The natural catastrophe module ended on a regime where prevention is not priced at all, the surcharge being uniform and indifferent to the care taken. The specialist market is the exact opposite pole: prevention is the main object of an underwriter's attention there, and it is priced. It is worth understanding what they look at, because the list of points examined and the list of points that actually weigh on the premium are not the same list, and the gap between them surprises everyone.
What a risk survey examines is fairly stable: the stability of the climate rather than its level, since it is humidity swings that work on panel and canvas far more than any average value; cumulative light exposure; the nature of hangings and mounts; racking and packing in store; water detection and its position relative to the works; security and access; and finally siting, meaning the room's elevation and its exposure to water ingress.
The mechanism that makes this module interesting rests on a mismatch. Almost all of these measures reduce the FREQUENCY of small damage: stable humidity avoids splits, good hanging avoids falls, correct packing avoids abrasion. Yet the underwriting decision on a collection is dominated by the SEVERITY of a single event at one place, as the mass loss module showed. A collector can therefore seriously improve everything that bears on frequency and see the premium move very little, which they experience as unfairness when it is an arithmetical consequence.
The two measures that actually move severity are precisely those collectors resist, and they should be named without being recommended: splitting the collection across several places, and getting out of a basement. One frustrates the pleasure of having the works together, the other is expensive and often means moving. Together they nonetheless weigh more than all climate improvements combined, because they act on the number of objects a single event can reach. The useful conversation with an underwriter therefore bears on geography, not on technique, which is never the conversation the collector prepared.
A point of law hides in the survey report and it changes the nature of everything above. An accepted recommendation can become a condition of the contract, and breaching a condition does not have the same effects as departing from good practice: depending on the drafting and the applicable law, it can defeat cover entirely. The difference between 'recommended' and 'required' in a survey report is therefore a difference of cover, written in a document nobody treats as contractual and that the collector often files without rereading.
One must then see how prevention meets proof, for that is the link this module adds to the previous ones. A stable climate slows deterioration, but above all it produces an object whose successive condition reports are comparable with one another. That is exactly what the depreciation module required: with no documented state before the damage, the gap cannot be measured and the head collapses for an evidentiary reason. Preventive conservation and provability of loss are therefore the same discipline seen from two sides, and a collector who holds one holds the other at no additional effort.
In practice three actions sum up the module. Treat the survey report as a contractual document and explicitly separate what is recommended in it from what is required. Know that the two most effective severity reducers are geographic rather than technical, even if one sets them aside with open eyes. And keep condition reports over time, because they serve twice, once to conserve and once to prove. None of these three requires an expert; all three require reading a document one already has.
A collection worth 8 million euros is kept in a basement store. The March risk survey makes four points: install floor water detection, raise the racking by twenty centimeters, stabilize humidity, and split the collection across a second location. The report classes the first three as recommendations and the fourth as a requirement, cover being granted subject to implementation within six months. The collector carries out the first three, judged useful and inexpensive, and sets aside the fourth. In November, rising groundwater damages forty works for 1.3 million euros.
The collector did what seemed reasonable and set aside the one point that engaged his cover, which is the exact error the survey report was meant to make impossible and which it instead makes likely by mixing two natures in one document. The first three points are recommendations: not doing them would have weighed on the premium at renewal, nothing more. The fourth is a requirement with a deadline, therefore a condition, and failing to perform it opens to the insurer an argument of an altogether different reach than the usual discussion about amount. That the collector did three things out of four compensates nothing, because the three he did acted on the frequency of small damage and the fourth was the only one acting on severity, that is, on the number of works a single event can reach. The loss that occurred is precisely the one the condition had been set against. Two qualifications deserve to be seen. The water detection installed in March may have shortened the immersion and therefore the extent of the damage, which can be documented and usefully argued. And the exact reach of the breach depends on the drafting and the applicable law, a condition not having the same effect everywhere; that is the first thing to have checked rather than the last. That leaves the lesson that holds for everyone: the line separating 'recommended' from 'required' sat in a technical report from March, filed without being reread, and it was worth 1.3 million.
- 01Almost every conservation measure reduces frequency, while underwriting a collection is dominated by severity at one place: the premium therefore moves little.
- 02The two severity reducers that really weigh are geographic, splitting the collection and leaving a basement, and they are the ones people resist.
- 03An accepted recommendation can become a condition: in a survey report, the gap between 'recommended' and 'required' is a gap in cover.
- 04A stable climate above all makes condition reports comparable over time, therefore depreciation provable: conserving and proving are the same discipline.
- 05The survey report is a contractual document, and it is the least reread document in the whole file.