Depreciation, or the loss that begins after the repair

In most lines, the loss ends when the repair is done. The pane is replaced, the roof is redone, and the property returns to its former state: that is what repair means. With a work of art it never happens. A canvas torn and then restored by the finest studio in Europe is not worth what it was worth the day before the accident, and the gap that remains is not a technical imperfection, it is a fact of the market. Insurance of valuables therefore invented a second head of indemnity, depreciation, which pays what stays lost once everything has been repaired.

It is worth understanding why the market penalizes a restored work, because the reason is not the one people assume. It is not that the restoration shows: good ones do not show to the naked eye. It is that any serious sale comes with a condition report, that ultraviolet examination reveals repaint, and that information circulates in a market where the same objects pass before the same experts. The buyer pays a premium for integrity, and withdraws it as soon as an intervention is documented. The consequence is disconcerting: between an excellent restoration and a merely competent one, the difference in value is smaller than one imagines, because what weighs is not the quality of the work but its existence.

From this comes a tension both parties discover mid-file. The insured has a duty to mitigate, therefore to restore. But a heavy restoration, reworking large areas, depreciates more than a minimal intervention that merely stabilizes. The act that reduces the first head increases the second, and both heads are paid by the same insurer under the same limit. There is therefore a point beyond which spending more on restoration costs more in total, and that point is not found by asking the restorer what they recommend, since they are the one participant in the file the trade-off does not affect.

The depreciation cap then complicates the picture rather than simplifying it. Many wordings bound that head, for instance at ten percent of the sum insured. On small damage the cap never bites and nobody notices it. On large damage it bites at exactly the moment depreciation becomes the principal loss: a severely damaged work can lose twenty-five percent of its value after restoration, and the insured recovers only ten. The cap therefore does not protect the insured against uncertainty, it transfers the tail of the distribution to them, which is precisely the part they believed they had insured.

That bound produces an inversion of interests one has to see coming. Beyond a certain level of damage, the restoration route becomes less favorable to the insured than a total loss, since the agreed value is paid in full whereas restoration returns only a fraction of the loss. Some wordings provide for exactly this, allowing the insured to claim a total loss and abandon the object where damage exceeds a proportion of the agreed value. Where that clause is missing, the parties negotiate in the shadow of the same arithmetic, and the negotiation bears less on the amount than on a question rarely asked aloud, namely who will keep the restored object.

Because there is a salvage, and it is forgotten because the word evokes crumpled metal. If the insurer settles a total loss, the damaged work belongs to it. A restored work with a documented history has a market: it is not worth zero, it is worth its value less the depreciation. The real cost of the total loss route is therefore not the agreed value, it is the agreed value less what the object will resell for. On a sought-after piece that salvage is far from negligible, and it brings the two routes much closer than intuitive arithmetic suggests. An adjuster who compares restoration against the agreed value without deducting salvage is comparing two things that are not comparable.

That leaves the point that decides everything and is settled before the loss. Depreciation is established by comparison, and comparables of restored works are rare by nature, since nobody willingly publishes the history of an accident. The only solid base is therefore the documented state of the work before the damage: a dated, photographed condition report with its ultraviolet imaging. Without it, the appraiser cannot measure a gap, having no starting point, and the head of loss the line invented to cover the irreparable collapses for a purely evidentiary reason. The insured is then indemnified for the restoration and for nothing else, on a loss that was nonetheless entirely real.