Guest Post: A Brief History of 35 U.S.C. § 284 — How Patent Damages Lost Their Way
A historical perspective on patent damages and the relationship between law, equity, and modern doctrine.
Editor’s Note
I’ve agreed to publish this guest post by Craig Edgar because it presents a provocative historical argument about patent damages under 35 U.S.C. § 284. The views expressed are the author’s and do not necessarily reflect my own.
The piece raises important questions about the relationship between traditional property-law damages principles and modern patent damages doctrine, and provides a masterful overview of the history of patent damages law more generally. I learned a great deal from reviewing it.
I’m publishing it to encourage discussion, even where readers may disagree. Some readers will agree with it; others won’t.
The full text (including endnotes) is below, and here’s a PDF
A Brief History of 35 U.S.C. § 284 — How Patent Damages Lost Their Way
Craig Edgar*
Much ink has been spilled on this Substack and LinkedIn regarding recent changes to post-grant procedure at the PTAB. Thomas Krause has called those changes “ridiculous” while his critics refer to the PTAB before the changes as the patent “death squad.” What if I told you that while PTAB procedure is important, it pales by comparison to another issue? What if I said eliminating weak patents through the PTAB is not the biggest concern? What if I said other concerns such as Section 101 patentability, patent thickets, and third-party funding, while important, are also mere sideshows compared to another issue? What if I said many patent cases that reached a verdict and awarded damages in approximately the past thirty years violate the governing patent damages statute – 35 U.S.C. § 284? That claim may sound implausible. But the roots of 35 U.S.C. § 284 lie in something that should be quite familiar to lawyers: the basic principles of property and trespass damages.
Remember when you were a 1L sitting in first semester property? The professor took out a bundle of sticks tied with a ribbon, and you learned about current property interests, like life estates, and future interests, like remainders. A property owner who owned the whole bundle could distribute some of his sticks (ownership interests in the land) to others, but when he held the entire bundle in fee simple, the property was his. An owner can exclude others and is entitled to sole possession. Of course, he must pay taxes or his property can be sold on the courthouse steps, but so long as he does, the government has no right to take it away without compensation.
The same can be said about a patent. It is a type of property: intellectual property. A patent has metes and bounds (claims) contained within so all can determine where it begins and ends. Anyone who infringes on the right of the patentee to exclude, is a trespasser. So long as the patentee pays taxes (maintenance fees), the property remains his until it expires.
Then, a few months into the class, your professor raises the issue of adverse possession. You mean Mr. Professor, that if an owner doesn’t use property and just lets it sit idle, a trespasser can come onto the land and use it? And, if he does so for a particular amount of time, say ten years, depending on the jurisdiction, the trespasser gets complete ownership of the land? This new information might take a while to accept, but eventually you grasp it. Makes sense – we want to reward economic use of land, and if a trespasser works that land for a sufficient amount of time, and the true owner doesn’t stop him, we want to award the trespasser full interest so he can sell it to a buyer, or pass it to his children upon his demise.
Next you discuss damages. If the length of time an adverse possessor must work the land, for example, is ten years, what if the true owner sits idly and allows the trespasser to clear the land, plant and tend a crop, and then, at harvest time, comes in and says: “This is my land, get off. I’ll harvest, sell the crop and you get nothing because you are a trespasser.” Well, that would be unfair; it’s unreasonable. Maybe the owner can indeed expel the trespasser, but when doing so he must compensate by giving the trespasser money to cover the time and effort he spent clearing and tilling, planting, and tending. Otherwise, the owner would benefit from the work of another. So indeed, if a true owner wants to expel the trespasser, he must pay for the increased value of his cleared land and the work associated with raising the crop. Most true owners don’t want to do that. So, most in that situation won’t eject a trespasser from the land at that stage (and there is no economic reason to do so because his work is increasing the land’s value). But what about the trespasser? He planted a crop on land he did not own. Well then, the owner is entitled to a reasonable rent for the use of the property. Now, that does not include profits from the harvest, but nevertheless, it is something because the trespasser wouldn’t have the harvest if he didn’t have access to the land. How could one determine reasonable rent? Well, it’s not easy because every piece of land is different. Different soils can be rocky, or covered with trees and brush, which require removal before applying the plow. But how to measure? Well, again it’s not easy, but perhaps we find other properties close to the one at issue, with the same sort of soil and same sort of brush, and look at the amount paid by that lessee to the owner. Or perhaps we take the purchase price of the land and come up with a lease amount based upon that. Might not be exact, might have to raise or lower the lease amount charged by a farmer two miles away because his land is closer to the river or has less limestone. However, a jury of farmers could likely look at evidence and, hopefully, find something “reasonable.” Regardless, the value of the untilled land should not reflect the profit made from selling the crop.
