The Accounting Runs Backward

Here is what most of us believe happens when a family estate is divided or a marriage is dissolved. Somewhere, a professional sits down with the file, lists everything the family owns, puts an honest value on each item, adds it all up, and divides the total according to each person's rights. The math produces the outcome. The document you sign at the end is the math, written down.

Hold that belief up to the light, because the most important discovery in Celine Bessiere and Sibylle Gollac's The Gender of Capital: How Families Perpetuate Wealth Inequality (Harvard University Press, 2023) is that, in practice, the process usually runs in exactly the opposite direction. The two French sociologists spent more than twenty years doing something almost no researchers manage: they got inside the closed offices where estates and divorces are actually settled, watched the professionals work, and asked them to describe their methods in their own words. The professionals, secure in rooms where no client would ever contradict them, obliged. What they described, the authors gave a name that belongs in every family's vocabulary: reversed accounting.

One clarification before the evidence, because it matters for reading everything that follows. The professionals in this book are mostly notaires, and the word does not mean what "notary" means to an American reader. A notary public in the United States mostly witnesses signatures. A French notaire is a licensed legal official the state requires you to use for property transfers, marriage contracts, and estates: a person who drafts and authenticates the deeds, values the assets, and advises the family, all at once. When a French family divides wealth, the notaire is the room the division happens in. And one note on names: the authors changed the names of the professionals and private families they studied, so the people quoted below are real, but pseudonymous.

The law describes five steps. The professionals described doing them in reverse.

On paper, the sequence is arithmetic and innocent, and it is worth stating cleanly because everything turns on its order: first the inventory of the property, then the valuation of each item, then the total and the calculation of each person's share, then the division into lots, and finally, if someone's lot is worth more than their share, a balancing payment to compensate the others. Count, value, divide, compensate. The math produces the outcome.

Now listen to the notaire the authors call Pierre Delmas, describing to a room of twelve jurists and sociologists how farms in his region have been passed to a single male heir: "The way we used to do it, it was very simple: by underestimating assets, by setting exaggerated deferred wages, and by using the disposable share of the estate, of course." Each tool he names is legal. Deferred wages exist so a child who worked unpaid on the family farm can be compensated at settlement; the disposable share is the slice of an estate a parent may freely direct. What is not in the law is the direction of travel. Delmas was equally plain about that: "In practice, we look at how much the person taking over the farm can give, and we make the succession fit. And we rarely go to court, because everyone accepts it."

Read that again. The starting point is not what the estate is worth. The starting point is what the chosen heir can afford to pay his siblings. Everything else, the valuation of the land, the size of the deferred wages, even what gets counted in the inventory at all, is then worked backward from that number until the paperwork appears to satisfy the law.

Divorce runs on the same reversed engine. Anne Prisot-Gallot, a lawyer who practiced family law in Paris for forty years, described her own work dividing couples' property in a written reflection she gave the researchers: the work consisted of gathering the elements for the accounting, finding an agreement between the parties, and, in her words, presenting things in such a way as to arrive at "the outcome upon which they had both agreed." The outcome comes first. The accounting is then presented, her verb, so the outcome looks like its conclusion.

Different professions, different bodies of law, one method. The family and its professional first reach a consensus about who gets the thing that matters. Then the inventory, the valuations, and the compensations are assembled backward until the document looks equal. The signature page shows you arithmetic. The arithmetic was reverse-engineered.

The prized asset is decided first, and the law has spent a century making that easier.

Reversed accounting is not a rogue habit; the legal system evolved to accommodate it. At the heart of every settlement, the authors show, sits what they call a structuring asset, the thing that must be kept: the farm, the business, the family home, the block of shares. France's original Civil Code of 1804 actually demanded equality in kind, each heir receiving assets of the same nature, and if an asset could not be split, it was to be sold. Across the twentieth century that principle was dismantled, one asset class at a time: modest housing from 1922, small farms from 1938, all farms and businesses by 1961, company shares in the early 1980s, each becoming eligible for "preferential allocation," meaning one heir takes the whole asset and owes the others a payment. By 2006, French law dropped any reference to equality in kind: shares need only be equal in value, on paper. And who receives the structuring asset is not random. In the French data, sons, especially firstborn sons, take the business, the land, and the house at consistently higher rates, while daughters are compensated in money, the adjustable variable that reversed accounting uses to balance its books. We unpack those statistics in It's Just Like Salic Law; here the point is simpler. The century-long drift of the law converted heirs into two classes: the one who receives the thing, and the ones who receive a number.

