Contemporary Views on Economics of Patents
Abstract and Keywords
A patent is a legal right to exclude granted by the state to the inventor of a novel and useful invention. Much legal ink has been spilled on the meaning of these terms. “Novel” means that the invention has not been anticipated in the art prior to its creation by the inventor. “Useful” means that the invention has a practical application. The words “inventor” and “invention” are also legal terms of art. An invention is a work that advances a particular field, moving practitioners forward not simply through accretions of knowledge but through concrete implementations. An inventor is someone who contributes to an invention either as an individual or as part of a team. The exclusive right, finally, is not granted gratuitously. The inventor must apply and go through a review process for the invention. Furthermore, a price for the patent being granted is full, clear disclosure by the inventor of how to practice the invention. The public can use this disclosure once the patent expires or through a license during the duration of the patent.
These institutional details are common features of all patent systems. What is interesting is the economic justification for patents. As a property right, a patent resolves certain externality problems that arise in markets for knowledge. The establishment of property rights allows for trade in the invention and the dissemination of knowledge. However, the economic case for property rights is made complex because of the institutional need to apply for a patent. While in theory, patent grants could be automatic, inventions must meet certain standards for the grant to be justified. These procedural hurdles create possibilities for gamesmanship in how property rights are allocated.
Furthermore, even if granted correctly, property rights can become murky because of the problems of enforcement through litigation. Courts must determine when an invention has been used, made, or sold without permission by a third party in violation of the rights of the patent owner. This legal process can lead to gamesmanship as patent owners try to force settlements from alleged infringers. Meanwhile, third parties may act opportunistically to take advantage of the uncertain boundaries of patent rights and engage in undetectable infringement. Exacerbating these tendencies are the difficulties in determining damages and the possibility of injunctive relief.
Some caution against these criticisms through the observation that most patents are not enforced. In fact, most granted patents turn out to be worthless, when gauged in commercial value. But worthless patents still have potential litigation value. While a patent owner might view a worthless patent as a sunk cost, there is incentive to recoup investment through the sale of worthless patents to parties willing to assume the risk of litigation. Hence the phenomenon of “trolling,” or the rise of non-practicing entities, troubles the patent landscape. This phenomenon gives rise to concerns with the anticompetitive uses of patents, demonstrating the need for some limitations on patent enforcement.
With all the policy concerns arising from patents, it is no surprise that patent law has been ripe for reform. Economic analysis can inform these reform efforts by identifying ways in which patents fail to create a vibrant market for inventions. Appreciation of the political economy of patents invites a rich academic and policy debate over the direction of patent law.
The word “patent” is used in many different ways. It sometimes refers to a government document, a memorialization of a grant from the government to the inventor of a novel, useful, and nonobvious invention. The government document contains a description of the invention that teaches the world how to practice it. “Patent” also refers to this disclosure. Finally, “patent” means a legal right to exclude that allows the right owner to keep others from making, using, or selling the invention set forth in the disclosure. If the patent has not expired, anyone can read the disclosure, but only those who are granted a license from the owner can practice the invention. “Look, but don’t touch” is the implicit warning attached to any patent when it is alive. But as soon as the patent expires, usually after a period of about 20 years (if periodic maintenance fees are paid to avoid an abandonment of the patent), anyone can look and touch: the invention is in the public domain for anyone to make, use, or sell. The ingenuity of the patent system is to promote accretion to practical knowledge through a system of time-limited property rights attached to an invention. For an analysis of maintenance fees, patent term length, and the use of fees to cross-subsidize different types of inventions, see Sukhatme (2014).
State intervention to define property rights is sometimes necessary to correct externalities. In the case of patents, externalities arise from the nonexcludability and nonrivalry of knowledge. If an inventor develops a new automobile engine to drive an electric car, the mechanics of designing the engine is knowledge that anyone who has the relevant technical skills can take to build or repurpose the invention. A patent provides a legal right to exclude that allows the inventor to gain rents from selling or licensing the invention. Notice, however, that the knowledge can still be shared, since the knowledge is disclosed in the patent document. The right to exclude extends only to the making, using, or selling of the knowledge. In short, the state by granting a patent makes the invention of the excludable while retaining the nonrivalry of the knowledge.
Once property rights are established, the owner of the right, initially the inventor, can market the invention disclosed in the patent. Much of the literature on patent rights focuses on how property rights facilitate the many markets for the invention. These markets would include licensing the patent for the purposes of manufacture or distribution, sales of the invention itself, uses of the invention to create final products or as components of a larger manufacture, and research uses of the invention for experimentation and study in universities and laboratories.
Kitch (1977) famously described the marketing function of patent rights in their prospect theory of patenting. As with claims over minerals or oil and gas deposits, patents serve to demarcate the landscape of knowledge by grants of exclusive rights in inventions. These grants allow the inventor to mine the field for all commercial applications of the invention. Under the prospect theory, internalizing the externalities from knowledge serves a largely commercial function, providing the inventor a safe harbor from competitive forces that would readily imitate or improve on the invention.
Although not directly defending prospect theory, the historical work in Khan (1995) supports a relationship between patents and economic development in the United States. A historical account of patent cases in the early 19th century shows that judges enforced patent rights on the theory that inventors responded to expected returns. Even in the 1850s, as changes occurred in patent law and economic development, judges repeated this rhetoric. Material incentives were recognized as one motivation for inventive activity, and the legal system buttressed the effectiveness of the patent system in stimulating technical change.
