Guest Post by Christopher A. Cotropia, Professor of Law and Austin Owen Research Fellow, University of Richmond School of Law; Jay P. Kesan, Professor and H. Ross & Helen Workman Research Scholar, University of Illinois College of Law; and David L. Schwartz, Associate Professor and co-Director of the Center for Empirical Studies of Intellectual Property at Chicago-Kent College of Law
Today, a certain type of patent litigant—the non-practicing entity (“NPE”), also known as a patent assertion entity (“PAE”), patent monetization entity (“PME”), or simply patent troll—is the target of much public debate, if not venom. Indeed, President Obama himself got involved in this debate, with his Council of Economic Advisers preparing a report this summer entitled “Patent Assertion and U.S. Innovation.” The Executive Summary of the President’s report sounds the following alarm about PAE suits:
Suits brought by PAEs have tripled in just the last two years, rising from 29 percent of all infringement suits to 62 percent of all infringement suits.
This asserted explosion in PAE-initiated litigation has fed into a wider perception that PAEs are out of control and need reining in by Congress. But is the factual assertion by the President’s report an accurate characterization of total PAE litigation activity?
We address this important issue in our new article, Patent Assertion Entities (PAEs) Under the Microscope: An Empirical Investigation of Patent Holders as Litigants. To investigate PAE litigation, we personally hand-coded all 7,500+ patent holder litigants in 2010 and 2012. In our coding, we finely classified the nature of the litigants, going beyond the simple PAE / non-PAE label. Specifically, we coded each patent holder as one of the following: (1) a University; (2) an Individual Inventor/Family Trust; (3) a large Patent Aggregator (e.g., Acacia); (4) a Failed Operating Company or Failed Start-up; (5) a Patent Holding Company that appears unaffiliated with the original inventor or owner; (6) an Operating Company; (7) an IP Holding Company affiliated with an operating company; or (8) a Technology Development Company (e.g., Walker Digital).
Based on our data, and contrary to the assertions in the President’s report, we do not find an explosion in PAE litigation between 2010 and 2012. In particular, the President’s report considered only the raw number of lawsuits filed in 2010 and 2012. By limiting its analysis to numbers of cases filed, rather than the underlying parties involved, the President’s report was incomplete and led to an erroneous conclusion.
To understand the reason for this error in the President’s report, one must consider an important change to patent litigation made by the America Invents Act (AIA). The AIA, adopted in 2011, changed the joinder rules relating to patent litigation by prohibiting patent holders from including multiple, unrelated defendants in a single lawsuit based on commonly-asserted patent(s). Instead, after the AIA, patent holders must file separate lawsuits against each unrelated defendant. Thus, rather than focusing on the raw number of lawsuits as the President’s report did, a more appropriate inquiry should focus on the total number of patent litigants, both patent holders and accused infringers, in order to gauge more accurately the actual number of patent disputes between parties. Using this approach, we found almost no difference between 2010 and 2012 in the number of litigants in patent cases. In other words, the “explosion” of PAE litigation between 2010 and 2012 is simply a mirage.
First, in terms of raw number of patent lawsuits, our study reveals that the number of lawsuits did increase between 2010 and 2012, with a substantial increase in the raw number filed by non-operating companies (by which we mean the sum of all categories except operating companies). Our first finding therefore is similar to the President’s report. For example, while what we coded as “Patent Holding Companies” (“PHCs”) suits comprised 16% (400) of cases filed in 2010, they comprised 38% (1,959) in 2012, and this is in comparison to “Operating Companies” suits comprising 69% (1,750) of cases in 2010 to 45% (2362) of cases in 2012. The bar chart below reports these findings along with those for all of the categories we coded:
However, when we evaluated the number of litigants, rather than just the number of cases, the picture changed substantially. We considered separately both the number of unique patentees alleging infringement in 2010 and 2012 and the total number of alleged infringers in those years. For both, we found little difference between 2010 and 2012. Our second finding therefore directly contradicts the statement in the President’s report that there was an explosion of PAE lawsuits in that time period.
Specifically, concerning the number of unique patentees starting litigation, the total number barely changed between the studied years, increasing from 1,610 unique patentees in 2010 to only 1,696 in 2012. Further, the distribution among types of patentees stayed almost the same, with the unique number of PHC patentees slightly increasing from 13% (214) of all patentees in 2010 to 16% (266) in 2012, while Operating Company patentees dropped slightly, from 72% (1,156) of all patentees in 2010 to 69% (1,165) in 2012. The graph below reports all of the unique patentee numbers for the various categories.
And the number of infringement defendants (perhaps the best measure of the impact of patent litigation) corresponding to each category of patent holder also did not vary much from 2010 to 2012. In particular, the overall litigation exposure from Operating Company suits compared to other types of patentee suits remained almost constant. The number of defendants in Operating Company suits slightly dropped from 49% (5,708) of the total number of defendants in 2010 to 48% (5,611) in 2012. And while the number of alleged infringers for other categories changed, the changes were slight. For example, the number of defendants in PHCs suits rose from 27% (3,179) of all defendants in 2010 to 32% (3,698) in 2012 while the number in Individual/Family Trust dropped from 11% (1,320) in 2010 to 8% (927) in 2012. The complete comparison of total number of parties excluding the patentee in 2010 versus 2012 is set forth in the graph below.
From this data, the only real change between 2010 and 2012, the increase in the raw number of lawsuits, is most likely driven almost entirely by a procedural rule change relating to joinder passed with the 2011 patent reform. These results call certain findings in the President’s report into serious question.
Information about lawsuit filings, obviously, cannot answer all questions about PAEs. There are other serious questions relating to PAEs that demand an empirical answer. For instance:
- What is the volume of patent demand letters? How many of those assert frivolous claims?
- Of the litigated cases, how many of the PAE (and non-PAE) cases are frivolous?
We do not know the answers to these questions (yet), and further rigorous empirical research is needed.
To aid the effort in understanding the patent litigation landscape, we have decided to release the underlying data—our specific classifications of the litigants in all patent lawsuits filed in both 2010 and 2012—to the public. We believe that releasing this data to the public, which unpacks the definition of PAE, provides better illumination to policy makers, researchers, and others interested in the patent litigation system. This data is available to everyone to examine and evaluate at www.npedata.com.
The full paper is available for download here. It provides a detailed analysis of the distribution of the 2010 and 2012 patent cases, including information based upon technology and litigant type.
We welcome any comments and suggestions on the article or the dataset.
 A more detailed description of the categories can be found on pages 13-14 of our draft article.