A young notaire the authors call Marc Pouget, 30 years old and newly a partner in his father's large rural firm, explained the resulting freedom with disarming cheerfulness, speaking as if planning his own estate: "Even if I have a family disagreement, I can institute one of my children as sole legatee in my will, if I have a problem with another, it's not a big deal. At least the one to whom I left everything, he'll take possession of the inheritance, and then, it's a matter of evaluating the assets, so, an economic negotiation."

Sit with the order of operations in that sentence, because it is the entire mechanism spoken aloud. First, possession: one child takes everything. Then, afterward, "a matter of evaluating the assets." Valuation is not the referee that decides who gets what. Valuation is the negotiation that happens after possession has already been decided, conducted between a sibling who now holds the asset and siblings who hold a claim, which is not a negotiation between equals.

Be careful to keep the professionals in focus for what they are. The notaires in this book are not cartoon villains stealing from daughters. They are consensus manufacturers. Their deepest professional instinct, documented across the book, is to keep families out of court and keep the kept thing in the family. When the family's unspoken preference is that the son takes the farm and the daughters accept payments and peace, the professional's craft consists of making that preference arithmetically presentable. The bias enters the room with the family. The professional's contribution is to dress it as math.

Our translation: the elders' meeting runs the same accounting, without the paperwork.

Everything above is French research, built on French civil law, and the brief statistics we cited are French. What follows is our translation, ours alone, into the settings LegacyPot writes for. The authors studied no African family and make no claim about one.

But test the shape against what you know. When an African family divides land or a business, the real decision is rarely made in any office. It is made in an elders' meeting, in a sitting room after a funeral, in a series of phone calls between brothers. Who will take the land is settled first, and it is settled by considerations that sound exactly like Delmas's region: who carries the name, who stayed, who can manage it, what the family has always done. Then, and only then, comes the accounting: the surveyor who somehow values the home plot modestly, the daughters' shares defined in cash or in a smaller plot elsewhere, the brother's years of living in the compound never counted as rent, the sister's years of remittances never counted as investment. If a lawyer or a land board is involved at all, it is usually at the end, to formalize what the meeting already decided. The sequence is reversed accounting without the notaire, and the diaspora member who wires money for years and then reviews the final "fair" document from abroad is its most reliable casualty, because they see only the arithmetic, never the meeting.

So treat every settlement document, every will, every valuation report that reaches you as the output of a process, and interrogate the process, not the arithmetic. The arithmetic will always check out; it was built to. Four questions do the real work. Which came first, the outcome or the math: was the person taking the main asset known before anything was valued? Who chose the valuer, and does that person benefit from a low number or a high one? What is missing from the inventory: the early gift, the business bought for one child years ago, the unpaid labor, the debt quietly forgiven? And what would each asset fetch from a stranger, on the open market, rather than in a transfer between relatives where the price is negotiable and the pressure is not?

Asking these questions out loud, in front of the family, is uncomfortable. That discomfort is the fee. The Pilon sisters in this same book, watching a bakery pass to their brother with a decade of free bread as their compensation, chose silence because, as one said, above all they did not want to start a quarrel. Their silence did not prevent the unfairness. It only prevented the unfairness from being discussed.

The defense against backward math is a forward inventory.

Here is where we go beyond the book, and we say so plainly: Bessiere and Gollac diagnose; the prescription that follows is ours. Reversed accounting has one structural weakness. It depends on the inventory being assembled late, after the outcome is chosen, by the people who chose it. An inventory that already exists, written down over years, witnessed by the whole family, cannot be quietly built backward. The pastry business bought for the son at 14 cannot vanish from an accounting it has been sitting in for thirty years. The daughter's decade of remittances cannot be reclassified as gifts if the ledger has called them contributions all along.

This is precisely what the Document Vault in LegacyPot exists to hold. Keep the running inventory there: every major asset, every title and deed, every significant gift to any child with its date and rough value, every family member's unpaid labor or cash contribution to a family enterprise. Not as an accusation, but as a shared memory that no single meeting can rewrite. When the division eventually comes, and it always comes, the family that opens a decades-old shared inventory has flipped the sequence back to the one the law intended: count first, value second, divide last.

The myth this article set out to break is the belief that the document is the decision. It is not. The document is the costume the decision wears. Your family will one day produce such a document, drafted by a professional or an elder, balanced to the last shilling or euro or dollar, and everyone will be invited to admire how even the numbers are. When that day comes, be the person at the table who asks the only question that matters: did we count first and decide after, or decide first and count backward? Families that can answer that question honestly divide wealth. Families that cannot, divide themselves, and call the paperwork proof that they didn't.