Prospect theory, however, has been the object of criticism. By emphasizing the commercialization of patents, the theory ignores the social functions of knowledge and the importance of science and information commons. But, even with the goals of commercialization, prospect theory is lacking. The theory assumes that the scope of patent rights has been correctly allocated. For reasons discussed throughout this article, that assumption is a strong one. More specifically, the argument assumes that patent rights are one-size-fits-all: what works for the chemical industry would work well for software or biotechnology. However, the product cycle and speed and rate of innovation vary across industries. Arguably, patent rights should be tailored to deal with these realities.
Furthermore, prospect theory ignores the process through which patent rights are obtained and enforced. How the process is structured can create incentives for rent-seeking by patent owners who seek broad rights and by granting agencies that may allocate patent rights to maximize their own interests. Some scholars argue that the process of obtaining patent rights leads to rent dissipation, dampening the incentives for rent-seeking. Current scholarship casts some doubt on the rent-dissipation hypothesis. Patent rights, in fact, may be far from correctly allocated. Furthermore, the patent system may have structural features that may favor overly broad grants of rights. This point will be examined more carefully through a consideration of the scholarly literature on obtaining patent rights, enforcing patent rights, the existence of worthless patents, the anticompetitive uses of patents, and proposals for patent reform.
Obtaining Patent Rights
By providing certain rights to exclude, patents create an environment for rent-seeking. Inventors invest resources in research and developing to produce inventive products that are brought to the patent office for consideration. Investment in research and development and application for the patent are social costs, borne in part by the inventor, that may be matched by social benefits in the form of innovation. Often, however, investment does not pay off. Sometimes, the payoff may be greater for the inventor than is deserved, for example through the grant of patent rights that are too broad. Other times, the patent grant may be too narrow, resulting in too many patents, with the resulting problem of patent thickets that lead to an anti-commons. See Heller (1998).
Grady and Alexander (1992) document the potential problems of rent-seeking in the patent system. Patent rewards might promote overinvestment in research and development, with the result of spurious invention. To echo Kitch’s prospect theory, rent-seeking may lead to overprospecting. Courts, the authors argue, can limit rent-seeking through judicious application of patent doctrine. For example, by identifying a broad scope of patent rights for pioneer inventions and a narrow scope for inventions that represent less of a breakthrough, courts can calibrate rents to the importance of the inventor’s work. Furthermore, by raising the standards for novelty or other patent requirements, courts can tailor the patent to the invention. By careful tailoring, rents from patenting can dissipate and overprospecting in certain fields of research and development can be controlled.
Although Grady and Alexander focus on courts, their arguments about rent-seeking and dissipation applies readily to administrative agencies. Across jurisdictions, inventors must apply for a patent and undergo a review process, called prosecution. A nation’s patent office has the authority to assess the application and render a judgment as to the grant of a patent. Denial of a patent can typically be appealed to the courts. Administrative gate-keeping, therefore, can also serve as a basis for rent dissipation. The office can create procedural and substantive rules that make it more difficult for inventors to obtain a patent. Substantive rules connect to the requirements for an invention to be patentable, such as novelty, usefulness, or inventiveness. Procedural rules entail fees, patent application forms, responses to agency decisions, appeals to the courts, and intervention by parties who might oppose the grant of the patent. The last procedural rule is called an opposition, about which much scholarly literature has been produced.
The pursuit of patents through administrative agencies can dissipate rents. But is a controversial question as to whether the agency seeks to dissipate rents or collect rents of their own. Courts as institutions do not collect rents. Judges do not get paid by the case, and their remuneration is not tied to particular judicial outcomes. Patent examiners, agents who are reviewing and making decisions about applications, often respond to incentives based on their judgments to reject or award a patent. Furthermore, the agency receives fees from the filing of applications. In addition, the agency obtains fees for maintaining a patent during the course of its life (typically a 20-year period). Therefore, there are incentives for the agency to increase its revenue from fees through a more lenient review process.
Scholarship has addressed the problem of rent-seeking within patent agencies. King (see Cohen & Merrill, 2003) finds that examiner hours have remained constant even as the number of patent applications have increased in the United States. He concludes that the quality of examination has not changed over time, although the pendency rate (the time for a patent to issue) has increased. He also reports a negative correlation between examination resources and litigation costs, suggesting that the costs of improving examination might be outweighed by the benefits from decreased litigation.
More recent research by Frakes and Wasserman (2013, 2014, 2015, 2016) suggests a bleaker story for rent-seeking within the patent office. Studying granting patterns by the United States Patent and Trademark Office (before and after 1991), the authors conclude that the Office’s fee structure biases the agency toward granting patents (Frakes & Wasserman, 2013). Specifically, the USPTO grants patents to technologies that have a high renewal rate, meaning that patent owners are willing to pay more to maintain the patent rather than abandon it through nonpayment of fees. These patents are typically in the area of information, computer, and health-related technologies, and such grants distort the patenting process and impose social costs in the form of higher prices for consumers through expanding patent protection for inventions that are not innovative. These incentives are exacerbated when the USPTO finds itself financially constrained through decreases in funding.
Frakes and Wasserman (2014) take their findings further by examining the structure of user fees in the USPTO. They identify a cross-subsidization in the fee structure through which inventors who have been granted patents subsidize the costs of current applicants. This cross-subsidy works through the maintenance and renewal fees paid by recipients of patents and the discounts given to current applicants, often based on entity size, but also attributable to the uncertainty in predicting examiner time to review an application. Because of this cross-subsidy, the agency has the incentive to grant patents on inventions that will have larger maintenance fees, as described in their 2013 article. A possible remedy is to base application fees on the true cost of administrative review rather than relying on the inherent cross-subsidy from successful applicants.
