Editor’s Note: This is part 1 of a 5 part series written by Steven J. Moore and with the assistance of Marvin Wachs and Timothy Moore, also of the Kelley Drye & Warren Patent Department.
PROLOGUE : On the Isles of the White Knights and the Lowly Bridge Trolls
Once upon a time, long, long ago, in a land far, far away, there existed an island kingdom called Inventis, a legendary land of technologically advanced people. Inventis was composed of many small islands, each separated by raging rivers of hot lava said to have spewed from the mouths of dragons. The only way to move from one island to the next was by way of a bridge of stone. Several bridges had been built in the past, in each case unlocking unanticipated riches on the bridged island. While it was thought that more riches were on the other islands, because of the effort and danger of building a stone bridge over hot rivers of lava, most of the islands of Inventis remained inaccessible. This was a problem for the King as the population of Inventis was growing rapidly, and the people simply needed more room. The King, looking to promote the progress of bridge building, promulgated an edict whereby any bridge builder was granted, for a time, the exclusive right to any new and useful discovery made on any previously isolated island. To assure that any bridge built would be sound, the King restricted such exclusive right to bridge builders who fabricated their bridges to exacting engineering specifications. He placed the responsibility to assure these provisos were met in the hands of his Ministry of Bridge Patency (“Patency Office”) which also provided administrative mechanisms for anyone wishing to challenge any asserted exclusive right.
On the main island there lived two distinct groups of artisans who were adept at bridge building. One group, the Commercializers of the Enclave of Corporate, sold many products and services to the public. Among their number were some large and sturdy White Knights. The other group, the Non-Commercializers of the Enclave of NPE, were known as dreamers who paid little immediate heed to making money through commerce. Although among their number were independent thinkers, new entrepreneurs, failed commercializers, university scholars and basic researchers, there were also some hideous, slinking creatures who had polluted waters of the mainland by making a living dredging up the wreckage of the past, and marketing it as valuable. Because of these slinking creatures, the Knights referred to all those coming from NPE as “Trolls.”
The Corporate Knights immediately began to build bridges, doing so in the dead of night so as to hide their fording techniques from others. In each case, the Knights found new succulent fruits, which the scientists classified as a species of Nirvan, which they sold to the public in the form of pies at a handsome profit. The Trolls, who were likewise prodigious bridge builders, were afraid of the dark of the night and thus decided to build their bridges in only shaded areas to conceal, as much as possible, their own fording techniques. Surprisingly, instead of finding succulent fruits on their islands, the Trolls commonly discovered special and distinct materials which the Knights uncannily very often found highly useful for the profitable commercialization of their discoveries.
The Knights, understanding the Trolls dread of the dark, one day decided to sneak across the Troll bridges in the dead of night to gather their advantageous materials without the Trolls’ permission. The Trolls, however, found out about such actions. They immediately went to the King’s Courts and demanded payment in gold from the Knights.
The Knights protested the Troll suits, asserting it was unfair that the Trolls, who never sold anything fabricated from their discovered materials to the public, should impede the Knight’s use of such materials. They argued that the Trolls’ newly discovered materials were unworthy of protection under the King’s edict, and that the Trolls’ discoveries were negated by failing to build bridges to the required engineering standards. The Trolls countered that the Knights had also built many similarly engineered bridges to islands where the Knights likewise held their discoveries in reserve for trade among competing Knights.
The Knights were incensed by the insolence of the Trolls and began to wage war in the press regarding the Trolls’ negative impact on the smaller Commercializers from whom tribute was also sought. They also sought expeditious redress from the Patency Office to confirm the invalidity of the Trolls’ claims, and an alteration in the law, known as the Anti-Troll Inventis Act (the “AIA”), ostensibly to aid small Commercializers in more easily challenging the Trolls, and to make it harder and more expensive for the Trolls to adjudicate their claims in the King’s courts.
Over time the public heard over and over again that the Courts and Patency Office were often rejecting the Troll’s claims. Thus, the public soon began to chime: “Strong Bridges built by the (K)night invariably lead to Nirvana, while rickety Troll bridges built where the sun does not shine always lead to all kinds of shit”(fertilizer that is!). And thus, the kingdom lived happily ever after (that is, until a huge tidal wave hit Inventis, causing it to be lost to the oceans – now known in antiquity as the Lost Isles of Inventis).