Okay you say, but what if the true owner is also a farmer? What if he grows corn, and his trespasser does too? Doesn’t that trespass cause the owner market harm? The true owner would have sold his corn at a higher price, if it hadn’t been for the trespasser. Well, okay, in that case, the owner gets some of the profit the trespasser made from his harvest to compensate for the market harm that resulted from the trespass. Well, how do we decide that? What if the trespasser raised alfalfa while the owner raised oats? Does that change things? Both may be used to feed dairy cows. What if the alfalfa raised by the trespasser diminished the oats price? This can become a very complex problem quickly. The owner does not have a legal right to a portion of the profit made from selling the crop, but allowing the trespasser to keep all the profit and only pay the value of the untilled land in this situation would be “unjust.” Perhaps foundational trespass legal damages (e.g. physical damage to property, loss of use, loss of market value) in some cases don’t fully compensate an owner so a court sitting in equity should award an equitable share of the trespasser’s profits to the owner to fully compensate for the harm caused by the trespass.
Property law and trespass damages theory was in the forefront of the minds of the Founders, and of Congress during most of our patent history. Early, patentees tried to bring an infringer’s profits into damages calculations. They were rarely successful.[1] It is understandable; from a purely emotional standpoint, shouldn’t a patent owner get some of the profit the infringer makes from use of the patent? Well, no because that profit has no bearing on the value of the unimproved patent, Just as the value of a crop depends upon the sweat of the farmer clearing the land and tending the crop, the value of an unpracticed patent that covers only a portion of a manufactured product is not reflected in the profit made from selling the product. The goal of trespass damages is to put the owner in the same position as he would have been absent the trespass. For an owner of untilled farmland, damages come solely from unimproved land. For most unpracticed patents, damages likewise come solely from the unimproved patent.
Initially, Congress limited damages for infringement solely to legal damages.[2] and, in the Patent Act of 1793, limited them to “established” license evidence.[3] If a patentee could show he licensed the patent, which is analogous to the cost of leasing farmland, the patentee would be entitled to the value of his license. That was often difficult to prove because unless the patentee had a license that covered the same patent that was infringed, it was difficult to get evidence that could be given to a jury. The Patent Act of 1800 expanded recoverable damages, to “actual damages.”[4] However, equitable remedies remained unavailable. Thus, in most cases, even after the Patent Act of 1800, a patentee without a license only received nominal damages because he had the burden of proof.[5] Furthermore, cases at law only allowed retrospective relief. Any patentee wishing to be compensated for ongoing infringement was forced to file multiple lawsuits.[6]
Congress changed the law in 1819 to enable equity courts discretion to issue injunctions.[7] The Patent Act of 1836 eliminated mandatory trebling, first instituted in the Patent Act of 1793, and made it discretionary.[8] Seymour v. McCormick was decided under the Patent Act of 1836.[9] Its dicta is worthy of quoting at length because it does a good job explaining recovery under the Patent Act with an understanding of foundational trespass:
It must be apparent to the most superficial observer of the immense variety of patents issued every day that there cannot, in the nature of things, be any one rule of damages which will equally apply to all cases. The mode of ascertaining actual damages must necessarily depend on the peculiar nature of the monopoly granted. A man who invents or discovers a new composition of matter, such as vulcanized India rubber or a valuable medicine, may find his profit to consist in a close monopoly, forbidding anyone to compete with him in the market, the patentee being himself able to supply the whole demand at his own price. If he should grant licenses to all who might desire to manufacture his composition, mutual competition might destroy the value of each license. This may be the case also where the patentee is the inventor of an entire new machine. If any person could use the invention or discovery by paying what a jury might suppose to be the fair value of a license, it is plain that competition would destroy the whole value of the monopoly. In such cases, the profit of the infringer may be the only criterion of the actual damage of the patentee. But one who invents some improvement in the machinery of a mill could not claim that the profits of the whole mill should be the measure of damages for the use of his improvement. And where the profit of the patentee consists neither in the exclusive use of the thing invented or discovered nor in the monopoly of making it for others to use, it is evident that this rule could not apply. The case of Stimpson’s patent [an improvement in making rail turnouts that enabled better switching. Stimpson was an NPE and sued multiple railroads for violating his patent][10] for a turn-out in a railroad may be cited as an example. It was the interest of the patentee that all railroads should use his invention, provided they paid him the price of his license. He could not make his profit by selling it as a complete and separate machine. An infringer of such a patent could not be liable to damages to the amount of the profits of his railroad, nor could the actual damages to the patentee be measured by any known ratio of the profits on the road.[11]
To summarize, for a patentee who benefits from holding his monopoly close, “[i]f any person could use the invention or discovery by paying what a jury might suppose to be the fair value of a license, it is plain that competition would destroy the whole value of the monopoly.” The McCormick Court therefore recognized that different types of patentees suffer different types of harm from infringement. A patentee who competes with an infringer may lose sales or market share. In that situation, allowing consideration of profits makes sense, just as a corn farmer may be harmed by a trespasser who also grows corn. Such an owner is not fully compensated by basing damages on the mere value of an unimproved patent (untilled value of land). However, a patentee who wishes “that all [manufacturers] should use his invention” cannot measure actual damages “by any known ratio of the profits.” Instead, these patentees are only entitled to “what a jury might suppose to be the fair value of a license.”
