Bsc(IM) Student: Niels Hald Kristoffersen Supervisor: John Howells An Assessment of the Evidence that “Patent Trolls” Have a Significant Impact on Innovation and an Exploration of the Proposed Policy Responses Aarhus University School of Business and Social Sciences Department of Business Administration May 2014 Characters Excluding Blanks: 109,123
54
Embed
An Assessment of the Evidence that “Patent Trolls” …pure.au.dk/portal/files/75088235/Patent_Troll_Thesis_Kristoffersen... · An Assessment of the Evidence that “Patent Trolls”
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Bsc(IM) Student: Niels Hald Kristoffersen
Supervisor: John Howells
An Assessment of the Evidence that “Patent Trolls” Have a Significant Impact on Innovation and an Exploration of the
Proposed Policy Responses
Aarhus University
School of Business and Social Sciences
Department of Business Administration
May 2014
Characters Excluding Blanks: 109,123
Niels Hald Kristoffersen
Page 1 of 54
Abstract
This thesis analyses synonyms of ‘patent trolls’, such as ‘Non-Practicing Entity’ and ‘Patent Assertion
Entity’, and finds that using them interchangeably with ‘patent troll’ is unwarranted. It then explores
some entities referred to as ‘trolls’, and concludes that the term can only be applied on a singular
entity basis, as it is solely the conduct of behaviour and not ownership status that can justify the
label. The thesis finds that, surprisingly, there is no existing evidence of harmful influence by actual
‘trolls’. Instead, evidence found on the erroneous synonyms is wrongfully credited to ‘trolls’ as well,
and the results of the studies analysing their impact are unreliable, and meaningless to the point of
being misleading.
Furthermore, it briefly looks at the Leahy-Smith America Invents Act, and examines some of the
proposed solutions to counter the issue, uncovering that the majority of them do not address the
problem it is claimed they address, but instead skew the balance of the patent system in favour of
wealthy or large corporations. Especially worth noticing, is the finding that the courts already have
tools to fight ‘trolls’ in court, and that future solutions, therefore, should focus on targeting the
behaviour of ‘trolls’ outside of court.
The thesis then goes on to discuss the findings and argues that the evidence does not support the
current notion of the problem, that policy change is unwarranted, and that pursuit of these is a
troublesome trend that needs to be brought to a hold.
This thesis is of relevance as much of the debate regarding patent trolls is based on misleading
information, and this thesis, unlike the current literature, provides a broad objective overview of the
issue. Due to the nature of the sources, and despite largely excluding court cases, this thesis as a
whole has come to resemble that of a literature review, and it can be used as such.
2. Exploration of the Term ‘Patent Troll’
Having established a patent as the right of a product, or a process, which can be used to exclude
others by law, what or who, then, is a patent troll?
The term ‘patent troll’ is legally not established, (Opitz & Pohlmann, 2013, p. 104) but was originally
coined by Peter Detkin when he was assistant general counsel at Intel. As he explained in a 2001
article:
“We were sued for libel for the use of the term ‘patent extortionists’ so I came up with ‘patent trolls,’
[…] A patent troll is somebody who tries to make a lot of money off a patent that they are not
practicing and have no intention of practicing and in most cases never practiced.” (Sandburg, 2001)
In short, the term ‘patent troll’ was a replacement for ‘patent extortionist.’ Therefore, the outset for
further exploration of the term is that a patent troll is a patent owner conducting some kind of
extortive behavior.
The phenomenon of ‘patent trolls’ is not new. Similar activities took place during the nineteenth
century: “Called ‘patent sharks,’ they bought dormant agricultural patents and then sued farmers
who were unknowingly using protected technology.” (Magliocca, 2007, p. 4) The debate about
patent trolls has garnered attention in recent years, and even the President, Barack Obama, has
taken a stand on the subject. When questioned about trolls he described them as “[…] trying to
essentially leverage and hijack somebody else's idea and see if they can extort some money out of
them.” (Masnick, 2013) A lot of literature has been done on this topic, though, unfortunately, as
Magliocca has noted “the debate on patent trolls is long on passion and short on proof.” (Magliocca,
2007, p. 3)This is well illustrated by Michael Rischs summary of the debate, in which he refers to
patent trolls as NPEs (Non-Practicing Entities):
“According to the detractors’ narrative, trolls are recent fly-by-night shops that assert business-
method and internet patents. Trolls assert low-quality patents in low-quality litigation. They obtain
patents from failed companies in fire sales. Worse, because trolls do not make anything, their
Niels Hald Kristoffersen
Page 9 of 54
patents do not provide anything of value to society. […] According to their proponents, NPEs create
patent markets, and those markets enhance investment in start-up companies by providing
additional liquidity options. NPEs help businesses crushed by larger competitors— competitors who
infringe valid patents with impunity. NPEs allow individual inventors to monetize their inventions.