Act 2: BACK TO REALITY – DESIGNING A STUDY TO TEST THE FABLE
And thus the fable goes – the patents (i.e., the bridges of the fable) of the large commercializers (the Knights) are strong and worthy, while patents asserted by other entities against the Knights, particularly those patents owned by an entity that does not commercially make use of at least one of their discoveries, are weak and unworthy. While this fable developed in 1993 has been stoked by a continually beating drum for 20 years, it has not been until just recently that the tools available to check out the validity of fable have become available to all . Given that this fable is being trumpeted as the basis for yet another round of large changes in the patent law, we feel the timing of our article could not be better.
When we began our study, we had no idea how far or in which direction it would lead us. Our original study, which in itself was a Herculean task, had very little, to do with Patent Trolls. This study thought only to probe the basis behind the USPTO’s high grant rate for reexaminations of the patents brought before it, something to our knowledge no one else had ever investigated. We had long ascribed to the position set forth by the USPTO, that is, that the USPTO rate of grant is high because only truly infirm patents were being challenged by reexamination. This seemed to answer why we had always seemed to find substantial invalidating art when clients hired us to file reexaminations, and, in truth, why we have had great success in such matters. However, we were intrigued by several authors who had begun positing more pedestrian bases for the high grant rate by the USPTO, such as the relatively high standard for non-obviousness that had developed after KSR, the very broad application of “broadest reasonable interpretation” test by the USPTO, or simply because the USPTO was looking at such petitions as a money making proposition.
To determine whether the high grant rate related to something other than just poor patents, we had decided to attempt to challenge at random patents from companies that are among the most vocal as to the strength of their patents. We fully anticipated that we would not be able to make out a case of obviousness with respect to claims of all the patents. To make the challenge interesting, the project was given to someone in our group who had never failed to cause significant damage to a patent owner in a filed reexamination, someone we euphemistically call the “Reexaminator.”
The project was conducted over several years in order to avoid excessive billable down time, and was rapidly finished before the new fees for ex parte reexamination went into effect. Probably spurred on by his record, the Reexaminator lived up to his reputation, each time upon taking up an assigned patent by finding significant prior art flaws with the claims. By the time the Reexaminator had reached the sixth patent, we were completely surprised to find all the patents selected at random seemed to have fatal prior art obviousness flaws in many, if not most, of their claims. This fact was confirmed when we filed six blinded ex parte reexaminations against each of the patents, and all requests were granted. The results of this study will be set forth in a future article. Our results from the challenge intrigued us, as it suggested that with enough thought and manpower the majority of patents might be found to be infirm in some manner. This seemed particularly true, as in each case, the Reexaminator had “wacked” the broadest independent claim. In an upcoming paper we will discuss our results, and our conclusions.
At about the time our results were coming in, there was a great deal of press relating to the poor patents of the so-called “Trolls” as compared to the patents of corporate America. These trolls were said to be raising havoc in the patent field. In particular the rhetoric regarding the “troll situation” was being raised many octaves by a paper published by Bessen and Neurer using RPX data, a company with a vested interest in promoting the troll story. This paper suggested that patent trolls were costing U.S. Companies nearly $30 billion a year in “direct costs”. See James Bessen and Michael J. Meurer, The Direct Costs From NPE Disputes (Boston University School of Law, Law and Economics Working Paper No. 12-34, June 28, 2012). While the foundation for such a study has been savaged by other academicians, See Schwartz and Kesan, Essay: Analyzing the Role of Non-Practicing Entities in the Patent System, Draft Paper July 8, 2013, To be published in the Cornell Law Review 2014 (noting in part that the Bessen and Neurer study was funded by corporate interests with a stake in the troll story and that the “costs” of their study actually related to transfers to patent holders without making any determination that such payments were in fact due or not), it is clear just from an simple internet search that the Bessen and Neurer assertions were heavily promoted in the press.
As new data bases and programs were coming available at the time, we thought it might be interesting to look at the patents of the so-called Trolls and those of corporate America to determine whether they truly fared differently from one. When we started peaking behind the curtain, as had been done to the dismay of the Wizard in the 1930’s movie the Wizard of Oz and the surprise of Dorothy, we started seeing things that seemed very outside what we have been told for years is hard fact.
