Abstract:
This chapter critiques widely accepted claims about the alleged failures of the patent system, such as the “patent explosion” and “patent thicket” theories. It examines the historical rise in patent applications and disputes the notion that increased patenting inherently harms innovation. Empirical evidence suggests that patent growth since the 1980s reflects globalization and stronger enforcement rather than declining patent quality. Nonetheless, claims of excessive litigation and low-quality patents were influential in shaping the America Invents Act (AIA) in 2011, despite weak supporting evidence. The Patent Trial and Appeal Board (PTAB), created by the AIA, has been widely used by large firms to challenge patents. While some view the PTAB as a quality control measure, others argue it disadvantages smaller innovators. Historical data shows that the U.S. patent system has evolved in response to market needs, not systemic failure. The author argues that policy decisions were shaped by ideological and economic interests rather than objective analysis. It concludes that strong patent enforcement has historically fostered innovation rather than hindered it.
Chapters in the Book:
- Part 3- Unmaking Patent Law
- Chapter 8- The Patent Litigation Explosion and Other Patent Horribles (p.164-190)
Summary
8. The Patent Litigation Explosion and Other Patent Horribles
In the following, four categories of “patent horribles” that describe the alleged failure of the patent system since the establishment of the Federal Circuit in 1982 will be discussed. These widely accepted claims have played a significant role in shaping patent policy, including the enactment of the America Invents Act in 2011. However, the empirical basis for these assertions was not thoroughly investigated and, upon closer examination, appears to be questionable. Despite contradictory evidence, these assumptions have remained influential and continue to shape patent policy across all branches of government.
8.1. The Patent Explosion
Starting in the early 2000s, academic and popular discussions of patent policy often referred to the “patent explosion.” Firstly, the “patent explosion” refers to the sharp rise in patent applications and grants starting in the early 1990s. In addition, the term is intended to convey the normative view that this increase is abnormally large. For example, Jaffe and Lerner argued in 2004 that many newly granted patents lacked merit. They claimed stronger patents encouraged patents acquisition rather than real innovation, increasing litigation and licensing costs. However, a closer analysis suggests these claims rely on weak or highly contestable factual grounds.
8.1.1. Examining the Evidence
Typically, the assertions of a “patent explosion” are based on absolute figures over a short period preceding and following the establishment of the Federal Circuit. For example, Jaffe and Lerner argued that post-1982 patent growth was unusually rapid and included many low-quality patents. They argued that this was accompanied by a proliferation of patent awards of dubious merit. However, their data is incomplete in two respects. First, it fails to distinguish between patent applications filed by foreign-resident and U.S.-resident inventors at the USPTO. (The USPTO classifies patents based on the stated country of residence of the first-named inventor on a patent application.) Second, the data does not consider patent applications based on changes in the size of the U.S. resident population. This has important implications. Since the mid-1970s, foreign-resident patent applications increased significantly, surpassing U.S.-resident grants by 2001. Thus, the rise in patents may reflect globalization and increased R&D activity rather than an excessive or unwarranted surge in patenting.
To provide a more accurate assessment of the relative increase in patenting activity, one should take into account patent applications and grants based on changes in the size of the U.S. population. Per capita patenting rates in the 1990s reveal that there was no abnormal increase in application. Instead, they are in line with historical levels seen in the late 19th and early 20th centuries. Similarly, as of 2002, U.S.-resident patent applications had not exceeded the 1923 peak, and grant rates were still below 1916 levels. When situated within this larger historical context and considered on a per capita basis, the post-Federal Circuit patenting surge does not appear as abnormal.
Additionally, Jaffe and Lerner assert that the increase in patenting during this period reflects a purported decline in USPTO examination standards. This claim lacks strong empirical support, as well. Historically, strong patent enforcement periods have correlated with increased patenting activity, while weak enforcement periods saw declines. The best explanation for the patent surge starting in the 1980s is not a largely undemonstrated decline in examination standards. Rather, it is due to changes in case law that increased the likelihood that a patent would be enforced. In other words, the rise in patenting reflected a rational market response, rather than excessive or low-quality patents. Thus, the patent surge after the 1980s was likely driven by legal changes that strengthened patent enforcement, not a decline in patent quality.
