BIO Report Indicts "Patent Reform" Proponents

    By Kevin E. Noonan --

The Biotechnology Industry Organization (BIO) earlier this week released a report on the purported bases of current patent "reform" efforts that have culminated in passage of a bill in the House of Representatives (H.R. 1908) and consideration of another bill by the Senate (S. 1145).  The Senate bill has been passed out of the Judiciary Committee and is said to be on the Senate's calendar for a floor vote this month, although growing, vocal opposition may derail these efforts.

The BIO report was prepared by Ann Mills, an economist and Assistant Professor, and Dr. Patricia Tereskerz, an attorney, epidemiologist, and Research Associate Professor in the Center for Biomedical Ethics at the University of Virginia.  The report, entitled "Proposed Patent Reform Legislation: Limitations of Empirical Data Used to Inform the Public Policy Debate," assesses the evidence underlying current patent "reform" efforts, and "a huge body of academic literature," and in particular, three studies, as forming the intellectual and evidentiary basis for these efforts.  The studies the authors identified are the Federal Trade Commission's (FTC) report: "To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy"; the National Academy of Sciences (NAS) Committee on Intellectual Property Rights in the Knowledge-Based Economy report: "A Patent System for the 21st Century" (which can be purchased or viewed for free here); and the National Research Council's (NRC) Committee on Intellectual Property Rights in Genomic and Protein Research and Innovation report: "Reaping the Benefits of Genomic and Proteomic Research: Intellectual Property Rights, Innovation, and Public Health" (which can be purchased or viewed for free here).  The authors reviewed the methodologies, evidence, and conclusions of each of these reports to assess whether the recommendations were in fact supported by the evidence presented.

The results constitute a scathing indictment of each of the reports, to different degrees, but with one clear conclusion:

[T]he empirical data which is being used to justify the need for overhaul either has serious methodological limitations or is non-existent.  Furthermore, and ironically, the limited studies which do exist, at least in the biotechnology industry, have not found that the patent system imposes serious impediments to innovation and successful commercialization.

Instead, the report illustrates that the different parties involved in the reports saw what they expected to see, a patent system in crisis, based on their own prejudices, interests, and political agendas, and then reported their results in the face of either no empirical evidence, or worse, evidence contrary to their conclusions.

The FTC report in some ways is the most benign, as it involved a survey of 300 participants across the broad spectrum of technology industries involved in the patent system.  These included biotechnology and the pharmaceutical industry; computer hardware manufacturers; and computer software developers and Internet companies.  The survey addressed two related questions:  first, whether the proliferation of patenting activities and patents in each industry had or was at risk for creating a "patent thicket" which would impair innovation; and second, whether the Patent Office was granting poor-quality patents.  Predictably, the results varied across the different industries.  For biotechnology and pharma, the patent thicket problem was not an issue; there may be an exception for research tools and for companies that produce DNA chips, for example, since such chips may require licensing of many different patents to the different genes (up to 10,000) on the chip.  However, in general, the biuotechnology and pharmaceutical industry did not report a concern about patent thickets, perhaps the authors noted because these industries had a lower number of patents.  Also, patents in these industries tend to be more discrete, with products not encompassing multiple patents from different patent-holders as may be common in other industries.

Similarly, poor quality patents were not reported to be a general concern.  Respondents did report that "incrementally modified" drug patents presented a potential problem, due to their capacity to extend patent protection that reduced the effectiveness of generic drug entry.  Litigation was also mentioned as a concern with regard to generics, as was the FTC's own "Generic Drug Study," which was directed at whether FDA policies were thwarting generic-promoting provisions of the Hatch-Waxman Act.  The Virginia authors note, however, that the concerns raised by the Generic Drug Study have been overcome by modifications of the FDA's policies.

