“Insurance” Provision in Transfer Pricing Agreement

ADIR v Apotex 2015 FC 721 Gagné J
            1,341,196 / perindopril / COVERSYL

This case is the accounting of profits portion of a bifurcated trial. In the liability decision, Laboratoires Servier v Apotex Inc, 2008 FC 825 aff’d 2009 FCA 222, Snider J held that the defendants had infringed the plaintiff’s ‘196 patent, and she allowed the plaintiffs to elect an accounting of profits [1]. This decision is primarily of interest for two points: the treatment of payments made on account of what was argued to be liability insurance provided by Apotex to its foreign purchaser to protect against infringement of a foreign patent, which is dealt with in this post, and the treatment of non-infringing alternatives, which I will consider in my next post.

Apotex Pharmachem manufactured infringing perindopril in Canada which it sold to Apotex Inc, which in turn formulated the API and sold the finished product in both Canadian and export markets [7]. The parties largely agreed as to the revenues from these sales, with one $22 million exception [16]. Apotex Inc sold its product for sale in the UK and Australia to affiliated companies, Apotex UK and GenRX, respectively. The transfer price agreements had two different prices defined, with a higher price payable by the purchaser when the product was considered a “Patent Challenge Product.” A product was defined as a Patent Challenge Product when a brand sold a competing product protected by a patent and there were no other generics on the market [38]. In such circumstances there was a substantial likelihood of patent litigation in the purchaser’s country. The same transfer price agreement also provided that Apotex would pay all legal expenses and indemnify the foreign purchaser against any such claims that might be brought [37]. Apotex argued that in effect Apotex Inc was providing the purchaser with litigation insurance and the higher price for a Patent Challenge Product was the premium for that insurance. Because it was a payment for insurance services, not payment for the infringing product itself, Apotex argued that the payment should not be considered revenue attributable to the infringement. The total of this putative premium was $22 million out of $51 million [17].

Gagné J accepted Apotex’s argument in principle [30]. However, she held it was not established on the facts that the increased payments for a Patent Challenge Product really were in the nature of insurance payments [67]. This was for a few reasons. One is that the price clause and the indemnification clause were severable [52]. Presumably this could be fixed with different drafting. Another problem, in Gagné J’s view, is that when a product is a Patent Challenge Product, not only will the risk of litigation be higher, the profits will also be higher because the market is less competitive. Gagné J held that “It is likely that the choice of a higher price in that context is, at least partially, triggered by the fact that Apotex wants to benefit from those favourable market conditions” [56]. I don’t find this point convincing. No doubt Apotex wanted to maximize its profits, but why a would a purchaser be willing to give it a larger share of the profits? The obvious answer is that Apotex Inc is willing to take on the risk – which is to say it is providing insurance. So I do not see the profit motive as distinct from the insurance argument.

A third reason is that the agreement provided not only that Apotex would bear the risk of legal action in the purchaser’s jurisdiction, but also that it would share in any settlement or award in the purchaser’s favour [60]. As discussed here and here, Servier sued Apotex in the UK and obtained an interlocutory injunction, but lost at trial. Servier was therefore liable to pay damages on the undertaking which it gave Apotex to obtain the interlocutory injunction. This means that Servier will have to pay the Apotex companies £17.5 million, which “is a significant consideration for the indemnity and legal services offered by Apotex to Apotex UK” [62]. In other words, the agreement can be seen as a risk sharing agreement rather than an insurance agreement. This point strikes me as more persuasive, but I’m not sure I’m convinced, as it seems to me that the share of that compensation can also be seen as just part of the notional insurance premium. Apotex Inc would be getting 90% of Apotex UK’s profits for a Patent Challenge Product [39], so it could hardly charge a higher share of the profits as a premium. But because a patentee’s profits which might be recovered in the event of litiation are normally substantially higher than the generic’s, and the risk of litigation was high, even 90% of Apotex UK‘s profits might not have been enough of a premium to make bearing the risk worthwhile. The provision for Apotex Inc to share in any award or settlement could be seen as a way of increasing the “insurance premium” on a contingent basis, rather than charging Apotex UK a premium that is higher than its entire profits.

A fourth point relates to the complex inter-jurisdictional nature of this dispute. There is an argument to be made that the amount awarded to Apotex in the UK litigation for having been wrongly kept out of the UK market is profit derived from Apotex’ infringement in Canada, as the product that would have been sold in the UK (where it was not infringing), would have been manufactured in Canada (where it was infringing). It is a bit odd to think of the notional product that was never made as being infringing, but this point was not addressed because of an agreement between the parties to the UK litigation. If the interlocutory injunction had not been granted, and Apotex had made sales in the UK, it would have made profits in the UK, but it would also have been liable to damages in Canada. In the UK litigation over damages on the undertaking, Apotex agreed that the damages that would have been awarded against it in Canada should be deducted from the profits that it would have made in the UK, to arrive at the amount owing by Servier on the undertaking. This is referred to as the “paragraph 26 concession” [64] (and see here and here for a discussion of the UK litigation). But the paragraph 26 concession was conditioned on the assumption that in the Canadian litigation, Servier would not seek compensation for the amount paid by it to Apotex on the undertaking in the UK litigation. If the UK profits and the Canadian damages in respect of those sales were exactly the same, then none of this would matter. If the Canadian damages were not deducted from the UK award, Servier could obtain fully offsetting damages in Canada. If the Canadian damages were deducted from the UK award, that award would be zero, and it would not matter whether Servier added it to the Canadian award. But if the UK profits are higher than the notional Canadian damages, then it does matter. If Apotex’s lost UK profits were $50m (to use an arbitrary number) and the Canadian damages were $30m, then Apotex should be entitled to retain $20m, which was the net amount attributable to the non-infringing UK sales. This one important reason why the EWCA held that ex turpi causa did not apply: applying ex turpi causa would defeat Apotex’s claim entirely, rather than reducing it according to the Canadian liability. But if Servier could then turn around and claim the $20m paid to Apotex in respect of the UK undertaking as damages in the Canadian litigation, Apotex would retain nothing as a result of the Canadian litigation. That is why Apotex agreed that the Canadian damages should be deducted from the UK undertaking only if Servier agreed not to make a claim in Canada for the amount paid on the UK undertaking.

This is all by way of background to Gagné J’s remark that “According to the same logic, I have to agree with the plaintiffs that instead of deducting $22 million dollars from the top line revenues from the sales of perindopril tablets the defendants should have asked to deduct from those revenues the costs incurred from litigating its dispute with Servier in the UK” [66]. I have to admit that I don’t quite understand this point. On Apotex’ argument, the $22m was an insurance premium which would cover the potential cost of a judgment against Apotex, so it is not clear to me why the UK litigation costs should be a substitute. Perhaps it is that interlocutory injunctions are readily available in the UK, so that Apotex’s real risk was not a liability judgment against it, because it would never be allowed to sell infringing product, but only the prospect of having to litigate. If so, then the argument may be that $22m is disproportionate to that risk, which is reflected in the litigation costs, so it cannot really have been intended as an insurance premium.

In any event, the nature of the payment turns on the facts and the interpretation of the contract, so I don’t want to second guess Gagné J. The main legal point of interest is that she did accept Apotex’s argument in principle. The details of her reasons for rejecting it on the facts may help others draft similar agreements which are effective in this respect.