Jonathan Barnett employs a similar metaphor in his article, “The Patent System at a Crossroads.” Though we don’t yet know the Supreme Court’s decision in Oil States Energy Service v. Greene’s Energy Group, he believes the decision may be a turn—or swing—away from the current course of patent reform—a reversal of his narrative’s sea change of coordinated judicial, congressional, administrative, executive, and market-based decisions. Thus, he ties together 20 years’ worth of changes in patent law with one neat bow and claims that Oil States puts us at a crossroads related to them all.

He sees patent policy as a zero-sum battle between forces that want “strong patents” and those that demand “weak patents,” but things are not that simple. He and others seek to cast Oil States as some watershed moment in an inexorable march against the fundamentally good, strong patent.

Doing so overstates patent law’s importance. Oil States isn’t about patents—it is the opening salvo in what promises to be a lengthy battle at the reconstituted Supreme Court between those who would seek to curtail the administrative state and those who would like to preserve its status quo. Four voices on the Court have made no bones about their defense of the old precedent and old system, most clearly in their defense of the many administrative law “deferences”—the Chevron, Auer, Skidmore, and other related doctrines. On the other side of the debate, new Associate Justice Neil Gorsuch has wasted no time planting flags, adopting a hard line against doctrines meant to preserve the independence of the “fourth branch” of government, the administrative state. He was critical of some key rulings in his confirmation hearings and in judicial opinions, and his writing and comments from the bench to date show those to be no flukes.

One focus of that criticism is the use of administrative schemes, tribunals, or other types of decisionmaking as an alternative to Article III courts. Examples include the “Vaccine Court” administered by special masters within the U.S. Court of Federal Claims, the September 11th Victim Compensation Fund, Workers’ Compensation, the Tax Court, the “Hatch–Waxman” Act and “Biosimilars” Act schemes, the bankruptcy courts, and other federal tribunals, schemes, and decisionmakers empowered by Congress. They are meant to be much more efficient than the adversarial system of the Article III courts in disposing of certain matters (much to the chagrin of counsel).

Enter, curiously, Oil States. There are a few very old Supreme Court cases, like McCormick, that taken out of context suggest that the federal grant of a patent must be properly adjudicated by (and only by) an Article III court. Would that it were so.

But it’s complicated, as it so often is, and the quotes Professor Barnett uses don’t represent a sustained set of rulings, scholarship, or thinking the way he suggests. It’s a hope tied post hoc to an unwieldy wagon, as he reveals with his vague, dreamy talk of “crossroads” and “erosion,” as he attempts to paint Oil States as emblematic of opposition to patent reform. (The cert grant did expose a noticeable void in the scholarship—one arguing that patents are an absolute property right—a gap the academy is now rushing to fill.)

In doing so, he has overstated the question granted. This is a case about using administrative tribunals to adjudicate rights arising at common law. It doesn’t directly bear on whether the right itself is theoretically private property; a temporary, tightly regulated federal license; or something in-between. It’s about the nature of the tribunal.

By way of explanation, the Supreme Court has a doctrine whereby “public rights” can be adjudicated by tribunals, whereas “private rights” that have sprung from the common law cannot be removed involuntarily from Article III courts (parties can, of course, resolve their differences). It is the doctrine best embodied in the concurrence in Murray’s Lessee v. Hoboken Land & Improvement Co., and the concurrence in Northern Pipeline, that a public claim arising in tort, equity, or admiralty cannot be withdrawn from the courts. It most recently arose in the form of a 2011 counterclaim in a bankruptcy court proceeding and it makes sense: you do not want to, and cannot, deprive an Article III court of a common-law claim for damages. That was the 5–4 ruling in Stern v. Marshall, authored by Chief Justice John Roberts (with Antonin Scalia concurring and Stephen Breyer authoring the dissent). It was and remains a close fight between two competing factions of the Court for whose test rules the analysis. So the appeal to this newly constituted Court should now, in hindsight, seem obvious.

