Innovation or Intellectual Scam: Patent Damages and Their Exploited History

With the U.S. Court of Appeals for the Federal Circuit’s revocation of the decision in the six-year-long case Caltech v. Apple and Broadcom (2022), Apple has been relieved of paying $1.1 billion in damage-based compensation demanded by Caltech, due to alleged infringement on the ‘710 and ‘032 patents held by the latter institution. [1] Given this was the highest patent-related damage amount demanded to date, there is perhaps no better time to highlight the financially exorbitant and highly exploitable nature of the patent law industry. [2] Given that universities attain about three thousand patents a year from either funding research or venture capital investment funds, this issue is only growing in prominence. [3] The denial of Caltech’s “two-tiered damage” theory garners support for a more careful revision of current patent law, specifically its overly-broad interpretation, that maintains its ability to promote innovation and better integrate technology.

Caltech v. Apple and Broadcom (2022) was originally filed by the plaintiff, Caltech, as a patent infringement case in May 2016. Broadcom chips, which are included in millions of Apple products, as they are essential to Caltech’s data transmission technologies—filed as ‘710 and ‘032 patents by the United States Patent and Trademark Office. [4] The case was first concluded in the U.S. District Court for the Central District of California, where a jury ordered Apple to pay Caltech approximately 837 million dollars and Broadcom to pay approximately 270 million dollars—a collective amount of approximately one billion dollars, one of the largest amounts ever awarded in patent-related damages. [5] This number was calculated according to Caltech’s proposed “two-tier damage” theory. [6] Essentially, Caltech accounted for both the temporal (year-aggregated) and physical (distributor and manufacturer) damages accrued in calculating how much it would have profited if Broadcom and Apple would have purchased the rights to the patents from the institution in December 2009. Caltech proposed a formula to account for the physical damages which combined profits made by Apple (the distributor) and Broadcom (the manufacturer), while the temporal damages were accounted for by calculating how much that profit would have multiplied itself in the corresponding years, adjusting for inflation. The Court of Appeals for the Federal Circuit, however, ruled this two-tier damage theory as “legally unsupportable,” stating that “Caltech’s damages model impermissibly applied two separate hypothetical negotiations for Broadcom and Apple for sales of the same chips.” [7] This lack of precedence, however,  does not yet spell out doom for Caltech’s demands: patent laws are infamously nebulous and open to different compensation models..

The grander discussion about patent-infringement cases has a quite simplistic origin: 35 U.S. Code § 284, otherwise known as The Patent Act of 1970. With regards to damage-based compensation, the law states that the amount awarded should be “adequate to compensate for the infringement, but in no event less than a reasonable royalty.” [9] According to this language, patent-related damages are compensatory, rather than punitive or prohibitive. In standard patent-infringement cases, reasonable royalties and lost profits are the two most common types of damages awarded; the former is defined as “the fair market value of a license” paid to the patent owner while the latter is “any money that they would have made if not for the infringement”. [10]

Two landmark cases that have been useful in clarifying the application of the language discussed above are Seymour v. McCormick (1854) and Panduit Corp. v. Stahlin Bros. Fibre Works (1978). In Seymour, the plaintiff, McCormick, held three patents which all related to a reaping machine. Through the progression of the case, McCormick amended his case and demanded that damage calculations be inclusive of the speculative profits that he would have received from his later improvements (which included parts of Seymour’s products that were not ruled as violating the original patent). The court ultimately rejected this request. The specific language used by the court concerned actualities: "[a]ctual damages must be actually proved, and cannot be assumed as a legal inference….” [11] One inevitable concern of this language is how it fits with the counterfactual model inherent in damage calculation—after all, courts only approximate. This seems to serve as a direct counterargument to Caltech’s idea of “modeling damages,” and rather suggests that the basis of damages should be based on actual past profits. However, due to the temporal antiquity of the case, one can argue that the Seymour ruling was the result of insufficient technology and subsequent inaccuracies. With modern improvements to economic and statistical modeling, then, current patent damage calculations are significantly less speculative and might avoid the problems faced by McCormick. 