The authors also cite the result of a study by Walsh and coworkers (John P. Walsh, Ashish Arora & Wesley Cohen, Effects of Research Tool Patents and Licensing on Biomedical Innovation, in Patents in the Knowledge-Based Economy (Wesley M. Cohen & Stephen Merrill eds), 285-340 (2003); which can be purchased or viewed for free here) that showed that the "tragedy of the anticommons" posited by Heller and Eisenberg (M. Heller and R.S. Eisenberg, "Can Patents Deter Innovation? The Anticommons in Biomedical Research," Science 280: 698-701 (1998)) had, in fact, not come to pass.  Indeed, the biotechnology industry has developed means to avoid patent issues, and the Utility and Written Description Guidelines for examining gene patents (among others) had "clarify[ied] patentability standards for emerging technologies, such as gene-related technologies."  Moreover, guidelines from the National Institutes of Health regarding disposition of materials (including cloned genes) obtained using government monies had also modulated any anticommons tendencies in this industry.

Finally, although some respondents voiced concerns about patent quality generally, the overall perception was that patent quality was high due to the attention this technology area had received from the Patent Office in promulgating examination guidelines.

The situation was not very much different in the computer hardware industry from the biotech and pharma areas, despite differing circumstances from the biotech and pharma areas.  Patent thickets were perceived to be more likely in this industry, because innovation tended to be cumulative, participants were more likely to patent defensively, and patents in this area were perceived to be easier to obtain.  The report noted that the technology in this industry was "complex, rapidly changing, and characterized by incremental innovation" as evidence of the threat that in the industry was being hampered by patent thicket problems (despite not receiving this response to this degree from study participants).

The report also noted the possible depredations of patent "trolls," based on a study that purported to support the conclusion that "Non-Practicing Entities" could contribute to the purported patent thicket through litigation (Bronwyn H. Hall and Rosemarie H. Ziedonis, "The Patent Paradox Revisited: An Empirical Study of Patenting in the US Semiconductor Industry," RAND Journal of Economics (32)1: 101-28 (2001)).  However, the BIO report authors note that the cited report was based not on an empirical study, but on a survey showing that one-third of the lawsuits filed by a group of 136 companies involved patents not invented by the company.  However, not all these patents were asserted by patent trolls, since the study did not take into account transfer of patents between "practicing entities."  Indeed, when the BIO authors reviewed the evidence "supporting" the study's conclusions, they found merely the observation that "[t]here was also an apparent rise in infringement suits brought by specialized 'patent licensing' companies."  Finally, while the study participants in this industry acknowledged the potential for patent thickets to arise, they also noted that "cross-licensing, patent pools, and the licensing requirements of standard setting organizations" have ameliorated this tendency.

The final group, the software developers and Internet companies, held "the most divergent" views about the role of patenting in their industry.  These participants noted that patents were less effective due to other forms of protection, such as copyright, the effects of open-source software, and the effects of competition as the main driver for innovation.  Like their computer hardware industry counterparts, these respondents recognized that innovation tended to be incremental.  In addition, more of these industry participants disputed that patents on software and business methods had a positive effect on innovation.  There was also recognition that the cycle of innovation in this industry was frequently out of sync with the timeframe required for patent protection, so the technology was obsolete before any meaningful patent protection could be obtained.  Because products could contain "dozens, if not hundreds, of patent[ed] . . . individual components" the possibility of patent thickets was high, as was the perceived negative effects of defensive patenting.

The BIO authors conclude that the FTC report "panelists across industries disagreed about the nature and degree to the patent system might be hindering innovation and successful commercialization specifically with regard to the issues of an anti-commons or patent thicket and questionable or poor quality patents," and that "in the biotechnology industry and the computer hardware industries, to the degree to which industry participants are experiencing patent thickets and facing issues associated poor quality patents, they are finding solutions or ways around these problems."