The terminology, unfortunately, is confusing enough that scholars and amici curiae soon zeroed in on patent grants as property—whether arguing they are “private property” or statutory grants given the “attributes of personal property,” as the 1952 statute says. That debate missed the point, though it did play to Gorsuch’s sensibilities. In their view, 200+ years of steady, quiet administrative patent grants, post-grant revisions, reissues, reexaminations, and other administrative actions by bureaucrats administering a complex system of hundreds of thousands of limited monopolies could be upended in part by a previously undiscovered fundamental right of our polity, buried deep within the Constitution, only now springing to life and importance.

The Importance of Patents

I will respond to Professor Barnett’s attempt to sow doubt in multiple areas of well-established, empirically supported observation. But let’s start with something his article lacks: context.

In 1984 the U.S. Patent and Trademark Office (PTO) issued just over 50,000 U.S. utility patents. By 2014 that number was 300,000, and by 2017 it had climbed above 320,000. In the past five years the number of utility patent applications has steadily risen by tens of thousands of applications a year—from 565,566 in 2012 to 650,411 in 2017. So at least inventors still find enough value in the system to apply for and earn patents at an astonishing, historic rate, with no end to the rise in sight. (Accordingly, the PTO’s budget has risen from just over $1 billion in 2001 to close to $3.5 billion this year.) This interest is at odds with Barnett’s claim of a substantially “eroded” “property-like nature of the patent right.” If they were of such little value, why would so many be paying so much for them?

He compares today to an earlier “nadir” in patenting from 1930 to 1970, when he says district courts were invalidating over 70% of patents. Yet from 1930 to 1970, American industry grew mightily, making us the richest nation in the world, with the most innovative companies across industries, from pharmaceuticals and medical devices to tractors and automobiles. Or that for the past five years—during what he calls a substantial “weakening” of the patent system—the stock market has steadily risen, we have emerged from a recession, growth is positive, and unemployment has fallen to almost 4%.

I am still waiting for the systemic death knell we keep being warned about to toll.

I am not suggesting correlation equals causation, as does Barnett. But the success of the U.S. economy during what he suggests are the weakest “nadirs” of the patent system at least begs two follow-up questions: just what do he and others mean when they advocate for strong patents generally; and if universally strong patents are a good thing, why, and to whom?

Question 1: What Do Commenters Mean by “Strong Patents”?

By strong patents, Barnett seems to mean making it easier to use any patent to obtain an injunction, win money damages, extract licensing fees, and withstand legal scrutiny, regardless of merit. He means making every patent more easily enforceable.

He can’t mean more patents and thus more patent protection—that’s already rising. He also can’t mean pendency of patent applications or the cost of patent prosecution—both are falling. And he can’t mean incidence of patent litigation—that, too, has seen historic highs in the past five years.

It’s likely he means patent valuation, which varies with what each patent owner can demand from others. Here, he’s on to something, in that the 2011 America Invents Act (AIA) has—without introducing new substantive legal grounds of patentability—made it cheaper and easier to apply preexisting legal requirements. Before the AIA, even the most improvidently granted patent would have cost at least $300,000 in legal costs and taken years to challenge. Thus, even the most worthless asset still had an inherent floor value untethered from merit—one based solely on legal leverage. Many paid rather than fight; many still do.

At bottom, the AIA has done exactly what Congress intended. What Barnett and others don’t like is just that. Less litigation, costing less money, resolved more quickly. Leverage reduced to around $100,000—a fair cost of legal fees associated with the new process. When he calls for stronger patents, he means giving all patents a higher valuation, one based on how easy it is to enforce them, regardless of merit.

Question 2: Are Easily Enforceable Patents Good for the U.S. Economy as a Whole?

Is there empirical evidence that, outside of the U.S. Food and Drug Administration framework and the Bayh–Dole Act established by Congress, easily enforceable patents are better for the American economy as a whole? Without question, a pharmaceutical company’s publicly listed, Orange Book–listed patents are essential to its millions or billions of dollars in revenue based on their interaction with statutory FDA approval processes and the marketability of drugs. And there is, as Barnett notes, the Bayh–Dole Act of 1980, which created the ex parte reexamination scheme (which he seems to take no issue with) and made at least universities much more profitable and motivated to spin out biotechnology start-ups. Bayh–Dole, while controversial, has been hailed as one of the more effective pieces of legislation in this area at achieving commercialization of inventions. True enough.

