Who Actually Controls It.

The Cayman Islands has just written that question into a filing obligation.

The jurisdiction’s beneficial-ownership framework — consolidated this year as the Beneficial Ownership Transparency Act (2026 Revision) (Act 13 of 2023, consolidated with Acts 3 and 11 of 2025; originally enacted 23 November 2023), revised as at 14 January 2026 and published as Supplement No. 4 with Legislation Gazette No. 4 dated 29 January 2026, and re-priced in March by gazetted amendment regulations — does something more interesting than the headlines suggest. The reporting most planners have seen frames it as a transparency crackdown — broader definitions, a paid access route, tighter deadlines. That framing is not wrong, but it misses the part that actually matters to anyone holding wealth through a Cayman vehicle. The reforms do not merely ask who owns a company. They ask, in operational detail, who can make it do things — and they require an answer, in writing, filed with the government, kept current.

For a structure that was built honestly, this is an administrative burden. For a structure that was built to obscure the fact that the settlor never really let go, it is something closer to a confession form.

The statutory test is set out at Section 4(1) of the Act. A beneficial owner is an individual who:

Read limb (a) and you have a familiar ownership threshold. Read limbs (b) and (c) and you have something else entirely — a factual inquiry with no percentage in it at all.

The Competent Authority’s own guidance, revised on 25 February 2026, is unusually candid about the reach of those limbs. Control, it says, “is meant to broaden the scope of analysis beyond direct or indirect ownership,” and can be exercised “through aggregate voting rights e.g., by siblings or connected persons” (Guidance on Complying with Beneficial Ownership Obligations in the Cayman Islands ¶ 2.6 (Competent Authority for Beneficial Ownership, issued 31 July 2024, revised 25 February 2026) (“BO Guidance”, v. 1.1)). It contemplates control exercised through “nominee appointments and powers of attorney” (BO Guidance ¶ 2.7). It contemplates control through “debt instruments or financing arrangements” — a lender with a contractual right to direct the company, or “a third party who can otherwise influence a shareholder by financial or relational means such that they have control over a Legal Person” (BO Guidance ¶ 2.9, which adds that “a regulated lending or credit institution providing financing to a Legal Person will rarely be considered as exercising control for the purposes of the Act as the lender or creditor is unlikely to have ultimate control”).

And then it says the sentence that ought to stop every planner in the field: “The exercise of control once is sufficient to indicate that it could be exercised again” — unless steps were taken to clarify ownership and control, or the control was only exercisable in circumstances that no longer exist (BO Guidance ¶ 2.8).

Once. A single exercise of control, on a single occasion, is treated as evidence of a standing capacity to control. That is not a reporting threshold. That is an evidentiary presumption, and it is the same presumption a court applies when it is deciding whether a trust is a trust or a costume.

The guidance illustrates the point with a worked example that reads like a case study in what not to build. An individual — “M” — holds no shares, holds no voting rights, and is neither a direct nor an indirect owner of the company. M controls only the appointment and removal of the directors. The guidance’s conclusion: “M must be identified in the BOR” (BO Guidance, Figure 2 and accompanying text). In a second example, a foreign investment manager who has “ultimate control to direct the CSP and the two directors” of a Cayman company is a beneficial owner of it, notwithstanding that a corporate service provider is the sole registered shareholder (BO Guidance, Figure 3 and accompanying text).

If your structure’s protection depends on the proposition that the person who picks the directors is not really behind the company, Cayman has now told you, in a published government document, that it does not accept that proposition.

It would be easy — and commercially tempting — to read all of this as bad news for offshore planning. We read it as almost exactly the opposite, and the reason goes to the heart of what we tell every client.

A structure that actually protects assets has four properties. It is seasoned: settled years before any claim was on the horizon, at a time when there was no creditor to hinder, delay, or defraud. It is irrevocable and discretionary: the settlor has genuinely parted with the property, and no beneficiary holds an entitlement that a creditor can attach. It is independently administered: a real trustee, exercising real discretion, who is capable of saying no to the settlor and occasionally does. And it is disclosed: it has nothing to hide, and it says so when properly asked.

Look at the Section 4(1) test again with those four properties in mind. A client who has genuinely relinquished control has a truthful and unremarkable answer to every question the register asks. The trustee is the trustee. The directors were appointed by the trustee. No one holds a power of attorney over the corpus. There is no side letter, no nominee arrangement, no informal understanding that the settlor’s wishes are instructions. The filing is boring — and boring is the entire objective.

The client who cannot file honestly is the client whose structure was never going to hold. A register did not create that problem. The register surfaced it, years before a creditor would have, and at a moment when the only cost is embarrassment rather than an avoidance judgment.