These functions, the proponents argue, justify the existence of NPEs. To be sure, whether an NPE
qualifies as a ‘troll’ depends on who is doing the name-calling.“ (Risch, 2012, p. 459)
2.1 Definition(s) of a Patent Troll As the above suggests, there appears to be no consensus as to what exactly a ‘Patent Troll’ is or how
to describe its activities, except that they are in some way ‘good’ or ‘bad’ actors. Many terms and
descriptions have been used in an attempt to describe these patent owners. In the following
sections, synonyms of ‘patent trolls’ in existing literature will be introduced and evaluated in order
to better define what can and cannot be justified as a ‘troll’.
2.1.1 Non-Practicing Entity (NPE)
The term ‘Non-Practicing Entity’, or in short NPE, and ‘patent troll’ are often used
interchangeably.(Hannon & Welsh, 2013) As the name suggests, it is an entity in possession of a
patent it does not practice, note that the word practice does not refer to the act of acquiring skill,
but to actions as designing and manufacturing products and processes. (Hannon & Welsh, 2013)
Among examples of definitions, Shrestha defines NPEs as “[…] firms that rarely or never practice their
patents, and instead focus on earning licensing fees.” (Shrestha, 2010, p. 114) And Bessen & Meurer
talk of them as “[…] individuals and firms
who own patents but do not directly use
their patented technology to produce
goods or services, instead asserting them
against companies that do produce
goods and services.” (Bessen & Meurer,
2014, p. 390) In the article “Extreme
Value or Trolls On Top?” Allison et al.
display a table of twelve different entity
descriptions, each describing a different
category of patent owners (see Table 1).
According to their account eleven of those twelve fit the label of a NPE, with number eight being the
only entity that practices its patents. (Allison et al., 2009, p. 11) A NPE can thus be classified as an
entity that does not practice its patents.
Table 1: Copy of Table by Allison et al., 2009, p. 10, Table 1
Niels Hald Kristoffersen
Page 10 of 54
2.1.2 Patent Assertion Entity (PAE)
Within the category of NPE definitions, a popular definition used to refer to patent trolls is ‘Patent
Assertion Entities’, in short PAEs. The term was coined by Colleen V. Chien (2010) in her article
“From Arms Race to Marketplace,” where she defines PAEs as “entities that use patents primarily to
get licensing fees rather than to support the development or transfer of technology. These entities
generally use their patents to sue, or threaten to sue, practicing companies. They are invulnerable to
patent counterattack and therefore have little to lose from patent litigation besides legal fees.” (p.
300) It has gained popularity since its conception, and it has been adopted for use by agencies such
as the White House, which used it in its “Patent Assertion and U.S. Innovation” report (June 2013),
and also by the Federal Trade Commission (2011, p. 8).
2.1.3 Patent Monetization Entity
The term ‘Patent Monetization Entity’, or ‘monetizer’ as a shortening, has been suggested as a fitting
label by Feldman et al., as it solely focuses on behavior where “patent rights are […] sold, traded,
grouped, regrouped, licensed, or repurposed, all for generating an income stream from the rights
themselves.“ (2012, p. 368) Along with NPE and PAE, Patent Monetization Entity was also used by
the United States Government Accountability Office (GAO) in its Intellectual Property report to
Congressional Committees (GAO, 2013)
2.1.4 Patent Shark
‘Patent Shark’ was, as described earlier, the term used in the nineteenth century. It was used to
describe the fight between small individual farmers being sued on their farming tools by ‘royalty
agents,’ the agents then being labelled as sharks. (Magliocca, 2007) The ‘Patent Shark’ term has also
been used more recently, this time in an opposite scenario, where the small patent holders are
described as the sharks, against the big manufacturers. Specifically by Reitzig et al. who denote them
as “[…] patent ‘sharks’ or ‘trolls’,” and describe them as “[…] patent holding individuals or (often
small) firms who trap R&D intensive manufacturers in patent infringement situations in order to
receive damage awards for the illegitimate use of their technology.” (Reitzig et al., 2007, pp. 134-
135)
2.1.5 Discussion of Synonyms
As a patent is the right of a product or a process, which can be used to exclude others by law,
enforcement of it cannot constitute a bad action in itself. If these patent trolls are conducting
extortive behavior, then it cannot be on the grounds of them enforcing their granted rights to
exclude. Therefore, what constitutes a troll cannot be the act of patent enforcement. The label
Niels Hald Kristoffersen
Page 11 of 54
‘patent extortionist’ must pertain to entities conducting something more than just patent
enforcement.