The Patent Act of 1870 was a major change. First, it allowed all patent actions to be “originally cognizable, as well in equity as at law, by the circuit courts of the United States . . . . .”[12] Secondly, and importantly for our purpose here, it distinctly separated the monetary remedies available at law and in equity. For a case at law, “the court may enter judgment thereon for any sum above the amount found by the verdict as the actual damages sustained, according to the circumstances of the case, not exceeding three times the amount of such verdict.”[13] For bills filed in equity, “the claimant [complainant] shall be entitled to recover, in addition to the profits to be accounted for by the defendant, the damages the complainant has sustained thereby, and the court shall assess the same or cause the same to be assessed under its direction,[14] and the court shall have the same powers to increase the same in its discretion that are given by this act to increase the damages found by verdicts in actions upon the case.”
Now we turn to 35 U.S.C. § 284 as passed in 1952. The current statute begins with: “Upon finding for the claimant the court shall award the claimant damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer.” Continuing to the second paragraph: “When the damages are not found by a jury, the court shall assess them.” The following sentence states: “In either event the court may increase the damages up to three times the amount found or assessed.[15] This language mirrors the Patent Act of 1870. A jury is responsible for “finding” damages (“found by verdict” in the Patent Act of 1870); a court sitting in equity “assess[es] damages.” In 1952, just as in 1870, Congress understood juries are limited to legal damages. If the claimant in entitled to profits in order to “adequately compensate” for the infringement, a court “shall assess” damages. Thus, the 1952 Patent Act continues the same processes used under the Patent Act of 1870 to determine damages. Equity is allowed, but only for instances in which the patentee is not adequately compensated by legal damages.[16]
For at least a few decades after 1952, we followed the statute. However, as time passed, many factors combined to cause courts to conflate the two systems. These factors include: the merger of equity and law which for patents began with the Patent Act of 1870, the adoption of the Federal Rules of Civil Procedure in 1938, which broadly expanded discovery in patent cases,[17] the decline of patent infringement actions filed at law versus equity that began in the Nineteenth Century,[18] a mistaken belief that equitable accountings overcompensated patentees,[19] and an increasing belief that all patentees were entitled to an injunction, which was ultimately corrected by the Supreme Court in eBay.[20]
Georgia Pacific, decided in 1970, expanded this confusion.[21] This seminal patent damages case was filed against USP, a competitor with Georgia Pacific. The court, citing 35 U.S.C. § 284, noted a special master had been appointed after validity and infringement were established “to determine the amount of damages to be awarded to USP under 35 U.S.C. § 284 (1952), which provides for ‘damages adequate to compensate for the infringement.’ The master, computing damages upon the basis of GP’s profits derived from the sale of the infringing article, awarded $685,837.00 to USP.” The trial court judge disagreed with the special master via an “exception.” Evidence relevant to damages was heard by the judge and the award was reduced. In essence, the judge sitting in equity apportioned the infringer’s profit to the famous GP factors. Rather than describing this process as equitable profit apportionment, the court framed this analysis as determining a “reasonable royalty.”[22] Over time, courts began applying the GP framework in nearly every patent case, which allowed evidence regarding the revenues and profits associated with the accused products. What had historically been equitable profit apportionment became the dominant method for calculating legal damages.[23]
The result is that a large portion of modern patent damages awards rely on methods that cannot be reconciled with an historical understanding of 35 U.S.C. § 284. This is particularly evident in litigation involving Patent Assertion Entities (PAEs) whose business model is buying unpracticed patents and then asserting them against multiple manufacturers. Confusion regarding the history behind 35 U.S.C. § 284 was a primary cause of the rise of PAEs, which began in earnest during the early 1990s. Discovery of profits are allowed in virtually all patent cases today and juries look at an infringer’s profits when an expert report uses the GP factors. Even when looking at licenses, in many instances those licenses were negotiated between competitors or, if not, are settlement agreements or negotiated agreements that arose either under threat of an injunction (which during most of our history were unavailable without showing entitlement to equitable jurisdiction), or under threat of injunction-type damages (consideration of infringer’s profits). Many of the current problems the Federal Circuit has with monetary damages arise from failing to apply foundational trespass methodology.[24]
Now, would the Supreme Court decide to enforce 35 U.S.C. § 284 as written if given an opportunity? That is an open question. Modern certiorari petitions typically filed against PAEs argue the jury misapplied damage methodologies rather than questioning whether those methodologies themselves are consistent with the statute’s historical structure.[25] After decades of applying the GP factors in the manner we have recently, returning to the earlier distinction between legal and equitable damages would be a major policy shift. However, it is also true that the Court hasn’t decided a pure patent damages case since 1964.[26] Further, the current patent system – marked by recurring disputes over apportionment and appellate reversals of large awards – has significant problems.[27] If the Court decides our practice during the last thirty years should continue, it would have to ignore not only historical trespass damages, but also hold that every previous Congressional patent statute and over 200 years of Supreme Court precedent should be forgotten. It will be indeed interesting to see how the Court approaches the problem if given a chance.
