The Trust That Isn’t

There is a moment that recurs in cross-border deals between London and Bucharest, and it is always quietly revealing. An English party reaches the point in the structure where, at home, they would say: it will be held on trust. The shares, the deposit, the security, the inheritance. And the Romanian side pauses — not because the idea is unfamiliar, but because the socket the English lawyer is reaching for does not exist in the wall.

This is not a language problem. Trust translates perfectly well into Romanian as fiducie. It is a concept problem — and concept problems are where the real money, and the real risk, live. Understanding why the trust has no native home in Romanian law, and what Romanian practice does instead, is one of the more useful things a dealmaker working across the two systems can carry in their head.

Why civil law has no socket for the trust

The English trust is a child of equity. For centuries England ran two parallel court systems, and out of that division grew something structurally peculiar: the idea that ownership itself can be split. The trustee holds legal title; the beneficiary holds the equitable interest. Two people, one asset, two genuine forms of ownership at once. To an English lawyer this is so ordinary it is invisible. To a civil lawyer it is close to a category error.

Romanian law, like the French law it descends from, rests on three pillars that together leave no room for that split. Ownership is unitary — proprietatea is one, full and indivisible, and cannot be carved into a legal half and an equitable half. The catalogue of real rights is a numerus clausus — fixed by law, so private parties cannot invent the beneficiary’s equitable interest as a new property right. And the classical theory of patrimony holds that each person has one and only one patrimoniu, a single estate answering to all their creditors — whereas a trust needs assets walled off into a sub-estate the trustee’s own creditors cannot touch.

Take the three together and the trust is not merely absent from civil law but structurally repelled by it. No Chancery, so no equity; no equity, so no beneficial ownership; one patrimony, so no ring-fence. The trust is not a missing feature — it is an answer to a question civil law decided long ago to answer a different way.

Romania imported a trust — then made sure no one would use it

Here is the twist most people miss. Romania does have a statutory trust. When the new Civil Code came into force on 1 October 2011, it imported, almost directly from the French fiducie, an institution called fiducia (articles 773–791). On paper it does what a trust does: a constituitor (settlor) transfers rights to a fiduciar (trustee), who holds them in an autonomous, ring-fenced patrimony for a beneficiar. The single-patrimony dogma was, for this narrow purpose, deliberately breached. The socket was installed.

And yet in fifteen years it has remained very largely unused — not for want of appetite, but because the legislator wrapped it in conditions that strangled its two most natural uses. Under article 776, a fiduciar can only be a bank, an investment or insurance firm, a notary or a lawyer; the trusted individual who in England would routinely hold shares on trust simply cannot. And article 775 voids any fiducia used to make an indirect gift — a single line that amputates the largest use of trusts in the common-law world, estate planning. Add notarised form on pain of nullity, registration duties, and a fiscal regime that never offered clear neutrality, and the market did the rational thing: it ignored the instrument. Avem fiducia, dar practica nu o folosește.

What practice does instead

When a tool is missing, transactional practice does not stop; it improvises. The key is to stop asking “where is the trust?” and start asking what function was the trust performing here? — because civil law delivers each function through a different, purpose-built instrument. Where common law divides ownership, civil law multiplies entities and obligations.

A deposit held against a condition becomes a notarial deposit or a bank escrow — the bank a contractual stakeholder, not an equitable owner. Security for a fluctuating lender syndicate, the job an English security trustee does, is rebuilt three ways: joint and several creditorship (solidaritate activă), where the agent holds a direct claim of its own; a plain security mandat; or, under foreign-law facilities, the parallel debt. An asset held for a hidden owner uses the mandat fără reprezentare or simulație with a secret counter-deed — effective, but the hidden party holds a claim against a person, not a right in the thing: a meaningful downgrade in insolvency. Ring-fencing is achieved not by splitting an owner but by adding one — the SPV. And succession routes around the trust entirely, through gifts with charges, the split of nuda proprietate and uzufruct, holding structures, and life insurance.

The cross-border sting

One trap surprises even sophisticated clients. Romania is not a party to the Hague Convention of 1985 on the Law Applicable to Trusts and on their Recognition. So a perfectly valid English or Jersey trust does not arrive in Romania with a recognition framework waiting for it. A Romanian court must characterise it through ordinary private international law — and may look through the trust to its underlying relationships rather than give effect to it as a trust. The beneficiary who assumed their equitable interest would be honoured may find the forum unable to locate the right into which to translate it. That is not a reason to avoid trusts cross-border; it is a reason to design them knowing where, and under whose law, they will need to bite.

General comparative information, not legal advice, and no lawyer–client relationship is created. The instruments above carry conditions, formalities and tax consequences, and cross-border recognition turns on the facts; any specific structure needs advice on its own facts and on the current state of the law.