Frakes and Wasserman integrate their empirical findings into a theoretical framework (Frakes & Wasserman, 2015). In addition to the cross-subsidization effects described in previous works, the authors also identify the inherent feature of the US patent prosecution process that a rejection of a patent application is never final. The applicant can appeal a rejection with the hope of a court reversal or can forestall a rejection through filing a continuation application. Identifying moments when the USPTO is financially constrained provides an experiment for the authors to study agency behavior. To quote, the authors form “a counterfactual through two dimensions of differentiation—in this case, by looking before and after negative resource shocks and by looking across technologies with different continuation proclivities” (Frakes & Wasserman, 2015, p. 640). Through this methodology, the authors are able to identify a tendency of the USPTO to overgrant patents in fields with high continuation proclivities during periods of financial constraints.
Because of these structural features of the USPTO, Frakes and Wasserman express skepticism about reform efforts to strengthen the legal standards for patentability. Patent examiners and the agency will seek to acquire rents even with a change in the ostensible legal standard. As support for this inertia in the prosecution process, the authors look at the behavior of “patent cohorts” in Frakes and Wasserman (2016). A patent cohort refers to the year in which a group of patent examiners was hired. The authors find that there are identifiable cohort effects in the decisions of a patent examiner. In other words, patent examiners are instilled with a particular culture that affects their professional behavior throughout their examination careers. Because of these cohort effects, changes in the patent law, such as the 2011 America Invents Act, may be implemented slowly, only as a new cohort of patent examiners enters the changed administrative culture.
While Frakes and Wasserman tell a story about the USPTO as an entrenched agency, other scholarship attempts to understand more specific administrative patent reforms. A common reform is to include an opposition system into the patent prosecution process. An opposition allows a third party, other than the patent examiner and the applicant, to participate in the decision to grant a patent. Without an opposition process, the decision to grant is one that is made solely with input from the application as reviewed by the examiner. An opposition allows any interested party, whether a competitor or a consumer group, to introduce prior art and other evidence that affects the granting decision. An opposition can arise either during the application process or, more typically, after the initial granting decision by the agency; hence the nomenclature of pre-grant and post-grant opposition.
Many countries have experimented with opposition systems. The United States implemented a system in 1980 called an ex parte re-examination, followed by an inter partes re-examination in 1999. The America Invents Act replaced the inter partes re-examination with a more expansive post-grant opposition system, which is currently under constitutional challenge before the United States Supreme Court. Experiments with oppositions have been fodder for scholarly studies of their effectiveness in opening up the prosecution process and potentially curing any tendencies to overgranting patents.
Graham et al., in their study published in Cohen and Merrill (2003), compare the opposition system of the European Patent Office (EPO) with the re-examination system in the United States. They find a strong contrast. Comparing matched pairs of patent applications between the two regions in the fields of biotechnology, pharmaceuticals, semiconductors, and computer software, the authors conclude that valuable patents are likely to be challenged in both jurisdictions but the rate of opposition in the EPO is more than 30 times higher than the rate of re-examination in the USPTO. Furthermore, while in the United States, re-examination leads to a cancellation of the patent in 10% of the cases, EPO opposition proceedings result in revocation of the patent in 35% of the cases and restrictions on the scope of the patent in 33% of the proceedings. The authors attribute these differences in outcomes to differences in the legal rules for bring the respective proceedings. In the United States, according to the authors, re-examinations are more costly to bring in terms of time and money than the European counterpart. Most often, a re-examination is brought by the patent owner herself in order to amend the claims. In other words, re-examinations are most often used preemptively to avoid potential challenges as to patent validity.
Levin and Levin, also in the Cohen and Merrill volume, present a formal economic model that helps to identify the benefits of an opposition system. They develop a model of patenting that focuses on the product market and the welfare effects on consumers. The authors identify several benefits from an opposition system, including the reduced costs of patent litigation (which absent an opposition system would be the primary method for challenging patents), the increased reward to valid patents, and the decreased reward to invalid ones.
In light of this empirical and theoretical work on opposition, it is not surprising that the United States introduced an expansive opposition system through the America Invents Act of 2011. Specifically, the Act provided for three new proceedings: inter partes review (IPR) and post-grant review (PGR). The first was implemented in 2012. In the same year, PGRs were permitted for certain business method patents. In 2013, PGRs were extended to any patent with a priority date after March 15, 2013. A PGR is allowed within the first nine months after a grant on any grounds for invalidating a patent. An IPR is available after that initial period, but solely on grounds of lack of novelty or nonobviousness. The AIA also allows a stay of litigation proceedings if any of these proceedings is pursued. The complexity of the new proceedings and their interaction with infringement actions make for potential for studying litigation and prosecution dynamics.
Vishnubhakat et al. (2016) is the first comprehensive study of the new procedures. The authors find that 70% of petitioners who initiate an IPR proceeding are engaged in a patent infringement lawsuit. This finding suggests that the IPR serves as a strategic counter to litigation with an alleged infringer bringing the administrative proceeding as a challenge to the validity of the underlying patent. Initiating an IPR can serve as a chip for settlement, a strategic maneuver to stay the litigation, or a deterrent to aggressive litigation tactics by the patent owner. What is interesting is that the remaining 30% of IPR petitions are not in the context of litigation. The authors find this a mystery and do not offer a solid explanation for what motivates these petitions. Perhaps they are brought preemptively; perhaps for reasons not tied to litigation threats but to policy concerns with patents in particular areas (such as healthcare or pharmaceuticals). In this pathbreaking study, the authors open up a rich mine for scholarship for studying the dynamics of administrative procedure and the dynamics between opposition and litigation.
