services execute transactions based on foundation-approved policy, separating policy-setting from execution and further reducing attribution of control.

None of this eliminates reporting obligations. CARF, local tax rules, and other regimes still apply. What this architecture does is concentrate those obligations inside the entity designed to bear them, ideally in a jurisdiction chosen intentionally rather than accidentally. A Cayman foundation controlling treasury presents a different compliance and enforcement landscape than a Delaware operating company controlling the same assets. A Swiss foundation with European-resident signers presents a different landscape than a Cayman foundation with globally distributed signers and outsourced execution. These distinctions shape real-world exposure.

This is the layer where structural service providers like Spindipper operate, long before any funds exist. Founders map intended treasury behavior to an entity stack, signer design, and grant flow as they choose where to incorporate. The goal is not to chase zero regulation, which does not exist, but to ensure that control, reporting, and enforcement surfaces align with the project’s actual risk tolerance and operating reality. Retrofitting this after a TGE, once habits, people, and flows are established, is materially harder and often politically painful.

Returning to the opening scenario, the most telling signal is not how large the treasury is, which chain it sits on, or which multisig tool is used. The telling signal is whether the founding team can answer cleanly and confidently whose balance sheet the treasury sits on. If the answer is vague or depends on informal understandings, the structure is already wrong. Treasury architecture is one of the few areas where early design decisions compound quietly but decisively. Getting it right upfront is far cheaper than discovering later that convenience turned into control.

This article is for informational purposes only and does not constitute legal or tax advice. Given the rapidly evolving nature of digital asset regulation, jurisdiction-specific professional advice should be obtained before implementing any of the structures discussed herein.

Have questions about treasury placement, signer architecture, or structuring a stewardship entity for your protocol? Feel free to get in touch for a friendly, no-pressure conversation.