What about the NPE definition, could those be the extortionists? Besides the above mentioned,
a problem with using NPE as a synonym for a patent troll is that it easily includes many entities that
might not be these bad actors, but simply own one or more patents on technologies that they do not
produce themselves. A major company like Apple Inc. could be classified as a NPE under some
definitions, as it greatly outsources its production (The Chosun Ilbo, 2012) and focuses on research
and development, i.e. Apple does not produce the goods to which they own the patented rights, and
could be characterized as a troll under these definitions (Quinn, 2013j). As is evident, categorizing all
NPEs as trolls, thus labelling any entity that does not manufacture a product as an extortionist is
outright wrong. Universities are an example of entities falling victim to the implication of defining
the trolls as NPEs, and it has been argued that as they by nature are devoted to research and
scientific progress, they should not in general be classified as these bad actors. (Lemley, 2008b;
Quinn, 2013h) To further illustrate the complexity of this issue, when Allison et al. list twelve entity
classifications they do not describe which, or if any, of the eleven NPE classes deserve the patent
troll label. (Allison et al., 2009, p. 11) Providing twelve patent ownership classes, and yet not being
able to identify a single of them as these patent trolls solely on account of ownership status, the
conclusion must be that being a patent owner cannot by definition make you a troll. As briefly
shown, even though NPE is often used to refer to these trolls it is a misleading association, as it does
not describe extortive entities. But what if an entity bases its business model on asserting patents,
could it then be fair to label it as a troll, i.e. extortive?
Just like NPE, using PAE to describe these trolls can hardly be correct either. As Feldman et al.
note, NPEs can have a business model of solely selling patents to third party entities, who then
assert them. Thus only focusing on the entities asserting the patents would not get to the root of the
problem, as the NPEs selling the patents can still create market distortions.(2012, p. 367-368) As
these trolls supposedly try to make a lot of money off a patent, which could be done through
asserting the patents, they could fall within the boundaries of PAEs. However, labelling PAEs
extortive by nature seems dubious, and it is not clear that using this term would necessarily include
the patent trolls that it is meant to describe. But as this term by definition could include anyone who
enforces a patent, labelling them all as extortive is incorrect.
One of the advantages of the Patent Monetization Entity term is, as Robin Feldman has
described, that “[i]t has the virtue of capturing the notion that the entity is specifically designed and
intended for monetizing patents, and it leaves out universities, whose core activities differ
Niels Hald Kristoffersen
Page 12 of 54
significantly from this group.” (Feldman, 2013, p. 266) And while that remains true, there is the
definitional problem of when exactly an entity would classify as being specifically designed for
monetizing patents. IBM is well known for its licensing of patents, which is in excess of $1 billion
annually, (The Economist, 2005a; Quinn, 2012) but would this classify it as a patent troll, a good or
bad, or even extortive actor under the monetizer term? And if not, then when would the amount of
its licensing revenues constitute a big enough percentage of its net revenue to label it a monetizer?
Earning in excess of $1 billion annually, it can hardly be claimed that IBM does not make money off
its patents. While possibly an indication, as to the activities of the entity, it still lacks clarity to
accurately define its boundaries, and that all entities monetizing patents are exercising extortive
behavior is doubtful. It is too broad a term to be used for this purpose.
Examples can be made of famous inventors who fit these degrading definitions under the NPE,
PAE, or monetizer terms, for example Thomas Edison. Edison was granted 1,093 patents throughout
his life, (Bellis, 2014) “primarily described himself as an inventor […] [and] made a fortune from many
patents he never practiced.”(McDonough, 2006, p. 198) “In 1915 he was awarded the Franklin Medal
in Engineering for ‘Discoveries contributing to the foundation of industries and the human race.’“(The
Franklin Institute) In 2010 Time Magazine published an article where he is referred to as the “[…]
greatest [inventor] of them all[.]”(Walsh, 2010) And yet, such an inventing and licensing scheme
would make him fit the definition of a patent monetizer, as well as a non-practicing entity. Can the
derogatory, extortionist synonym, ‘patent troll’ be meant to include a man such as Edison? In the
words of Gene Quinn: “if Thomas Edison is a patent troll then we really should want to encourage
patent trolls, right?” (Quinn, 2014b)
Upon closer inspection the ‘Patent Sharks’ definition by Reitzig et al. also suffers from the flaw of
vilifying patent enforcement, and Reitzig et al. themselves even note that the “R&D intensive
manufacturers” illegitimately use the technology. When this terminology is to be used for identifying
actors causing problems in the patent system, labelling the infringed the sole problem does not
seem accurate. They further describe their view of the problem as:
“[…] concerns of today’s leading R&D multinationals [are] potentially overlooking these (often
small) inventors’ patents and being caught in the trap of inadvertent infringement. Today’s patent
‘sharks’ or ‘trolls’ seem to place their bets on […] corporate ‘negligence’ or monitoring deficiency.
Sharks or trolls have no intention of engaging in the production of the technology underlying their
patents, but instead make money from royalty payments they obtain directly from their licensees or
indirectly in terms of damage awards. […] trolls intend to take their victims by ‘surprise’ to facilitate
their attempts to force manufacturers into unexpected licensing fees.” (Reitzig et al., 2007, p. 135)
Niels Hald Kristoffersen
Page 13 of 54
The implication of this definition would be that any patent holder enforcing a patent against
infringers while claiming compensation is conducting some kind of negative or “bad” behavior. In
other words, on the basic level, enforcement of a patent results in the holder being labelled patent
shark. But since a patent specifically grants the right to exclude by enforcement, then this shark
label seems undeserved, simply labelling anyone who enforces a patent as a patent troll cannot be
correct if it is supposed to describe entities conducting extortive behavior.
