Much Ado About Injunctions: Redux

Patent policy debates have raged since 2006 when the Supreme Court decided eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006). Critics of eBay frequently cite the decision as having effectively ended the ability of patentees to obtain injunctions, vitiating the “right to exclude” afforded by U.S. patents, gutting the value of U.S. patents, devastating U.S. innovation,  and making the United States a patent backwater compared to the European Union and the People’s Republic of China.[1] Remarkably, for such a lynchpin of numerous patent policy arguments, relatively little hard data about the actual impact of eBay has permeated the debates. Perhaps that is because the data over the past 20 years tells a strikingly different story: on average, post-eBay courts routinely grant well over 80% of permanent injunction requests in patent infringement cases.

Seven years ago, an August 2019 post on Patent Progress reported data on grant rates for permanent injunctions in patent cases both before and after the eBay decision. Based on that data, the post concluded that eBay, in fact, had little impact on the ability of operating companies to obtain an injunction, but that non-practicing entities (NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it.) had a reduced likelihood of obtaining an injunction (although still able to do so in the majority of cases).

Newly available data from cases since 2019 demonstrates that those conclusions still hold water and provides greater clarity on exactly how eBay has affected different types of patent owners. In particular, the data shows that even some non-practicing entities (NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it.) that play constructive roles in the U.S. innovation ecosystem—such as research universities—are still able to obtain injunctions in most instances, similar to operating companies. Thus, eBay has been an effective tool to distinguish between patentees who are not protecting innovation and are pure rent-seekers, on the one hand, and patentees who contribute to innovation and should be eligible for injunctions on the other. If anything, the new data reveals that perhaps the worst patent assertion entities (PAEs), the true “patent trolls,” are granted injunctions too often even under eBay.

Background

Among the remedies potentially available to an owner of a U.S. patent when enforcing the patent against an infringer is a permanent injunction blocking the infringer from further infringing activities (e.g., selling infringing products).[2] Such an injunction may only be granted “in accordance with the principles of equity.”[3]

Prior to eBay, patentees enjoyed a presumption that, upon a finding of patent infringement, they would be entitled to a permanent injunction except in “unusual” cases where “exceptional circumstances” are present or “in rare instances” to protect the public interest.[4] The Supreme Court rejected this categorical rule, and held instead that permanent injunctions in patent cases are subject to the same legal standards as any other type of case, namely satisfying the traditional four equitable factors:  (1) irreparable harm; (2) inadequacy of remedies at law, such as monetary damages; (3) balance of hardships; and (4) the public interest.[5]

The crux of the eBay decision and the principles of equity required by the statute are that injunctive relief is a remedy intended to address harms that cannot be addressed through monetary compensation.

The years following eBay spawned a 20-year effort to undo the Supreme Court’s decision. This effort has included proposed legislation, most recently in the form of the RESTORE Patent Rights Act. Oceans of ink have been spilled to advance the view that eBay has had a tremendously negative impact on U.S. innovation.[6] The U.S. International Trade Commission (ITCInternational Trade Commission) has seen a massive growth in its Section 337 docket because it can grant an injunction-like exclusion orderAn order issued by the U.S. ITC as a result of a 337 action, excluding from entry into the United States goods found to infringe a U.S. patent..[7] Many believe that European courts, like the Unified Patent Court (UPC), are superior venues for patent litigation and enforcement (particularly for standard essential patents (SEPs)) because they grant injunctions virtually automatically.[8]

The core assumption underlying all of these efforts is that eBay has significantly reduced the ability of patent owners to obtain injunctions in U.S. courts. In fact, it is not uncommon to hear in some quarters that it is impossible or near-impossible to obtain injunctions after eBay.[9]

The data, however, tells a far different story.