Property rights provided by patents may facilitate the development of new inventions and their propagation. As Kitch suggested, property rights aid prospecting, whether in the search for natural resources or for innovation. But these property rights come with costs. Obtaining property rights can lead to rent-seeking by inventors and by the agency granting the rights. Grady and Alexander suggest that courts can dampen the effects of rent-seeking through rent-dissipating judicial doctrines. However, courts become involved in patent law only when there is private party litigation, which as will be seen in the next section imposes its own costs.
Enforcing Patent Rights
Patent litigation can arise in two ways. A patent owner might initiate a lawsuit claiming that a party has infringed its patent rights—namely, used, made, or sold the patented invention without the permission of the owner. Alternatively, a party that might be allegedly infringing a patent initiates a lawsuit seeking a declaratory judgment from the court as to the patent’s validity. These variations can be complicated through the filing of counterclaims. For example, in a suit initiated by a patent owner, the alleged infringer can raise a counterclaim that the patent owner has violated antitrust laws. In addition, in a declaratory judgment suit brought against the patent owner, the owner might will likely counterclaim patent infringement. Litigation introduces complex strategies as patent owner and other parties seek to assert their rights and gain whatever benefits they can from the dispute. See Chien (2009) and Harhoff (2009).
As the discussion on obtaining rights sets forth, challenges to patent validity in the patent office can interact with litigation strategies. More broadly, patent quality can affect the dynamics of infringement, litigation, and settlement. If competitors or other users of an invention think the patent was improperly granted, they are more likely to engage in infringing activity. Consequently, patent owners are more likely to suspect patent infringement and initiate litigation. Anticipating a patent infringement suit, a potential defendant acts defensively by initiating a declaratory judgment suit. Fears of liability and patent invalidity in turn feed incentives for settlement either before or after the initiation of litigation.
Meurer (1989) presents a theoretical model that demonstrates how uncertainty over patent quality supports both more patent infringement and litigation. Bessen and Meurer (2006) introduce monitoring by the patent owner and the decision to imitate or invent around by a potential infringer in a model of patent litigation. Their theoretical work supports their notion that in a world of clear property rights, there would be no patent lawsuits. Instead, vague boundaries introduce uncertainty for firms, resulting in patent lawsuits in an equilibrium. The authors predict that clearer property rights as promised by the introduction of the United States Court of Appeals for the Federal Circuit in 1982 and the shift from claim construction by judges, rather than juries in 1995 should lead to a reduction in patent litigation. Empirical scholarship, discussed below, addresses these and other predictions.
Theoretical scholarship on patent litigation supports other institutional features of the patent system. Crampes and Langinier (2002), on whose work Bessen and Meurer build through the introduction of the strategy of “inventing around,” find that patent holders may prefer settlement of litigation to even relatively high patent damage awards because of the costs of monitoring. As a result, patent damages may have counterintuitive effects on entry of imitators, with more entry arising even with more threatened litigation. Under the authors’ analysis, more threatened litigation makes it less likely that any specific infringer will be found liable, because of the costs of monitoring. Therefore, imitators might find the risk of liability to fall as litigation threats increase. Cramper and Langinier’s work suggests that solutions like clear property boundaries may not be sufficient for limiting patent infringement and the attendant legal disputes. Aoki and Hu (2003) consider the effects of imitation and litigation lags on patenting and innovative behavior. Higher pecuniary imitation costs favor the patent owner over the imitator, but faster rates of imitation may stimulate ex ante licensing, and slower imitation reduces the settlement fee paid by the imitator. In short, timing can affect litigation dynamics as much as pecuniary costs. This set of findings buttresses the role of patent litigation as a tool for transmitting information about the validity of the contested patent to potential market entrants, as analyzed in Choi (1998). Depending on the strength of patent protection, potential competitors can delay entry in order to free ride on information that might arise from litigation against the first entrant. The patent owner, in turn, might choose to litigate even a weak patent to avoid information about the patent from divulging through the litigation process. Patent litigation is a waiting game for both the patent owner and potential competitors, creating the possibility for complex dynamics. Among these dynamics is the possibility of settlement by the patent owner to avoid negative information about a weak patent from being revealed.
Cross-licensing and patent pools are also an important part of patent litigation dynamics. Choi (2010) shows that patent owners have incentives to form patent pools and cross-licensing agreements as a resolution to complex patent litigation. Although there is an argument that such arrangements are efficient as a means to settle ongoing disputes and as integrating complementary technologies, the author points out that such agreements can also shelter invalid patents from challenges. He suggests that antitrust law should limit patent pools if the parties have not been able to fully litigate the validity of patents, when litigation costs are not too large. Tandon (1982) presents an interesting model in which patent life and compulsory licensing are determined simultaneously. The model shows that in some situations an optimal patent will last forever with compulsory licensing at the optimal rates. Patents with compulsory licensing can improve social welfare when compared with a patent system without compulsory licensing, even for patents that have a shorter duration.
Empirical work on patent litigation adds further complexity to our understanding of the dynamics between patent owners and potential infringers. Lanjouw and Schankerman (2001) start with the premise that litigation costs lower the return on investment in research and development. They study a wide range of patent litigation and find there is wide variation among patents in terms of their exposure to litigation risk. For example, chemical patents create low litigation risk, while patents for drugs and health have high litigation exposure. Not surprisingly, the most valuable patents, as determined through number of forward and backward citations, are most at risk for litigation, creating concerns for investment in research and innovation, especially by small firms. Their work supports industry-specific considerations for patent litigation reform and judicial review. Henry and Turner (2006) review appellate outcomes at the Federal Circuit, the 1982 appeals court created to review patent trials. They examine decisions from 1953 to 2002 (pre- and post-establishment of the Federal Circuit). Comparing Federal Circuit decisions with those of its predecessor appellate courts, the authors find that the Federal Circuit has been less reluctant to affirm trial court rulings of patent invalidity but more willing to affirm rulings of noninfringement. Lunney (2003) also found this pattern and concluded that the Federal Circuit’s approach to invalidity and infringement “will tend to limit the patent system’s ability to ensure the expected profitability, and hence the existence of desirable, but high cost innovation” (Lunney, 2003, p. 1).
