What Does It Matter If There Is No NIA?

Airbus Helicopters S.A.S. v. Bell Helicopter Textron Canada Limitée 2017 FC 170 Martineau J
            2,207,787 / helicopter landing gear

The hot topic in Canadian remedies law lately has been the role of the “non-infringing alternative” or “NIA” in assessing monetary remedies. The legal test for an NIA was set out in Lovastatin 2015 FCA 171 (discussed here and here), and modified by Perindopril 2017 FCA 23 (here). Consequently, it is not surprising that Martineau J’s analysis began with an extended discussion of the issue, concluding on the facts that there was no valid NIA [171]-[214]. But it is a bit odd that in his actual assessment of damages, reviewed in yesterday’s post, his conclusion that there was no valid NIA concept played no particular role in his analysis. In this post I’ll suggest that when reasonable royalties are claimed, there is always a valid NIA; the NIA is, by definition, whatever the infringer would have done in the ‘but for’ world. This means that whether or not an NIA is explicitly identified, we can always work backwards from the damages assessment to figure out the implicit NIA. I’ll also suggest that it probably is not usually helpful to start by applying the modified Lovastatin test to identify the legally available NIA. Rather, it is normally better to start with the key factual question – what would the infringer have done but for the infringement? – and then apply the Lovastatin test to ask whether that NIA-in-fact is also an available NIA-in-law.

Turning to the facts, in the initial section on “Determining the existence of any valid non-infringing alternative,” four possible NIAs were considered [171]-[214]. Two, a special purpose wheeled landing gear and an “I-Beam” landing gear, which was innovative but largely untested, uncertified, and previously rejected, were not really serious contenders on the facts [200], [206]. Another candidate was the Conventional gear, which had long been used by Bell in prior helicopters. Martineau J held that the Conventional gear was not a “true alternative,” because it was not economically viable [198]. The final contender was the Production gear, which was actually used on the helicopters sold by Bell. Martineau J held this was not a valid NIA either, because it could not and would not have been sold on the eve of the first infringement (the third and fourth criteria of Lovastatin FCA), because it did not exist at that time [214]. Consequently, “there were no valid NIA on the eve of first infringement meeting the ‘could and would test’” [259] (see similarly [302]).

I have suggested that in its Perindopril Accounting decision, Apotex Inc v ADIR 2017 FCA 23, the FCA relaxed the requirement in Lovastatin FCA [79], that the NIA be “available on the market instantaneously,” so it might be suggested that the Production gear should have been considered an NIA even though it was not “available” on the eve of the first infringement. But it is not necessary to pursue this line of reasoning. There is a bigger question: what difference does it make? In both Lovastatin (see here) and the old UK case of United Horse Shoe & Nail Co v Stewart & Co, (1888), 5 RPC 260 (HL), it mattered because if the NIA could not be considered as a matter of law, then the patentee was entitled to claim lost profits, on the basis that it would have made the sales that were actually made by the infringer. But in this case Airbus did not even make a claim for lost profits.

So, what difference did it make on the facts of this case that there was no available NIA? Not much, as it turned out. As Martineau J emphasized, the fundamental question in assessing ‘but for’ causation is to “assess a hypothetical world where the defendant’s impugned conduct did not take place” [180]. In the reasonable royalty calculation, the infringer’s maximum willingness to pay is determined by the difference in the infringer’s position between the actual and ‘but for’ worlds, so the object of the exercise is to determine “whether Bell would have any economic benefit to enter in a license agreement to use the Legacy gear” [182]. This requires the court to determine, as a matter of fact, what probably would have happened but for the infringement. Martineau J undertook exactly this inquiry (as described in yesterday’s post), and there was no point where he held that something probably would have happened but for the infringement, and yet he should ignore it in assessing the benefit to the infringer because that something did not pass the test for being a valid NIA. Consequently, the fact that there was no valid NIA played no role in his damages determination.

There is a puzzle here. If, as Martineau J found, there is no legally available NIA, where does this leave Bell during a hypothetical negotiation? You can’t sell a helicopter without landing gear, and if there truly was no economically viable alternative to the infringing gear, then it follows that Bell could not have sold any helicopters at all but for the infringement. If, in the ‘but for’ world, Bell would have been entirely excluded from the market, then a large proportion of the sales that Bell actually made would have gone to Airbus instead, given that Airbus is a major player in the market [122], and Airbus should have been able to prove lost profits, based on those lost sales. But Airbus didn’t even try to prove lost profits from lost sales [113], and Martineau J certainly did not award damages based on lost sales. This implies there must have been an NIA; if Airbus could not prove lost profits, that must be because Bell would have been in the market in the ‘but for’ world, selling a helicopter with some kind of non-infringing landing gear. In other words, even if an NIA is not explicitly identified, there is always an implicit NIA; it is whatever the court finds the infringer would have done but for the infringement.