Updated Data on Permanent Injunctions

The August 2019 post on Patent Progress examined data on injunction rates and noted that, before eBay, contested motions for permanent injunction were granted about 89% of the time following a plaintiff victory in a patent infringement suit.[10] After eBay, permanent injunctions were granted about 85% of the time, slightly lower but hardly the catastrophic collapse that eBay detractors advertise.[11]

Further, the post examined different types of plaintiffs. Specifically, using high-volume plaintiffs (HVPs) as a proxy for NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it., the data showed that HVP entities received injunctions 89% of the time before eBay, whereas they were granted injunctions about 63% of the time after eBay.[12] In contrast, non-HVP entities’ injunction rates shifted only slightly from about 89% to about 87%.[13] This data is shown in the following table:

This data seems to show first and foremost that eBay did not have the cataclysmic impact on the ability of patentees to obtain injunctions that critics claim. In fact, even NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. (or at least HVPs) were significantly more likely than not to successfully obtain an injunction even after eBay, though injunction rates did decline somewhat. Operating companies (or at least non-HVPs) experienced almost no change in injunction rates after eBay, successfully obtaining an injunction in the overwhelming majority of cases.

Newly available data on cases since 2019 sheds further light on these issues. This data was obtained from the Docket Navigator database, a comprehensive database of federal civil litigation. From a dataset of all patent cases from January 1, 2019, through December 8, 2025, all decisions on motions for permanent injunction were isolated. To focus on the effect of eBay, decisions on purely procedural, or other non-substantive  or non-patent-related bases, were excluded (e.g., denial for mootness), as were Hatch-Waxman ANDA cases and so-called “Schedule A” cases. Finally, the type of entity requesting the injunction was coded using the Stanford NPENon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. Litigation Database classificationAll patents are categorized into areas of technology with a system of classes and subclasses established by the USPTO. scheme. This data is shown in the following table:

The data is stark. From 2019 to 2025, when district courts evaluated a plaintiff’s request for a permanent injunction on the eBay factors following a finding of infringement, district courts granted injunctive relief 92% of the time (88% in full, 4% in part), including both contested and uncontested motions.[14] When considering only contested motions, district courts granted injunctions 84% of the time (76% in full, 8% in part), very consistent with the data reported in August 2019 (85%). Again, compared to a roughly 89% grant rate prior to eBay, it is clear that the Supreme Court’s decision had a relatively minor impact on overall availability of injunctions.

Compared to August 2019, this new data is more precise as to the types of entities requesting these injunctions. The Stanford NPENon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. Litigation Database[15] divides patent plaintiffs into 13 different classifications, which are further placed into three major categories (except classificationAll patents are categorized into areas of technology with a system of classes and subclasses established by the USPTO. no. 10, “undetermined” entities):[16]

1. Practicing Entities (classificationAll patents are categorized into areas of technology with a system of classes and subclasses established by the USPTO. no. 8): companies that make or sell products or services
2. Non-PAE NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. (classificationAll patents are categorized into areas of technology with a system of classes and subclasses established by the USPTO. nos. 2, 3, 6, 7, 9, 11-13): NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. with a primary purpose other than asserting patents in litigation, including universities, individual inventors, and technology development companies
3. PAEs (classificationAll patents are categorized into areas of technology with a system of classes and subclasses established by the USPTO. nos. 1, 4, and 5): NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. with the primary purpose of litigating patents, such as entities that acquired patents from others for litigation purposes

For practicing entities, which made up the vast majority of injunction requests, courts granted injunctions 94% of the time (89% in full, 5% in part) for contested and uncontested motions, and 85% of the time (77% in full, 8% in part) when looking just at contested motions. This data also is very consistent with pre-eBay rates (89%) and the August 2019 data (87% for non-HVPs, contested motions only).

For NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. that are not PAEs, courts granted injunctions 86% of the time (84% in full, 2% in part) for contested and uncontested motions, and 78% of the time (74% in full, 4% in part) for contested motions only. Interestingly, these rates are only slightly lower than those for practicing entities.