Interestingly, a true “patent explosion” in patent applications and grants occurred after the early 2000s. Per capita U.S.-resident patent grants exceeded the 1916 peak in 2012 but have remained mostly stable since 2015, with a slight rise in 2019. While pointing in the same direction, this observation is stronger for patent applications than patent grants. In general, one can say that patenting activity significantly increased after 2005. However, these trends appear to be stabilized, although at levels that are still historically high.
8.1.2. Interpreting the Evidence
It is important to keep in mind that the fact that patenting rates have increased is inherently ambiguous as a matter of social welfare without further analysis. For example, a rise in patents can indicate opportunistic litigation. But it may also reflect increased innovation or business models that support R&D investment and thus market entry and competition. However, critics of the post-1982 patent system mainly focused on litigation and licensing costs while largely ignoring long-term benefits. Publications such as the Jaffe and Lerner book significantly influenced academic and policy discussions, shaping the perception of a “patent explosion” as harmful. Ultimately, the view that excessive patents burdened innovation and consumers became widespread despite weak or unexamined empirical support. In the end, the “patent explosion” critique led to even broader claims that the patent system was “broken” and required urgent reform.
In conclusion, the “patent explosion” thesis was widely adopted in legal and economic scholarship without rigorous empirical validation. In its aftermath, this thesis led to three related claims: the “patent thicket,” “junk patents,” and “patent litigation explosion” theories. In the same fashion, these claims argue that strong patent protection hinders innovation and benefits entrenched incumbents. Despite empirical analysis revealing that these assertions rest on weak evidence, much of conventional U.S. patent policy wisdom is based on these claims.
8.2. Patent Thickets
Conventional wisdom holds that high patent issuance creates a “thicket” of litigation and licensing that discourages innovation. This claim emerged after the 1980 Bayh-Dole Act, which allowed research institutions to patent federally funded inventions. Moreover, Diamond v. Chakrabarty which affirmed patent rights for biotechnology, contributed to the discussion. The “thicket” thesis argues that IP rights restrict the free flow of scientific discoveries, potentially hindering innovation. This is because biomedical research had formerly operated in a mostly IP-free environment in which ideas and discoveries could flow freely. The logic is asserted as IP rights would hinder innovation since IP right holders would have economic incentives to restrict access to discoveries. However, an alternative view suggests that IP rights attract investment, accelerating the commercialization of academic R&D into medical therapies. This is underscored by historical evidence which highlights that limiting patents on federally funded research regarding their assignment before Bayh-Dole hindered industry partnerships. This, the benefits of commercializing federally funded research may outweigh any increase in transaction costs due to expanded patent rights.
8.2.1. Examining the Evidence
The most influential paper advancing the patent thicket thesis was a 1998 publication by Michael Heller and Rebecca Eisenberg, focused on biomedical research. Crucially, this paper is often treated as an authoritative statement, despite being a theoretical concern not supported by empirical evidence. Since its publication, a substantial body of empirical evidence has allowed for a factual evaluation of the patent thicket thesis. The evidence generally shows that the patent thicket thesis has not significantly hindered biomedical research.
In particular, repeated surveys of biomedical researchers by John Walsh found little evidence that patents or licensing issues have materially delayed research projects. In 2006, a review concluded that the anti-commons problem[1] predicted by the patent thicket theory was not supported by available data. Furthermore, in another study from 2021, no evidence was found to support the theory that patenting research inputs in biotechnology negatively impacted innovation.
Researchers have shown that a particular patent-related dispute, the Harvard oncomouse, may have slowed research for a certain period of time. However, the broader trend shows patents complement open science.
In a similar fashion, Heller and Eisenberg have acknowledged the gap between anti-commons theory and real-world conditions in biomedical research. The authors have concluded that the empirical evidence for the anti-commons effect is inconclusive. Despite this, the patent thicket thesis continues to be referenced as an established risk in IP policy discussions. The conclusion reached by a review of the policy literature in 2006 is still relevant: policymakers more often rely on anecdotes rather than systematic data.
8.2.2. Interpreting the Evidence
If a hypothesis fails to align with available evidence, it is essential to consider alternative explanations that better match the data. Given the weak support for the patent thicket thesis, it is worth exploring the alternative hypothesis that patent protection has not hindered research and innovation. Instead, IP enforcement may foster economic gains by promoting the commercialization of scientific research.