Wikipedia tells us that the term “troll” was first popularized in 2001 by Peter Detkin, former Assistant General Counsel at Intel. Detkin is said to have later distanced himself from the term as started to be used others to refer to patent holders who were not asserting patents that they themselves were not practicing. The latter is said to have occurred because Detkin recognized that Intel itself often asserted patents that it had purchased but were not practicing. See, Wikipedia “Patent Troll.”
Our first move was to understand who was being identified as “Trolls” by the authorities who have been writing articles so long on the topic. We were surprised to find upon extensive searching on the Internet that we could not find a single research paper which provided any exhaustive list of the huge numbers of “Trolls” which were said to be stalking corporate America. Instead we found many papers making their analyses based upon proprietary databases that were not available to the public, and thus the veracity of the designation “Troll” was not subject to any validation. We decided to do some investigative reporting, contacting a number of the authors of the so-called troll stories. We were surprised over and over again to be given the same explanation – the lists of trolls they were referencing in the articles were considered “trade secrets”.
This got us to wonder – if Congress is taking credence in these reports, shouldn’t these authorities be upfront about whom they were characterizing as a “Troll”? Months after our calls, we were able to finally identify, for the first time, an article published identifying an extensive list of so called “monetizers,” albeit the list was limited to entities asserting patents in classes 370 (telephonic communication and/or 455 (telecommunication) alone. See Robin Feldman, Tom Ewing & Sara Jeruss, The AIA 500 Expanded: The Effects of Patent Monetization Entities, (UCLA Journal of Law & Technology, Research Paper No. 45), available at http://ssrn.com/abstract=2247195. However, it was still unclear to us whether we should be calling any entity wishing to monetize a right it holds to exclude others equated to the despicable “Troll” of the fairy tale.
As noted by Professor Mark Lemley’s group in the University of Pennsylvania Law Review article “Extreme Value or Trolls on Top? The Characteristics of the Most-Litigated Patents,” Vol. 158, p. 1 – 37, if one calls a “patent troll” any company enforcing a patent that covers inventions that they do not practice, the number of trolls goes dramatically up, while if one counts “only companies enforcing patents that cover inventions they did not themselves develop, the number of patent trolls drops dramatically” (from their 2010 data to 12% of the patents and 7% of the lawsuits).
Given the difficulty of identifying a “Troll” without a universally accepted definition, we decided in designing our study of the fable of the patent troll to jettison the whole idea of the “boogey man” and instead to look at the entities actually asserting and facing the challenges, in terms of whether an entity was or was not set up to sell or provide to the public goods and/or services (whom we characterized as a “Producer”). We further investigated whether an entity was an independent inventor or simply an alter ego of the independent inventors listed on a patent (that is the company’s principals were the inventors on the patent being asserted), whether the entity was an emerging or research based company (an entity that does not yet sell a product but is expending its funds predominantly on research and development of a product), whether the entity was a university, or whether the entity seemed to be only engaged in licensing and litigation activities.
The Troll story has been iterated over and over again. We are told that they have a “great, very smart plan” settling for low amounts of money “that hiring a lawyer to just look at the case would be a lot costlier than giving up and paying right away.” See Jesus Diaz, Patent Troll is Suing Small Hotels and Coffiee Shops for Offering Wi-Fi, Gizmodo, October 4,2011, http://gizmodo.com/5846419/patent-troll-is-suing-small-hotels-and-coffee-shops-for-offering-wi+fi. Three questions immediately jump out regarding such depiction of the troll: First, isn’t it very rare to obtain a patent that covers the products or services of a large number of companies? Second, doesn’t it cost more to file suit and communicate with defendants through an attorney than a few thousand dollars? Third, if the patents were objectively without doubt obvious, why did so many defendants acquiesce to settle rather than file an ex parte reexamination of the patent in the past, when the cost of such filing was only $2500? Although we did not quantize the same it was pretty clear to us at the dockets of numerous NPE and Producer cases, that there seemed to be little support for the rapid settlement scenario story as made out by many with respect to the “Trolls,” at least with respect to the non-independent inventor related NPEs.
