Acknowledging the Limitations of the FTC’s PAE Study

Cross-posted from the Center for the Protection of Intellectual Property (CPIP) blog.

The FTC’s long-awaited case study of patent assertion entities (PAEs) is expected to be released this spring. Using its subpoena power under Section 6(b) to gather information from a handful of firms, the study promises us a glimpse at their inner workings. But while the results may be interesting, they’ll also be too narrow to support any informed policy changes. And you don’t have to take my word for it—the FTC admits as much. In one submission to the Office of Management and Budget (OMB), which ultimately decided whether the study should move forward, the FTC acknowledges that its findings “will not be generalizable to the universe of all PAE activity.” In another submission to the OMB, the FTC recognizes that “the case study should be viewed as descriptive and probative for future studies seeking to explore the relationships between organizational form and assertion behavior.”

However, this doesn’t mean that no one will use the study to advocate for drastic changes to the patent system. Even before the study’s release, many people—including some FTC Commissioners themselves—have already jumped to conclusions when it comes to PAEs, arguing that they are a drag on innovation and competition. Yet these same people say that we need this study because there’s no good empirical data analyzing the systemic costs and benefits of PAEs. They can’t have it both ways. The uproar about PAEs is emblematic of the broader movement that advocates for the next big change to the patent system before we’ve even seen how the last one panned out. In this environment, it’s unlikely that the FTC and other critics will responsibly acknowledge that the study simply cannot give us an accurate assessment of the bigger picture.

Limitations of the FTC Study

Many scholars have written about the study’s fundamental limitations. As statistician Fritz Scheuren points out, there are two kinds of studies: exploratory and confirmatory. An exploratory study is a starting point that asks general questions in order to generate testable hypotheses, while a confirmatory study is then used to test the validity of those hypotheses. The FTC study, with its open-ended questions to a handful of firms, is a classic exploratory study. At best, the study will generate answers that could help researchers begin to form theories and design another round of questions for further research. Scheuren notes that while the “FTC study may well be useful at generating exploratory data with respect to PAE activity,” it “is not designed to confirm supportable subject matter conclusions.”

One significant constraint with the FTC study is that the sample size is small—only twenty-five PAEs—and the control group is even smaller—a mixture of fifteen manufacturers and non-practicing entities (NPEs) in the wireless chipset industry. Scheuren reasons that there “is also the risk of non-representative sampling and potential selection bias due to the fact that the universe of PAEs is largely unknown and likely quite diverse.” And the fact that the control group comes from one narrow industry further prevents any generalization of the results. Scheuren concludes that the FTC study “may result in potentially valuable information worthy of further study,” but that it is “not designed in a way as to support public policy decisions.”

Professor Michael Risch questions the FTC’s entire approach: “If the FTC is going to the trouble of doing a study, why not get it done right the first time and a) sample a larger number of manufacturers, in b) a more diverse area of manufacturing, and c) get identical information?” He points out that the FTC won’t be well-positioned to draw conclusions because the control group is not even being asked the same questions as the PAEs. Risch concludes that “any report risks looking like so many others: a static look at an industry with no benchmark to compare it to.” Professor Kristen Osenga echoes these same sentiments and notes that “the study has been shaped in a way that will simply add fuel to the anti–‘patent troll’ fire without providing any data that would explain the best way to fix the real problems in the patent field today.”

Osenga further argues that the study is flawed since the FTC’s definition of PAEs perpetuates the myth that patent licensing firms are all the same. The reality is that many different types of businesses fall under the “PAE” umbrella, and it makes no sense to impute the actions of a small subset to the entire group when making policy recommendations. Moreover, Osenga questions the FTC’s “shortsighted viewpoint” of the potential benefits of PAEs, and she doubts how the “impact on innovation and competition” will be ascertainable given the questions being asked. Anne Layne-Farrar expresses similar doubts about the conclusions that can be drawn from the FTC study since only licensors are being surveyed. She posits that it “cannot generate a full dataset for understanding the conduct of the parties in patent license negotiation or the reasons for the failure of negotiations.”

Layne-Farrar concludes that the FTC study “can point us in fruitful directions for further inquiry and may offer context for interpreting quantitative studies of PAE litigation, but should not be used to justify any policy changes.” Consistent with the FTC’s own admissions of the study’s limitations, this is the real bottom line of what we should expect. The study will have no predictive power because it only looks at how a small sample of firms affect a few other players within the patent ecosystem. It does not quantify how that activity ultimately affects innovation and competition—the very information needed to support policy recommendations. The FTC study is not intended to produce the sort of compelling statistical data that can be extrapolated to the larger universe of firms.

FTC Commissioners Put Cart Before Horse

The FTC has a history of bias against PAEs, as demonstrated in its 2011 report that skeptically questioned the “uncertain benefits” of PAEs while assuming their “detrimental effects” in undermining innovation. That report recommended special remedy rules for PAEs, even as the FTC acknowledged the lack of objective evidence of systemic failure and the difficulty of distinguishing “patent transactions that harm innovation from those that promote it.” With its new study, the FTC concedes to the OMB that much is still not known about PAEs and that the findings will be preliminary and non-generalizable. However, this hasn’t prevented some Commissioners from putting the cart before the horse with PAEs.