But outside of those sectors, there is very little evidence that patents generate wealth for the economy as a whole. Rather, they shift it. This seems obvious. They are, after all, a legalized, temporary monopoly—a serious restriction on free-market trade—and they do not intend to (or include a mechanism to) independently generate wealth for the country absent other statutory help. Do they generate wealth for individual companies at the expense of others? Sure. But their primary use is to concentrate wealth in the hands of the patentee for a limited period as a reward for publishing useful innovation.

This can generate licensing fees for companies that may have innovated in the past but no longer turn a profit on products—i.e., it allows older companies or those without manufacturing capacity to hold on longer, to retain revenue, and in some notable cases to not manufacture anything at all. That is perfectly fine and legal as long as those patents cover what parties say they do, and as long as they were inventive when granted and remain so. Patents reallocate wealth as a reward for inventive public disclosure. Thus, they should only be granted when they produce a societal benefit that outweighs those costs, as scholars like Stanford law professor Lisa Larrimore Ouellette note. Ostensibly, American industry and innovators use patent disclosure to advance the progress of the useful arts. Monopoly causes harm to competition and thus the free market, but that’s intended where justified and part of the deal in some cases, assuming a frictionless system. But what about transaction costs?

Out-Of-Proportion Transaction Costs

Enter the lawyers. Patent law is unquestionably one of the highest-priced fee-for-service professions. As the American Intellectual Property Legal Society shows in its annual anonymized survey of its wide-ranging membership, litigation costs where $10–$25 million is at risk averaged $2 million in 2017, down from their peak in 2013 at $3.325 million. Professor Barnett knows this, asserting in a 2009 article that “median patent discovery and litigation costs [were then] $2.5 million and $4 million respectively.” It now costs on average $5,000–$8,000 to apply for a patent, with no guarantee of success, and many companies are prosecuting tens of thousands each year.

It is a highly lucrative career, and has turned the patent bar associations into powerful, well-heeled lobbies with tens of thousands of patent lawyers invested in the continued existence of those high costs. One can see why amici support for reversing the “pendulum” of 2011’s modest procedural reforms that reduced litigation costs overall might be plentiful and vociferous. Patent lawyers as an industry have a stake in this one, and the continued success of those reforms could hit counsel in the wallet.

Not that there’s not plenty of money still to be made. Here again, context matters; in any other area of the law, $2 million in litigation expenses would be extreme. Here, that’s supposedly a cause for concern. But that is a debate for another day. What matters is the context of a system where legal fees routinely top $1 million—and where hundreds of thousands of new causes of action—new patents—issue every year. The system siphons off money for legal fees to the tune of millions of dollars a year for American businesses. And it does so, in many cases, at the behest of wrongful claimants seeking nothing more than to extract fees. It needs to be carefully policed. That’s where the real debate lies. As Federal Circuit Judge Kimberly Moore once noted, “Can the patent system flourish if the scope of the patentee’s property right is wrongly assessed one-third of the time?”

A convenient new dialectic / Professor Barnett and others, playing to the imagined conservative and neo-originalist sensibilities of Justice Gorsuch, are suddenly painting a picture of a battle between libertarian strong-patent-rights advocates versus liberal administrative wonks. But the patent system has long been, for better or worse, a complex regulatory framework where patent monopolies exist only by the grace of Congress and the executive. Until recently, strong-patent-rights advocates full-throatedly supported that framework.

Consider reexamination, another post-grant procedure available since the passage of the Bayh–Dole Act in the early 1980s. It was initially proposed by patent owners as a means to fix errors in too-narrow patents. Or reissue proceedings, now nearly 200 years old, where entirely new claims can reissue out of a previously issued patent. And there are certificates of correction, derivations, interferences, inter partes reexaminations, supplemental examination, and more. None of which Oil States’ counsel had any issue with; in briefing and during oral argument, counsel conceded the constitutionality of these other post-grant procedures.

