Measuring the loss: euro for euro, or multiplied

The breach is established, the cover responds, disclosure does not defeat it. That leaves the question that decides the amount, and it has not one answer but two, separated by a factor that commonly reaches five or ten. The first measure is the cost of putting it right, euro for euro: the liability was unrecorded to the tune of five hundred thousand, the loss is five hundred thousand. The second is the diminution in value of what was bought: if the price was set at nine times earnings and the breach reveals earnings overstated by five hundred thousand, the buyer paid four and a half million too much. Both measures are defensible, and most disputes about quantum are in reality disputes about which one applies.

One must first understand why the multiple is not abusive, because an underwriter hearing it for the first time sees an attempt at enrichment. The buyer did not buy a balance sheet, it bought a stream of earnings, and it paid a multiple of that stream. If the stream was misstated, the price was wrong by that same multiple: it is the arithmetic of its own offer. The seller's answer is symmetrical and serious: the multiple expressed the buyer's view of the future, its synergies and its competition in the auction, and the seller never warranted that view. The debate therefore concerns the nature of the multiple, market fact or buyer's judgment, and it is not settled in the abstract.

What settles it in practice is not a rule of law but a question of fact about the business, and this is the most useful point in the module. A one-off breach, a tax assessment on a closed year, a single piece of litigation, a forgotten liability, is a balance sheet item: it does not recur, it does not affect the earnings the buyer capitalized, and it is indemnified euro for euro. A breach affecting recurring earnings, revenue durably overstated, a permanent cost understated, a structurally lower margin, touches exactly the quantity of which the price was a multiple, and diminution in value is its proper measure. Classifying the breach as recurring or one-off is therefore worth more than any other argument in a quantum discussion.

Wordings have come to address the subject, and it should be checked at placement rather than discovered at the claim. Some policies expressly confirm that loss includes diminution in value, computed on a multiple where appropriate. Others just as expressly exclude loss computed on a multiple. The presence or absence of one sentence changes the order of magnitude of every future claim under the contract, and it is negotiated at a moment when nobody has a claim in mind, between two points that look more urgent. It is the same shape of trap as the tax gross-up clause, and it is worth handling them together since they are negotiated at the same moment with the same counterpart.

A series of adjustments then turns a simple figure into an accounting exercise, and they all run the same way. Deducted are sums recovered from third parties, indemnities received under another policy, tax savings arising from the charge itself, and any benefit the buyer derives from the remedial work where it leaves the business in a better state than it should have been. Each of those items is legitimate and each opens a technical discussion of several weeks. That is where most of the elapsed time of a settled file is consumed, long after the principle of cover has ceased to be contested.

Proving a diminution in value finally raises a difficulty of a particular kind: one must reconstruct the negotiation that would have taken place had the buyer known. It is a counterfactual, as in political risk, but the world to be reconstructed is not an economic environment, it is a discussion between two parties one of which had competitors. The documents establishing it all exist, and they belong to the buyer: its valuation model, its investment committee papers, its successive price ranges, its negotiation notes. They will be produced.

That last point deserves to be retained as a rule of conduct rather than as a remark. The valuation model and the investment committee paper are claim documents from the day they are written, months before any claim exists. If they show the buyer's analysts valuing the business in a range of seven to nine times, and nine being adopted to win an auction, the insurer and the seller will argue that the real diminution in value is computed at the bottom of the range. The buyer's own internal documents thus set the ceiling on its own claim, and nobody, at the moment of writing them, writes them with that in mind.