In fact, the very call for the FTC to institute the PAE study started with its conclusion. In her 2013 speech suggesting the study, FTC Chairwoman Edith Ramirez recognized that “we still have only snapshots of the costs and benefits of PAE activity” and that “we will need to learn a lot more” in order “to see the full competitive picture.” While acknowledging the vast potential benefits of PAEs in rewarding invention, benefiting competition and consumers, reducing enforcement hurdles, increasing liquidity, encouraging venture capital investment, and funding R&D, she nevertheless concluded that “PAEs exploit underlying problems in the patent system to the detriment of innovation and consumers.” And despite the admitted lack of data, Ramirez stressed “the critical importance of continuing the effort on patent reform to limit the costs associated with some types of PAE activity.”

This position is duplicitous: If the costs and benefits of PAEs are still unknown, what justifies Ramirez’s rushed call for immediate action? While benefits have to be weighed against costs, it’s clear that she’s already jumped to the conclusion that the costs outweigh the benefits. In another speech a few months later, Ramirez noted that the “troubling stories” about PAEs “don’t tell us much about the competitive costs and benefits of PAE activity.” Despite this admission, Ramirez called for “a much broader response to flaws in the patent system that fuel inefficient behavior by PAEs.” And while Ramirez said that understanding “the PAE business model will inform the policy dialogue,” she stated that “it will not change the pressing need for additional progress on patent reform.”

Likewise, in an early 2014 speech, Commissioner Julie Brill ignored the study’s inherent limitations and exploratory nature. She predicted that the study “will provide a fuller and more accurate picture of PAE activity” that “will be put to good use by Congress and others who examine closely the activities of PAEs.” Remarkably, Brill stated that “the FTC and other law enforcement agencies” should not “wait on the results of the 6(b) study before undertaking enforcement actions against PAE activity that crosses the line.” Even without the study’s results, she thought that “reforms to the patent system are clearly warranted.” In Brill’s view, the study would only be useful for determining whether “additional reforms are warranted” to curb the activities of PAEs.

It appears that these Commissioners have already decided—in the absence of any reliable data on the systemic effects of PAE activity—that drastic changes to the patent system are necessary. Given their clear bias in this area, there is little hope that they will acknowledge the deep limitations of the study once it is released.

Commentators Jump the Gun

Unsurprisingly, many supporters of the study have filed comments with the FTC arguing that the study is needed to fill the huge void in empirical data on the costs and benefits associated with PAEs. Some even simultaneously argue that the costs of PAEs far outweigh the benefits, suggesting that they have already jumped to their conclusion and just want the data to back it up. Despite the study’s serious limitations, these commentators appear primed to use it to justify their foregone policy recommendations.

For example, the Consumer Electronics Association applauded “the FTC’s efforts to assess the anticompetitive harms that PAEs cause on our economy as a whole,” and it argued that the study “will illuminate the many dimensions of PAEs’ conduct in a way that no other entity is capable.” At the same time, it stated that “completion of this FTC study should not stay or halt other actions by the administrative, legislative or judicial branches to address this serious issue.” The Internet Commerce Coalition stressed the importance of the study of “PAE activity in order to shed light on its effects on competition and innovation,” and it admitted that without the information, “the debate in this area cannot be empirically based.” Nonetheless, it presupposed that the study will uncover “hidden conduct of and abuses by PAEs” and that “it will still be important to reform the law in this area.”

Engine Advocacy admitted that “there is very little broad empirical data about the structure and conduct of patent assertion entities, and their effect on the economy.” It then argued that PAE activity “harms innovators, consumers, startups and the broader economy.” The Coalition for Patent Fairness called on the study “to contribute to the understanding of policymakers and the public” concerning PAEs, which it claimed “impose enormous costs on U.S. innovators, manufacturers, service providers, and, increasingly, consumers and end-users.” And to those suggesting “the potentially beneficial role of PAEs in the patent market,” it stressed that “reform be guided by the principle that the patent system is intended to incentivize and reward innovation,” not “rent-seeking” PAEs that are “exploiting problems.”

The joint comments of Public Knowledge, Electronic Frontier Foundation, & Engine Advocacy emphasized the fact that information about PAEs “currently remains limited” and that what is “publicly known largely consists of lawsuits filed in court and anecdotal information.” Despite admitting that “broad empirical data often remains lacking,” the groups also suggested that the study “does not mean that legislative efforts should be stalled” since “the harms of PAE activity are well known and already amenable to legislative reform.” In fact, they contended not only that “a problem exists,” but that there’s even “reason to believe the scope is even larger than what has already been reported.”

Given this pervasive and unfounded bias against PAEs, there’s little hope that these and other critics will acknowledge the study’s serious limitations. Instead, it’s far more likely that they will point to the study as concrete evidence that even more sweeping changes to the patent system are in order.

Conclusion

While the FTC study may generate interesting information about a handful of firms, it won’t tell us much about how PAEs affect competition and innovation in general. The study is simply not designed to do this. It instead is a fact-finding mission, the results of which could guide future missions. Such empirical research can be valuable, but it’s very important to recognize the limited utility of the information being collected. And it’s crucial not to draw policy conclusions from it. Unfortunately, if the comments of some of the Commissioners and supporters of the study are any indication, many critics have already made up their minds about the net effects of PAEs, and they will likely use the study to perpetuate the biased anti-patent fervor that has captured so much attention in recent years.

Share on FacebookShare on Google+Tweet about this on TwitterShare on RedditPrint this pageEmail this to someone