In designing our study we decided to look behind the patents that were being asserted in litigation to determine whether such patents were initially assigned to a company selling or providing goods and/or services, to an independent inventor, to an emerging or research based company, or to a university. This seemed to make sense to us, as more and more articles were being written about NPEs being set up or contracted by large and small producing corporations to monetize their patents. It appeared to us to incongruous to append the word “Troll” to an entity which was a producing company using a proxy to bring suit, but not appending such word to the same producing company if it brought suit in its own name. That is, why would one treat differently an entity set up by a large producing company to provide value from its patents (or an entity which was sharing in any valuation with a large producing company when obtaining value on litigated patents previously held by them) differently from the actual company that was seeking value? We also felt that it was time to separate out suits brought by individual inventors, who under Hewlett Packard v. Acceleron, 587 F.3d 1358 (Fed. Cir. 2009) were being forced to file suit rather than sending a licensing letter to a company to adjudge interest in their patents (in order to avoid a potential declaratory judgment action in some far flung jurisdiction).
In terms of entities involved in a challenge, we understood that the patent office has long segregated filers into so-called “small” and “large” entities depending on whether they have more or less than 500 employees, based on a definition from the Small Business Administration. To make our data somewhat comparable, we investigated how the different entities involved in patent challenges characterized themselves when they filed their patent applications, that is whether they filed under “large” or “small” entity status (of course, this allowed us to determine which of these entities did not even participate in the patent realm). Furthermore, we used the Forbes Global 2000 list of the largest public companies in the world (2012 – basically companies having a valuation of 1 billion or more, or sales over 1 billion dollars), along with the 250 largest private companies as reported by Fortune (companies with net sales over 2 billion dollars), which we reference herein as the “Global 2000+” to distinguish the truly large companies from other Producers. We note that our Global 2000+ list underestimates full range of the world’s largest corporations as information on private companies is difficult to obtain, and the list of private companies we could find did not account for any private companies that had annual sales between 1 billion dollars and 2 billion dollars.
Our study looked at what actually happened when patent suits were brought by each of the different entities identified above. To do this study, we recognized that we would need to look at cases brought at least 5 years ago, to maximize the possibility that the cases were actually resolved through litigation and/or parallel reexamination. This study involved following hundreds (267) of randomly selected patent litigation cases filed between 2006 and 2008 (“pre-AIA”) through termination, and through any intervening reexamination. We also looked at 428 random cases filed after the AIA to check on the statements being made about the “DEVELOPING” “Troll” problem.
The results set forth below required extensive searches into among other things, the commercial status of parties involved in any challenge, the size status of each party involved in the challenge, the assignment history of any asserted or challenged patents in the challenge, the initial and ultimate outcome of any challenge, time between patent issuance and filing of suit, the number of unique challengers with respect to a data set, the corporate structure of the challenger, and information available concerning licensing activities of the challengers. In terms of coding, we truly feel that our coding was completely unbiased. In truth no one cared one way or the other how things turned out, we just wanted to understand what was really happening given all the conflicting rhetoric. While our group has represented companies in litigations and reexaminations against what some have characterized as the larger “trolls,” our group has not represented any of these so-called “troll” entities, and we did not feel any compunction about shading the data one way or the other.
We warn that while our conclusions are based on significant sample sizes which we believe will reflect the larger data available, they are not based on extremely large sample sizes. Thus, it is possible that one or more of our conclusions may be contradicted in the future when looking at larger samples. Unfortunately our work had to be limited given that the entire study was self-funded both in time and expense not by some large institution, but by a single department in a law firm. What we truly hope this study will do is to impel others, without an agenda in the outcome (i.e., a group without any support from large corporate America or the patent assertion entities), who have significant pecuniary deep pockets to probe further our findings.
Lastly, we understand that the conclusions of our study may be found by some to be controversial, in being contrary to long held beliefs that have been held over the years by numerous parties, including most recently by some members of Congress, the Executive Branch and Judicial Branch (are there any other branches of government??!). Indeed we expect some would publish these results with significant trepidation. However, given that our group has never been accused of timidity in representing clients before the courts or the USPTO, as demonstrated by our legal representation of Dr. Tafas in his famous challenge of the USPTO rules in Tafas v. Dudas (with others, except for GlaxoSmithKline, standing on the sidelines), we set forth the results of our study below.
 The Stanford IP Litigation Clearinghouse, directed by Mark Lemley and Joshua Walker, was launched in December 2008, but permitted only limited access to the general public at large. Today, it is said to include more than 100,000 patent infringement outcomes, each as determined by members of the Stanford group. Granular access to the database has been traditionally limited to the government, academicians, and select others. Limited access is provided by subscription through Lex Machina. Although we subscribe to Lex Machina, we were unable to use the service to check outcomes against the Lexis database we utilized (Lexis Total Patent).