Download PDF

Note from the Editor:

This article criticizes Justice Clarence Thomas’ opinion in Oil States. It is the third in a series of articles by Professor Epstein about the issues at stake in Oil States Energy Services, LLC v. Greene’s Energy Group, LLC.

The Federalist Society takes no positions on particular legal and public policy matters. Any expressions of opinion are those of the author. Whenever we publish an article that advocates for a particular position, as here, we offer links to other perspectives on the issue, including ones opposed to the position taken in the article. We also invite responses from our readers. To join the debate, please email us at [email protected]


In Oil States Energy Services, LLC v. Greene’s Energy Group, LLC,[1] the Supreme Court, by a seven-to-two vote, upheld the controversial provisions of the America Invents Act (AIA) that allow for inter partes review (IPR) before the Patent Trial and Appeal Board (PTAB) after the final issuance of a patent.[2] Petitioners argued unsuccessfully that the AIA’s IPR procedures violated Article III and the Seventh Amendment of the Constitution. The majority opinion was written by Justice Clarence Thomas on behalf of Justices Kennedy, Ginsburg, Breyer, Alito, Kagan, and Sotomayor. Justice Neil Gorsuch wrote a short dissent in which Chief Justice Roberts joined.[3]

As should be evident from the lopsided majority in Oil States, patent law has become a field in which a broad left-right coalition on the Supreme Court supports a comprehensive vision of patent law that couples weak property rights with high-level administrative control.[4] From the right, Justice Thomas has written major opinions on such critical topics as injunctive relief[5] and the exhaustion doctrine,[6] all of which have narrowed the rights of patentees. From the left, Justice Breyer has aggressively applied Chevron deference[7] to increase the influence and power of the PTAB,[8] to which the head of the U.S. Patent and Trademark Office (USPTO) has delegated all administrative review of patent cases.[9] This combination of narrowing rights for patentees and increasing administrative power is the greatest long-term peril facing patent law. The trend promises to undermine further the effectiveness of the patent system, a process that accelerated with the ill-advised revisions of the Patent Act of 1952 by the AIA.[10]

At issue in Oil States were provisions of the AIA that authorize any party sued for infringement in federal court to initiate IPR before the PTAB after a final patent has been issued.[11] Justice Thomas stresses the narrowness of our holding,given that it only deals with IPR.[12] But his flawed reasoning has powerful implications for the future direction of both the patent system and administrative law in general. As described by Professor Gregory Dolin, the AIA’s procedures are the final stage of a transformation in patent practice that began more than three decades ago.[13] Until 1981, administrative review was only allowed before a final patent had been issued. Once the patent issued, it counted as a completed grant that, like other grants of property, could only be modified or set aside in an Article III judicial proceeding, in which the challenger could show that the grant was improperly made or that the grantee had failed to comply with some of its terms. Historically, these standard rules of property applied with equal force to patents and land grants, since both are constitutionally protected forms of property.[14] In 1981, as part of the Bayh-Dole Act, Congress adopted a procedure that allowed any member of the public to demand that the USPTO take a “second look” at the validity of any patent already issued.[15] That review was not mandatory, and it was restricted to a reexamination on two key questions: novelty and nonobviousness. Nevertheless, it changed the fundamental balance between the Executive and the Article III courts, paving the way for the AIA’s novel IPR practice. The 1981 reforms did not lead to a massive level of patent innovation, so Congress passed a second reexamination statute in 1999[16] that persisted until its repeal and replacement by the AIA, which instituted the current form of IPR. Oil States continues this trend by expanding the power of administrative agencies at the expense of Article III courts in adjudicating patent disputes, which has the effect of narrowing the intellectual property rights of patentees. I have already commented on how this decline in the doctrine of separation of powers will work to undermine the regime of stable property rights.[17]

This article evaluates Justice Thomas’ reasoning in Oil States by following the structure of his opinion. In Part I, it examines and rejects his claim that patents fall squarelywithin the domain of public rights for which adjudication before an Article III court is not necessary.[18] In Part II, it examines the effect of IPR on the patent system, with an emphasis on the separation of powers issues that Oil States raises. It concludes with a brief comment on how various parties—including Congress and intellectual property practitioners—might address the problems with the patent system that the Court failed to correct in Oil States.

I. Are Patents Public Rights?

A. Stern and Northern Pipeline: Limiting the Domain of the Bankruptcy Courts

Justice Thomas begins his analysis with a discussion of the distinction between private rights, which are causes of action between parties, and public rights, which involve public issues in which the government has a direct stake. He starts by referencing the two key modern cases—Stern v. Marshall and Northern Pipeline Construction Co. v. Marathon Pipe Line Co.—to make two points: that the Courts public rights jurisprudence has not been definitively explained,and that its precedents have not been entirely consistent.[19] But he does not mention that both cases emphatically denied the power of Congress to remove certain causes of action from the jurisdiction of Article III courts by declaring the rights at issue to be public rights. In Stern, Chief Justice John Roberts stressed that the bankruptcy court—a non-Article III court—did not have constitutional authorityto enter judgment on common law claims, including damages actions for patent infringement, which have been regarded as actions at common law since the Statute of Monopolies of 1623.[20] He further added that Congress may not withdraw from judicial cognizance any matter which, from its nature, is the subject of a suit at the common law . . . .’[21] Under Stern, suits at common law, including disputes over patent rights, must be adjudicated in Article III courts.