* Craig Edgar is a staff attorney of the Intellectual Property Litigation Division at the Kansas City, Missouri office of the law firm of Shook, Hardy & Bacon L.L.P. Mr. Edgar earned a B.S. degree in Chemistry from the University of Missouri-Kansas City in 1986, a J.D. from Washburn University Law School in 1995, and an LL.M. in Environmental & Natural Resources Law from the University of Utah in 1998. The opinions expressed are solely the author’s. Nothing herein reflects the views of the firm of Shook, Hardy and Bacon, L.L.P., or its clients.
[1] See Seymour v. McCormick, 57 U.S. (15 How.) 480, 489-91 (1853). See also City of New York v. Ransom, 64 U.S. (23 How.) 487, 490-91 (1859).
[2] “Congress deliberately chose to exclude equitable jurisdiction and remedies from early patent disputes. Early patent acts, for example, did not confer equitable jurisdiction or authorize injunctions.” Christopher S. Storm, A Series of Historical Accidents, 31 Mich. Tech. L. Rev. 73, 77 (2024). The First Congress specifically considered, and rejected, allowing patentees to file bills in equity under the Patent Act of 1790. Id. See also James Ryan, A Short History of Patent Remedies, 6 CYBARIS: An Intell Prop. L. Rev. 150, 158 n.44 (2015).
[3] See Patent Act of 1793, Ch. 7, 1 Stat. 318-323 (Feb. 21, 1793) (“a sum, that shall be at least equal to three times the price, for which the patentee has usually sold or licensed to other persons, the use of the said invention; which may be recovered in an action on the case founded on this act”) (emphasis added). An “action on the case” is a case tried at law to a jury. Note too that trebling in cases under the Patent Act of 1793 was mandatory.
[4] “[T]hree times the actual damages sustained by the patentee.” Act of Apr. 17, 1800, ch. 25, 2 Stat. 37, 38, 3 (emphasis added).
[5] This was during the era of precision. “Precision was the controlling concern in the 19th century,” not reasonableness. Ryan, supra note 2, at 196. Absent precision, patentees got nominal. Id. at 165. Reasonableness in legal damages did not arrive, at least in earnest, until the early Twentieth Century. Id. at 194. It is important here to note that equity did not require precision to be awarded. Id. at 192.
[6] Craig Edgar, Struggling Against Entropy: Monetary Patent Infringement Damages After Ebay, 66 Drake L. Rev. 45, 55 (2018).
[7] See Christopher S. Storm, The Patent Industry Versus eBay, 130 Penn. St. L. Rev. 73, 105 (2025) (“courts ‘upon any bill in equity . . . shall have authority to grant injunctions, according to the course and principles of courts of equity’”) (quoting Patent Act of 1819, ch. 19, 3 Stat. 481) (emphasis original to Storm). This language is virtually unchanged today. See 35 U.S.C. §283.