Panduit Corp. v. Stahlin Bros. Fibre Works (1978), on the other hand, concerned specific calculations and their validity under patent law. The case itself primarily concerned a set of calculations conducted by a court-designated commission, which was later appealed. The defendant, Stahlin Bros. Fibre Works, cut the price of the product involving the infringed patent in the middle of the suit and demanded new calculations of profits based on decreased prices. In its verdict, the court also states that, to obtain lost profits, “a patent owner must prove: (1) demand for the patented product, (2) absence of acceptable non-infringing substitutes, (3) his manufacturing and marketing capability to exploit the demand, and (4) the amount of the profit he would have made.” [12] Taking these four premises collectively, awarding damages should heavily fall on manufacturers over distributors. It is not difficult to see how this is also the intuitive route, as, in its initial suit, Caltech itself also called for damages based solely on profits made by Broadcom in its transactions with Applet. 

Indeed, these standards presuppose an intentionally flexible standard mostly materialized through expert testimonies and mathematical intricacies, rendering damage calculations highly dependent on modeling techniques and expert testimony, rather than any legal presumptions. Given the ever-evolving nature of the technology scene and government desires to encourage innovation have rendered the granting and defending of patents a highly profitable and vulnerable practice. As such, patent laws have long been targets of exploitation by Patent-Assertion Entities (PAEs) which are roughly defined as shell companies or institutions that only hold patents but do not produce products and derive profits solely from patent infringement cases. Stanford Law noted that PAEs comprise fifty-nine percent of all patent defendants. [13] One of the more infamous cases for PAEs is  NTP, Inc. v. Research In Motion, Ltd. (2005) which resulted in a settlement for approximately six hundred-twelve million dollars despite the United States Patent and Trademark Office reviewing the questionable patents from NTP and invalidating some of them. [14] This case is demonstrative of multiple problematic trends in patent law cases: most of them tend to end in settlements due to a long time frame (Caltech has lasted about 6 years) and are only practical for large institutions. Thus, such cases are practically useless for small innovators whose profits are most disproportionately affected by a patent infringement. The issue of over-specific—and thus inherently uninnovative and purely financial—patents also lends itself to the problem, since they allow PAEs to file more cases and levy more financial burden on their opponents. Patent infringement, then, becomes a pure numbers game, evidenced by how they can often end up with invalidpatents. 

These specific loopholes that already exist within the field of patents itself are only amalgamated by the issue of overcompensation in patent damages. The concept of “entire market value” was first developed in Garretson v. Clark (1884) and was later reinforced in Westinghouse Electric & Manufacturing Co. v. Wagner Electric and Manufacture (1912) and Leesona Corp. v. United States (1979). In Garretson v. Clark, the Supreme Court allowed for damage calculations based on the “entire market value” of the product if the plaintiff is able to demonstrate by “reliable, tangible proof that the value of the machine … is due to the use of his patented invention.” [15] In Westinghouse Electric & Manufacturing Co. v. Wagner Electric and Manufacture, the Supreme Court reaffirmed the idea of presupposing the whole value unless proven otherwise: the plaintiff is entitled to profits made by the infringer even through reselling—and even when the patent itself is an improvement upon other previously established patents. [16] In Leesona,the United States Court of Claims reinforced that plaintiffs are entitled to “reasonable and entire” profits from the government. [17] This principle, then, not only expanded the field of potential patent complaints but also elevated the baseline compensation. Its implications have already been felt; in Apple Inc. v. Samsung Electronics Co. (2016) the Supreme Court ruled that a patent infringement of one part of the product makes an infringer liable “to the extent of his total profit.”  [18] Returning to Caltech, given that the patent violation was Apple’s usage of Broadcom chips, the lost profits could be interpreted as either the entire profits that Broadcom received from contracts with Apple or the Apple’s entire profits from any product that contained the Broadcom chips. Caltech’s current calculation is, in fact, an accumulation of both.  

More charitably, the two-tiered theory can even be supported by viewing the entire market value through a temporal lens, in which the distributor’s profit would be intuitively included in the calculation. Temporality in general markets is tightly associated with the linear structure of manufacture, distribution, and retail. One could reasonably argue that it is not until this course of transaction is completed that all potential profit has been extracted. Still, the simultaneity critique that Caltech proposes in terms of its two-tier model is poorly warranted. There is a notable difference between temporal aggregation—the profit as made by the exchange and distribution of the product as one succeeds the other—and simultaneous aggregation—e.g., negotiations with Apple and Broadcom starting at the same time.

