On Wednesday, I testified before Congress about the prevalence of patent assertion entity (PAE) cases at the ITC. For as long as I have been an academic, I have been studying the ITC, and my testimony drew upon the scholarly literature about the International Trade Commission including three review articles — Patently Protectionist, Patent Holdup, the ITC, and the Public Interest, Protecting Domestic Industries at the ITC, a separate amicus brief, and an ITC treatise in development that I have authored or co-authored.
As detailed in my written submission, my research has found that in the last 18 months (Jan. 2011-Jun 2012), PAEs brought more than a quarter of Section 337 patent cases, and nearly half of the total respondents appearing before the ITC were there because of a PAE-initiated case. (The submission details my methodology and reports the results of an independent parallel analysis that has been conducted by other attorneys, which find higher shares than I report for licensing-based investigations. It also reports my analysis that, in most cases, the PAE patent was purchased from its original owner.)
About a month ago, the ITC released its own official statistics. I regret that I only became aware of it yesterday, after my testimony, but I applaud the ITC for its responsiveness and engagement on this issue, given their full docket of other concerns, not the least of which includes carrying on their tradition of efficiently and fairly resolving trade disputes. But the heart of academic discourse – the discussion and examination of data, ideas, and assertions – is particularly important in the context of policy-making. Thus, I am thankful that the ITC chose to provide this data, particularly in a form detailed enough to enable a comparison to my data.
Though some may interpret the ITC data as a refutation of my claims, I have a different view. My view is that the data are closer together than they are farther apart.
My definition of PAE differs slightly from the ITC’s definition of NPE (which they divide into “NPE1” and “NPE2”), my data extends to 2Q 2012 whereas their’s stops as 1Q 2012, and they appear to have included non-patent (e.g. TM) investigations in their non-NPE investigation numbers, which I exclude. Also, because their report doesn’t include numbers, I had to estimate them from the graphs. Nevertheless, the numbers, at least in my opinion, are similar.
- 47% vs. 43% respondents: I claimed that in the last 18 months, 47% of the total (patent) respondents (332 out of 701) in the ITC were there because of a PAE-initiated investigation. The ITC’s chart on page 4 contains the comparable data on NPEs, regarding the last 15 months of investigations.
It appears to report that in 1Q2012, NPE1 + NPE2 respondents numbered about 45+70 = 115 and non-NPE respondents numbered 110. That makes a ratio of 110/(110+115) = 49%. When 2011 data (205/(205+300)) is added in, the total percentage comes to 43%.
- 28% vs. 19% investigations: I reported that about 28% (23 out of 81 through 2Q2012) of patent investigations are NPE investigations. Based on the ITC’s chart on page 2, the ITC is reporting about 15 NPE cases through 1Q2012, out of about 77 cases or 19%. This number is farther off from mine than the respondent number, so further investigation is warranted. However, it appears that the main source of discrepancy comes from 2Q2012 where my data reflects that 50% of the investigations come from PAEs.
Reasonable minds may disagree about how significant the gap is. Indeed, it could fairly be said that the investigation figure I report is 50% higher than the ITC’s (28% v. 19%), and that the sample size is too small to draw any statistical conclusions. Still, reporting the results together, the ITC and I agree that:
- 43-47% of respondents at the ITC in the last 15-18 months are respondents named in NPE/PAE investigations.
- 19-28% of investigations at the ITC in the last 15-18 months are NPE/PAE-initiated investigations.
Also, my testimony did not cite any numbers regarding the claim that eBay has caused PAEs to file their cases at the ITC at a higher frequency. However, the ITC’s graphs appear to report that 1 out of 13, or 8% investigations in 2006 and about 5 out of 75, or 7% of respondents in 2006, were NPE investigations. This establishes a baseline, of about ~ 7% of respondents and ~8% of investigations in 2006, from which change can be measured.
Again, I applaud the ITC’s willingness to provide statistics on this important issue. In the coming weeks, I will try to compare datasets with them so that any discrepancies and remaining questions can be answered.