As readers may already be aware, a bill pending in the U.S. House and Senate, titled the “Realizing Engineering, Science and Technology Opportunities by Restoring Exclusive (RESTORE) Patent Rights Act” would add the following provision to Patent Act § 283: “Rebuttable Presumption.—If, in a case under this title, the court enters a final judgment finding the infringement of a right secured by a patent, the patent owner shall be entitled to a rebuttable presumption that the court should grant a permanent injunction with respect to that infringing product.” By requiring the defendant to come forward with reasons why an injunction should not issue, the bill would reverse the burden of proof established in eBay Inc. v. MercExchange L.L.C., 547 U.S. 388 (2006), which states that the prevailing “plaintiff must demonstrate: (1) that it has suffered an irreparable injury; (2) that remedies available at law, such as monetary damages, are inadequate to compensate for that injury; (3) that, considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and (4) that the public interest would not be disserved by a permanent injunction” (emphasis added). Though most observers believe the bill has little chance of passing this term, it has received considerable attention as of late, including in this thoughtful blog post by Dennis Crouch, this Law360 article from last week (which quotes me), this article in Bloomberg Law, and the recent FedSoc webinar in which I also participated. Although my views on the matter are widely known, I thought it might be helpful to summarize them here.
First, there is little doubt that courts award prevailing patent owners permanent injunctive relief less frequently after eBay than they did before. The exact percentage varies from one year to another, and in some years is actually a high percentage (90% plus) of all cases in which the prevailing patentee requests a permanent injunction. But there is some self-selection going on; litigants know what the standards are, and if there is little chance of obtaining a permanent injunction you’re not going to request one. Practicing entities, particularly competitors of the infringer, still have a very good chance of obtaining a permanent injunction other than in cases involving FRAND-committed SEPs; patent assertion entities (PAEs) do not. Imposing a rebuttable presumption in favor of injunctive relief would, presumably, increase the number of injunctions to some degree, and accordingly enhance the bargaining power of patent owners (including PAEs) by enabling them to make a more credible threat of obtaining injunctive life; how much things would change, however, would depend on how courts interpret the change, in particular how difficult the case law makes it for defendants to rebut the presumption. That would remain to be seen.
Purely as a matter of economic theory, the changes wrought by eBay could be either good or bad. Protecting patent rights (or any sort of rights) through injunctive relief eliminates the need for a court to determine damages going forward, and encourages the parties to negotiate their own deal. That may be advantageous if we think that courts are likely to err in setting damages appropriately, and that the errors are likely to be biased in favor of defendants and material (in the sense of negatively impacting the patent incentive). Also, some but not all business models rely on exclusivity, as opposed to indiscriminate licensing. On the other hand, according courts the discretion to deny injunctive relief and award ongoing royalties reduces the risk of patent holdup. Patent holdup simply means that a patentee who can credibly threaten an injunction against an implementer can extract a royalty, ex post, that greatly exceeds the ex ante value of the invention over the next best available nonfringing alternative—a royalty that reflects some portion of the implementer’s nonredeployable sunk costs and its higher ex post switching costs. If and when patent holdup arises, it’s bad for consumers and it’s potentially bad for innovation as well, because it distorts the patent incentive by overvaluing the invention at issue. As a matter of theory, we should be particularly concerned about the risk of patent holdup in cases involving complex products, where even the unintentional infringement of a single patent in a complex device embodying hundreds or thousands of patents can threaten to shut down an entire product line. The poster child here is, of course, NTP v. Research in Motion, the Blackberry patent case that was decided shortly before eBay.
That’s theory. As an empirical matter, I’m aware of three peer-reviewed economic studies— Mezzanotti & Simcoe (2019), Mezzanotti (2021), and Helmers & Love (2024)—that report finding no negative impact on U.S. innovation resulting from the eBay decision, and (according to the last two of these papers) a marginally positive effect. Moreover, it seems inevitable that if eBay were overturned, PAEs would have enhanced leverage. Again, people may disagree over whether that would be good or bad, but there is empirical literature suggesting that PAE lawsuits have a negative impact on startups (see, e.g, Cohen, Gurun & Kominers (2019)). This evidence may not be dispositive; all studies have their limitations, as the authors recognize. But I have yet to see any serious effort on the part of eBay’s critics to engage with these empirical studies and show where they have gone wrong.
There are a couple of other points that are worth making. First, critics of eBay such as Adam Mossoff point out that, as a matter of history, in the nineteenth century U.S. courts almost always awarded permanent injunctive relief to the prevailing patent owner who sought such relief in a court or equity. Of course, the world is a very different place than it was in the nineteenth century; the framers and the generations immediately following them obviously had no concept of smartphones or connected cars or the various other complex technologies we rely upon today. But putting that aside, one thing to note is that up until the early twentieth century, it wasn’t altogether clear that U.S. courts had the authority to award reasonable royalties even for past infringement, let alone ongoing royalties for future infringement. (There were cases authorizing patentees to seek “established royalties” for the duration of the patent term, in the form of a lump-sum, but only if the patentee’s practice was to license all takers on similar terms, and such cases were rare. There also was precedent authorizing the federal government to take a nonexclusive license for a reasonable royalty. See discussion here at pages 1095-98 of my article on nominal damages.) So, if a court were to deny a patent owner an injunction in, say, 1850, there would have been nothing to stop the infringer from starting up again, and forcing the patent owner to file another lawsuit. In that sense, the future harm facing the patent owner was in some meaningful sense irreparable. Today, if we have the option of awarding ongoing royalties (and prejudgment interest and, in appropriate cases, enhanced damages and attorneys’ fees for willful infringement) then doctrinally, in many cases the future harm facing the patent owner is reparable through monetary damages. People may still disagree about how accurate those damages determinations are, but if there is any good evidence that U.S. patent damages are on average undercompensatory, I have yet to see it.
Second, in most other countries at present, injunctive relief remains the norm; even in other common law countries such as the U.K., where courts have discretion to deny injunctive relief, the burden of showing why that should happen rests with the defendant, and in most (though not all) cases these courts still grant injunctions. However, it is important to note that there also may be other rules and practices in place that appear to keep PAE lawsuits in check to some degree, including shifting some portion of attorneys' fees to the prevailing party; there also tends to be less extensive pretrial discovery, and there are no juries in patent infringement actions, both of which features add to the cost of litigating in our system. In any event, as I stated above, if we were to move away from the eBay standard, we should expect more lawsuits by PAEs and more demands for injunctive relief by owners of SEPs. Again, in the absence of evidence that these costs are necessary to ensure innovation, I think this would be a risky option for the U.S. to undertake at this time.