MWA and MWP in Compensatory Damages

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

The 787 patent claims a type of helicopter land gear known as a “moustache” landing gear [3]. As discussed here (FCA) and here (FC), in Eurocopter v Bell Helicopter Textron Canada Limitée 2012 FC 113 aff’d 2013 FCA 219, Martineau J held one claim of the 787 patent, owned by Eurocopter (now the helicopter division of Airbus) to be valid and infringed by Bell. This decision is the damages phase of that bifurcated trial [1]. In the result, Martineau J awarded $500,000 in compensatory damages and $1,000,000 in punitive damages, plus pre-judgment and post-judgment interest [2], [108]. The decision is interesting and important in several ways. It is significant as one of very few Canadian patent cases to award punitive damages. The issue of compensatory damages is also important, because it deals with the inherently difficult problem of assessing reasonable royalty damages for a complex product.

Martineau J’s reasons are correspondingly lengthy and wide-ranging, and a discussion of every issue raised would require a law review article. On the topic of law review articles, Martineau J cited several of my articles, including an article which I co-authored with Tom Cotter, “A New Framework for Determining Reasonable Royalties in Patent Litigation, (2016) 68 Fla L Rev 929, which is so recent that I only received my hard copy in the mail in the same week that the decision was released. (Martineau J evidently cited to the SSRN version.) It is of course very gratifying to me personally to be cited, but I’m particularly excited because Martineau J cited a range of scholarship, including empirical articles on behavioural psychology [107]. In an era when some courts are finding law review articles to be increasingly unhelpful, it is very encouraging to find courts that are willing to engage with the latest scholarship, and scholars who are doing work that the courts do find useful.

Turning back to the decision, I will limit myself to commenting on some of the fundamental points. The first couple of posts will consider compensatory damages. This post provides an overview, and considers the treatment of the hypothetical negotiation, and particularly the concept of the patentee’s minimum willingness to accept and the infringer’s maximum willingness to pay. The next post will consider the problem of the non-infringing alternative, again restricting myself to compensatory damages. A third post will consider the issue of punitive damages.

Complex Products
On the whole, my view is that Martineau J’s approach to compensatory damages was fundamentally sound, though I do have a few quibbles about some aspects of the analysis. That is not surprising, as the case presents a problem that is inherently very difficult, in two different ways. First, the patented technology was one aspect of a complex product, which is to say that the patented landing gear contributed only part of the value of the end-product, the helicopter itself. This is in contrast with a patent for a new pharmaceutical, for example, which is ‘simple’ in the sense that the patented technology contributes substantially all of the value of the end-product. At the other extreme are many of the patents in the ICT sphere, in which many of the thousands of patented technologies, such as an ‘overscroll bounce,’ individually contribute only a small part of the total value of a smart phone. It is typically much harder to assess damages in complex product cases, because we only have good information about the value of the end-product itself, and while we normally know that the patented technology contributed something of value, quantification of that value is difficult. In a complex product case, it may be difficult or impossible to prove lost profits, even when the patentee and the infringer are direct competitors, as in this case. In principle, the patented feature probably tipped the balance in favor of the infringer’s product rather than the patentee’s, for at least some consumers, and the patentee should be entitled to lost profits on those sales, and a reasonable royalty on the rest: see e.g. Jay-Lor 2007 FC 358; Mentor Graphics v EVE-USA (Fed Cir 2017), slip op 21-22. However, the evidence is not always sufficiently granular to allow lost profits to be proven. This case falls part way between pharmaceuticals and smart phones, though it is closer to the complex products end. The patented landing gear was only one part that contributed value to the helicopter, but it was one of a handful of key technologies that marked Bell’s 429 model, and it was featured in advertising and promotion. Even so, the patentee might have been able to prove lost profits, except for a second complicating factor: no helicopters with infringing gear were ever sold by Bell. Bell had copied the infringing “Legacy” gear used in its original design from Eurocopter, and used this for testing, certification, and promotion; but in response to this litigation, Bell designed around the patent with the “Production” gear, which was ultimately held not to infringe. Only helicopters equipped with Production gear were ever sold, and, presumably for this reason, Airbus did not claim any lost profit damages [113].