What is the implicit NIA? Consider in turn the three heads of benefit to the infringer, as found by Martineau J (and discussed in yesterday’s post):

Saved Development Costs [308]-[325]
The Production gear was based on the infringing Legacy gear, and consequently, the cost to Bell of developing the Production gear was less than it would have been but for the infringement. Martineau J referred to this as “the incremental cost [to Bell] of developing a ‘clean sheet’ non-infringing alternative” [301]. This means that in the ‘but for’ world, Bell would have developed a ‘clean sheet’ non-infringing alternative. The implicit NIA is the ‘clean sheet’ development of the non-infringing alternative – not the alternative gear itself. It is right to reject the Production gear itself, as Martineau J did, because the benefit to Bell from the infringement is precisely that it saved costs in developing the non-infringing alternative. If we take the Production gear itself to be the alternative, then we are ignoring the very benefit to the infringer. The facts here are different from the situation in Lovastatin, where Apotex had actually developed the NIA at the time of the infringement. It is also different from the situation in Grain Processing, 185 F.3d 1341(Fed Cir 1999), where the infringer had not actually developed the NIA, but on the facts it had a very high confidence of being able to do so successfully almost immediately, so the non-infringing product itself, and not its development, was the NIA. On the facts in this case, in contrast, developing a commercially successful alternative would have been an uncertain and expensive process, which was substantially aided by the infringement.

But even if it was the development of non-infringing gear which was the NIA, and not the gear itself, it still matters what non-infringing gear was developed, because different types of gear, would, presumably, require different development cost. It was apparently not the Conventional gear. In evidence referred to by Martineau J [324], Airbus’ expert testified that it would have cost $250,000 to develop “conventional” gear rather than the Legacy gear [239], but in that hypothetical world Bell would be selling helicopters with Conventional gear, not Production gear, which would never have been developed. Because Conventional gear isn’t as good, Airbus would also presumably gain more sales, but that isn’t reflected in the award. Rather, Martineau J refers to unspecified gear “which could compare favourably with the patented Moustache gear” [325]. I don’t have the benefit of the full record, but I find it hard to understand how the experts could have put a price on designing landing gear which has never actually been developed. Knowing that it has to “compare favourably” with the patented gear makes the problem worse, because the patented gear was inventive, and so this notional NIA would likely also have to be inventive, but inventive in some different, non-infringing way. What makes more sense to me is to use the development of the Production gear as the NIA; on its face the non-infringing alternative which compares favourably to the infringing gear which would be most likely to have been developed in the ‘but for’ world, is the non-infringing alternative which compares favourably to the infringing gear that was actually developed in the real world. This also has the practical advantage that the cost of development is more easily ascertainable; the inquiry is how much the use of the Legacy gear saved Bell in developing the Production gear, rather than in developing gear that is undefined and never actually designed or produced. Perhaps Martineau J felt constrained not to say that the Production gear was the gear which would have been developed, because he had already held that it was not an NIA because it was not actually available on the eve of the first infringement [214]. But if it is impermissible to use development of the Production gear as the NIA for that reason, surely this same point applies with even more force to an unspecified landing gear which not only did not exist at the time, but which does not exist even now. In any event, on my view, it was not the Production gear itself but the development of the Production gear that is the NIA. The alternative that was actually available to Bell at the time of the first infringement was not to use the Production gear, but to begin the process of developing it.

So, to repeat, it seems to me right to say, as did Martineau J, that the Production gear itself was not the NIA because it was not available at the time. The real NIA was the development of the Production gear.

Promotion of Legacy Gear [326]-[354]
Another head of benefit identified by Martineau J is that Bell was better able to promote its new Model 429, and in particular, it benefitted by attracting deposits on Letters of Intent (LOI). He attributed 13% of the deposits to the infringement. What is the implicit NIA here? Perhaps because the evidence only permitted a rough apportionment, Martineau J did not specify explicitly how Bell would have attracted LOIs in the ‘but for’ world. It seems unlikely that Bell would have attracted 87% of actual LOIs if it had advertised helicopters without any landing gear at all. In this context it doesn’t make sense to say that Bell would have advertised the Production gear, given that the advantage of the infringement was the head start in actually developing the Production gear. The most plausible alternative is that Bell would have advertised helicopters with the Conventional gear. This is consistent with the view that the real NIA was the development of the Production gear. The Production gear would not have been available during the development period, which would have coincided with the marketing period, so the next best alternative after that would have been used during that period, which was probably the Conventional gear.

Risk Premium
As discussed in yesterday’s post, I don’t see reduced litigation risk as being a source of benefit to Bell, so I don’t see any implicit NIA. The reduced development risk, on the other hand, reflects exactly the same implicit NIA as the reduced development costs, namely the development of the Production gear.

Summary
In summary, Martineau J was right to hold that none of the four alternatives he considered were the NIA, not because there was no NIA at all, but rather because the true NIA – the development of the Production gear – was not put to him directly. I would also suggest that to avoid this kind of confusion, rather than starting with the legal question of “what is the NIA?”, it is better to start with the factual question of “what would have happened but for the infringement?” That inquiry itself will implicitly define a non-infringing alternative, and once that is established as a matter of fact, we can then look to see whether it is acceptable as a matter of law. That way, there is no risk of missing the true NIA, and there is no need to waste time on legal analysis of alternatives that are not relevant on the facts. (Whether the Lovastatin test in particular is better seen as a legal test, which may operate to exclude alternatives which would in fact have been used, or rather as a guide to the factual inquiry, is a question best left for another day.)