A question that should be raised for all these degrading definitions is: “when infringement takes
place, how come the ‘bad actor’ is the infringed and not the infringer?” In the aforementioned
‘patent shark’ scenario the R&D intensive manufacturer has violated the rights of the patent holder.
Yet claiming that the infringed is the bad actor would imply that the infringer is doing nothing wrong,
even when stealing or copying patented technology that it did not develop itself. Basically the R&D
intensive manufacturer can then benefit from ignoring patents, licensing, and royalties when
violating the rights of the patent holder. It seems that President Obamas notion of someone “trying
to essentially leverage and hijack somebody else's idea” is more fitting for the R&D intensive
manufacturer than the patent holder in this scenario. While the ethical fairness in the notion of
“hiding” to later reap damage awards for infringement by unknowing entities is questionable, is it
fair to hold them responsible for taking manufacturers by ‘surprise’ if it is due to their negligence or
poor monitoring of existing technology? If negligence and poor monitoring is the cause, then vilifying
any patent holder, some of whom might not be hiding at all, hardly seems correct, and labelling
them as patent sharks seem outright wrong. This would imply that the R&D intensive manufacturers
could just deliberately ignore or negligence monitoring - it has been argued that they already do
(Feldman & Ewing, 2012, p. 19; Lemley, 2008a) - and then call “shark!” or “troll!” upon infringement
claims by any patent owner. This way it would benefit greatly at no expense for stealing technology,
while the patent holder would be vilified. Essentially this would mean that a complete abolishment
of patents would be required to avoid the existence of patent trolls. The logic of these patent troll
definitions are suspicious, and puzzling to say the least. It is clear that they cannot adequately be
used to define bad actors exploiting the patent system, especially not when used to describe
infringed entities.
2.1.6 So Far
A reasonable assessment of the terms at this point is that there exists no sufficient explanation as to
what defines these extortionists, or who they are. These attempts simply cannot suffice. It seems
strange that these trolls gain so much attention, yet no better definition exists? The label ‘patent
extortionist’ must pertain to something else than these vague overly inclusive definitions. An
investigation of some of the “entities” labelled trolls seems appropriate.
Niels Hald Kristoffersen
Page 14 of 54
2.2 Entities Labelled Patent Trolls Even though the term was coined in 2001, it did not receive the public’s attention until 2006 where
NTP, Inc. fought Research In Motion Limited (RIM), the maker of BlackBerry, in a fight which ended
up costing RIM $612.5 million in settlement (Kelley, 2006; Quinn & Brachmann, 2013b). As described
in a Leader in The Economist (2005b), in public RIM was seen as a victim and NTP as a patent troll,
though NTP requested RIM to license its patents which RIM was infringing with their BlackBerry
products, but RIM refused to license. This brought a legal battle and threat of injunction upon RIM,
in other words, it was “[…] the chief author [o]f its own fate.” (The Economist, 2005b)
An entity often pointed to as a patent troll, is Acacia Research (Quinn, 2014c). It is the biggest
publicly traded patent-licensing company and has been a plaintiff in 280 lawsuits. (Feldman & Ewing,
2012, p. 15) Then CEO of Acacia, Paul Ryan, in a 2012 interview described the business model as
follows: “Acacia Research partners with patent owners to get them fair compensation for their
inventions and splits the net revenue with them typically on a fifty-fifty basis. 95% of [our 250
portfolios][…] are partnering deals fifty-fifty with the patent owner.” (Quinn, 2013b)
Just like Acacia, Intellectual Ventures is commonly seen as a patent troll. Listed as the biggest
patent holding company by Business Insider in a 2012 article, they call Intellectual Ventures “[…] the
mother of all alleged patent trolls.”(Fuchs, 2012) Intellectual Ventures describe itself as an
“invention capital company,” (Intellectual Ventures, 2014a) and its activities as “IP-for-defense,” “IP
Monetization,” “Litigation Clearance,” “Sponsored Invention,” and “Strategic Buying.” (Intellectual
Ventures, 2014b) The CEO and cofounder Nathan Myhrvold said, in 2006, that he hoped to never file
a patent lawsuit, (Varchaver, 2006) yet in 2010 Intellectual Ventures started doing just that. (Quinn,
2010) Although as some had speculated, it had been using third parties to carry out its litigation
activities up until that point. (Feldman & Ewing, 2012, pp.13-14) Intellectual Ventures is closely
studied by Feldman & Ewing in their article “The Giants Among Us” (2012), they uncover that
Intellectual Ventures is in possession of a total patent portfolio between 30.000-60.000 patents and
applications as of May, 2011 (p. 5) and that it has at least 1276 shell companies and entities related
to its activities. (p. 5) A very interesting observation is who the investors behind Intellectual Ventures
are, as they count corporations such as Amazon.com, Apple, Cisco Systems, Google, Intel, Microsoft,
Nokia, Sony, and Yahoo!.(Feldman & Ewing, 2012, p. 44)
Some other entities that have been portrayed as trolls worthy of mentioning are Eon-Net,
(Quinn, 2011b) Lodsys, LLC, (Mullin, 2013; Worstall, 2013) as well as Rockstar Consortium which is
backed by Apple, Microsoft, BlackBerry and others (Finley, 2013; McMillian, 2012).