This is the same lesson the courts have been teaching from the other direction. Where a debtor moves assets on paper while keeping them in his orbit in fact, the transfer does not defeat the creditor; it hands the creditor a second cause of action and a set of badges of fraud to prove it with. Retained control has always been the flaw that collapses a structure in litigation. What Cayman has done is make retained control a disclosure event as well as a litigation risk — the same defect, now visible at two moments instead of one.

There is a hard-edged corollary that clients rarely want to hear. If you cannot honestly complete the beneficial-ownership filing for your own structure, you do not have a protection problem you can solve with better paperwork. You have a structure problem, and the paperwork is telling you the truth.

The Act’s treatment of trusts deserves particular attention, because it converts something we have always argued for into something Cayman now records.

Where a trust would meet the beneficial-owner definition if it were an individual, Section 4(3) requires that a trustee of the trust be identified as the contact person, and that trustee “may be contacted to provide information on the BOs of the Legal Person” (BO Guidance ¶ 4.1; Act § 4(3)). A corporate trustee incorporated in Cayman is itself a reportable legal entity and must be included in the register (BO Guidance ¶ 4.3: “A corporate trustee that is incorporated, established or registered in the Cayman Islands is an RLE in accordance with the Act and should therefore also be included in the BOR.”). A foreign corporate trustee must be named, together with the name, email address, and telephone number of an individual within it — and the guidance specifies that this must be “an individual of authority who is able to provide BO Information relating to the foreign corporate trustee if requested, such as the senior managing official” (BO Guidance ¶ 4.4).

Note what that structure assumes. It assumes the trustee is a real institution with real officers who can actually answer questions about the trust. A trustee who is genuinely independent has no difficulty here; naming the person who can speak for the trustee is simply naming a fact. But a “trustee” who is in substance the settlor’s instrument now has to be named, identified, and made reachable by a government authority — while the individual trustee test at Section 4(2) of the guidance runs the same ultimate-effective-control analysis right back through the trust itself (BO Guidance ¶ 4.2, identifying an individual trustee as contact person where the trust meets one of the Section 4(1) conditions).

Independent administration has always been the line between a trust and a sham. Cayman has now put a phone number next to it.

The reform that has drawn the most commentary is the “legitimate interest” access route, and it is worth correcting the record, because the secondary coverage has garbled the chronology.

The route is not new in 2026. It was created by the Beneficial Ownership Transparency (Legitimate Interest Access) Regulations, 2024, which came into force on 28 February 2025 (see Conyers, Cayman Islands: Legitimate Interest Access of Beneficial Ownership; BO Guidance at 36 (FAQ 4)). What happened in 2026 was a re-pricing, and nothing more. The Beneficial Ownership Transparency (Legitimate Interest Access) (Amendment) Regulations, 2026 (SL 8 of 2026) were made in Cabinet on 21 January 2026, laid in Parliament on 6 March, and affirmed on 11 March 2026. They amend a single regulation of the 2024 Regulations — regulation 4 — to do two things: raise the fee for a single search from thirty dollars to seventy-five dollars, and add a new fee of two hundred and fifty dollars “for administrative services in respect of the application if the application is for multiple applications to be made within a period of one year in relation to any number of legal persons” (SL 8 of 2026, regs. 1–2, amending reg. 4(2)(b)(iii) of the 2024 Regulations; published as Supplement No. 3 with Legislation Gazette No. 15 dated 18 March 2026. The Regulations state the figures in “dollars”; Cayman Islands coverage reports them as CI$75 and CI$250 respectively).

That second figure has been widely described in the trade press as an “annual subscription” to the register. It is not. It is an administrative charge on a bulk-application route — the difference matters if you are trying to understand how much friction actually stands between a hostile party and a filing.

Access is not open to the world. An applicant must fall within one of three categories — a journalist or bona fide academic researcher; a person acting for a civil society organisation engaged in combating money laundering or terrorist financing; or a person seeking information in the context of a potential or actual business relationship or transaction with the legal person — and must additionally demonstrate a legitimate interest in the information for the purpose of preventing, detecting, investigating, combating, or prosecuting money laundering, its predicate offences, or terrorist financing (Legitimate Interest Access Regulations, 2024, reg. 3; see Conyers, supra).

Two limits matter enormously for planning, and both are routinely misunderstood.

First, the register cannot be searched by person. Searches must be directed at a named legal entity; as Conyers puts it, “No applications for searches are permitted on the basis of an individual’s name.” The Registrar may refuse an application where not satisfied that a legitimate interest has been established, and must retain application records for six years. A hostile party who does not already know which Cayman entity to look at cannot use this route to go fishing for one. The register is a confirmation tool, not a discovery tool.

Second, what is disclosed is deliberately partial. For an individual, the accessible particulars are name, country of residence, nationality, month and year of birth, and the mechanism of control. Percentage ownership is withheld where control is through shareholding — as Conyers notes, “If the mechanism of control is through ownership of shares, the individual’s percentage ownership will not be accessible” — and information on minors is excluded altogether.

This is a design choice, and Cayman has defended it on constitutional rather than commercial grounds. The government’s position rests on the Court of Justice of the European Union’s Grand Chamber decision in Joined Cases C-37/20 and C-601/20, WM and Sovim SA v Luxembourg Business Registers, ECLI:EU:C:2022:912 (CJEU, Grand Chamber, 22 Nov. 2022), which held that general public access to beneficial-ownership registers under the Fifth Anti-Money Laundering Directive was a disproportionate interference with the privacy and data-protection rights guaranteed by Articles 7 and 8 of the EU Charter, and invalidated the relevant provision — Article 30(5), first subparagraph, point (c) of Directive (EU) 2015/849 as amended, insofar as it provided for general public access. Writing on the Oxford Business Law Blog in April 2026, Andrew James Perkins observed that Article 9 of the Cayman Islands Constitution offers privacy protections analogous to Article 8 of the European Convention on Human Rights, and argued that pressure for a fully public register “risks privileging political considerations over legal coherence” (Andrew James Perkins, Beneficial Ownership Transparency in the Cayman Islands: Balancing Privacy, Access, and International Pressure, Oxford Business Law Blog (8 Apr. 2026)).

We take no position on that political argument. The planning point is narrower and more durable: the trend line runs one way. Cayman has held the line at legitimate-interest access rather than a public register, but the definition of who must be reported has broadened, the access route exists, the fees are modest, and the competent authorities — the Royal Cayman Islands Police Service, the Financial Reporting Authority, CIMA, the Anti-Corruption Commission, the Tax Information Authority, and, on a request basis, their foreign counterparts — have had direct access all along under Section 22, unencumbered by the legitimate-interest test entirely (BO Guidance at 35–36 (FAQ 4)). Any plan whose viability depends on the register staying narrow is a plan that has bet on the direction of policy. That is not a bet we advise clients to place.

The practical guidance follows directly, and it is uncomfortable only for the wrong kind of plan.

Stop treating disclosure as the enemy. The premise that protection comes from a creditor never finding the asset was already obsolete; automatic exchange of information and beneficial-ownership reporting finished it. Protection comes from the asset being genuinely, lawfully, and durably beyond the client’s control — which is a fact about the structure, not a fact about the creditor’s knowledge. A structure that survives only while it is secret is not protected. It is merely undiscovered, and undiscovered is a temporary condition.

Audit the control question before someone else does. Every client with a Cayman vehicle should be able to answer, today, on paper: who appoints and removes the directors? Who can direct the corporate service provider? Does anyone hold a power of attorney, a nominee arrangement, a veto, or a financing right that amounts to control? Has anyone ever exercised control informally — because under paragraph 2.8 of the guidance, once is enough to raise the question, and the only cure is to have clarified ownership and control afterwards.

Fix defects while they are still administrative. There is a narrow and closing window in which “the settlor has more control than the documents admit” is a governance problem you can remedy through proper restructuring, on a clear day, with no claim pending. Once a creditor exists, the same remedy becomes a transfer made under the shadow of a claim — and the badges of fraud that follow are considerably more expensive than a corrected filing.

Choose architecture, not opacity. Where genuine creditor resistance is the objective, it comes from substantive law — a robust charging-order remedy, a real limitation regime, an independent fiduciary — and from timing. It has never come from the hope that no one would ask. Cayman is asking. Every credible jurisdiction is asking, or will.

The instinct in our industry is to read every transparency reform as an erosion of what we do. That instinct is a century old and it is wrong. What Cayman’s 2026 framework erodes is not asset protection. It is the counterfeit of asset protection — the structure that looks independent on the organisational chart and takes its instructions from the settlor over dinner.

A register that asks “who controls this?” is only frightening to the client whose honest answer is “I do.” For everyone else, it is a form. And the difference between those two clients was never going to be settled by a filing deadline. It was settled years earlier, on a clear day, when one of them built something real and the other built something that merely looked real.

The Act has simply moved the moment of reckoning forward. We would argue that is a mercy. It is far better to discover that a structure does not hold when the consequence is a corrected register entry — than to discover it when the consequence is a judgment.