The reason why the patent thicket thesis lacks empirical evidence, might break down to two reasons. First, without the legal exclusivity provided by patents, outside investors would have no incentive to invest significant funds in developing technological innovations. This was well-known prior to the Bayh-Dole Act. At that time, federal agencies struggled to attract commercial interest in innovations from government labs due to legal restrictions preventing patent assignments. Second, patents can lower transaction costs by allowing innovators to disclose their valuable assets to potential business partners. Therefore, cooperative agreements are facilitated, because each partner’s rights in the commercial development process are clearly defined.
Without patents or other forms of intellectual property protection, smaller start-up firms would be vulnerable to expropriation by larger, established companies. By reducing the risk of expropriation, patent protections enabled the formation of business partnerships. Ultimately this integration of complementary innovation and commercialization capacities led to quicker and more cost-effective market introductions.
The biotech industry’s success since Bayh-Dole and Chakrabarty, along with increased patent enforcement, aligns more with the “patent enablement” thesis than the patent thicket thesis. Following the patent enablement thesis, the Bayh-Dole Act, Chakrabarty decision, and the Federal Circuit jurisprudence fostered biotech commercialization by securing property rights. The industry’s success aligns with this thesis, demonstrating real-world performance and the role of patents in promoting commercialization. Looking at data from 1980 to 2003, over 470 biotech firms went public, with patents increasing venture capital (VC) investments in the sector. This aligns with evidence collected by Adelman and DeAngelis. They showed that patents increased biotech firms’ chances of receiving VC funding, especially among universities. Thus, there was a shift in innovation toward universities and start-ups. The Bayh-Dole Act spurred this structural change in biotech, making R&D investments more attractive and fueling innovation and commercialization.
To concretely illustrate the value-enhancing function of patents in the biotech ecosystem, consider the Cohen-Boyer patent. It was one of the first biotech patents, supporting the creation of synthetic human insulin. The biotech product was developed and marketed jointly by the start-up Genentech and Eli Lilly. Crucially, the patent was licensed widely at rates below market value. This meant that the patent did not hinder innovation. Instead, it facilitated the development of medical innovations by enabling broad licensing and collaboration. By licensing the patent, many firms gained access to the technology and secured VC funding and partnerships with pharmaceutical companies to bring innovation to the development phase. This model of patent-driven commercialization has been replicated multiple times, leading to the development of many new drugs in biotech.
8.3. Junk Patents
The patent explosion and patent thicket assertions reflected concerns that there was an “excessive quantity” of patents being issued by the USPTO. This could potentially hinder innovation through high licensing and litigation costs. Critics claimed the USPTO had lowered its standards for issuing patents, which led to an influx of low-value patents that failed to meet patentability thresholds. Specifically, agencies like the FTC and National Research Council warned that the increased quantity of patents reflected reduced standards of utility, novelty, and non-obviousness. In 2009, a report released by the Brookings Institution suggested that the “junk patents” are a widespread problem, stating that the USPTO was flooding the market with low-quality patents.
The popular belief was that the USPTO was acting as a rubber stamp. This led to the 2011 America Invents Act (AIA), establishing the Patent Trial & Appeals Board (PTAB). The AIA created a “two-front litigation” scenario. Now, patent owners faced increased litigation costs with challenges to patent validity in both, courts and the PTAB.
8.3.1. Examining the Evidence
Remarkably, the empirical literature had not provided compelling evidence to support the junk patents thesis at the time of the enactment of the AIA in 2011. For example, Frakes and Wasserman found no compelling evidence that the USPTO practices were systematically driving patent over-issuance. It should be noted that the authors present evidence showing an examiner bias toward patent over-issuance. However, this result does not rely on any elements of the patent examination process that are specific to the strengthening of patent rights in the early 1980s.
On the other hand, the “junk patents” issue is frequently only highlighted by anecdotal examples. Crucially, it was not backed by rigorous empirical evidence showing systematic low-quality patent issuance by the USPTO. For these concerns to be valid, a rigorous empirical analysis would have needed to demonstrate a decline in patent examination standards over time, particularly since the 1980s. Like the patent explosion thesis, widespread adoption of the junk patents thesis can be traced back to imperfections in an academic publication. In 2001, a study by Quillen and Webster claimed an implausibly high patent grant rate of 97%, leading to the theory of “junk patents.” While the study was later revised and the figure was reduced to an 85% grant rate, the estimates fueled concerns over patent over-issuance. Both studies were referenced in the 2003 FTC and 2004 National Academy of Sciences reports, which raised alarms about low-quality patents but lacked conclusive evidence. However, both reports acknowledged the uncertainty in evidence regarding low-quality patents and highlighted contested methodologies. Other studies that applied more nuanced methods estimated patent grant rates between only 60% and 76% in 1981-1998. A subsequent study found the grant rate for patents filed from 1996-2005 had declined to 71%. to put this in context, this means that the USPTO had likely returned to a stricter standard by the time the AIA was enacted and the PTAB was established.