Then note, as Justice Anthony Kennedy did, that Congress can change the term of patents to whatever lawmakers think is appropriate. Patent law is inextricably bound up in the administrative—i.e., bureaucratic—process of applying for and granting patents. It’s invariably a lot of paper. Once granted, patents are subjected in court to administrative deference per se—there is a presumption of validity in District Court of all patents arising from an expert agency expected to do its job. Trying to cast one administrative procedure aside as “administratively” minded and another as “property” minded is dishonest; all patents spring from a bureaucratic framework and represent a very temporary and very tightly regulated constraint on free trade. To suggest otherwise is to oversimplify. It can be both.

Is property-rights erosion really happening? / Professor Barnett claims that “it has now become popular in certain business and policy circles to endorse so-called ‘efficient infringement’ strategies.” He provides no support for this allegation; for my part, the only place I have seen that term used is as a relatively new talking point on blogs—Google it and see for yourself—and at events as a means to rhetorically discredit further patent reform.

The “use now, pay later” argument he and others make is that “efficient infringement” is premised on—and founded in—the Supreme Court’s 2006 eBay decision, which made it much harder for courts to issue preliminary injunctions, at least as interpreted by lower courts. That, in turn, changed most of patent litigation leverage into a fight over potential damages or costs. Fair enough. But assuming anyone would so wantonly infringe as a business strategy ignores Halo and the Supreme Court’s recent treble damages decisions, decisions that give district court judges wide latitude in assessing up to three times the assessed damages for willful infringement and to impose fees. If as a patent owner you notify others of infringement and they willfully continue, as long as you trust the validity of your patents, you’ve got quite a case on your hands. I don’t know a company or counsel that would sign off on that risk; there’s no way to budget for that kind of pain. That’s certainly not an invitation to efficiently infringe anything. Even if it were, Barnett’s argument here is just a criticism of the Supreme Court in eBay circa 2006, nothing more.

Are “patent trolls” really a problem? / Professor Barnett and others have recently begun suggesting that non-practicing entities (NPEs)—or pejoratively, “patent trolls”—don’t even exist. They have taken to speaking of the “patent troll narrative”—or even the inartfully redundant “false patent troll narrative”—to suggest it is a story, i.e., one made up, unsupported, and inaccurate.

Yet, Congress, the Federal Trade Commission, the Department of Justice, the PTO, and the courts have written about, catalogued, and studied the problem at length. (That’s not to mention the hundreds of companies that have been targeted.) For instance, the FTC conducted an extensive, well-researched study on NPEs using its subpoena powers. It carefully documented the problem, releasing the years-long study in 2016. Barnett fails to explain or acknowledge the wealth of evidence—published, public, and obvious—demonstrating the practice. I will.

While shell companies make it easy to hide ownership, nonetheless NPEs are generally pretty brazen. Some are so successful that they are publicly traded Fortune 500 companies—companies like Marathon Patent Group, Acacia Research, and others. Indeed, they are known as “publicly traded intellectual property companies” (PIPCOs). Others stay private but are nonetheless prolific, litigating through alter egos—IP Edge, Dominion Harbor, Blackbird Technologies, IP Valuation, and the many limited liability companies controlled by Leigh Rothschild, Nick Labbitt, Brian Yates, or any of the companies once associated with Erich Spangenberg of the moribund IP Nav. And then there are the Hawk Technologies of the world, which has brazenly sued 200+ entities, including grocery stores, transit authorities, municipalities, and even multiple branches of the nonprofit Goodwill for their use of closed-circuit surveillance video.