Turning to the NAS report, the BIO authors noted that this report was more concerned with theoretical considerations raised by academic commentators on the possibility that patents, particularly poor-quality patents, could stifle innovation.  Focusing on patent quality, the NAS report assessed three "direct: measures of quality:  "1) the ratio of invalid to valid patent determinations in infringement lawsuits; 2) the error rate in USPTO quality assurance reviews of allowed patent applications; and 3) the rate of claim cancellation or amendment or outright patent revocation in re-examination proceedings in the USPTO."  The BIO authors note that, while seemingly benefiting from a patina of objective, quantifiable data, the actual numbers of patents that "qualified" under the NAS criteria were low:  "the litigation rate was just over 1 percent; only 2-3 percent of a year's  patents are reviewed [for quality] by the USPTO; and re-examined patents represent about 0.3 percent of the total number of patents."  Yet, "despite the lack of evidence showing a systematic decrease in the quality of patents issued – and some evidence showing that the quality of patents has not declined but is actually improving, the NAS believes that the USPTO is issuing more poor quality patents" according to the BIO report authors.  The NAS cited four reasons for their conclusions:  the failure of the Patent Office to automate filing and processing applications; the relatively higher patent approval rate in the U.S. as compared with other countries; the difficulties in examining genomic and business methods patents (despite successful Patent Office efforts specifically relating to each) and finally a perceived "dilution" in the non-obviousness standard.  The BIO authors note that the evidentiary basis for the second of these factors is, to put it kindly, questionable:  while Quillen and Webster estimated that the approval rate is between 85 percent and 97 percent, others put the approval rate at closer to 75 percent, and the USPTO reported an approval rate of 54% in 2006.

The NAS report also raised the issue of patent thicket and anticommons effects in biotechnology, despite the results of the Walsh study to the contrary (and ironically, it was the NAS that commissioned the Walsh study, only to ignore its conclusions when they did not "fit" the expected outcome).  Finally, the NAS report addressed the potential for patents to limit access to clinical testing, based on a study by Merz and colleagues (J. Merz, F.D.G. Kriss, D.D.G. Leonard, and M.K. Cho, "Diagnostic Testing Fails the Test," Nature 415: 577-79 (2002)).  The study reported that 25% of laboratory physicians reported abandoning a clinical test due to patent royalties, but the BIO authors noted that clinical labs typically charge insurers for such tests, in contrast to laboratories engaged in non-commercial research.  Moreover, the NAS study found no evidence that patients lacked access to needed clinical tests as the results of patents.

Finally, the BIO report considered the NRC's study, which was the most limited in scope and was directed to "trends in patenting genomic and protein related inventions."  Surprisingly, the numbers of these patents actually declined in the 2000-2001 timeframe, no doubt due in part to application of the new utility and written description guidelines on applications submitted before the guidelines were promulgated by the Office.  In addition, a survey of 30 U.S. universities and other academic institutions (19 of which responded) owning 75% of the granted genomics patents (2,700 patents) showed their licensing practices "broadly consistent" with the NIH guidelines (including the Guidelines for Licensing Genomic Inventions).  Finally, the NRC commissioned a second Walsh study, directed to a sampling of 1,124 persons that included university investigators, government laboratory workers, and 563 industry scientists.  There were also 299 researchers in "the fields of biomedical sciences of high commercial interest (because of their association with normal and disease-associated cellular processes)."  Thirty-three percent of the sampling responded.  The results of this Walsh study were strikingly similar to the results of the first study.  Of a total of 274 persons whose research goals were "drug discovery" "basic research" or "other," "the top five reasons for project abandonment, were, in order of frequency, 'lack of funding' (62%), 'conflict with other priorities' (60%), 'a judgment that the project was not feasible' (46%), 'not scientifically important (40%),' and 'not that interesting' (35%)  -  in short, scientific reasons having nothing to do with patenting.  Only 10% of all respondents cited 'unreasonable terms for obtaining research inputs' and only 3% cited 'too many patents covering needed research inputs' as reasons for abandoning a research project.

The BIO authors, in summary, report that despite the different scope, goals, and perspectives of the reports and their authors, all of the reports conclude that patents pose a danger to innovation due to patent thickets, anticommons effects, and excessive litigation, and that patent quality is poor.  The three reports also have similar recommendations, including increased post-grant review and other measures to challenge granted patents, a higher non-obviousness standard, and ensuring adequate funding for the Patent Office.  (Paradoxically, this last recommendation has in fact been achieved, as Patent Docs learned from John White on Monday; see "PLI's John White Discusses Tafas/GSK v. Dudas.")

