Ownership and risk: owning ruins somebody else must raise

Several modules of this course rest on a notion none of them sets out: knowing who owns the works while they are being built, and who bears the risk. The two questions look like one and are distinct. Ownership says who holds; the burden of risk says who loses if the thing perishes through nobody's fault. On a construction site they almost always separate, and it is that separation which explains otherwise incomprehensible situations, such as a project owner having to pay for works destroyed before it received them.

The general-law principle is that ownership of works built on land follows ownership of the land, by accession: what is incorporated into the ground belongs to the owner of the ground, as incorporation proceeds. The project owner therefore becomes owner of the foundations the day they are poured, with no deed, no acceptance, and no payment. That point surprises parties from other legal systems, where ownership transfers only on payment or delivery, and it has a direct consequence: the property insured by the construction policy already belongs, in large part, to the party that has received nothing.

The burden of risk follows another logic, a contractual one. In a works contract the contractor is generally bound to deliver completed works, and therefore bears the risk of loss until acceptance: if the works perish before it, even through no fault of its own, it must rebuild them to be paid. The module on acceptance described that switch from the angle of warranties; it must be added that it also moves that burden. Owner from incorporation, the project owner nonetheless does not bear the risk before acceptance, and it is precisely the gap between the two that organizes the project.

That configuration explains the structure of the policy, and notably a point that looks odd: the insured with an interest in the cover is not always the owner. The contractor, which will have to rebuild at its own cost, has an obvious insurable interest in works that do not belong to it. The project owner, the owner, has a lesser interest before acceptance and a considerable one after. That is why insured status, treated in its own module, is read according to the moment as much as to the person, and why a single policy bought by the project owner in practice covers a risk falling first on somebody else.

Payments blur that neat scheme, and that is where projects get complicated. Interim payments, payments on materials on account and retention create situations where the project owner has paid for items it does not yet own, because they are in the shop or in transit, and the module on value outside the fence showed what happens to them. A title transfer clause on materials on account, frequent on major equipment, therefore advances ownership without advancing the burden of risk, creating a third interval where the two notions are misaligned.

The case of works built on another's land deserves isolating, because it undoes the accession mechanism. Under a concession, a building lease or an occupation of public property, the works may belong to the builder for the whole term of the contract and revert to the landowner only at its end, sometimes thirty or fifty years later. The construction policy then insures property whose owner is not the one assumed, and the question of who receives the indemnity, treated in the lenders module, arises here for an entirely different reason.

What must be done lies in a question asked once, at start-up, whose answer fits in three lines. Who owns the works during construction, and by what mechanism. Who bears the risk before acceptance, and does the contract depart from the principle. Finally, are there items paid for but not incorporated, and where are they. Those three answers change no cover, but they say who has an interest in what, and that interest is the very condition of insurance: one cannot insure what one has no interest in, and on a construction site the interest moves while one builds.