8.3.2. Interpreting the Evidence
Even if one assumes that the grant rate increased under the Federal Circuit, it does not necessarily indicate a decline in patent quality. Two studies that claim over-granting bias, attribute it to factors that were present before and after the 1980s reduction in standards. Other research from 2004 and 2006 suggested a lack of examination rigor at the USPTO relative to the European Patent Office (EPO). However, the lower grant rates of the USPTO can be attributed to EPO applicants withdrawing applications. Furthermore, a 2008 study concluded that procedural and substantive differences between patent offices complicate comparing grant rates and attribute any differences to examination rigor. In short, at the time of the AIA and PTAB’s creation, empirical evidence on patent quality showed no proof of a decline in examination standards. Specifically, there was no clear evidence that the increase in patent applications and issuance from the 1980s was caused by a decline in the USPTO examination quality. It may nonetheless be the case that the USPTO had issues with under screening low-quality patents. However, finding evidence for that is impossible without defining objective measures of patent quality. It may also be the case that mechanisms can be designed to improve patent quality without weakening patents as incentives or monetization tools. However, at the time of the enactment of the AIA and the creation of the PTAB, policymakers relied on the unsupported “junk patents thesis.”
8.4. Patent Litigation Explosion
In addition to a flood of low-value patents, the purported explosion in patent issuance was thought to have caused a surge in patent litigation. Both claims were derived from the underlying claim that the Federal Circuit had lowered the standards that patent owners must meet. This increase in litigation was assumed to result in opportunistic entities acquiring patents to secure large settlements, harming innovation and undermining the patent system. Similar to the previous claims, the patent litigation explosion was a plausible claim based on stylized facts[2]. During the 1990s, the number of patent infringement suits doubled, continuing to rise through 2011, with around 3,500 suits filed that year. Policy commentators, such as Jaffe and Lerner (2004) and Bessen and Meurer (2009), noted the increase in patent litigation. They attributed it to the expanded patent subject matter and enforceability. A 2013 White House report highlighted the increase in patent suits, contrasting it with the “relatively rare” infringement suits of the 20th century. However, when considered with the number of patents issued or in force, the increase in patent litigation appears less exceptional. As noted by Congress, the number of patent infringement suits filed as a percentage of patents remained relatively constant from 1993 to 2006.
Subsequent empirical research showed that the absolute rise in the number of patent infringement suits starting in the 1990s was historically unexceptional. Katznelson’s research from 1923-2013 revealed patent litigation intensity was not higher post-Federal Circuit compared to earlier periods, except for 2011-2013. The 2011-2013 surge was due to a change in federal civil procedure limiting consolidation of multiple defendants in one case. In historical research, a study discovered that the highest rate of patent litigation occurred in the mid-to late 19th century, not after the Federal Circuit’s establishment. While patent litigation did rise in absolute terms post-Federal Circuit, this increase was not exceptional when considered with the rise in issued patents. Therefore, the claim of an unprecedented “patent litigation explosion” was largely unfounded.
8.5. Policy Payoff
A stylized fact argument will persist when it supports policy objectives favored by a strong coalition of constituencies and is not resisted by an equally strong opposing coalition. The “patent thicket,” “junk patents,” and “litigation explosion” assertions, which are all extensions of the underlying “patent explosion” thesis fulfil these conditions. They served as useful tools for advocating policies to reduce issued patents and discourage patent litigation. As a result, there was little interest in closely examining the reliability of the factual claims underlying the “patent explosion” thesis. In the end, patent-skeptical groups advocating the patent explosion thesis achieved a landmark policy victory with the enactment of the AIA in 2011. The AIA sought to address patent quality concerns and prevent opportunistic litigation. This was achieved through the creation of the Patent Trial and Appeal Board (PTAB), a new tribunal within the USPTO to challenge patent validity.