For PAEs, district courts granted injunctions 72% of the time (all in full) for contested and uncontested motions, and 71% of the time (all in full) for contested motions only. Although notably lower than for practicing entities, the fact that even PAEs receive injunctions 72% of the time they request one is shockingly inconsistent with the narrative that eBay critics continue to try to advance.

Conclusion

Twenty years after the eBay decision seemingly upended patent law, empirical data makes clear that it did not. Forcing plaintiffs to prove their need for equitable relief (rather than presuming it) has, at most, modestly reduced injunction rates. Indeed, for operating companies that need to protect their products (and the innovative technology embodied in them) from infringing competitors, eBay had essentially no effect—they continue to have the ability to use patents to defend their innovation and exclude infringers in most cases.

For NPEsNon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. that play a constructive role in our innovation ecosystem—research universities, companies that develop their own innovative technology to license to others, etc.—they, too, have seen minimal impact to their ability to access injunctions. With grant rates of about 80% or higher, these entities are able in most cases to protect their investments in R&D and their ability to control how their innovations are commercialized, equities that can go beyond mere pecuniary value. For example, a university or research lab can use patents to preserve the value of their R&D and engage in tech transfer with a startup or small business as an exclusive licensee to bring their invention to market, even after eBay.

For PAEs, entities that exist primarily to maximize monetary revenue via broad licensing of patents under the threat of litigation, it is understandably harder to establish that harm by another’s use of the patented technology cannot be compensated by money. And yet, even PAEs are successful at obtaining injunctions well over half the time. Clearly, exaggerated claims of eBay eliminating injunctions or causing patent values to plummet as a result have little basis in empirical fact. Just as those misleading claims should be disregarded, the policy prescriptions they are used to justify should be questioned as well.

As the August 2019 post on eBay noted in its closing, calibrating injunctions carefully is critical, especially in industries where products and services incorporate hundreds or even thousands of individual technologies, because injunctions are typically all-or-nothing. On the one hand, if a tablet manufacturer offers an innovative patented feature that grants its tablets a competitive advantage and forms a major part of the tablet’s market demand, an injunction may be appropriate to prevent a competitor from stealing that advantage via the sale of infringing competitor tablets.

On the other hand, if a PAEPatent Assertion Entity.  A narrower term for trolls that focuses on the core business model rather than whether the entity is actually making use of the patented technology ("working the patent"). seeking a royalty payout on a patent that contributes $0.10 of value in a minor component of a $1000 tablet is able to block the entire product from the market with an injunction, they have been gifted far more leverage than their technology warrants. A PAEPatent Assertion Entity.  A narrower term for trolls that focuses on the core business model rather than whether the entity is actually making use of the patented technology ("working the patent"). is not protecting its product or market share but rather is seeking to maximize licensing revenue (and blocking the product actually would produce no licensing revenue). Thus, the PAE’s injury is better addressed through a monetary award.

Based on the empirical evidence, it seems clear that current law under eBay is effectively drawing those distinctions. As the saying goes, there is no need to fix something that is not broken.