Niels Hald Kristoffersen
Page 15 of 54
Contrary to the formerly mentioned trolls, an entity that is not often labelled as a troll is Google.
Like Apple and IBM, Google is an entity that fits the descriptions of a NPE, and acquires patents that
it asserts, just like NTP. Actually, Google has been found guilty of unlawful conduct by a jury verdict.
It has been argued that this in fact makes Google a convicted troll. (Mueller, 2013)
An important entity labelled a troll is MPHJ Technology Investments, LLC, as it has been “caught”
by New York Attorney General Eric T. Schneiderman. Described by Schneiderman as a patent
assertion entity, it has been labelled a troll because it has conducted the following behavior:
“MPHJ, have […] a strategy of targeting small and medium-sized businesses[,][…] send[ing] deceptive
and abusive letters […] in an effort to extract small, often nuisance-value license payments from
them. […] [I]t told […] businesses that they ‘likely’ infringed its scanner-related patents, creating the
impression that MPHJ had conducted a meaningful, individualized analysis of the targeted company’s
business. In fact, MPHJ merely sent form letters to companies of a certain size and industry
classification. In addition, MPHJ falsely told businesses that most other businesses it had previously
contacted had acquired licenses when in fact only a handful of businesses had done so. MPHJ also
provided misleading information about the fees that the (few) prior licensees had paid.” (New York
City Press Office, 2014)
The “small, often nuisance-value payments” demanded was a license payment of $1,200 per
employee. (Assurance No. 14-015, 2014, p. 3)
2.2.1 Closer Examination of the Entities
Starting with the entities widely considered patent trolls, it appears to be a matter of personal
belief rather than evidence as to what constitutes a troll. In the NTP case, the troll label appears to
be justified by acquisition and enforcement of a patent, (Quinn, 2014c) but this cannot be
considered as extortive behavior when RIM was infringing and refused to license, RIM did not leave
NTP with much else to do but litigate to enforce its granted rights. Thus despite fitting the NPE label,
NTP can hardly be considered an extortive actor, nor patent troll.
Regarding Acacia, it “reported gross revenues of just $34 million in calendar year 2006, about
what IBM makes in a week [of] licensing […].” (Hosie, 2008, p. 3) More recently, the About Us section
of Acacias webpage states as of April 5th, 2014, that it has “generated over $1,000,000,000 revenue
to date, and has returned more than $455,000,000 to our patent partners.”(Acacia, About Us) When
looking at the testimonials done by the partners of Acacia on its website (Acacia, Patent Partner
Testimonials), if making the assumption that they are genuine, the vast amount of testimonials
made and gratitude expressed does seem to back up the aforementioned business strategy
Niels Hald Kristoffersen
Page 16 of 54
described by Paul Ryan. It is therefore hard to conclude that Acacia can be seen as one of these
extortionist actors, despite clearly pursuing an aggressive litigating approach. When the reason for
doing so is defending the rights of infringed patent owners against infringers, the conclusion must be
that it should not be labelled a patent troll, as it is merely enforcing the granted patent rights. As
Spencer Hosie notes: “[…] Acacia is an extortionist and IBM an icon. What gives?” (Hosie, 2008, p. 3)
Taking any stand on Intellectual Ventures is difficult, as noted by Feldman & Ewing, much about
it is shrouded in secrecy (2012, p. 3). One can only speculate about to which extent the company
monetizes its massive portfolio, and whether it creates market distortions or not. Nevertheless,
labelling Intellectual Ventures a troll on the grounds of their patent portfolios size and assertion of
them is not sufficient, with no concrete evidence of Intellectual Ventures conducting extortive
behavior it cannot be labelled a troll.
In short, MPHJ engaged in abusive pre-litigation tactics with the intent to deceive, and provided
misleading information, to scare targets into settling.(Quinn, 2014a) It is hard to describe this as
anything but extortionist behavior. MPHJ appear to be the only entity so far that fits the patent
extortionist label. Though MPHJ fits the descriptions of PAE, and monetizer, none of these
descriptions define it as an extortionist, the only thing that identifies this is the deceptive and
abusive tactics it used in its assertion.
It can thus be concluded that we cannot identify a patent troll simply on whether it owns,
asserts, or monetizes patents. And that labelling any entity a troll on those unfounded criteria’s is
misleading. The troll label should only be reserved for those who conduct unlawful, abusive, and
extortive behavior, as the MPHJ case showed. As any entity, in any industry, in possession of any
kind of patent can conduct unlawful, extortive behavior, no matter the entity size, there is no way to
classify these trolls broadly. This implies that the troll label can only be applied on a singular basis.