These firms often play name games that seem comical—until you’re facing down service. IP Edge controls or was the genesis of litigious alter egos Anuwave, Autoloxer, Banertek, Bartonfalls, Carnition, Drogo IP, eDekka, Finnavations, HelioStar, Kevique Technology, Kobace, Long Corner Consumer Electronics, Loramax, MagnaCross, Mod Stack, Mozly Tech, NovelPoint Security, Oberalis, Opal Run, Olivistar, Orostream, Peppermint Hills, Reef Mountain, Ruby Sands, Serenitiva, Somaltus, Vaultet, Venus Locations, and Wetro LAN, among others. (The Rothschild entities, for their part, are transparent enough to generally include the Rothschild name, suggesting their alter egos are created more out of a legal or fiscal strategy than as a means of obfuscating ownership).

If parties were truly interested in treating patents as property, they would be advocating vigorously for mandatory recordation of assignment and transfer and further regulation of the sale and transfer of those rights, just as land, automobiles, and other types of property are heavily regulated in transfer or sale. Indeed, a voluntary database already exists, it just doesn’t punish those actors that choose to keep transfers and thus ownership secret, increasing legal cost accordingly. It avoids press and obscures what NPEs are doing. You don’t hear anyone pushing mandatory recordation. (Though recently the lack of it came back to bite Intellectual Ventures when it was revealed—after the firm filed an International Trade Commission complaint on five patents, shortly thereafter dismissed—that the firm wasn’t the true owner of the asserted assets.)

They might also advocate for seeking out and punishing the worst abusers of the system—those that so clearly and demonstrably exist. They could nominate someone at the DOJ or FTC to issue a public report detailing the holdings and practices of those few that are so unquestionably abusive we can all agree we’d be better off without them. But those changes wouldn’t ease assertion.

Are “bad patents” really a problem? / Concerning “bad patents,” Professor Barnett omits discussing the many changes in invalidity law over the past 20 years. Many of the patents litigated today were granted in the late 1990s or early 2000s. Since then, many facets of patent validity have changed dramatically per Supreme Court precedent, including the legal tests for invalidity, obviousness, subject matter eligibility, anticipation, and even claim construction and the standards of validity in general.

As Congress correctly noted during the passage of the AIA, many of the millions of patents that have already issued would not have issued if examined today. Without passing judgment on the propriety of these intervening rulings, it is objectively clear that some patents have issued under old standards that would not under new ones; they are thus unpatentable.

But even putting that aside, with 320,000+ patents issued and more than 650,000 applications filed each year, and with grant rates that we can all agree are at least above 50% despite the quibbles Barnett raises, plenty of patents will issue that contain defects, errors, or otherwise invalid claims. Even ostensibly “good” patents can have “bad” claims, i.e., invalid ones—which was the whole point behind creating the patent-owner-driven ex parte reexamination process. And no one, not even Oil States’ counsel, is advocating holding ex parte reexamination or the 200-year-old reissue process unconstitutional.

That, in turn, means that Barnett and others aren’t upset about the fact of administrative reexamination itself—i.e., post-grant cancellation. Rather, they’re concerned with the form of it being too effective at lowering the transaction costs once associated with patent litigation. But that’s explicitly why Congress created it.

“Royalty stacking” is beside the point / Professor Barnett seeks to minimize the problem of royalty stacking by citing studies and evidence of lower single-digit royalty percentages. But multiple per-device royalties of “just” 3% or 5% on billions of dollars is a significant thumb on the scales. He also makes the correlation/​causation leap again, suggesting that because smart phone prices are dropping generally, stacking must not be a problem. It’s an argument currently spilling out in epistolary form before the Justice Department Antitrust Division.

The more reasoned approach to royalty stacking is to acknowledge that there are two main academic camps arguing over the empirical evidence of royalty stacking, that those two camps tend to line up on the same side of most other issues, and that they are rather beside the point. Jorge Contreras of the University of Utah has done just that in an upcoming paper, preaching a Gordian knot approach. Noting that any individual instance of stacking or hold-up would still be wrong and worth regulating regardless of empirical evidence (or lack thereof) of a widespread problem, he suggests moderation, continued vigilance by government agencies, but certainly no activist pressure based on theories of widespread stacking problems. The middle view here might not be as dramatic or make either side happy, but that may be a sign that it’s the more reasonable approach.