Critics of eBay Inc., et al. v. MercExchange, L.L.C., 547 U.S. 388 (2006) argue this Supreme Court’s decision misinterpreted both this language and historical precedent, arguing most if not all patentees should receive injunctions upon proof of validity and infringement. For example, see Dennis Crouch, Patently-O, Back to 1789: How Founding-Era Equity Could Resurrect NPE Injunctions, July 7, 2025 (“eBay itself made historical errors that make it difficult to know whether the court intended to apply a straight historical analysis, or one that is more fictionalized”); Presumption of Injunction: How the RESTORE Act Aims to Re-Empower Patent Holders, July 31, 2024 (“[f]or the vast majority of American history, a judgment of patent infringement (by a court sitting in equity) led almost directly to injunctive relief barring ongoing infringement”). See also Adam Mossoff, The Injunction Function: How and Why Courts Secure Property Rights in Patents, 96 Notre Dame L. Rev. 1581 (2021). Mostly, these critics both get the history wrong and misapply Cont’l Paper Bag Co. v. E. Paper Bag Co., 210 U.S. 405 (1908). See Storm, supra at 113-20. Both patentee and infringer in Cont’l Paper Bag Co. were competitors. See Edgar, supra note 6, at 94 n.290. “This type of patent holder benefits from holding his monopoly close, even if he never uses or licenses the patent, because he stops his competitors from using the technology.” Id. Such a patentee is entitled to an injunction under both foundational equity and English Chancery rules.
This belief eBay is incorrect comes from multiple places, but probably the most significant is that modern practitioners deny Congressional intent to separate patent practice in the United States from English practice. See Dennis Crouch, Patently-O, Intitled to Tie Him Up: Can 18th-Century Chancery Practice Restore Patent Injunctions?, March 2, 2026. Professor Crouch believes modern practice should follow the practice of English Chancery when determining equitable relief in patent cases. However, the First Congress was aware of English Chancery practice and wanted no part of it. See supra, note 2. Instead, it wanted to follow “traditional” or “established” equity. See Storm supra at 106-08. Traditional equity never allows equity to be utilized when there is an “adequate remedy at law.” Traditional trespass common law and equity is summarized and outlined by the first few paragraphs of this post.
Much of the Court’s discussion in Cont’l Paper Bag Co., is dicta. See Storm, supra, at 117. Modern readers should appreciate Cont’l Paper Bag Co., was decided in 1908, a period where the public, having recently lived through the Robber Baron Era, was not enamored with monopolies of any sort. The Court might have believed awarding an injunction to a Non-Practicing Entity (NPE) would not be particularly well-received by an early Twentieth Century public, necessitating what Storm calls “dictum [that] helped transition patent industry talking points into the modern age.” See Storm, supra at 117. eBay critics misapply this dicta to NPEs who are not competitors. See Crouch supra, Patently-O, Back to 1789, How Founding-Era Equity Could Resurrect NPE Injunctions (expanding the Court’s holding that “the remedy of injunction to prevent infringement of his patent will not be denied merely on the ground of nonuser of the invention” to the eBay Court’s “positive cit[ation]” of Cont’l Paper Bag Co. “[I]n its statement that the district court’s categorical denial of injunctive relief was improper. . . the eBay Court did not contend with the new tension it implicitly created”) (emphasis added). There is no “tension” between the actual holding of Cont’l Paper Bag Co. and current practice under eBay. Edgar, supra note 6, at 94 n.290.
[8] Edgar, supra note 6, at 56.
[9] Id. at 57.
[10] See Stimpson v. W. Chester R.R. Co., 45 U.S. (4 How.) 380 (1846).
[11] McCormick, 57 U.S. (15 How.) at 489.
[12] Patent Act of 1870, 16 Stat. 198-217, § 55. It is true the Judiciary Act of 1789 allowed federal circuit courts [trial courts] to hear both cases at law and bills in equity. However, when doing so the judge wore a different “hat.” So, patentees wishing to recover both legal and equitable remedies, had to file both in law and equity. The Patent Act of 1870 eliminated this requirement. See Ryan, supra note 2, at 189-90.
[13] Patent Act of 1870, 16 Stat. 198-217, § 59 (emphasis added).
[14] Patent Act of 1870, 16 Stat. 198-217, § 55 (emphasis added). Three issues arise here requiring discussion, even in a shortened blog post, which necessarily oversimplifies, rather than a law review article. First, the term “profit” in the Patent Act of 1870 is sometimes misunderstood to include lost profit cases. Lost profits are legal damages and were historically proved to a jury. See Ryan, supra note 2, at 174. The need to prove legal damages to a jury is the reason behind the “strict standards of proof” required to receive an award of lost profits. See Mark A. Lemley, Distinguishing Lost Profits from Reasonable Royalties, 51 Wm. & Mary L. Rev. 655, 661 (2009). In essence, any plaintiff bringing a common law trespass action must show proximate causation to gain an award from a jury. Unless a patentee can show he would have made sales, “but for” the infringement, he cannot qualify for lost profits. The confusion regarding a jury considering “lost profits” and consideration of an infringer’s profits in equity is a prime cause of what Lemley describes as “the growing practice of patentees opting for a distorted measure of royalties [reasonable royalites] that is greater than the profits they actually lost.” Id. at 673. In the old days, a patentee brought a lost profits case to obtain a jury trial and avoid the length and expense of an equitable accounting. Now, there is little benefit to following the “strict standards of proof” in lost profits and patentees just “opt” for reasonable royalty since today they get a jury trial in any event.
