Here is a piece of arithmetic almost no family runs before it is too late, because running it feels disloyal.
Here is a piece of arithmetic almost no family runs before it is too late, because running it feels disloyal.
A founder builds an estate worth, say, 500 million shillings: a business, two rental properties, a farm. She has five children and, being fair, leaves each an equal fifth. Each child now holds 100 million. Respectable, but already not the kind of holding that changes a life; it is the kind that gets consumed by it. Now let each of those five children have four children of their own and repeat the equal split. Twenty grandchildren each inherit 25 million, minus the funeral costs, the succession disputes, and the properties sold cheap because five owners could not agree. In two generations of perfectly fair division, a fortune became school fees. Run it once more and the great-grandchildren inherit a story.
Nothing went wrong in this family. Nobody stole, nobody gambled, nobody married badly. The estate was destroyed by division itself, by the simple fact that families multiply faster than assets do. Call it the sibling multiplier: every generation, the denominator grows. Fabian Pfeffer and Alexandra Killewald's "Generations of Advantage" (Social Forces, 2017) found that direct gifts and bequests carry only 12.3 percent of the wealth correlation between generations, well behind homeownership at 28.4 percent and education at 25.5 percent. Part of the reason the bequest channel is so weak is exactly this: what finally gets bequeathed has usually been divided until it cannot do anything.
Families have faced this math for as long as families have owned things, and they have found exactly three responses.
This is the default, and it feels like love. Equal shares for every child, each free to do as they wish. Its virtue is real: it prevents the resentment that unequal division breeds, and it is what most legal systems will impose anyway if you die without a plan.
Its cost is the arithmetic above. Splitting shrinks each share, and it also destroys the properties that made the asset productive. A business divided five ways loses coherent management. A building with five owners cannot decide on a repair. A farm split into strips loses the scale that made it a farm. Past a certain point the asset stops being the kind of thing that generates wealth at all. Split and scatter is the fastest route from an estate to a memory, and it is what happens when a family makes no decision, which is why it is the most common outcome on earth.
The old answer concentrates instead: one heir, usually the eldest son, takes the productive asset whole, and the others adjust. It solves the arithmetic perfectly. The denominator stays at one, forever.
But it solves it by manufacturing injustice on schedule. Every generation, all children but one are disinherited by birth order, a rule that selects for neither competence nor commitment, and the excluded siblings do not simply vanish; they carry the grievance into the family's next fifty years. Much of the world has also legislated against it: statutory intestacy regimes across Africa now divide estates among widows and all children, so the eldest's automatic claim is often legally void the moment the founder dies without documents. Concentration by birth order buys the asset's survival at the price of the family's, and frequently ends up with neither.
The third way severs the assumption both other responses share: that inheriting means receiving a piece you can carry away. Instead, the productive asset is never divided at all. It is placed in a structure, a family company, a holding, a trust, and what the heirs inherit is shares in the structure and the income those shares pay. Everyone owns; nobody carves. The farm stays one farm with one manager and twenty shareholders, instead of twenty strips with twenty owners and no farm.
The family business literature has mapped this path in detail. Kelin Gersick, John Davis, Marion McCollom Hampton, and Ivan Lansberg, in Generation to Generation: Life Cycles of the Family Business (Harvard Business School Press, 1997), built on the three-circle model that Renato Tagiuri and John Davis developed at Harvard Business School in 1978, which separates family, business, and ownership into overlapping but distinct systems. Their developmental insight is the one that matters here: ownership must change form as the family multiplies. Stage one is the controlling owner, the founder. Stage two is the sibling partnership, a handful of brothers and sisters holding jointly. Stage three is the cousin consortium, dozens of shareholders across branches who may not even know each other well. Each stage needs governance the previous one did not: a sibling partnership can run on a phone call, a cousin consortium needs a shareholders' agreement, a board, a dividend policy, and rules for selling shares. Families that fail, in this framework, are usually families that reached stage three while still governing like stage one, with assets held loosely in a dead founder's name and decisions made by whoever shouts loudest at the burial.
The most complete working example in the corpus is the Mulliez family of northern France, owners of Auchan, Decathlon, and Leroy Merlin. Around 994 family shareholders own a portfolio of roughly 150 companies employing over 615,000 people, and no member inherits a company or a building. They inherit shares in the family association, tradable only inside the family at a formula-set price. Five children or ten, the denominator can grow without the assets ever fragmenting, because what multiplies is the shareholder register, not the asset. The Mulliez did not escape the sibling multiplier. They rehoused it where it cannot do damage.
If the arithmetic sounds abstract, drive through any densely settled farming region of East Africa and look at the field boundaries. Land fragmentation is the sibling multiplier made physical: each generation, the family plot is subdivided among heirs, and the strips narrow until a farm becomes a hedge dispute.
The World Bank studied this directly in Benoit Blarel, Peter Hazell, Frank Place, and John Quiggin's "The Economics of Farm Fragmentation: Evidence from Ghana and Rwanda" (World Bank Economic Review, 1992), which documented how widespread multi-parcel, subdivided holdings had become in Sub-Saharan Africa, driven in large part by inheritance and population pressure. The authors were careful, and we should be too: fragmentation is not pure loss, since scattered plots let farmers spread risk across soils and seasons. But the direction of travel has alarmed governments enough to legislate. Rwanda, the continent's most densely populated mainland country, wrote the arithmetic into law: its 2005 Organic Land Law prohibited subdividing agricultural parcels below one hectare, a rule its later land legislation retained. When a state has to ban families from splitting land among their own children, the sibling multiplier has stopped being a private planning question and become national policy. Kigezi in Uganda, the Kenyan highlands, Malawi: the pattern repeats wherever fertile land meets large families and equal division.
Notice what the land case proves: response one is not a neutral default. Left alone, the math runs to completion. Every generation that says "we will divide it fairly and let everyone farm their piece" is one generation closer to pieces that cannot be farmed.
You do not need Auchan to use the third way. The structure scales down further than most families believe.
A family farm can become a family company: the land titled to the company, heirs holding shares, one member or a hired manager farming it whole, profits paid as dividends per share. A matrimonial home plus two rentals can sit in a holding whose shares pass by will while the buildings never split. Even the family's shared savings pots are the same logic in miniature: nobody carves out their fraction of the education pot; members hold claims on a whole that stays whole. The essential moves are always the same three: put the productive asset inside a structure, convert heirs from owners of pieces into owners of shares, and write the governance now, while the founder is alive, including the rule the Mulliez consider sacred, a formula price and an internal market so a member who wants out can sell without forcing a breakup.
What the third way costs is honesty. The family must say aloud that the farm will never be cut, that no child will get "their piece," and that fairness will be delivered in income and shares rather than in fence posts. Some families cannot say this. Their grandchildren will inherit the hedges.
Count your heirs. Not abstractly: write the number of children, and under it your best guess at the number of grandchildren, and divide your estate by the second number. That figure, minus disputes and forced sales, is what split and scatter delivers.
Then decide which response your family is actually running. If you have made no decision, you have chosen response one, because it is the default the law and the funeral will execute for you. The alternative must be built while you are alive: the structure, the shares, the written rules. Keep the asset whole and share the income, or accept, with open eyes, that you are the founder of an estate designed to disappear by division, on schedule, two generations from now.