End Notes

1. E.g., Berkeley Ctr. for Law & Tech., How US Patent Enforcement Decline Elevated the UPC as the Global Forum for Injunctive Relief (Nov. 20, 2025), https://www.law.berkeley.edu/research/bclt/bclt-legal-analysis/upc-p1/; Press Release, Office of Sen. Chris Coons, Senator Coons, Colleagues Introduce Bipartisan, Bicameral Bill to Restore Injunctive Relief for Patent Infringement (Feb. 26, 2025), https://www.coons.senate.gov/?p=11326; Market Institute, The Market Institute Supports the Reintroduction of the RESTORE Patent Rights Act (Feb. 26, 2025), https://marketinstitute.org/the-market-institute-supports-the-reintroduction-of-the-restore-patent-rights-act/; Alliance of U.S. Startups & Inventors for Jobs, USIJ Applauds Re-Introduction of the RESTORE Patent Rights Act (Feb. 25, 2025), https://usij.org/usij-applauds-re-introduction-of-the-restore-patent-rights-act/; Council for Innovation Promotion, Council for Innovation Promotion Applauds Reintroduction of the RESTORE Patent Rights Act (Feb. 25, 2025), https://c4ip.org/council-for-innovation-promotion-applauds-reintroduction-of-the-restore-patent-rights-act/; Innovation Alliance, Innovation Alliance Applauds Reintroduction of RESTORE Patent Rights Act (Feb. 25, 2025), https://innovationalliance.net/from-the-alliance/innovation-alliance-applauds-reintroduction-of-restore-patent-rights-act/; Paul R. Michel, To Win the Tech Race Against China, Restore the Power of the U.S. Patent, Townhall (June 21, 2025), https://townhall.com/columnists/paul-michel/2025/06/21/to-win-the-tech-race-against-china-restore-the-power-of-the-us-patent-n2659176; Conservatives for Property Rights, Injunctions: Too Much and Too Little (Oct. 17, 2019), https://www.property-rts.org/post/injunctions-too-much-and-too-little; Gene Quinn, Seeds of demise were sown when SCOTUS removed exclusivity from the patent bargain (SepStandard-Essential Patent.  A patent that a participant in standards development process declares to be essential to the practice of the standard.. 19, 2018), https://ipwatchdog.com/2018/09/19/seeds-demise-scotus-removed-exclusivity-patent-bargain/; Richard A. Epstein, The Disintegration of Intellectual Property? A Classical Liberal Response to a Premature Obituary, 62 Stanford L. Rev. 455 (2010).
2. 35 U.S.C. § 283.
3. Id.
4. eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388, 393-94 (2006).
5. Id. at 391-92.
6. Supra note 1.
7. 19 U.S.C. § 1337; Colleen V. Chien & Mark A. Lemley, Patent HoldupAn attempt by a patent owner, by threatening litigation, disproportionate damages, or injunction/exclusion of a product, to extract payment from a producing company greater than the value of the patented technology., the ITCInternational Trade Commission, and the Public Interest, 98 Cornell L. Rev. 1 (2012), https://scholarship.law.cornell.edu/clr/vol98/iss1/1/.
8. Injunctions in Patent Law: Trans-Atlantic Dialogues on Flexibility and Tailoring (Jorge L. Contreras & Martin Husovec eds., 2022), https://doi.org/10.1017/9781108891103; Giuseppe Colangelo, Lost in Translation? Injunctions and Patent Enforcement in a Transatlantic Perspective, J. World Intell. Prop. (2026), https://onlinelibrary.wiley.com/doi/10.1111/jwip.70023; Vincent Angwenyi et al., Unified Patent Court Advantages Leave US Trailing Behind, Law360 (June 22, 2023), https://www.law360.com/amp/articles/1690339. 
9. See supra note 1.
10. Josh Landau, Much Ado About Injunctions, Patent Progress (Aug. 1, 2019), https://patentprogress.org/2019/08/much-ado-about-injunctions/.
11. Id.
12. Id.
13. Id.
14. For uncontested motions, courts did not simply rubber stamp them all because the movant has the burden of proof, instead applying eBay and still denying an injunction when the movant had failed to satisfy the four-factor test, albeit at an expectedly low rate given the lack of opposition (4% full denial, 2% partial).
15. Stanford NPENon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. Litigation Database, https://npe.law.stanford.edu/.
16. Shawn P. Miller et al., Who’s Suing Us? Decoding Patent Plaintiffs since 2000 with the Stanford NPENon-Practicing Entity.  A broad term associated with trolls but now disfavored because it includes universities and legitimate technology developers that seek to license technology in advance rather than after a producing company has independently developed it. Litigation Dataset, 21 Stanford Tech. L. Rev. 235, 244-46 (2018), https://law.stanford.edu/publications/whos-suing-us-decoding-patent-plaintiffs-since-2000-with-the-stanford-npe-litigation-dataset/.