When the family office sees everything and says nothing

There is a recurring moment in succession planning meetings that anyone who has watched a few industrial families from the inside recognises immediately. Around the table sit the seventy-year-old founder, two or three children, the head of the family office with the team, occasionally an external legal adviser and the banker holding the principal relationship. The structure of the generational transfer is discussed: trust, holding companies, share distribution, governance of the vehicle, and the conversation is usually technical and precise. Yet beneath the technical conversation a second conversation runs silently, the one each person in the room is having inside their own head, for instance the fact that the designated successor has neither the interest nor the substance for the role, the founder knows this perfectly well but will never admit it, the founder’s wife disagrees with the proposed distribution but will not say so in front of the others and the head of the family office has been watching the cracks open for two years without finding the right moment to raise them.

The meeting closes with next steps defined, timelines set, the follow-up booked for the month after, and everyone leaves satisfied with the technical progress and no one has said what needed saying. Three years later, when succession activates in earnest, the legal structure holds up perfectly and the family finds itself in mediation.

This pattern is not an accident, not a failure of character and not professional negligence either, but a property of the family office’s position relative to the system it serves, and this is where it is worth pausing for a moment because the right diagnosis changes the prognosis.

The British reader will recognise something familiar here. The wave of dynastic disputes that has reached the public record over the past decade, from the Murdoch settlements to the more recent Singapore and Hong Kong wealth battles covered at length in the FT and the Economist, almost never turns on technical failure. The trusts were drafted properly, the holding structures were sound, the tax position was defensible, and what broke was something the technical work could not address, and was never asked to address, and the architecture that produced the technical work could not have addressed even if asked.

The family office, in its typical configuration, is built to optimise what is measurable, meaning portfolio returns, tax efficiency, the quality of the legal structure, the coherence of consolidated reporting and so on. These are crucial dimensions and an office that does not look after them is not doing its job, because the problem is not what the office does but what its structural architecture tends to leave outside the analytical perimeter, and what it leaves outside almost always concerns succession.

Three asymmetries operate beneath the surface and reinforce one another. The first is temporal: the strategic choices of a family office are evaluated in annual or quarterly reporting windows, while the consequences of succession choices emerge ten or twenty years later, when those who made the decisions are often no longer operational. The decision-maker does not pay the cost of the error and whoever will pay the cost was not at the table when the decision was taken. A flawless legal structure architected today may turn out to be a straitjacket in fifteen years, but the head of the family office who designed it will have retired or moved to another mandate.

The second asymmetry is epistemic: the family office values by construction what is measurable and tends to treat what is not measurable as noise, for instance the dynamic between the founder and the first wife, the accumulated resentment of the cadet branch, the psychological fragility of the next generation playing manager at twenty-five without being one, are all structural facts of the family system with enormous patrimonial consequences, but because they do not fit on a spreadsheet they do not enter the decision-making process either.

The third asymmetry is the most delicate and deserves to be handled without rhetoric. The family office operates inside an incentive structure where the person who pays, the person who is served and the person who appraises performance are often the same person or variations of the same person. Telling the seventy-year-old founder that he is unconsciously sabotaging the succession because he does not really want to let go of control, or telling the patriarch that the designated son lacks the capacities being attributed to him, is technically possible but professionally costly. Not because the head of the family office is weak in character, but because relational equilibrium with the principal is the condition for keeping the mandate and uncomfortable candour is not rewarded by the system; rather, it is punished with a speed that anyone who has worked in the role knows well. Silence is not collusion but the rational equilibrium of an agent operating inside a system where the principal’s embarrassment has immediate consequences for the relationship, while the consequences of unspoken truths emerge years later and tend to be attributed to other factors. Over the years I have seen this dynamic operate almost identically in London, Geneva, Milan and Miami, with different cultural inflections but the same architecture underneath.

The recent public data tells the same story in aggregate. Only just over half of global family offices have a formalised succession plan and of those only a portion is actually written down rather than verbal. Sixty-two per cent of families who initiate a succession process fail to complete it within the planned timeline and forty-three per cent of those who do complete the technical transfer report significant family discord within three years. These are not numbers that can be explained by technical incompetence on the part of the offices involved, given that the professionals who operate in this segment are on average very capable and the legal and tax structures they produce are often impeccable. They are explained by a wrong sequence, meaning that planning systematically begins from the measurable, fiscal and structural dimension and defers or omits altogether the diagnosis of the family system. The work begins where the light is on and leaves in the dark the part that actually matters.

There is then a fourth dynamic worth naming because it is the least discussed in the literature of the sector. The family office is inside the system it serves, lives in daily proximity to the family, knows the grandchildren by name, attends the Christmas dinners, is invited to the funerals. This proximity is an enormous relational asset but it is also an analytical constraint, because whoever is inside does not see the system, they inhabit it, and the question here is not whether the head of the family office is sufficiently intelligent, but rather whether his position structurally allows him to bring to the principal a perspective that requires distance, a perspective that implicates the principal himself as part of the problem to be analysed. More clinically, no system can govern itself fully from within without a separate level of analysis, or to put it differently, in successions what is needed is someone who does not carry the structural constraint of remaining in equilibrium with the principal, because the problem to be diagnosed is precisely that principal and the way he is managing or avoiding his own exit from the stage.

This does not mean the family office should be replaced; on the contrary, it means that the family office alone is not sufficient for the class of problems that complex successions pose. The distinction is important and must be stated clearly: the family office is the necessary daily presence, knows the family, holds the institutional memory of decisions, runs the operational structures. What it cannot structurally do, because its position does not allow it, is bring the external perspective that treats the family system as an object of analysis in its own right, with the capacity to tell the founder the things that whoever lives within his perimeter cannot afford to say. These are two different levels of work, complementary, and conflating them is expensive.

Whoever runs a family office and reads these lines probably recognises at least one concrete situation in which they thought “this should be said” and chose, reasonably, not to say it. That choice was not wrong given their position but it was wrong given the absence of a second level that could say it in their place. The successions that work over the next twenty years will be the ones where someone, from a position of sufficient externality, will have done the work of naming what the family office sees and cannot say. Whoever fails to build that second level will find themselves structurally exposed to a specific kind of failure that is technically impeccable, humanly disastrous and patrimonially costly. It is the failure the numbers are already describing and it will continue to repeat itself for as long as succession is thought of as a problem to be solved within the perimeter of the family office, rather than through an architecture of analysis that includes it and exceeds it.