One would expect this empirical picture of invalidity and infringement rules to affect the behavior of patent owners and imitators as litigants. Somaya (2003) confirms this intuitive prediction. He posits two hypotheses for the dynamics of patent litigation, the strategic stakes hypothesis and the mutual hold-up hypothesis. According to the strategic stakes hypothesis, patent litigation is a mechanism to protect valuable strategic assets. According to the mutual hold-up hypothesis, patent litigation is a defensive mechanism to obtain access to external technologies. Using a sample selection probit model of litigation, he finds support for the strategic stakes hypothesis but inconclusive evidence in favor of the mutual hold-up hypothesis. Breaking his study down by industry, the author finds that nonsettlement in patent lawsuits involving research medicines is increased by strategic stakes. However, lawsuits involving computer hardware show support for both the strategic and hold-up uses of litigation. In the early part of the product life-cycle, patent owners are more likely to litigate in order to obtain exclusivity (strategic stakes). However, patent owners are willing to settle computer patent litigation disputes in order to avoid substitution to competing technologies by infringers (mutual hold-up).
Considering patent litigation in the aggregate over the last two decades leads to two competing hypotheses. One is that courts in the United States have become friendlier towards patent owners. The second is that increased investment in research and development has led innovating firms to pursue litigation more vigorously. Cook (2007) finds evidence for both hypotheses. “While we find that both court outcomes and patent grants may have played a role in patent litigation, our results indicate that an increase in patent grants was more likely the larger factor in the patent litigation explosion” (Cook, 2007, p. 70). Courts, however, do affect the filing of litigation; therefore, reform of how courts operate can be a meaningful way to address negative implications of the patent litigation explosion, such as creating impediments to competition and innovation.
Empirical studies support the finding that patent quality can affect litigation dynamics, which in turn affects innovation and the diffusion of technology. Patent infringement remedies provide incentives for litigation and settlement. Remedies include injunctions and damages. An injunction is a court order that requires the losing party to refrain from acting or to act affirmatively. The actions include not making, selling, or using an infringing product or possibly affirmatively redesigning a product to avoid the infringement. Injunctions potentially limit competition in the marketplace through restraining the choices of a competitor. These restrictions are necessary to protect the rights of the patent owner, but if the injunction is too broad, the rights of consumers and competitors are affected. Determining the proper scope of a patent injunction is a challenging judicial task. Even if courts can impose the correct injunction, the problem remains of settlement in light of an imminent injunction. The patent owner might be able to expand its rights through settlement from a defendant competitor who seeks to avoid the weight of injunctive relief.
Damages also create difficult incentive problems. Patent owners who succeed in litigation can obtain lost profits caused by the infringement with a floor of a reasonable royalty. A reasonable royalty is what the parties would have negotiated if the patent had been licensed rather than infringed. Therefore, a successful patent owner can expect to obtain, at a minimum, what it would have obtained had the infringer chosen to license the patent, and could obtain more if there is adequate evidence of lost profits. Furthermore, a patent owner can obtain lost collateral revenues from the infringement. For example, if the patent owner provided services to support the patented infringement or sold unpatented component parts that fit the patented invention, the lost revenues from these services and components can also be recovered as a part of patent damages. In addition, if the patent owner can show that the infringer acted willfully, patent damages can be trebled. Patent damages can be sizeable, and they create strong incentives for patent owners to bring an infringement suit.
The rationale for patent damages is compensation for the violation of patent rights. Injunctive relief also has a compensatory feature by affirming the patent owners’ rights of exclusivity in the using, making, and selling of the patented invention. Furthermore, both also have a punitive component in deterring infringing activity. In assessing patent remedies, the question is how they work to both compensate for and deter infringement while not creating strategic incentives for litigation and settlement.
Lanjou and Schankerman, published in Cohen and Merrill (2003), confirm these predictions about litigation at a general level. They find that patent litigation is heterogeneous with litigation behavior varying across industry and technological arts. Specifically, they find that industry and technology effects influence the decision to litigation but not post-initiation outcomes such as settlement or win rates. Settlement occurs quickly after the suit is filed, with little expenditure of judicial resources. The bad news, as the authors state, is that individual inventors and small firms bear the costs of patent litigation as much as larger firms. The authors propose that patent litigation reform efforts should focus on the threat of patent litigation rather than the duration of litigation, since most lawsuits settle fairly quickly.
Ziedonis, published in the same volume, focuses on litigation in the semiconductor industry and finds that in the 1990s patent litigation increased as patents became broader in scope. This pattern contrasts with that during the early growth of the semiconductor industry in the 1970s, when firms worked to stave off litigation and engage in more cooperative, cross-licensing endeavors. Ziedonis concludes that patent strength, as determined by scope, seems to be an important predictor of litigation, with narrower patents providing some incentive for pooling and licensing rights in semiconductor inventions.
Some patents are more valuable than others. This statement is not surprising, but the reality which it describes is quite striking. Only a small fraction of the over 150,000 patents issues narrowly are litigated. Even fewer are licensed or turned into a commercial product (Allison et al., 2004; Moore, 2005). The patent system is essentially a lottery, with only a handful hitting it big, not many hitting commercial pay dirt at all, and most not receiving any returns for the investment. The skewness of patent rewards makes some scholars question the efficiency of the patent system.
