The Family Office Versus the Trustee

The question that exposes the arrangement is never a large one.

It is not who should have sold the shares, or whether the property was undervalued. It is smaller and more ordinary than that. Who renewed the insurance on the warehouse in Ikeja? The trustee understood that the family office attended to the properties. The office understood that anything held in the trust was the trustee's to administer. Both were reasoning correctly from their own engagement. The policy lapsed in March, and nobody discovered it until August.

Neither institution failed. The space between them did.

This is the characteristic pathology of a family that has built both institutions and never defined the boundary between them — and it is worth saying at the outset that it is a pathology of the well-organised. A family with no trustee and no office does not have this problem. It has a worse one.

The essay on the office built it and deferred this question. The essay on the council dealt with a related and more dangerous version of it — the council that begins to direct the trustee. This essay completes the pair: what each institution actually is, the three zones in which their functions overlap, the five dysfunctions that overlap produces, and the protocol by which a Nigerian family makes the two work as one system.

The instinct is to see the trustee and the office as two service providers of the same species, differing in what they do. They are not the same species, and almost every practical consequence follows from that.

The trustee holds legal title. It is the owner of the trust property in law, and its obligations arise from the deed and from equity: to act in the interests of the beneficiaries, to keep the property safe, to invest within its powers, to keep accounts, and to exercise its discretions honestly and on proper considerations. A beneficiary who believes the trustee has failed has a remedy against it. The relationship is fiduciary, it is not terminable at will, and in Nigeria a corporate trustee acting in that capacity operates under a licensing regime that exists precisely because the role carries other people's property.

The family office holds nothing. Its authority comes from an engagement letter, its duties are those it has contracted to perform, and its client is whoever signed — which may be the founder, the trust, the council, or a holding company. A beneficiary dissatisfied with the office has, in the ordinary case, no claim against it at all. It can be replaced next quarter without anybody's consent but the signatory's.

That asymmetry produces the operative rule, and it is worth stating in one line. Duties can be delegated; responsibility cannot. Where the office performs a function that belongs to the trust, it is acting as the trustee's agent, and the trustee's obligation to supervise what its agent does is not discharged by the fact that the family chose the agent, likes the agent, or pays the agent directly.

Three practical consequences follow immediately. A trustee that permits an office to make decisions the deed reserves to the trustee has not saved itself work; it has acquired an exposure. An office that accepts responsibility for a trust function without a written instruction has taken on an obligation it is not paid to carry and cannot be sued upon, which is the worst combination available. And a family that pays for both and defines neither is paying twice for a supervision that is not occurring.

The asymmetry runs to how each ends, which is where families are most often surprised. Replacing an office is a letter and a handover; replacing a trustee requires the mechanism the deed provides, the appointment of a successor, and the transfer of legal title to every asset held — a process measured in months and, where Nigerian land is involved, requiring consents that may take longer still. The two institutions therefore have very different life expectancies within the same structure. The office will probably be replaced two or three times in the life of a trust, which is a decisive argument against allowing the family's only authoritative record to live inside it.

The trustee owns and answers to beneficiaries. The office administers and answers to a signature. Every dysfunction in this essay is a consequence of one behaving as though it were the other.

The boundary is clear at the extremes. Nobody doubts that executing a transfer of trust property is the trustee's act, or that chasing a tenant's service charge is administration. The difficulty lives in three zones where both institutions have a legitimate claim to be involved.

The first is the record. Both keep one, and they keep it for different purposes: the trustee because it must be able to account to beneficiaries and to a court, the office because it administers assets day to day. Two records maintained honestly will still diverge — different valuation dates, different treatment of a capital expense, a disposal recorded in one and pending in the other — and the divergence is discovered at the worst moment, usually when a beneficiary compares two documents that should say the same thing.

The second is the asset itself. Who decides that the property should be let rather than sold? Who selects the tenant, negotiates the terms, and signs? Who monitors the manager? The decision is the trustee's; a great deal of what surrounds the decision is administration; and the line between the two runs directly through the middle of every ordinary transaction in the life of a Nigerian property portfolio.

The third, and the most sensitive, is communication with beneficiaries. The office is close to the family, is trusted, and is usually the first to be telephoned. The trustee carries the legal obligation to inform, and its statements have consequences. When the office answers a beneficiary's question about entitlement, it is speaking on a matter that is not its own, and a reassurance given casually in a corridor has a way of becoming the beneficiary's settled expectation and, eventually, a complaint.

A fourth zone deserves separate mention because it is where the largest sums are decided. Who selects an investment manager, who monitors performance, and who is answerable for an allocation that proves unwise? The investment power is the trustee's, constrained by the deed and by the statutory framework governing trustee investments, and it cannot be handed over informally to an office because the office is closer to the family and has views. The office may source, may monitor, may report, and may present a recommendation. The decision to place trust capital in an asset remains a fiduciary act, and it should look like one in the file.