Secondly, the “in addition” language is often misunderstood as allowing a patentee in equity to “double dip.” That is, to get both “profits” and “damages.” See Kori Corp. v. Wilco Marsh Buggies & Draglines, 761 F. 2d 649, 654 (Fed. Cir. 1985) (“[p]rior to 1946, the patent laws provided for recovery of both the patentee’s damages and the infringer’s profits”). This is incorrect. The prime directive of equitable jurisdiction is that it is available only if legal recovery is inadequate. See Root v. Lake Shore & M.S. Ry. Co., 105 U.S. (15 Otto) 189, 207 (1881). Likewise, if equitable remedies fully compensate patentee, no legal damages are awarded. See Edgar, supra note 6, at 66 n.102. The “in addition” language only allows courts to use legal remedies, equitable remedies, or a combination of both in order to “adequately compensate” a patentee for infringement. See Ryan, supra note 2, at 190 n.241.
Thirdly, patent historians often argue equitable monetary remedies are unavailable today. Storm argues the Patent Act of 1946 eliminated them. See Storm, supra note 2, at 96. Ryan disagrees, arguing “whether this was Congress’s intent is debatable.” Ryan, supra note 2, at 191. Regardless, both argue that they were eliminated either by statute, case law, or both. Even the Supreme Court believes the Patent Act of 1946 and Aro eliminated accountings. See SCA Hygiene Products Aktiebolag v. First Quality Baby Products, 137 S.Ct. 954, 964-65 (2017). In dictum, it states, “[t]he equitable remedy of an accounting, however, was not the same as damages. The remedy of damages seeks to compensate the victim for its loss, whereas the remedy of an accounting, which Congress abolished in the patent context in 1946, sought disgorgement of ill-gotten profits.” Id. (citing Aro Manufacturing Co. v. Convertible Top Replacement Co., 377 U.S. 476, 505 (1964) (emphasis added)). But see Edgar, supra note 6, at 86-87 n.234. If Congress eliminated accountings in 1946, the Supreme Court in 1964 would not have been concerned its decision might result in “lengthy proceedings before a master” upon remand. Aro Mfg., 377 U. S. at 502 n.18. The “profit disgorgement” language in Aro flows from the mistaken belief that the Patent Act of 1870 allowed recovery of equitable monetary remedies “in addition” to legal, thus overcompensating patentees. Id. at 505. Believing equitable accountings overcompensate patentees is based on this misreading of the Patent Act of 1870 and misreading Westinghouse Elec. & Mfg. Co. v. Wagner Elec. & Mfg. Co., 225 U.S. 604 (1912). See infra, note 19.
[15] 35 U.S.C. § 284 (emphasis added). The “in either event” language is a reference to the historical mandatory trebling at law that began in the Patent Act of 1793 and became discretionary with the Patent Act of 1836.
[16] Equitable accountings had many flaws. The biggest problem was that they took a very long time to conduct. See Ryan, supra note 2, at 190.
[17] The Judiciary Act of 1789 allowed for some minimal discovery in legal matters. However, early discovery was very unlike modern discovery. Instead of exchanging information pre-trial, a defendant was required only to bring his papers to court just before trial, allowing opposing counsel just a few moments to view before trial began. See Carpenter v. Winn, 221 U.S. 533 passim (1911). Equity courts were able to order equitable discovery under some circumstances and then review a defendant’s books and depose witnesses pre-trial. However, early, that type of discovery was only allowed when the defendant had a fiduciary relationship with the plaintiff. See Joel Eichengrun, Remedying the Remedy of Accounting, 60 Indiana L.J. 463, 468 (1985). When reading Nineteenth Century cases today, we often see language that an infringer is holding profits, as a “trustee” of a patentee. This language was invoked because courts found themselves ordering accountings to assess damages. Edgar, supra note 6, at 70-71 n.132.
Minimal discovery suffices in lost profit cases because most of the information needed is already in the patentee’s possession. For instance, McCormick already knew the profit he made from selling each reaper. He only needed to know the number of infringing reapers Seymour sold (300) to complete his lost profit damage theory. See McCormick, 57 U.S. (15 How.) at 487.
[18] See Christopher Beauchamp, The First Patent Litigation Explosion, 125 Yale L.J. 848, 913-14 (2016).