From the perspective of prospect theory, the result is not striking. Only so many land claims result in the striking of oil or other valuable natural resources. Prospects are gambles. Therefore, one should expect many losers. The efficiency of the system rests on the social benefits from the valuable inventions that the patent system promotes and supports through commercialization. Within the parameters of prospect theory, the relevant question is what would be the social benefits and costs of having no patent system. As Kitch and other proponents emphasize, the benefits would not be as sizable and the costs would be increased secrecy and a lower level of innovation.
But there are important costs of so-called worthless patents. Rational choice might suggest that if an invention proves worthless, inventors should ignore it and move on to another project. While the costs of worthless patents are sunk, patent owners seek to recoup their investment by selling the worthless patents to other companies who might seek to enforce them for settlement value. Patent owners are selling worthless patents for scrap value, and the purchasers see the patents as investment in possible revenue streams from licensees and those who seek to avoid the costs of litigation. This phenomenon is sometimes referred to, in the popular press, as patent-trolling, but this blanket term masks the variety of entities that buy up patents for licensing or litigation purposes. These entities are labelled “patent assertion entities” or “nonpracticing entities.” The former is a more accurate description of entities that actively assert patent rights, while the latter describes entities, like some universities, that acquire patents but do not manufacture or practice the protected invention.
Kesan and Gallo (2006) present a game-theoretic model to explain why bad patents persist in the marketplace despite the existence of judicial review. Their model rests on asymmetric information about the quality of patents and the dynamics of settlement. They proposed, six years before the passage of the America Invents Act, an administrative opposition system that would allow parties to challenge the validity of patents. The authors modeled their proposed system on the opposition scheme existing in Japan at the time. Kesan and Gallo explain one aspect of worthless patents elegantly, and their work invites further empirical investigation to understand the interplay of persistent invalid patents and litigation. An open question is why there so many patents that are valid but nonetheless worthless. Is this a failure of markets or a failure of the granting agency?
Several scholars have addressed the litigation dynamics of patent assertion by acquiring companies. Lemley and Melamed (2013) paint a comprehensive picture of the trolling and patent assertion entities. The entities that purchase patents for enforcement purposes are varied, with different effects on social costs and benefits. Many of the problems stem from the disaggregation of patent portfolios and the hold-up costs of litigating under a set of uncertain and manipulable patent standards. Cotropia et al. (2014) categorize different types of patent assertion entities (PAEs) but find that there has not been an explosion in lawsuits brought by PAEs, as reported by the press and other scholars. Instead, they attribute any increase in the number of lawsuits to changes in procedural rules making it more difficult to join parties. Therefore, the scholarly literature casts doubt on the characterization of patent assertion entities, or trolls, and their behavior. FTC (2016) and Thumm and Gabison (2016) provide empirical studies on patent assertion entities as a guide to policymaking.
Patents and Competition
Greenleaf (2011, p. 87) reports that “in the United States, the use of patents to stifle competition dates from the mid-nineteenth century. . . After 1865, corporations in various sectors of the American economy exploited basic and tributary patents.” Examples include the sewing machine industry, harvesters, the cotton gin, arc lighting, and shoe manufacture. Usselman (2002) examines how patents, especially Westinghouse patents on braking technology, shaped the railroad industry. During this period, courts developed the doctrine of savings, which measured patent damages on the amount of R&D expenditure saved by the infringer. According to Usselman (p. 138), “Westinghouse had found a way to extract large profits from the market for patented railroad technology. A bevy of other inventors, aided by the doctrine of savings, appeared poised to duplicate the threat.” But patent disputes in the railroad industry gave way to a new era of railroad management that focused on administration rather than patenting. “Railroads,” Usselman continues, “increasingly perceived the patent system as an anachronism whose hidden liabilities might undermine efforts to provide more efficient transport. . . [T]he men responsible for managing the railroads would find ways . . . to create alternative pathways of innovation in their industry.”
Fast-forwarding more than a century, many of the concerns over patent and competition persist in the Joint Department of Justice and Federal Trade Commission Report on antitrust and patent law (Department of Justice and Federal Trade Commission, 2007). The report reflects a century’s evolution in legal and economic thinking about the anticompetitive uses of patents. Among ongoing concerns are both anticompetitive litigation and licensing practices, with an emphasis on the latter. Specific areas where antitrust scrutiny might apply are unilateral refusals to license in situations where technologies are complementary or where there is a significant threat of hold-up by the patent owner; standard-setting involving essential technologies that are subject to patent; tying arrangements; and practices that potentially expand the scope of patent rights. More recent concerns include reverse payment settlements involving the entry of generic drug competition, a topic that the report only mentions briefly but that has received greater attention in the courts. Also percolating through the courts and scholarship is the use of FRAND licensing to resolve antitrust concerns in markets for technology standards; on this issue, see Contreras (2015). For those interested in contemporary debates over patents and competition law, the DOJ/FTC report is an important starting point for assessing contemporary scholarship and policy thinking.
