Lex Machina Patent Litigation Damages Report; Allison/Lemley/Schwartz Paper

1.  As I noted yesterday, the Lex Machina Patent Litigation Damages Report "surveys the landscape of patent litigation damages awarded in U.S. District Court cases filed from January 1, 2000 through December 31, 2013."  Here are some highlights:

First, "[d]amages for patent infringement in U.S. District Court cases are awarded infrequently. Out of 36,629 patent cases filed and terminated from 2000 through 2013, only 708 cases (1.9%) involved compensatory damages awards" (p.i).

Second, "[w]hile relatively few patent cases involve damages awards, total awards made in these cases involve significant dollar amounts. Over $15 billion in damages, fees, costs and interest have been awarded over the past 14 years. This total includes over $13 billion in compensatory damages, almost $1 billion in enhanced damages and over $200 million in attorneys’ fees" (id.).  The $13 billion in compensatory damages consists of a little over $8 billion in reasonable royalties; a little under $3 billion in lost profits; and a little over $2 billion in "compensatory lump," that is, where "the specific sub-type (reasonable royalties or lost profits) is not specified or the apportionment of the award between sub-types is not specified" (pp. v, 1).  Note, however, that the "report is limited to the actions of the district courts to best help readers understand the actual (rather than the legally proper) behavior of those courts. Therefore, this report does not contain data on whether or not damages awarded by district courts were ultimately modified (or overturned) on appeal, or on outcomes in the district courts on remand" (p. vi).

Third, although the top three compensatory awards over the relevant time period were $1 billion or more (not including Apple v. Samsung, which the authors excluded because the damages were "awarded on the basis of both trade dress claims and patent infringement claims without apportionment," and this "coded . . . as 'cross-category' damages, which are excluded from this report unless explicitly stated otherwise"), the vast majority are for much lower amounts.  The median compensatory award for 2013 was $688,000.  The median reasonable royalty award for 2013 (based on 25 cases) was $403,000; the median lost profit award (based on only four cases) was $5,000,000; and the median compensatory lump, based on 26 cases, was $669,584.  Two awards of enhanced damages over the relevant period exceeded $100,000,000 (including last year's Stryker v. Zimmer case, which I blogged about here), but the median enhanced damages award for 2013 (based on six cases) was $698,965.  The top district in terms of compensatory damages totals over the relevant time period was the Eastern District of Texas (over $5 billion).

2.  While we're on the topic of statistics, I thought I would also note that John Allison, Mark Lemley, and David Schwartz have just published a paper titled Understanding the Realities of Modern Patent Litigation in 92 Texas Law Review 1769.  Here is a link to the paper, and here is the abstract:

Sixteen years ago, John Allison and Mark Lemley published the first detailed empirical look at patent litigation. In this Article, with the help of Professor Schwartz, they update and expand the earlier study with a new hand-coded data set. They evaluate all substantive decisions rendered by any court in every patent case filed in 2008 and 2009—decisions made between 2009 and 2013. They consider not just patent validity but also infringement and unenforceability. Moreover, they relate the outcomes of those cases to a host of variables, including variables related to the parties, the patents, and the courts in which those cases were litigated. The result is a comprehensive picture of the outcomes of modern patent litigation, one that confirms conventional wisdom in some respects but upends it in others. In particular, they find a surprising amount of continuity in the basic outcomes of patent lawsuits over the past twenty years, despite rather dramatic changes in who brought patent suits during that time.

Among the more interesting findings are an overall invalidity rate over the relevant time period of 42.4%, and an overall patentee win rate of 26%.  However:

If a case reached the trial stage, patentees fared much better. Overall, patentees won 60.7% of the trials, which included prevailing on 59.4% of patents decided by juries and 63.9% of patents decided by the bench. And it bears repeating that many cases are settled after a denial of summary judgment and before trial. These patents are not included in our statistics on definitive rulings, and many presumably involve a monetary payment to the patentee. These selection issues should be taken into account when considering the statistic that accused infringers win approximately three quarters of the patents that end with a definitive ruling. Patentees often get paid even without a definitive ruling.

Another finding that may be of particular interest to readers of this blog is that "patents of foreign origin . . . were much more likely to prevail in court in a merits decision than those issued to domestic inventors. In addition to being more likely to prevail overall, foreign inventor patents were less likely to be held invalid, less likely to be held invalid on summary judgment, and less likely to be found obvious," a finding suggesting "that there is no bias against foreign inventors, though it may be driven in full or in part by selection effects."  Finally:

the observable characteristics of the patents don’t seem to have much, if any, bearing on the outcome of the cases involving those patents. Neither the number of adjusted citations received nor the number of prior art references have any significant correlation to overall win rates, validity, or infringement outcomes.  Citations seem to tell us nothing about whether patents are valid or whether they are likely to be infringed. That is remarkable given how much effort economists have spent measuring the value of innovation by patent citation counts.