All usage of terms as NPE, PAE, or monetizer can only be used for an over inclusive generalization of
a group of patent holders, or used as a scapegoat for other purposes, possibly to divert attention
from being an infringing entity. Unfortunately, most literature and debate on this issue assume the
trolls to be one of those patent owner classifications. As is evident, what constitutes a troll is an
entity conducting unlawful and deceptive behavior and tactics, interestingly, this goes well in line
with the tactics that have been attributed to trolls under the misleading and degrading synonyms
previously explained.
Niels Hald Kristoffersen
Page 17 of 54
2.3 Tactics Employed by Trolls Assumptions exist about the tactics of these trolls, but not many ring true when investigated. For
example, a common assumption is that these trolls “use weak and vague patents to threaten
product manufacturers and extract excessive licensing fees or engage in frivolous infringement
litigation.” (Shrestha, 2010, p. 115) And this was backed up by Allison et al. (2011), however, Allison
et al. only studied 106 NPE patents, (p. 680) and studies employing larger samples have found that
the patents brought by trolls are “[…] of equal or higher quality,” (Risch, 2012, p. 481) that trolls
generally “[…] hold high value patents[,]” (Shrestha, 2010, p. 150) and the findings of Fisher &
Henkel (2011) “contradicts […] that patent trolls concentrate on enforcing low-quality patents.” (p.
19) Therefore, a shift of focus to how they assert them seems more appropriate if we are to label
these entities “extortive”. Interestingly, this has been defined much more accurately, but does not
generally apply to the very entities that they are attempting to vilify. For example, the White House
had adopted PAE as the synonym for a patent troll in its report, yet it defines their behavior as
“[…] threatening to sue thousands of companies at once, without specific evidence of infringement
against any of them; creating shell companies that make it difficult for defendants to know who is
suing them; and asserting that their patents cover inventions not imagined at the time they were
granted.” (June 2013, p. 1)
These tactics do come off as extortive, and the tactic of suing a vast number of companies without
specific evidence of infringement fits well with the behavior of MPHJ mentioned earlier, but claiming
that every entity fitting the PAE classification is conducting this behavior by nature seems dubious to
say the least.
Another asserted tactic that is worth mentioning, is attempting “[…] to shake down small businesses
by bringing specious patent infringement lawsuits knowing that it will be far cheaper for the
defendants to settle with a licensing fee than it will be to litigate, even if there is clearly no
infringement.” (Quinn, 2011d) This tactic has also been described by Rader et al. (2013), and fits the
case of MPHJ, the case of Eon-Net, (Quinn, 2011b) and it can be argued to pertain to the behavior of
Lodsys. (Lee, 2013; Mullin, 2013; Quinn, 2013c)
2.4 Conclusion of Patent Troll Exploration A ‘Patent Troll’ is an extortive entity conducting, deceptive, abusive, and possibly unlawful behavior
in connection with patent litigation. Many definitions have been proposed in an attempt to identify
these ‘patent extortionists,’ but being a patent owner and enforcing, asserting, or monetizing a
patent cannot in itself be extortive behavior. Therefore, none of them can be used to accurately
capture a patent extortionist scenario. The closest to a definition of a real troll that can be made, is
Niels Hald Kristoffersen
Page 18 of 54
entities that, in connection to patent litigation, by use of unlawful, extortive, abusive, or deceiving
behavior and tactics, use patents as a means to extort money from other entities. This can only be
determined on a case-by-case basis of entity behavior. Finding out who is a real troll is thus very
difficult, and it cannot be generalized. Therefore, any existing work or analysis done on this topic, on
the foundation of the erroneous definitions of NPE, PAE, monetizer, or the like, has to be treated
carefully. The results of those may vary highly depending on use of terminology, and will most likely
be subject to wrong and misleading results that will not be based on, or targeting, the real patent
trolls. So when the debate targets the wrong entities, which measures have then been taken to
counter them?
3. Measures Taken to Counter Trolls
Even though some states, such as California and Vermont, have their own local statutes aiming to
fight patent trolls, (Martinez, 2013) the only action that has been taken nationwide is the Leahy-
Smith America Invents Act.
The Leahy-Smith America Invents Act, or AIA, was the first significant change to the U.S. patent
system since 1952 (Goldman, 2011). Its purpose was to amend title 35 of the United States Code in
order to provide patent reform, (Leahy-Smith America Invents Act, 2011) and was signed into law by
President Barack Obama on the 16th of September 2011 (The White House, 2011). The changes
brought by the AIA are many as it contains no less than 37 Sections, (Leahy-smith america invents
act.2011) and it is far beyond the scope of this thesis to account for them all, yet while it did bring
many changes to the patent system, it did not do much to counter the real troll issue. During the
debate prior to the signing, various members of Congress had expressed their displeasure over NPEs,
(Feldman et al., 2012, p. 359) yet despite containing 37 sections, the majority of the changes in the
Act did not target NPEs, or real trolls for that matter. There is, however, one specific provision in the
AIA which was aimed directly against a “NPE tactic” of joining large numbers of defendants in one
lawsuit. (Feldman et al., 2012, p. 360) Before the AIA, a patent owner could sue any number of
defendants, sometimes well over one hundred, who offered different products and were
geographically scattered across the country, in a single infringement suit, by alleging that their
commonality was infringing the same patent. (Quinn, 2011a) The AIA changed this by requiring that
accused infringers can only be joined in one suit, if:
“(1) A claim is made against parties jointly or severally, or a claim arises out of the same transaction,
occurrence, or series of transactions or occurrences; and (2) questions of fact common to all
defendants or counterclaim defendants will arise in the action.” (Quinn, 2011a)
Niels Hald Kristoffersen
Page 19 of 54
The AIA was thus the first step taken by Congress in modern times to address this patent troll issue.