The problem (and the scandal) is that the BIO authors show that there is no empirical evidence for the need for any of these recommended changes.  They do not mince words in this regard:  "the call to overhaul the patent system to date is based more on conjecture, anecdote, and individual publicized cases, rather than upon reality."  They cite the "coincidence" of the timing of these reports and the "ethical, legal, and economic concerns" over "patents in the biotechnology industry such as those involving genes," where "policy has largely been driven by social unease, preliminary data, and literature on adverse practical ramifications, as well as several highly publicized patent protection controversies."  Indeed, the only empirical studies cited by the reports, the two Walsh studies, provided conclusions opposite to the conclusions contained in the reports.  The BIO report concludes with a consideration of the eBay v. MercExchange and KSR Int'l Co. v. Teleflex Inc. cases as evidence that some of the patent quality and patent "troll" concerns raised by reports have been already addressed.

As stated in the BIO report, "there is a lack of evidence that justifies overhauling the patent system in a way that could potentially disrupt the incentives of industries that rely on patents to innovate."  Paradoxically, the BIO authors find evidence in each of the three reports that the current patent system is working, and working well, particularly for the biotechnology industry.  These three reports, along with several academic polemics supported by dubious premises and containing insufficiently-supported conclusions, are the intellectual basis for the patent "reform" efforts currently before Congress.  It is unfortunate that the House and Senate have labored on this legislation without the benefit of the BIO report's analysis of the "unclothed emperor" characteristics of these reports, and the extent to which the evidence does not support the rabble's chanting that the patent system is broken and in need of extensive overhaul.  But it would be even more unfortunate if we didn't all take the time to write our Senators and be sure they know that we know that the patent "reform" movement is without empirical basis, and that we expect them to vote accordingly.

For additional information on this and other related topics, please see:

• "Millennium Pharmaceuticals Spent $1.28 Million on Lobbying in 2007," February 8, 2008
• "Patent Reform and Infringement Damages: Some Economic Reasoning," February 5, 2008
• "Department of Commerce Sends Letter on Patent Reform to Senator Leahy," February 4, 2008
• "Biotech and Pharma Opposition to Senate Patent Reform Bill," February 3, 2008
• "The Letters Keep Coming Over the Senate Transom," January 30, 2008
• "U.S. Senate Mailbox Filling with Letters against Passage of Patent 'Reform' Bill: An Update," January 23, 2008
• "U.S. Senate Mailbox Filling with Letters against Passage of Patent 'Reform' Bill," January 18, 2008
• "Patent Reform Discussed on Senate Floor," December 21, 2007
• "Enjoined New Rules and Patent Reform Finally Appearing on Biotech Industry's Radar," December 20, 2007
• "Chinese IP Judge Discusses Implications of U.S. Patent Reform Bill and Two Congressmen Heed Warning," December 17, 2007
• "IPO President Seeks Deletion of Patent Reform Provision," December 12, 2007
• "Senate May Act on Patent 'Reform' Bill in the New Year," December 2, 2007
• "The Wall Street Journal Gets It Half Right," November 5, 2007
• "BIO CEO Provides Briefing on Follow-On Biologics and Patent Reform," September 18, 2007
• "Patent 'Reform' Bill Passes House of Representatives," September 9, 2007
• "Reversal in Microsoft Case Weakens Patent Reform Argument," August 7, 2007
• "San Francisco Chronicle Opines on Patent Reform," August 6, 2007
• "Patent Reform Bill to Be Delayed?" June 12, 2007
• "Senate Judiciary Committee Holds Hearing on Patent Reform," June 10, 2007
• "Could Creating a U.S. 'Utility Model' Patent Fulfill the 'Need' for Patent Law Reform?" May 21, 2007

Comments

[jwint]

Great analysis, Kevin.

[EG]

In my comments to my two Senators urging them to vote against S. 1145, I specifically referred to this BIO Report to show how those supporting the so-called Patent Law Reform have no factual case to support any alleged "rampant abuse" of the patent system that would require such so-called "reform". I urge others who realize S. 1145 and it's "evil twin" H.R. 1098 are "shams" to do likewise.