[19] This arose primarily from misreading Westinghouse Elec. & Mfg. Co [Westinghouse]. Ryan opines that Westinghouse fails to follow McCormick’s dicta and exemplifies a “nightmare scenario of a patentee gaining entire profits of a business for small improvements” in equity. Ryan, supra note 2, at 177. Storm notes that “multiple House witness” during debate over the bill that ultimately became the Patent Act of 1946 criticized Westinghouse for overly compensating patentees in equity. Storm, supra note 2, at 89. Actually, Westinghouse doesn’t award a patentee 100% of an infringer’s profit in an equity case. Instead, it merely rules the burden of proof on damages should not solely lie on the patentee. It found that the plaintiff argued that the “non-infringing changes were not, in fact, improvements but instead, “added nothing to the profits, but, on the contrary, had [made the product worse].” Westinghouse, 225 U.S. at 616-17.
Westinghouse notes the master agreed with the patentee and awarded 100% of the profit. Id. at 622. However, the trial judge disagreed, finding the changes were improvements, and the burden was entirely on the patentee to apportion. Id. The patentee could not do so and thus received nominal damages. Id. The Westinghouse Court remanded the case to the trial court to apportion with the burden shared. “Neither the court nor the master discussed the question of apportionment, and the record does not afford satisfactory data for entering a final decree. This no doubt arises from the fact that both parties relied so entirely upon their theory that the burden was on the other that facts were not proved which might otherwise have been established.” Id. Even at best, the infringer’s changes were likely so miniscule that they didn’t add much to the value of the infringing product. Further, the infringer’s comingling, was likely an intentional attempt to make apportionment impossible for the plaintiff. (Infringer “inextricably commingled and confused the parts composing it”). Id. at 618. In a choice between either awarding nominal damages or full profit to the patentee “[t]he loss had to fall on the innocent or the guilty. In such an alternative, the law places the loss on the wrongdoer.” Id. at 619. Thus, Westinghouse follows McCormick’s dicta which provides a patentee who “is the inventor of an entire new machine” can be awarded full profits because he can “supply the whole demand at his own price.” McCormick, 57 U.S. (15 How.) at 489. Further, it would be inequitable to award nominal damages to a patentee victim of a bad actor infringer.
[20] See Storm, supra note 7, at 117.
[21] Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y. 1970).
[22] Id. at 1117-18. Courts sitting in equity never had to be precise with apportionment. See supra, note 5. Thus, what the court did was find a “reasonable” award rather than precise one. However, the mistaken belief that precision was required in both fora led to the rise of reasonable royalty. See infra note 23.
[23] To be fair to the GP court, it was following language from Dowagiac Mfg. Co. v. Minnesota Moline Plow Co., 235 U.S. 641 (1915). Dowagiac was an equity case. An injunction issued; an equitable accounting ordered. The master awarded nominal damages because the patentee had not met his burden of proof for apportionment. Id. at 643-44. The Supreme Court reversed, holding that although damages were not “substantial,” it was inequitable to award nominal. Id. at 647. It remanded and ordered the master to try and find a “reasonable royalty” because “mathematical exactness” is not required to establish damages when conducting an equitable accounting. Instead, a “reasonable approximation” will suffice. Id. at 647. The Court carefully noted a “reasonable approximation” at equity was not quite the same is in law. It distinguished its holding from Coupe v. Royer, 155 U. S. 565 (1895). “In that case – an action at law – there was no proof of what would have been a reasonable royalty, but only of what the defendant had made or might have made out of the infringement, and all that the court held was (a) that the damages were not to be measured by what the defendant had gained or might have gained, but by what the plaintiff had lost, and (b) that, as the evidence disclosed “no license fee, no impairment of the plaintiff’s market, in short, no damages of any kind,” the verdict could not exceed a nominal sum.” Dowagiac Mfg. Co., 235 U.S. at 649.
Note too Dowagiac Mfg. Co., favorably cites Hunt Bros. Fruit-Packing Co. v. Cassiday, 64 F. 585 (9th Cir. 1894). In Hunt Bros., a case at law, the patentee attempted to show an established “royalty” by offering the amount of profit he made from each sale of his machines. The trial court noted that the patent had two claims, and only one at been infringed. So, it reasoned the sale price of the “entire machine” did not apportion between the two claims and awarded nominal damages. The Ninth Circuit reversed, finding that in cases like this “there are necessarily no data from which the value of the royalty can be calculated with mathematical certainty.” Id. at 587. “The plaintiff was clearly entitled to damages for the infringement. If there had been an established royalty, the jury could have taken that sum as the measure of damages. In the absence of such royalty, and in the absence of proof of lost sales or injury by the competition, the only measure of damages was such sum as, under all circumstances, would have been a reasonable royalty for the defendant to have paid. This amount it was the province of the jury to determine. Id (emphasis added).