Martineau J began his analysis with a discussion of key principles. This included the concept of a hypothetical negotiation in which the parties split the difference between the infringer’s maximum willingness to pay (“MWP”) and the patentee’s minimum willingness to accept (“MWA”) [109]-[170]; and a discussion of the availability of a non-infringing alternative (NIA), in which he concluded that there was no valid NIA [259]. He then rejected both of the main models presented by the experts for each side: [241]-[253], [254]-[278]. All of these raise interesting issues, but rather than starting with an extended discussion of what Martineau J didn’t do, I’ll start at the end, with what he did do.

As Martineau J emphasized, whether the question is an accounting of profits, lost profits damages, or reasonable royalty damages, “the Court’s task stays the same: assess a hypothetical world where the defendant’s impugned conduct did not take place” [180]. This is the starting point for all these remedies because they are all based on ‘but for’ causation. In an accounting of profits, the court assesses the difference in the infringer’s position between the actual and ‘but for’ worlds to determine the profit caused by the infringement. In assessing damages, the court looks at the same comparison, but from the patentee’s side: “[t]he defendant is only required to make good any loss which would not have been suffered by the plaintiff ‘but for’ the defendant’s actions” [109]. When the claim is to lost profits, the court directly assesses how much better off the patentee would have been but for the infringement.

When the claim is to a reasonable royalty, as in this case, the usual approach, used by the parties in this case, is to consider a hypothetical negotiation between the parties, in which the infringer pays a royalty equal to some part of its gain caused by the infringement, on the theory that the patentee’s loss is the foregone royalty. The basic idea is that in a negotiation both parties expect to come out better off, or there will be no trade. If I’m buying a chair, the most I am willing to pay depends on how comfortable the chair is, how attractive, how durable, etc., all of which contribute to its value to me. For a poorly made wooden stool, the value to me might be $10, and for a luxury office chair that I will be sitting in for most of the day, it might be $2,000. But for any product, no matter how nice, there is a price at which I would prefer to walk away empty handed. That is known as my “maximum willingness to pay,” or “MWP” [116]. The MWP represents the economic value of the product to the buyer. If I would be willing to pay $2000 for a chair, and I get it for $500, the value to me is still $2000, but I have gotten a net surplus of $1500. On the other hand, there is a minimum price, known as the “minimum willingness to accept,” or “MWA,” at which the seller would prefer to walk away without having made a sale [116]. The difference between the buyer’s MWP and the seller’s MWA is the bargaining surplus, also known as the available gains from trade. In standard bargaining theory, the final price will fall somewhere between the MWA and the MWP [223-25]. This model means that the loss to the patentee turns on the benefit to the infringer, because that determines the MWP. As Martineau J said, the court must determine “whether Bell would have any economic benefit to enter in a license agreement to use the Legacy gear” [182] (and similarly [345]).

Once the MWA and MWP have been determined, there is a conceptually difficult question as to how to split the difference. One well-known theoretical model, the Nash Bargaining Solution, says that if parties have equal bargaining power, the final price will split the surplus between the parties, which means that the final price will be half way between the MWA and the MWP. What, exactly, is meant by “bargaining power” is not very clear. The economic literature tends to treat it as a label for a result – whichever party gets a bigger share of the surplus must have had more bargaining power – rather than an analytical tool in its own right. In any event, on this model we can at least say that the reasonable royalty will never be higher than the MWP or lower than the MWA. If the reasonable royalty is exactly equal to the MWP, then reasonable royalty damages amount to the same thing as an accounting of profits; the infringer disgorges its entire economic benefit.

While determining the split is conceptually difficult, the biggest practical problem is determining the MWP, which is to say the economic benefit to the infringer. Conceptually that is easy enough – it is equal to the value of the invention in the hands of the infringer, as compared with the value of the NIA. But making that determination on the facts is normally very difficult in the context of complex products. That was the main challenge facing the parties, and Martineau J, in this case.

In the end, Martineau J assessed the reasonable royalty as a lump sum of $500,000. This was evidently notionally a split between Bell’s maximum willingness to pay (“MWP”) of $525,000 and Airbus’ minimum willingness to accept (“MWA”) [116] of $475,000 [379]. This implies a 50/50 split of the surplus between the parties, but with the MWP and MWA so close, the “broad axe” rather than any formal analysis was sufficient.