4. To Which Extent is there an Impact from Patent Trolls?
When this is the only measure implemented so far, what evidence exists of patent troll activities and
its impact? Is the need for further solutions urgent? Are these trolls truly wreaking havoc through
the patent system? While a lot of literature exists, the amount of actual evidence is surprising. No
studies have investigated the true trolls conducting abusive, extortive behavior, so a look at the
existing evidence for the derogative synonyms will be investigated in the following section.
4.1 Existing Evidence
4.1.1 Recent Surge in Patent Litigation
Just a quick glance at Chart 3, 4 and 5 listed in section 1.2 leaves no doubt that there has been a
patent litigation increase in recent years. In 2012 it reached “the highest number ever recorded”
(PwC, 2013) and claims such as “An ‘explosion of patent litigation’ greater than any in history is
imposing an unwarranted burden on industry and diverting resources better spent on innovation” are
“[…] almost universally accepted as true[.]” (Kline & Cassidy, 2014) While the increase in numbers
cannot be disputed, Kline & Cassidy (2014) show that from 2002 to 2012 the patent litigation rate
was 1.57 percent, whereas the corresponding patent litigation rate from 1790-1860 was 1.65
percent. Furthermore, Khan show that, from 1840-1849, the patent litigation rate was 3.6 percent.
(Khan, 1995, p. 63) The rate of litigation is thus not historically high. Astonishingly, the number of
suits during today’s smartphone patent wars are: “less than one-quarter the number of patent suits
filed during the first ‘Telephone Wars’ […] [where] the American Bell Telephone Company and its
successor, AT&T, litigated […] 587 patent cases alone.” (Kline & Cassidy, 2014) Judging from this, it is
hard to agree that this is an explosion greater than any in history. As it turns out, approximately 96
percent of the rise in litigation since 1991 can be explained by the rise in the amount of patents
granted. (PwC, 2013, p. 6) In fact, there has been an average of 99.9 patent trials every year since
1980, (Quinn, 2011c) which is remarkable given the increase in the number of patents issued over
the same time period. A study conducted by the GAO (Governmental Accountability Office)
supported these findings, though it reported that from 2010 to 2011, stakeholders had said the
increase was most likely influenced by the anticipation of changes in the AIA. (GAO, 2013)
Niels Hald Kristoffersen
Page 20 of 54
The AIA is a very important element to include when looking at the latest rise in patent litigation,
because of the recent changes caused by its enactment. As described in Section 3, it prohibits
litigants from suing multiple defendants in one suit, unless specific requirements are met. As
Martinez (2013) points out, on September 15, 2011, 54 cases were filed against 800 defendants, if
that were to happen now, after the change of law, the 800 defendants would be sued separately,
and it would thus be 800 cases, not 54. (Martinez, 2013) Basically, due to the way the AIA was
worded and constructed, it has most likely helped cause an increase in the sheer number of
litigation suits filed. As can be seen from Chart 6 there was a massive spike in litigation up to the
implementation of the AIA, as litigants “[…] rushed to the courthouse to get their lawsuits filed
before the Act became effective.” (Feldman et al., 2013, p. 69)
Furthermore, as noted by Katznelson (2013a) in his article, “The America Invents Act at Work – The
Major Cause for the Recent Rise in Patent Litigation,” Title 35 Section 325(b) of the United States
Code, after the works of the AIA, states that the court may not stay a motion for preliminary
injunction on the basis that Post-Grant Review was filed or instituted, if the patentee’s suit is filed
within three months of patent issuance. It has therefore caused patentees to identify infringers and
file suits within the three months of issuance, in order to preserve the patentee’s preliminary
injunction power. He, therefore, argues that the reason behind the surge in patent litigation for
2012 is caused by litigation of newly issued patents. (Katznelson, 2013a) This change can clearly be
seen on Chart 7 on the following page.
Chart 6: Patent Litigation Spike Prior to the Passing of the AIA. (Chart copied from Feldman et al., 2013, p. 68)
Niels Hald Kristoffersen
Page 21 of 54
Chart 7: Patent Age at Filing of Infringement Lawsuit (Chart copied from Katznelson, 2013, “The America Invents Act at Work - The Major Cause for the Recent Rise in Patent Litigation”.)
Clearly there is a surge in patent litigation, but it is caused by an increase in the number of
patents issued and the changes brought about by the AIA, as the AIA actually encouraged and
fostered this increase in litigation. Therefore, it cannot be considered problematic when the
litigation amounts to the regular hundred trials per year. Thus the evidence does not point to these
patent extortionists, i.e. trolls, as the explanation for this recent surge. When the current rates of
litigation are not deviating from the historical norms, are these trolls then really on the rise? Do they
bring the majority of suits? Or is some other entities responsible for the litigation increase?