In biotech, “patent thickets” have never been much of a problem / Professor Barnett, scare quoting the well-documented “patent thicket” problem, uses as his counterfactual the only field where volume patenting isn’t widespread (yet): biopharmaceuticals. But that phenomenon is widely understood and easily explained by the need for biopharmaceutical patents to be listed in the FDA’s Orange Book. In short, pharmaceutical patents are tightly constrained by the need for approval of the underlying drugs on which they read through rigorous scientific testing and a complicated statutory schema—the Drug Price Competition and Patent Term Restoration (Hatch–Waxman) Act and the related Biologics Price Competition and Innovation (Biosimilars) Act. Yet even that field occasionally suffers from the problem, as with certain patent portfolios surrounding drugs that now top 100 patents and counting.

In the high technology area, the problem is documented and widespread. To illustrate, recently a company called Provenance Asset Group (PAG) purchased some 5,600 patent families from Nokia. PAG spent millions in legal fees analyzing the first 2,600 and proudly announced that it had found 56 that it thought strong enough to assert. The number isn’t notable; what’s notable is the millions of dollars in legal fees and hundreds of hours of lawyer time it took to get to that answer. Think about those millions in fees—and this is before any court costs picking through the thousands of legal claims—as a barrier to entry and you get why a patent thicket is such a problem. Even valueless or irrelevant assets, when grouped into a subject matter area, a standard, a portfolio, a pool, or a tranche for purchase or assertion, present a sizable barrier to challenge and a resulting inherent value.

I’m reminded of a footnote in a David Foster Wallace book where he likens thousands of pages of bureaucratic public documents containing valuable, if obfuscated, information to “the giant solid-gold Buddhas that flanked certain temples in ancient Khmer. These priceless statues, never guarded or secured, were safe from theft not despite but because of their value—they were too huge and heavy to move.” With 6,000+ patents issuing from the PTO weekly and emerging portfolios and pools topping thousands of assets, the patent thicket problem explains itself.

Conclusions

In short, Oil States is about administrative law; it doesn’t represent a change in patent policy writ large. If there is some broad, systemic shift in efforts at reasonable reform, then Barnett’s article, other commentary, and the intense lobbying pressure from industries that have been displaced by those reforms are what will have caused it.

Readings

• “A Market Test for Bayh–Dole Patents,” by Ian Ayres and Lisa Larrimore Ouellette. Cornell Law Review 102(2): 271–331 (2017).
• “Are District Court Judges Equipped to Resolve Patent Cases?” by Kimberly A. Moore. Harvard Journal of Law & Technology 15(1): 2–39 (2001).
• “Debugging Software Patents after Alice,” by Jonathan Stroud and Derek M. Kim. South Carolina Law Review 69: 1–33 (2017).
• “Has Patent, Will Sue: An Alert to Corporate America,” by David Segal. New York Times, July 14, 2013.
• “How Often Do Non-Practicing Entities Win Patent Suits?” by John R. Allison, Mark A. Lemley, and David L. Schwartz. Berkeley Technology Law Journal 32: 237–307 (2017).
• “Meet America’s Most Prolific Patent Troll,” by Kevin Drum. Mother Jones, Oct. 27, 2016.
• “Much Ado about Hold-Up,” by Jorge Contreras. Working paper, Feb. 13, 2018.
• “Patent Holdup, the ITC, and the Public Interest,” by Colleen V. Chien and Mark A. Lemley. Cornell Law Review 98: 1–45 (2012).
• “ ‘Patent Trolls’ and Patent Remedies,” by John M. Golden. Texas Law Review 85: 2111–2161 (2007).
• “Public Enforcement of Patent Law,” by Megan M. La Belle. Boston University Law Review 96: 1865–1928 (2016).
• “The PTAB at Five: Reduced Leverage for NPEs,” by Jonathan Stroud. IAM Magazine, November/​December 2017.
• “The Surprising Resilience of the Patent System,” by Mark A. Lemley. Texas Law Review 95: 1–57 (2016).
• “Worthless Patents,” by Kimberly A. Moore. Berkeley Technology Law Journal 20(4): 1521–1552 (2005).