In what follows, I will argue that Martineau J’s analysis of the key problem – the economic benefit to Bell from the infringement – was correct as a matter of law and principle. My quibble is that he carried it out under the rubric of Airbus’ MWA, rather than Bell’s MWP.

Bell’s MWP
Martineau J held that Bell’s MWP was $525,000 [377]. This was calculated as the unit price of the infringing Legacy gear, $25,000, for all 21 infringing Legacy gears that were manufactured for the development of the Bell 429 [376]. In effect, Bell’s maximum willingness to pay was considered to be what it would have had to pay to buy the infringing gear. If there had been an actual legitimate market for the Legacy gear, in which Bell would have been entitled to purchase the gear and use it however they saw fit, this would make sense, as Bell would choose to purchase in that market rather than license from Airbus. But there was no such actual market for the infringing gear. Airbus never licensed core technologies and never licensed technologies to competitor companies [145], [149], [358]. The figure of $25,000 was notionally determined at the liability phase, on the basis of evidence which is entirely obscure: 2012 FC 113, [415]. 

While the court referred to this as the “monetary value” of the gear, the question is the monetary value to whom? The figure might have been suggested as the price at which Airbus would have sold gear to a purchaser of one of its helicopters if the existing gear had been damaged. That would presumably have been higher than Airbus’ MWA, but there is no reason to think it would be equal to the maximum willingness to pay of a helicopter owner in need of new landing gear – would Airbus be out to gouge its existing customer? – and moreover, a party who needs landing gear for a helicopter they already own is in a very different position from Bell, who needed the landing gear for development and certification of their new model, but not for actual sales. For example, if the use of the gear allowed Bell to enter the market three years early and sell twenty helicopters that it would not otherwise have sold (to be clear, I’m not suggesting this was true on the facts), then the value to Bell could have been far more than this notional $25,000. If a chair sells for $500, but it is worth $2,000 to me, my MWP is still $2,000 – I have just gotten a good deal.

Moreover, even if the monetary value of the gear itself is $25,000, that’s beside the point, since Bell is not notionally purchasing landing gear, it is purchasing a licence to allow it to manufacture its own non-infringing gear. The bottom line is that the figure of $525,000 bears no particular relationship to Bell’s MWP.

Airbus’ MWA
Martineau J calculated Airbus’ MWA as having three components [378]:

            $250,000 for the saved costs in development of the Bell 429 [325]
            $100,000 promotion of Legacy gear [354]
            $125,000 risk premium [370]

I’ll consider these in turn.

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 should have a significant impact on the determination of a reasonable royalty” [301]. The benefit to Bell from the infringement is the difference between the amount it actually cost to develop the Production gear, and the amount it would have cost to develop a non-infringing alternative, but without using the infringing Legacy gear. (See similarly, [310], [312], [323]). On the facts, Martineau J found that the saved development cost was, at a conservative estimate, $250,000 [239], [324]. In my view, this is a straightforwardly correct application of ‘but for’ causation in determining the benefit to Bell from the infringement.

The only problem with the analysis is that the benefit to Bell from the infringement determines Bell’s maximum willingness to pay, not Airbus’ minimum willingness to accept. Bell would have been willing to pay at least $250,000 for a licence to use the infringing Legacy gear in development, as this is the amount it would save in development costs. But Martineau J used this figure to determine Airbus’ MWA, on the view that because Bell saved at least that much, Airbus “could reasonably refuse” any offer below that amount [325]. That is a reasonable statement in its own right, but Airbus’ minimum willingness to accept is not determined by what a reasonable offer would be; it is defined as the least Airbus could accept while still coming out strictly better off, if only just barely, than if it refused to license at all. The hypothetical negotiation model which the parties invoked, and Martineau J accepted, starts by first determining the outer limits of the bargain that might be struck. At that point there is no question of reasonableness; on the contrary, the question on both sides is what is the worst deal that would nonetheless make each party better off, though only by $1. Reasonableness does come in to the equation, but only at the next step, of dividing the surplus. By using the benefit to Bell as the basis for determining Airbus’ MWA, Martineau J in effect conflated the two stages of the analysis.

Promotion of Legacy Gear [326]-[354]
Another real world benefit gained by Bell is that “In the real world, Bell benefited from the infringement, inasmuch as during the infringing period it advertised and promoted a Bell 429 helicopter equipped with the infringing Legacy gear” [329]. On the facts, this was a real benefit. Bell had had a somewhat stodgy reputation, and the Model 429 was an attempt to establish itself as an innovator [152]. The Legacy gear was new, eye-catching, and better. It was not the only feature that would have attracted the attention of potential buyers, but it was one that was advertised by Bell, and would have been noticed, and was a significant part of the package that Bell was promoting [156].

The difficulty is how to quantify this, given that no helicopters were actually sold with infringing gear. The approach accepted by Martineau J was to look to the interest earned by Bell on deposits made on refundable Letters of Intent [LOIs]. The idea is that without the Legacy gear, some customers would not have made deposits – and indeed, many deposits were ultimately cancelled once the production aircraft were released – so, but for the infringement, Bell would not have had the use of that money [342]. Martineau J found that the total benefit to Bell from the use of all the LOI deposit money was $763,558, taking the cost of capital, conservatively (that is, minimizing Bell’s liability), as that proposed by Bell’s expert [343]. This seems to me to be entirely correct on the principle of ‘but for’ causation.

The difficulty is that we don’t know how many LOIs were attracted by the landing gear, or how many cancellations were due to the use of Production gear rather than Legacy gear. The basic problem here is the same as in trying to assess lost profits; it is probable that some part of the LOI deposits is attributable to the infringement, but how much [329]? Bell’s expert proposed an arbitrary number of 5%, which Martineau J considered too low in light of the considerable emphasis put on the Legacy gear in Bell’s promotional materials [344-49], but he also rejected the 100% figure proposed by Airbus’ expert. Ultimately, Martineau J settled on a figure of about 13% [354]. Given the overall benefit from the use of the LOI deposit money was $763,558, this means that the advantage to Bell from the infringement, in this respect, was about $100,000. I would point out that there is no specific evidentiary basis for the 13% number. This is not to criticize Martineau J. The problem is that with a complex product like a helicopter, it is almost impossible to determine how much interest was attracted by the landing gear in particular; the use of a “broad axe” is inevitable.

Overall, this seems to me to be an entirely correct application of but for causation to determine the benefit to Bell from the infringement, in the specific form of the use of money it would not otherwise have had. Again, this raises the problem that the deposits from the LOIs were a benefit to Bell, as Martineau J explicitly recognized [329], [343], and yet Martineau J used it to establish Airbus’ MWA.
 

Risk premium [355]-[370]
The most theoretically interesting point raised by Martineau J is the risk premium. He considered what I consider to be two distinct sources of risk, namely litigation risk and NIA development risk. While Martineau J treated these factors together, as both being sources of risk, in my view they are quite distinct.

Litigation risk
The basic idea here is that Bell would have been willing to pay a premium to avoid the prospect of litigation (see [358], [364], [365]), but beyond that, I must admit that I’m not sure I completely follow the point.

One idea is that Bell would be willing to pay more to avoid the risk inherent in litigation. But this point cuts both ways; litigation would be risky for Airbus as well. Consider how the negotiation would play out if the parties were negotiating over the $250,000 in development costs that Bell would save by using the patented technology. If the patent was sure to be valid – maybe it had already been tested in litigation against another party at the time of the negotiation – then Bell would be willing to pay up to that full amount for a licence, because if it walked away from the negotiation without a licence, it would have to spend an extra $250,000 to develop the production gear. But suppose instead that the patent had not yet been tested. With most patents there is a significant risk of invalidity. After all, even in this litigation, a number of claims in the 787 patent were held invalid, and while Airbus was successful in its argument that the Legacy gear infringed, it was unsuccessful in arguing that the Production gear also infringed. If Airbus demanded $240,000, Bell could say to Airbus “Let’s litigate - if you win, you’ll get $250,000, but if you lose you’ll get zero, and there’s only a 50/50 chance that you’re going to win, so at this point, all you can expect from litigation is $125,000.” In other words (neglecting litigation costs), the infringer should be able to get a discount for the probabilistic nature of the patent rights. This suggests that, if anything, the litigation risk should reduce the royalty that would be paid, rather than increasing it: see Lemley & Shapiro, “Probabilistic Patents,” 19 J Econ Perspectives 75 (2005).

Another angle is that the risk relates to the extra litigation costs that will be incurred, and which could be avoided by settling: see esp. [364] (referring to “costs of litigation”). But the difficulty with this is that, again, both parties face litigation costs. The standard view is that litigation costs do not affect the royalty unless those costs are asymmetrical.

On the whole, I don’t see how the litigation risk could increase Bell’s MWP, but I must acknowledge that the discussion was brief and I may have missed the point.

NIA development risk
The other factor considered by Martineau J is that “Bell would have been prepared to pay more in order to avoid the risks and uncertainties associated with the development of the Bell 429 with an unknown NIA rather than using the already tested and proven patented [infringing] gear” [366] (and similarly [359], [364], [367]). The idea here is that we now know that Bell was able to develop the Production gear at an incremental cost of $250,000, but at the time of the infringement, Bell didn’t know that it would be able to do so. Even if, at the time of the infringement, Bell had correctly estimated the incremental cost of developing the Production gear at $250,000, it would have been willing to pay more than that to avoid the risk that development would cost more, or worse, delay launch.

In my view, Martineau J is correct in principle on this point. My article with Tom Cotter, “A New Framework for Determining Reasonable Royalties in Patent Litigation, (2016) 68 Fla L Rev929 was cited by Martineau J for the proposition that development and commercialization risk should be taken into account [365]. Our main argument in that article is the parties to a hypothetical negotiation should be assumed to negotiate with all information, including ex post information (i.e. information arising after the putative time of the hypothetical negotiation). This presents a bit of a puzzle for how to address risk-shifting, because parties to an actual negotiation may use the license terms to shift risk, but if we use ex post information, how can we say there is any risk at all, if we already know what happened? Our argument is that we should look to see, with the benefit of hindsight, who actually bore the risk. In this case, Bell did use the Legacy gear to market the Model 429, and so it did in fact benefit from reduced risk as opposed to having to develop its own landing gear, and therefore it is appropriate, as Martineau J held, that Bell should have to pay a risk premium to reflect the fact that the infringement reduced its risk.

Airbus’ MWA Redux
To repeat for convenience, Martineau J held that Airbus’ MWA was $475,000, which was made up of three components [378]:

            $250,000 for the saved costs in development of the Bell 429
            $100,000 promotion of Legacy gear
            $125,000 risk premium

He held that Bell’s MWP was $525,000 [378]. He awarded $500,000, which splits the difference, which is justifiable either as being the Nash Bargaining Solution to the division of the surplus, or simply on the “broad axe” principle [379].

I have argued that the figure of $525,000 is basically irrelevant, as it is neither Bell’s MWP nor Airbus’ MWA. I have also argued that the figure of $475,000 is not Airbus’ MWA, but rather a conservative estimate of Bell’s MWP – conservative, because the first figure in particular, was conservatively estimated.

Then what was Airbus’ MWA? In bargaining theory generally, the MWA is usually taken to be the seller’s marginal costs. If it costs $100 to make a chair, the seller who accepts an offer of $101 is still better off (by $1), than if she walks away without making a sale at all. In the patent context, where the licence is not to purchase a physical thing, but rather a licence to use the patented technology, the patentee’s marginal cost is normally zero. To grant a license to someone else allowing them to make my invention costs me nothing, so I am better off to take $1, if the alternative is for that party not to take a licence at all, and to wander off and invest in real estate instead.

The hypothetical negotiation model is quite intuitive when the infringer sold into a market that the patentee could not possibly have captured. Suppose for example, that the infringer made infringing product in Canada and then sold it in an export market where the patentee did not sell at all. In that case it is reasonable to suppose the parties could have arrived at a win-win agreement, where the patentee would have licensed the infringer to sell into the export market in return for a share of the profit. In that context it is also intuitive to suppose that the patentee’s minimum willingness to accept would be $1, keeping in mind that the patentee does not bear any of the manufacturing costs, and that the alternative is for it to get nothing at all, as it can’t exploit the market itself. Of course the patentee would not be happy about getting $1, and in the normal outcome of the hypothetical negotiation it would get more, because the MWA only sets the absolute floor on the reasonable royalty. If the infringer’s MWP is substantial, then the royalty will also be substantial, even if the patentee’s MWA is zero.

The model is much less appealing if, on the facts, the patentee would never have licensed to the infringer, as in this case. This is commonly the case when the parties are competitors selling into the same market. The ideal solution in such a case is that the patentee will claim lost profits. In that case, there is no need to resort to an implausible hypothetical negotiation; the patentee would not have licensed to the infringer, because it would have made the sale itself. But it is not uncommon to have a case, such as this one, where in fact the patentee would not have licensed, and yet the patentee is not in a position to prove lost profits, perhaps because sufficiently granular evidence is not available.

In principle, even though we know that Airbus would not in fact have licensed, if Airbus could not prove lost profits on the facts, then for the purposes of the hypothetical negotiation we have to accept that Airbus would not have made any of the sales that Bell actually made. If that is so, then if Bell walked away from the hypothetical negotiation without a license and did not use the patented gear, Bell would have been worse off, but Airbus would have been no better off. That means that Airbus would have been better off giving a licence for $1, rather than walking away empty handed. It is as if Bell’s sales were made into a foreign market that Airbus could not have tapped.

This gives rise to a tension in the hypothetical negotiation model: in principle Airbus’ MWA should be taken to be zero, even though we know that in reality Airbus would not have licensed, and would have sold into the same market as Bell.

What flows from that? If we take Airbus’ MWA to be zero, and Bell’s MWP as $475,000, and apply the Nash Bargaining Solution, the result would be a royalty half way between the MWA and the MWP, or about $237,000.

That seems wrong, and I think it is wrong. Why should we apply the Nash Bargaining Solution, which assumes equal bargaining power? Martineau J explicitly found as a fact that Airbus was in a much stronger bargaining position [170], [378], and consequently, “the amount of compensation should come very close to Bell’s MWP” [378]. That is a correct application of standard bargaining theory. But note that Martineau J, in his explicit reasoning, did not really apply this. He viewed Bell’s MWP as $525,000, Airbus’ MWA to be $475,000, so his award of $500,000 just split the difference, which implies equal bargaining power.

On the other hand, if we take Airbus’ MWA to be zero, and Bell’s MWP to be at least $475,000, and apply unequal bargaining power, as found by Martineau J, then we should end up with a royalty that is very close to Bell’s MWP. And we also need to keep in mind that, as I understand Martineau J’s factual findings, the MWP for Bell of $475,000 was a conservative estimate; Bell’s MWP was at least that much. That implies that a reasonable royalty of $500,000 is justifiable on the facts found by Martineau J, even taking Airbus’ MWA to be zero, so long as we also take account of the fact that Airbus had superior bargaining power.

There is an important conceptual point here. Martineau J stressed at many points that Airbus had greater bargaining power because in reality Bell and Airbus were competitors selling into the same market: [364] [378]. I noted above that the hypothetical negotiation model is particularly strained, if unavoidably so, when the patentee would not have licenced and would have competed, and where the real harm is probably lost sales, which cannot be proven because of evidentiary difficulties. Perhaps that is a situation where it is generally appropriate to award a greater part of the surplus to the patentee. Martineau J suggested as much in his risk premium analysis, when he noted that “Eurocopter would also reasonably have considered additional risks associated with granting a worldwide license to its direct competitor and might have reasonably required an additional amount (a risk premium) inasmuch as this risk is not accounted for already in the royalty formula contemplated by the parties on the eve of first infringement” [364]. This resolves the tension that arises in the hypothetical negotiation model when the patentee is in fact a competitor selling into the same market, who simply cannot prove lost profits on the evidence: we acknowledge that when the patentee cannot prove lost profits, and so its MWA in the hypothetical negotiation is zero, but we also acknowledge the reality that it would not have licensed by giving it the lion’s share of the surplus attributable to the infringement. I don’t want to commit too firmly to that view, as I haven’t really considered the issue in this way before, but the insight suggested by Martineau J’s reasons certainly deserves to be developed further.

I do have one more quibble. At the end of the day, the reasonable royalty assessed by Martineau J resulted in Bell having to disgorge essentially the entirety of the economic benefit that it gained from the infringement. This amounts to the same thing as an accounting of profits. If Martineau J had granted an accounting of profits in the first place, the result would have been the same, and all of the confusion relating to the MWA and the MWP would have been avoided. I’m not saying that an accounting should have been awarded for that reason alone, but perhaps the reasons Martineau J gave for holding that Airbus had more bargaining power in the hypothetical negotiation, in particular the point that Airbus would be licensing to a competitor, is also a consideration that would support the granting of an accounting of profits.