Many legal scholars have taken issue with patent law and proposed alternative ways of  reimagining patent law in order to avoid exploitation. One notable theory by Ted Sichelmen,Professor of Law, the Director of the Center for Intellectual Property Law & Markets, and the Executive Director of the Center for Computation, Mathematics, and the Law at the University of San Diego, proposes that the public should view patent laws as a public regulatory system rather than private law. The private character of the law, as Sichelman argues, can be seen in its language of “compensation” and “harm,” and its previously implicit tort nature that had to be explicitly rejected in Leesona, in which the plaintiff demanded damages from government agencies instead of private personnels. [19] In this public view, a patent would allow patentees to sue and receive damage (in full) on behalf of the nation, rather than as an individual inventor, for the broad societal purpose of maximizing innovation incentives rather than as compensation for personal financial injuries. [20] This “attorney general status” would allow patentees to receive compensation not only on the basis of lost profits but also societal losses;  damages become a representation of the state’s valuation of intellectual property, which reflects the state’s emphasis on intellectual progress. This shift in the rhetorical framework could potentially standardize the formula for damage calculation, based on a unified notion of “the loss of the state” rather than financial demands based on the subjective profitability of the product alone.    

The explicit rejection of the $1.1 billion damage as being “legally unsupportable” in Caltech represents a long-enduring discontent held by the judges toward the scene of exploitation painted above. Caltech’s case, though itself not a standard patent-infringement fraud case, contains almost all the elements—the extortionary amount and the arithmetically obscured calculation methods—that would lead one to harbor frustration towards them.  The ambiguities of patent-infringement cases (and their future) pose an obvious threat to innovation, as they expose anyone with an idea to the grave danger of a drawn-out, highly technical, and potentially bankruptcy-inducing lawsuit from some of the best legally-armed entities. 

edited by Isabella Souza

Sources: 

[1] The California Institute v. Broadcom Ltd., Fed. Cir., No. 20-2222 (2022).

[2] Blake Brittain, and Jonathan Stempel, “Apple, Broadcom Win New Trial in $1.1 BLN Caltech Patent Case,” Reuters ( February 4, 2022),https://www.reuters.com/technology/apple-broadcom-get-new-damages-trial-11-billion-caltech-patent-case-2022-02-04/.

[3] Samantha Stainburn,  “Who Owns Your Great Idea?” The New York Times (December 24, 2008), https://www.nytimes.com/2009/01/04/education/edlife/whoseidea-t.html.

[4] The California Institute v. Broadcom Limited, Govinfo  (February 4, 2022), online at https://www.govinfo.gov/app/details/USCOURTS-ca13-21-01527/USCOURTS-ca13-21-01527-0 (visited February 8th, 2022)

[5] Id at 1.

[6] Id at 3.

[7] Brittain and Stemple, “Apple, Broadcom.”

[8] Id

[9] 35 U.S. Code § 284

[10] “Damages in Patent Infringement Cases,” Justia (October 18, 2021), online at https://www.justia.com/intellectual-property/patents/infringement/damages-in-patent-infringement-cases/#:~:text=The%20two%20main%20types%20of,the%20facts%20of%20the%20case (visited March 31st, 2022).

[11] Seymour v. McCormick, 57 U.S. 480 (1853).

[12] Panduit Corp., Plaintiff-appellant, v. Stahlin Bros. Fibre Works, Inc., Defendant-appellee, 575 F.2d 1152 (6th Cir. 1978).

[13] Shawn Miller et. al., “Who's Suing Us? Decoding Patent Plaintiffs since 2000 with the Stanford NPE Litigation Dataset,” 21 Stanford Technology Law Review 234, 235 (September 21, 2018), https://doi.org/10.2139/ssrn.3216095. 

[14] NTP, Inc. v. Research In Motion, Ltd.  418 F.3d 1282 (Fed. Cir. 2005), online at https://www.lexisnexis.com/community/casebrief/p/casebrief-ntp-inc-v-research-in-motion-ltd.

[15] Garretson v. Clark, 111 U.S. 120 (1884), online at https://supreme.justia.com/cases/federal/us/111/120/.

[16] Westinghouse Elec. Co. v. Wagner Elec. Co., 225 U.S. 604 (1912), online at https://supreme.justia.com/cases/federal/us/225/604/.

[17] Leesona Corp. v. United States, 599 F.2d 958 (Fed. Cir. 1979), online at https://casetext.com/case/leesona-corp-v-united-states.

[18] Samsung Electronics Co. v. Apple Inc., 580 U.S. ___ (2016), online at https://www.supremecourt.gov/opinions/16pdf/15-777_7lho.pdf.

[19] Ted Sichelman, "Purging Patent Law of Private Law Remedies," 92 Texas Law Review 516, 517-572 (February 2014).

[20] Id.