8.5.1. The PTAB Action
Among the mechanisms provided by the PTAB to challenge an issued patent, the most commonly used is the inter partes review (IPR) mechanism. It allows any party to challenge a patent’s validity based on lack of novelty or non-obviousness due to prior art. Hence, unlike federal court litigation, the IPR mechanism removes the standing requirement, allowing challenges from entities that might not otherwise have legal standing. Furthermore, IPR challenges can be filed at any time during the life of a patent. Moreover, no statutory presumption of validity applies in IPR, thus making it easier to invalidate patents.
Since its 2013 implementation, the PTAB has become a powerful tool for firms facing or expecting patent infringement lawsuits. This is because defendants in federal court can use IPR petitions as a “second front” strategy to increase the cost of litigation for patent owners, often discouraging lawsuits. This became obvious in 2022, when over 80% of IPR petitions were filed alongside pending district court cases, making it a standard litigation tactic. As a consequence, economic pressures may force settlements favorable to infringers or deter smaller patentees from filing suits in the first place. Data on PTAB outcomes suggest that it is often prudent for a patentee to settle rather than proceed to a fully adjudicated PTAB proceeding. During the life of the PTAB through 2022, around 8,500 patents were challenged. A third of which were partially invalidated and 10% were fully invalidated. In other words, among all patents targeted by petitioners, only 57% survived with all claims intact.
8.5.2. Evaluating the PTAB’s Performance
The implications of the PTAB’s performance to date are ambiguous. The high success of PTAB petitioners’ rate may indicate effective patent quality control on one hand or strategic misuse by incumbents on the other. It is nonetheless possible to gain some insight by considering the composition of the pool of entities that regularly file challenges at the PTAB. In recent years, large firms such as Samsung, Apple, Google, Intel, and Microsoft have increasingly dominated PTAB petitions, with their share rising to a quarter in 2021. It is notable that many leading PTAB petitioners were strong advocates for the AIA statute that established the PTAB. Interestingly, most of these firms are common defendants in patent infringement litigations. Hence, one would naturally expect these firms to be among the most frequent users of the PTAB mechanism. Thus, concerns may arise that large firms may use the PTAB to block legitimate claims from smaller innovators. It is worth noting that, during 2014-2022, slightly more than half of PTAB petitions filed each year challenged patents held by operational entities (instead of non-practicing entities). In addition, concerns about the PTAB misuse further grew as major tech firms oppose due-process-based reforms aimed at correcting procedural biases favoring challengers over petitioners.
In two decisions in 2018 and 2020, the PTAB ruled that the USPTO director could decline IPR petitions if a related district court case was pending. Subsequently, the USPTO director Andrei Iancu exercised his statutory authority to designate those PTAB decisions as precedential. In response, Apple, Cisco, Google, and Intel sued the USPTO director, claiming he exceeded his authority. The lawsuit was dismissed in 2021. However, the tech firms pushed for the “PTAB Reform Act” in 2022, aiming to reverse these procedural changes. And while the legislation failed, the USPTO director Katherine Vidal changed policy in 2022 anyway. Yet, it should be noted that these policy changes had little practical impact. In summary, based on existing evidence, it remains unclear how many PTAB petitions deter nuisance lawsuits versus how many consider legitimate infringement claims. However, what can be observed is that many PTAB petitions do not target non-practicing entities. This casts doubts about whether opportunistic patent litigation was ever a widespread problem. Given these empirical uncertainties in PTAB operations, doubts remain about whether it genuinely improves the U.S. patent system.
8.6. Closing Thoughts
The U.S. patent system is often portrayed as being captured by opportunistic entities that harm innovation. Critics cite issues like patent explosions, thickets, junk patents, and excessive litigation. However, empirical evidence for these claims is weak or ambiguous. Patent growth since the late 1980s is historically not abnormal. Market participants usually resolve patent thickets through contracts. There is no credible evidence that examination quality declined at the USPTO. The increase in patent litigation is the naturally expected result of the increase in patent issuance. As will be discussed, the revival of the patent system in the 1980s likely benefited, rather than harmed, U.S. innovation.
[1] The anti-commons problem suggests that excessive patenting can hinder innovation by creating overlapping rights, making it difficult for researchers to access essential knowledge and technologies.
[2] Stylized facts refers to arguments that may be theoretically plausible, but they are supported by a very limited number of empirical proof. More often, they rely on anecdotes.