4.1.2 Who Litigates?
To investigate this, a look at the evidence on the amount of litigation brought by trolls is needed. In
the following, the nature of patent plaintiffs found by four different prominent studies will be
examined.
“Patent Trolls by the Numbers”, by Chien (2013)
These findings presented by Chien is based on data provided by RPX Corporation, a company
described by Chien as “a publicly-traded company that provides solutions to troll threats[.]” In her
findings, she observes that the percentage of litigation brought by trolls from 2010 to 2012
increased from 29 to 62 percent of all suits filed. (C. Chien, 2013) Recall that Chien defines a troll as a
PAE (See section 2.1.2).
Niels Hald Kristoffersen
Page 22 of 54
“The AIA 500 Expanded: Effects of Patent Monetization Entities on US Litigation”, by Feldman et
al. (2013)
Recall from Section 2.1.3 that Feldman et al. use the definition Patent Monetization Entities to
define what constitutes a patent troll. What they find is that lawsuits filed by these patent
monetizing entities have increased from 24.6 percent in 2007 to 40.4 in 2011 and 58.7 percent of
cases filed in 2012. (Feldman et al., 2013, p. 54) See Chart 8 below for their graphical illustration of
their findings.
Chart 8: Patent Litigation Brought by Entities, 2007-2011 (Chart copied from Feldman et al., 2013, p. 55)
“Assessing Factors That Affect Patent Infringement Litigation Could Help Improve Patent Quality”,
by the Government Accountability Office (2013)
Based on the same data as the original study by Feldman et al. (2012), the report from GAO
surprisingly report different findings. While GAO also reports that there was a decrease in share of
suits brought by operating companies, it reports that the share brought by PMEs only increased to
24 percent in 2011, and that this was a statistically insignificant increase. Another finding made is
that for the period measured (2007-2011), operating companies litigated against 1.9 defendants per
suit on average, while PMEs litigated against 4.1 defendants per suit on average. (GAO, 2013, p. 21)
Niels Hald Kristoffersen
Page 23 of 54
“Unpacking Patent Assertion Entities (PAEs)”, by Cotropia et al. (2013)
An important study on this topic is the one by Cotropia et al. 2013. It is especially noteworthy in the
aspect that, contrary to the 3 previous studies, the data has been made publicly available, at
www.npedata.com, which naturally makes the study a lot more trustworthy, as its conclusions can
be replicated and verified, and we are not forced to trust the authors words that the findings are, in
fact, true.
In their study, Cotropia et al. (2013) deliberately exclude calendar year 2011 from their litigation
analysis due to the AIA and the litigation spike it brought. (p. 12) They do not provide a classification
of who is or is not a troll; instead they divide PAEs into eight different categories. (p. 15) The ones
emphasized here, are category (3), (5) and (6). They describe Category (3) as “large patent
aggregator: a company with a large patent portfolio whose primary business is enforcing patents
and numerous others. This includes Acacia companies and Intellectual Ventures.” (p. 17) Category (5)
as: “patent holding company: companies […] that appear to have been formed solely to hold and
enforce a patent or small portfolio of patents. […] Frequently, these companies were formed shortly
before litigation was commenced.” (p. 17) And category (6) as: “operating company: companies that
manufacture products or deliver services (other than licensing patents). (p. 17) What they find is that,
for 2010, there were 1610 unique patent litigants, whereas the corresponding number for 2012 was
1695. (p. 21) Also, the total number of defendants was 11,671 in 2010 and 11,603 in 2012. (p. 24)
The total number of litigants and defendants thus almost stayed constant between 2010 and 2012.
As is evident from Chart 9, 10 and 11 in the following, category (6) Operating Companies made up
the largest percentage of cases filed even though it decreased from 2010 to 2012, with a respective
share of 69.5 percent in 2010, and 43.5 percent in 2012, but at the same time the percentage of
Operating Companies as defenders also decreased, from 71.8 percent to 68.6 percent. Contrarily,
category (5) Patent Holding Company went from a 15.9 percent litigation share in 2010 to 37.7
percent in 2012, and an increase as defendant from 13.3 percent to 15.7 percent. The authors
conclude that in the aggregate “Large Aggregators and Patent Holding Companies comprise,
together, 31.59% (3,684) of the total number of parties in suits excluding the patentee in 2010 and
37.74% (4,380) in 2012. Adding Individual/Family Trust cases increases the total number of parties in
suit excluding the patentee to 5,004 (42,91%) in 2010 and 5,307 (45.74%) in 2012. This is in contrast
to Operating Company cases which involved 49.48% (5,708) of the parties in 2010 and 48,35%
(5,610) in 2012.” (pp. 35 & 36) It is thus clear that non-practicing entities have an increasing
presence from 2010 to 2012, though operating companies still make up the majority of cases filed.