Lim (2017) provides a comprehensive overview of economic analysis of patents and competition. The author summarizes what is labelled the neoclassical approach to patents, based on models of monopoly and oligopoly and investment in risky research and development. Such models support a more limited role for antitrust intervention into patent licensing and litigation practices, on the theory that more aggressive enforcement can undermine incentives for invention and innovation. The author, however, turns to the behavioral economics literature to suggest areas where antitrust authorities may provide more scrutiny in order to protect consumer interests and improve efficiencies in technology markets. Behavioral economics, as a rubric, includes information economics, theories of strategic behavior, and psychological critiques of the rational actor model. Specifically, Lim points to information asymmetries in contracting and systemic misunderstanding about uncertainty and risk assessment as reasons why competitive markets might fail in creating efficient private orderings for licensing and selling patented technologies. Consumers may be subject to lock-in effects that patent owners can exploit in order to increase the reach of patent rights to tie in sales of unpatented goods and services. In addition, confirmation bias may move litigants to more aggressively pursue or defend lawsuits, resulting in increased patent litigation and opportunistic settlement. Lim’s article is an important starting point for understanding traditional terms of the antitrust-patent debate and assessing alternative heterodox theories.
Most interesting about behavioral economics in general and Lim’s discussion in particular is the recognition of how psychological limits might also affect the judiciary and legislators. As a decision-maker frames issues squarely within terms of incentives to invent and innovate, it becomes more difficult to identify how the patent right can be used anticompetitively. These psychological limitations can become entrenched in the case law because of precedent and can linger in legislative debates, given the dynamics of interest-group and electoral politics. Behavioral economics advocates like Lim urge identification of these limitations and often recommend structural reforms, such as oversight and substantive rules of decision, to dampen them. Patent and antitrust do not remain fixed categories around which to frame policy, but as interacting legal regimes both aimed at promoting competition and innovation.
Issues about patent reform percolate throughout this discussion, whether about patent remedies or about the patent prosecution process. Bessen and Meurer (2008) have presented a provocative agenda for patent reform drawing on their work on information externalities and the problem of uncertain patent boundaries. Their research is filled with industry anecdotes drawn from litigation records and data on litigation costs, supported by event studies analyses. The take-home message from this work is that “patents are a net disincentive.” But, as Lunney (2008) points out, they do not adequately estimate the costs of litigation, whether social or private. Event studies, by focusing solely on the impact of discrete events on stock prices, do not take into consideration the more complex dynamics of patent litigation and settlement or the social benefits and costs of obtaining and defending patents. Lunney also contests support for their information externality explanation as the cause of increased patent litigation. He shows, for example, that increases in patent litigation have been as large in industries where patent boundaries are murky (software) as in industries where they are clearer (chemical and pharmaceuticals). He concludes: “I have little doubt that the patent notice problems that Bessen and Meurer have identified are a part of the problem with the patent system . . . I am not persuaded that they have identified the problem with the system.” As the range of scholarship discussed in this article should show, to reduce patent law to one problem may well be a misguided venture, ignoring the richness of the landscape consisting of invention, innovation, and competition.
Ghosh, S. (2004). Patents and the regulatory state: Rethinking the patent bargain metaphor after Eldred. Berkeley Technology Law Journal, 19, 1315–1388.Find this resource:
Lim, D. (2013). Patent misuse and antitrust law: Empirical, doctrinal, and policy perspectives. Cheltenham, UK: Edward Elgar.Find this resource:
Maskus, K. (2012). Private rights and public problems: The global economics of intellectual property in the 21st century. Washington, DC: Peterson Institute for International Economics.Find this resource:
Scotchmer, S. (2004). Innovation and incentives. Cambridge, MA: The MIT Press.Find this resource:
Sherman, B., & Bentley, L. (2008). The making of modern intellectual property law. Cambridge, UK: Cambridge University Press.Find this resource:
Allison, J. R., Lemley, M., Moore, K., & Trunkey, D. (2004). Valuable patents. Georgetown Law Journal, 92, 435–486.Find this resource:
Allison, J. R., Lemley, M., & Schwartz, D. (2014). Understanding the realities of modern patent litigation. Texas Law Review, 92, 1769–1799.Find this resource:
Aoki, R., & Hu, J.-L. (2003). Time factors of patent litigation and licensing. Journal of Institutional and Theoretical Economics, 159, 280–301.Find this resource:
Bessen, J. E., & Meurer, M. J. (2006). Patent litigation with endogenous disputes. The American Economic Review, 99, 77–81.Find this resource:
Bessen, J. E., & Meurer, M. J. (2008). Patent failure: How judges, bureaucrats, and lawyers put innovation at risk. Princeton, NJ: Princeton University Press.Find this resource:
Chien, C. (2009). Of trolls, Davids, Goliaths, and kings: Narratives and evidence in the litigation of high tech patents. North Carolina Law Review, 87, 1571–1615.Find this resource:
Choi, J. P. (1998). Patent litigation as an information-transmission mechanism. The American Economic Review, 88, 1249–1263.Find this resource:
Choi, J. P. (2010). Patent pools and cross-licensing in the shadow of patent litigation. International Economic Review, 51, 441–460.Find this resource:
Cohen, W. M., & Merrill, S. A. (2003). Patents in the knowledge-based economy. Washington, DC: The National Academy Press.Find this resource:
Contreras, J. (2015). A market reliance theory for FRAND commitments and other patent pledges. Utah Law Review, 2015, 479–558.Find this resource:
Cook, J. (2007). On understanding the increase in U.S. patent litigation. American Law and Economics Review, 9, 48–71.Find this resource:
Cotropia, C. A., Kesan, J., & Schwartz, D. L. (2014). Unpacking patent assertion entities. Minnesota Law Review, 99, 649–703.Find this resource:
Crampes, C., & Langinier, C. (2002). Litigation and settlement in patent infringement cases. The RAND Journal of Economics, 33, 258–274.Find this resource:
Department of Justice & Federal Trade Commission. (2007). Antitrust enforcement and intellectual property rights: Promoting innovation and competition. Washington, DC: U.S. Dept. of Justice and Federal Trade Commission.Find this resource:
Federal Trade Commission. (2016). Patent assertion entity activity: An FTC study. Washington, DC: Federal Trade Commission.Find this resource:
Frakes, M. D., & Wasserman, M. F. (2013). Does agency funding affect decisionmaking? An empirical assessment of the PTO’s granting patterns. Vanderbilt Law Review, 66, 65–149.Find this resource:
Frakes, M. D., & Wasserman, M. F. (2014). The failed promise of user fees: Empirical evidence from the U.S. Patent and Trademark Office. Journal of Empirical Legal Studies, 11, 602–635.Find this resource:
Frakes, M. D., & Wasserman, M. F. (2015). Does the U.S. Patent and Trademark Office grant too many bad patents? Stanford Law Review, 67, 613–676.Find this resource:
Frakes, M. D., & Wasserman, M. F. (2016). Patent office cohorts. Duke Law Journal, 65, 1601–1656.Find this resource:
Grady, M. F., & Alexander, J. I. (1992). Patent law and rent dissipation. Virginia Law Review, 78, 305–348.Find this resource:
Greenleaf, W. (2011). Monopoly on wheels: Henry Ford and the Selden automobile patent. Detroit: Wayne State University Press.Find this resource:
Harhoff, D. (2009). Economic cost-benefit analysis of a unified and integrated European patent litigation system. Munich: Institute for Innovation Research, Technology Management and Entrepreneurship.Find this resource:
Heller, M. (1998). The tragedy of the anticommons: Property in the transition from Marx to markets. Harvard Law Review, 111, 621–688.Find this resource:
Henry, M. D., & Turner, J. L. (2006). The Court of Appeals for the Federal Circuit’s impact on patent litigation. The Journal of Legal Studies, 35, 85–117.Find this resource:
Kesan, J., & Gallo, A. A. (2006). Why ‘bad’ patents survive in the marketplace and how should we change? The private and social costs of patents. Emory Law Journal, 55, 61–134.Find this resource:
Khan, B. Z. (1995). Property rights and patent litigation in early nineteenth-century America. The Journal of Economic History, 55, 58–97.Find this resource:
Khan, B. Z. (2009). The democratization of invention: Patents and copyrights in American economic development 1790–1920. Cambridge, UK: Cambridge University Press.Find this resource:
Kitch, E. W. (1977). The nature and function of the patent system. The Journal of Law and Economics, 20, 265–290.Find this resource:
Lanjouw, J. O., & Schankerman, M. (2001). Characteristics of patent litigation: A window on competition. The RAND Journal of Economics, 32, 129–151.Find this resource:
Lemley, M. A., & Melamed, A. D. (2013). Missing the forest for the trolls. Columbia Law Review, 113, 2117–2187.Find this resource:
Lim, D. (2017). Retooling the patent-antitrust intersection: Insights from behavioral economics. Baylor University Law Review, 69, 124–193.Find this resource:
Lunney, G. S., Jr. (2003). Patent law, the Federal Circuit, and the Supreme Court, a quiet revolution. Supreme Court Economic Review, 11, 1–80.Find this resource:
Lunney, G. S., Jr. (2008). On the continuing misuse of event studies: The example of Bessen and Meurer. Journal of Intellectual Property Law, 16, 35–56.Find this resource:
MacLeod, C. (2002). Inventing the Industrial Revolution: The English patent system 1660–1800. Cambridge, UK: Cambridge University Press.Find this resource:
Merges, R. P. (1988). Commercial success and patent standards: Economic perspectives on innovation. California Law Review, 76, 803–876.Find this resource:
Merges, R. P., & Nelson, R. R. (1990). On the complex economics of patent scope. Columbia Law Review, 90, 839–916.Find this resource:
Meurer, M. (1989). The settlement of patent litigation. RAND Journal of Economics, 20, 77–91.Find this resource:
Moore, K. (2001). Forum shopping in patent cases: Does geographic choice affect innovation? North Carolina Law Review, 79, 889–938.Find this resource:
Moore, K. (2005). Worthless patents. Berkeley Technology Law Journal, 20, 1521–1550.Find this resource:
Oddi, A. S. (1996). Un-unified economic theories of patents: The not-quite Holy Grail. Notre Dame Law Review, 71, 267–326.Find this resource:
Schmookler, J. (1972). Patents, invention, and economic change: Data and selected essays. Cambridge, MA: Harvard University Press.Find this resource:
Sherman, B., & Bently, L. (1999). The making of modern intellectual property law. Cambridge, UK: Cambridge University Press.Find this resource:
Somaya, D. (2003). Strategic determinants of decisions not to settle patent litigation. Strategic Management Journal, 24, 17–38.Find this resource:
Sukhatme, N. U. (2014). Regulatory monopoly and differential pricing in the market for patents. Washington & Lee Law Review, 71, 1855–1922.Find this resource:
Tandon, P. (1982). Optimal patents with compulsory licensing. Journal of Political Economy, 90, 470–486.Find this resource:
Thumm, N., & Gabison, G. (2016). Patent assertion entities in Europe: Their impact on innovation and knowledge transfer in ICT markets. Seville, Spain: European Commission, Joint Research Centre.Find this resource:
Usselman, S. W. (2002). Regulating railroad innovation. Cambridge, UK: Cambridge University Press.Find this resource:
Vishnubhakat, S., Rai, A. K., & Kesan, J. (2016). Strategic decision making in dual PTAB and district court proceedings. Berkeley Tech. Law Journal, 31, 45–122.Find this resource: