How Accurate are Damages Calculations?

TearLab Corp v I-Med Pharma Inc 2016 FC 350 Russell J

2,494,540 / TearLab System / i-Pen System

This cases raises, albeit indirectly, the interesting question of how it could be shown that damages are inadequate for the purpose of establishing irreparable harm.

Russell J refused to grant an interim injunction in favour of TearLab which would have prevented I-Med from launching its allegedly infringing product pending a motion for an interlocutory injunction. (TearLab is the exclusive licensee of the University of California's ‘059 patent [6].) Russell J’s reasons followed established Canadian law. To get an interim or interlocutory injunction, the patentee must show irreparable harm, and the standard for showing such harm is stringent: the evidence of harm must be “clear and not speculative”, the harm itself must flow almost inevitably, and mere difficulty of calculating damages does not constitute irreparable harm, “provided there is some reasonably accurate way of measuring those damages” [43]. In this case, Russell J held that the evidence presented by TearLab did not satisfy these requirements.

Russell J’s conclusion was not very surprising on the facts. TearLab relied on the evidence of Mr Tierney, the retired business director of Allergen Eye Care, to the effect that there was no good model for quantifying TearLab’s losses [38]. But as Russell J pointed out, Mr. Tierney was not an expect in market forecasting or quantification of damages. And in any event, the courts have long acknowledged that damages assessment is a “broad axe” not a “rapier point,” and mere difficulty of calculating damages does not establish irreparable harm. There is an argument to be made that “meaningful doubt as to the adequacy of damages” should be sufficient to constitute irreparable harm, at least as a part of a balancing analysis, but that view has not carried the day in the Federal Courts.

This raises the question of whether it is possible, even in principle, for a patentee to establish to the satisfaction of the Federal Court that the difficulty in assessing damages is sufficient to constitute irreparable harm. The legal question is whether “there is some reasonably accurate way of measuring those damages.” On its face this seems to be an objective question capable of being addressed with the right kind of evidence. But question may be circular. The difficulty is that, so far as I know, the damages assessment itself is the best method we have of establishing the true loss. In s 8 and pharma damages cases, for example, sophisticated economic models are used to assess the difference between pharmaceutical prices with and without generic entry. In the absence of some more accurate method of assessment then it will be impossible to prove that damages are not reasonably accurate. But in light of the widespread judicial acknowledgement of uncertainties in damages calculations, even the best possible estimate of damages is not necessarily reasonably accurate. There is a difference between saying that damages are $1,000,000 plus or minus $1,000, and saying damages are $1,000,000 plus or minus $1,000,000. 

How accurate are damages assessments? In principle a formal error analysis of existing methods of damages assessment could address this question. However, that would be demanding from an evidentiary perspective, as the patentee and alleged infringer would be not only arguing over the quantum, but also over the accuracy of the quantum, which adds a layer of complexity to an already complex inquiry. And from a strategic perspective, the patentee would be caught in a squeeze, as it would want to argue that the errors are large in order to establish that damages assessments are not accurate in order to obtain an interlocutory injunction, but that would mean conceding the possibility that its subsequent damages claim might fall in the lower end of the range. My point isn’t that it is unfair to put the patentee in this squeeze, but rather that there are practical reasons why evidence of the accuracy of damages assessment is unlikely to be forthcoming. Another practical problem is that undertaking an error analysis in order to make the case for an interlocutory injunction would be very expensive and likely to be worth the cost only in pharma cases (if even there). Given that pharma has the NOC system which provides a statutory interlocutory injunction, it is hard to imagine a case in which it would be worthwhile for a plaintiff to incur this expense. Perhaps this is an issue that could be addressed by an academic economist.