Each of the following arises in the overlap zones. Each has a correction, and none of the corrections is expensive.

Everything above is prevented by a short document that almost no Nigerian family has: a written protocol between the trustee and the office, executed by both, and reviewed annually. It need not be long. It must be specific.

It states the authoritative register and its custodian. It sets out the calendar of recurring obligations with a named responsible institution against each, so that no line is unattributed. It defines the information flow in both directions — what the office sends the trustee and by when, what the trustee sends the office, and what each is entitled to ask for. It fixes the form and frequency of reporting to the family, so that a beneficiary receives one document rather than two that disagree.

It records the decision boundary explicitly: which matters the office may settle within its mandate, which it must refer, and the value or category thresholds at which referral is triggered. It provides an escalation path for disagreement between them, which will happen and should not have to be resolved by a family member. And it maps the fees, so that the family can see what each institution is paid for and satisfy itself that nothing is being paid for twice.

One provision deserves particular attention: who speaks to beneficiaries about entitlement. The right answer is almost always the trustee, in writing, with the office facilitating rather than answering. The office may confirm what has been paid; only the trustee should characterise what is due. That single line prevents most of the misunderstandings that later present as grievances.

And it should contain a short schedule for the event the whole architecture exists to survive: the death of the founder or of a principal beneficiary. Who is notified, by whom, within what period. Who secures the documents and the premises. Which institution instructs counsel on the estate as distinct from the trust. Who communicates with the family, and when the first meeting is convened. Thirty days of clarity here is worth more than any other provision in the protocol, because it is the only period in which both institutions will be operating simultaneously under maximum pressure, with the person who used to resolve their disagreements no longer available to do it.

A trustee without an office administers badly, at trustee rates. An office without a trustee administers property that somebody still personally owns. Two of them without a protocol supervise nothing, twice.

The protocol costs an afternoon. It is the cheapest document in this entire architecture.

A last practical question: which comes first, and does a family need both?

Nigerian families almost invariably acquire the trustee first, because settling a trust is the visible, advised, ceremonial step, and administration is neither. The result is a trustee performing — or more often not performing — an administrative function nobody engaged it for, at rates set for fiduciary work. The family concludes that the trust is expensive and does little, which is an accurate observation about the arrangement and an unfair one about trusts.

For most families the better sequence is the reverse of the customary one. Build the administrative capability first, in the virtual form described in the essay on the office — one competent coordinator, an inventory, a calendar, a reporting pack. Then settle the trust, with the office already able to tell the trustee exactly what it is receiving. A trustee handed a complete inventory on day one behaves differently for the following twenty years than one handed a bundle of documents and a promise that more will follow.

And both are genuinely needed only above a certain threshold of complexity. A family with one property, one company and two beneficiaries does not need two institutions; it needs a trustee that has been engaged properly, with an administration mandate written into its terms and priced openly. The two-institution structure earns its cost when there are multiple entities, multiple asset classes, foreign elements, or a beneficiary class wide enough that somebody must be answerable to the family in a way a fiduciary cannot be.

Two questions should be asked openly before either appointment, and Nigerian families ask neither often enough. The first is about fees: how each institution is charged — a percentage of assets, a fixed retainer, time costs, transaction charges — and what specifically falls outside the quoted figure. An itemised map of both, side by side, is a reasonable thing to request and an informative thing to read, because it usually reveals which functions nobody is being paid to perform.

The second is about the trustee itself, and the useful questions are unglamorous. How many trusts does the officer who will handle this one currently carry? What is the average tenure of that officer? Can the trustee produce a specimen of the reporting a beneficiary will receive? What has it done, in a real case, when a family instructed it to do something it considered improper? The last question is the most revealing, and a trustee that cannot answer it has probably never been in the position — which, over the life of a structure, is not a recommendation.

That completes the nine. The essay on the second-generation question names the failures; the seven institutional essays build what prevents them; and this one settles the boundary between the two that are most easily confused.

It is worth noticing what the sub-cluster as a whole has and has not required. Not one of the nine institutions demands a change in Nigerian law, an offshore structure, or a fortune. Each of them requires a document, a schedule, and somebody whose job it is. The whole architecture is, in the end, a set of arrangements for ensuring that nothing important depends on a particular person continuing to remember it — which is the only durable answer anybody has ever found to the second-generation question.

The trustee owns, and answers to beneficiaries. The office administers, and answers to a signature. Duties may be delegated; responsibility may not. All five dysfunctions are prevented by one register, one calendar, one reporting line, and a protocol nobody has written.

Families rarely lose wealth in the space where an institution failed. They lose it in the space between two institutions that each believed the other was watching.