Thus, Dowagiac Mfg. states simply that mathematical exactness is not required either at law or in equity. An award under both just needs to be “reasonable.” Neither system however, allows “actual damages” to be based on an infringer’s profits “in the absence of . . . injury by competition” unless there is a license. See Dowagiac Mfg. Co., 235 U.S. at 649. Hence, neither law nor equity would allow a PAE to base damages upon an infringer’s profits, unless the patentee invents an “entire[ly]’ new machine.” See McCormick, 57 U.S. (15 How.) at 489. The fact that Hunt Bros. significantly expands the definition of established royalty over the elements outlined by Rude v. Westcott, 130 U.S. 152, 165 (1889) (for those elements see Ryan, supra note 2, at 191) is a feature, not a bug. Courts in the Twentieth Century tried to make things easier for patentees to prove damages at law to minimize the need for equitable accountings. However, that effort certainly did not extend to considering what the infringer “made or might have made” as a criterion of damages in a non-competitor case.
This effort to move away from “mathematical certainty” in both law and equity was the driver behind the rise of reasonable royalty. In the modern era, we read statements in the legislative history of the Patent Act of 1946 describing the bill as “making it unnecessary to have proceedings before masters and eliminating the element of profits except as an element of general damages,” as eliminating accountings (emphasis added). However, the legislative witnesses were not referring to eliminating accountings, but reducing the need to have one by making them “unnecessary” in many cases, and, if ordered, making them easier to conduct by adding more court oversight. See Edgar, supra note 6, at 81-82. Additionally, Hunt Bros. shows courts were trying to find ways to make it easier for patentees to bring cases at law by minimizing the strict rules of lost profits and established royalty. Both patentees in Hunt Bros. and Dowagiac Mfg. Co. were in competition with infringers. In the last thirty years or so, we have completely lost the entire plot of patent damages because we do the equivalent of an equitable accounting (base reasonable royalty in cases where the patent is only a small part of a product on the sales price and profits gained by the infringing product) even when the patentee is a PAE. See Edgar, supra note 6, at 91. Further, we do this “at law” by allowing juries to hear this evidence. Id. To further compound our error, instead of making it easier, we made it even harder to bring a lost profit case. See supra, note 14.
[24] For example, see Dennis Crouch, Patently-O, The Remedies Remedy is Almost Complete: EcoFactor v. Google, May 21, 2025. Professor Crouch calls the “whole setup and issue” regarding the rejection of the license by the en banc Federal Circuit Court “pretty ridiculous.” I agree partially with Professor Crouch when he writes the “undermining of injunctive relief in eBay” was the cause of the “tightening of requirements for proving monetary damages.” However, by using the term “undermining,” he implies eBay was wrongly decided.
Another example is Uniloc USA, Inc., v. Microsoft, Inc., 632 F.3d 1292, 1319-20 (Fed. Cir. 2011). This PAE case eliminated the 25% Rule of Thumb using the Entire Market Value Rule (EMVR) as justification. Id. The 25% Rule of Thumb was used to allocate profits in patent infringement lawsuits for 50 years. Robert Goldscheider , The Classic 25% Rule and the Art of Intellectual Property Licensing, 2011 Duke L. & Tech Rev. 115, 118.
The true culprit is that awarding monetary damages reflecting a right to an injunction when the patentee is unentitled to an injunction is, inherently, unreasonable. As long as federal courts continue to apportion profits in matters where the patentee gets no benefit from holding his monopoly close, monetary damages will remain hopelessly muddled.
[25] See e.g., Apple Inc. v. VirnetX Inc., et al., No. 19-832, Petition for Writ of Certiorari, U.S. 2019.
[26] See Aro Mfg. Co., 377 U.S. 476 (1964).
[27] See e.g., Mark A. Lemley, Ignoring Patents, 2008 Mich St L Rev. 19, 21-22. Instead of “promot[ing] the Progress of Science and useful Arts,” for quite some time technology companies have instructed their engineers to not even look at patents for fear of being accused of willful infringement. See id. Furthermore, the Federal Circuit continues to struggle to find “inventive” ways to keep large patent awards from “surviving.” A big factor regarding whether a verdict “survives” appellate review is its size. Bowman Heiden, How Large Patent Damages Awards Actually Play Out, Law360, Dec. 3, 2025. As Heiden puts it “the bigger the number, the shakier the outcome.” We’re reaching a stage where, much like in the old days when we did accountings, justice delayed is justice denied. See Ryan, supra note 2, at 190. Switching back to doing accountings again, with the understanding that courts will no longer merely transfer everything over to a master, but retain control, might well be better than our current method.
Editor’s Note (closing)
I’d be interested to hear how others think about the relationship between § 284 and modern damages practice—particularly whether the historical framing here resonates.