Earlier, in Northern Pipeline, Justice William Brennan anticipated the centrality of Article III courts when he refused to allow their ouster by revised bankruptcy courts. After his own exhaustive account of the separation of powers, he concluded:

In sum, our Constitution unambiguously enunciates a fundamental principle—that the “judicial Power of the United States” must be reposed in an independent Judiciary. It commands that the independence of the Judiciary be jealously guarded, and it provides clear institutional protections for that independence.[22]

He noted that [a]s an inseparable element of the constitutional system of checks and balances, and as a guarantee of judicial impartiality, Art. III both defines the power and protects the independence of the Judicial Branch.”[23] That impartiality is wholly lacking in cases tried before PTAB judges; the Director of the USPTO can expand or alter the composition of a panel at will in order to reach a decision to his or her liking.[24] By contrast, judges who sit on Article I courts dealing with taxation or bankruptcy have long, fixed terms and use neutral assignment systems to avoid the possibility of strategic misbehavior.[25]

B. Crowell and Murray’s Lessee: Distinguishing Public and Private Rights

Next, Justice Thomas discusses a snippet from Crowell v. Benson.[26] Crowell was a difficult case in which Chief Justice Charles Evans Hughes upheld the Longshoremen’s and Harbor Workers’ Compensation Act of 1927 against a constitutional challenge.[27] This federal statute was modeled on state workers’ compensation systems that presented few difficulties under state constitutions. The hard question in Crowell was whether their novel form met the more stringent federal due process and separation of powers requirements for Article III courts. The Supreme Court held that the employer must be allowed to obtain de novo review in district court on the factual jurisdictional question of whether the claimant was an employee of the defendant.[28] The case was a delicate accommodation because it stretched the principles governing the constitutional position of Article III courts to fit a new structure—a more rapid and certain administrative workers’ compensation scheme that displaced factual resolution of tort claims in court—without abandoning judicial oversight altogether.

Justice Thomas writes, referring to Crowell, Our precedents have recognized that the [public-rights] doctrine covers matters which arise between the Government and persons subject to its authority in connection with the performance of the constitutional functions of the executive or legislative departments.’[29] This sentence distorts Hughes’ opinion. First, it omits the beginning of the quoted sentence, inverting its meaning. The full sentence reads: As to determinations of fact, the distinction is at once apparent between cases of private right and those which arise between the Government and persons subject to its authority in connection with the performance of the constitutional functions of the executive or legislative departments.”[30] Crowell came out as it did because Article III courts retained control over all questions of law. Chief Justice Hughes worried about how workerscompensation claims were tried in the first instance before a commission on which no Article III judge sat. But he thought the statute overcame this difficulty by making it clear that [r]ulings of the deputy commissioner [who oversees these cases] upon questions of law are without finality. So far as the latter are concerned, full opportunity is afforded for their determination by the federal courts through proceedings to suspend or to set aside a compensation order. . . .”[31] Crowell’s reasoning offers no warrant for the PTAB, which has virtually final authority over all questions of both fact and law insofar as it receives full deference from reviewing courts.[32] There is no doubt that PTAB judges count as “inferior officers” under the appointments clause, which gives some indication of the breadth of their powers.[33]

Crowell points to Murray’s Lessee v. Hoboken Land & Improvement Co. to explain why the deputy commissioners—non-Article III judges—do not adjudicate public rights, further weakening Justice Thomas reliance on the case.[34] Chief Justice Hughes first noted that a large number of cases dealing with “interstate and foreign commerce, taxation, immigration, the public lands, public health, the facilities of the post office, pensions and payments to veteransinvolved public rights.[35] In stark contrast, he concluded that the statutory scheme at issue in Crowell could not be properly considered a public rights case. Accordingly, he held that workers’ compensation claims that concerned only ordinary individuals involved only private, not public rights. When these actions were brought at common law, Hughes continued, the right to jury trial was preserved in an Article III case by the Seventh Amendment, but in cases that arose either in equity or admiralty, Congress could change how factual questions in an Article III court could be resolved, so that the use of special masters and commissions was wholly proper.[36] Crowell thus explicitly rejects the notion that Congress may lodge dispositive control over new statutory causes of action—whether they involve legal or equitable remedies—in an administrative tribunal instead of an Article III court. Instead, Article III gives Congress discretion only as to how factual issues should be resolved in equity and admiralty cases.[37]

That result is borne out by recalling that Murray’s Lessee goes to great lengths to explain that Congress cannot encroach on the jurisdiction of Article III courts.[38] Murray’s Lessee upheld an 1820 statute that provided for administrative resolution of a dispute over money that customs collectors owed to the United States; it did so because the Court found that the monetary dispute in question was in the nature of a public right. But Murray’s Lessee is easily distinguishable from Oil States. Murray’s Lessee sustained the statute against a due process challenge in part because the disputed accounting system it put into place to determine the net payments owed had been in widespread use in both England and the American states since the earliest days of the nation. The new IPR under the AIA has no such pedigree, which undercuts any claim to institutional legitimacy. Murray’s Lessee also stressed the routine administrative nature of the accounting, which is a far cry from the power to act as the final decider of patent validity in the novel regime imposed by the AIA without due regard for nearly 60 years of practice under the Patent Act of 1952.[39]

C. Ex parte Bakelite: Public Rights in Claims Against the United States

To further bolster his position, Justice Thomas references Ex parte Bakelite Corp. for the proposition that Congress may decide cases arising between the government and others, which from their nature do not require judicial determination and yet are susceptible of it.[40] No one disputes this proposition; the only question in Oil States was whether post-grant disputes over patent rights fit this description. But Justice Thomas does not mention that Bakelite understood questions of law to require adjudication in an Article III Court:

The [relevant] section provides that, to assist the President” in making decisions thereunder, the Tariff Commission shall investigate allegations of unfair practice, conduct hearings, receive evidence, and make findings and recommendations, subject to a right in the importer or consignee, if the findings be against him, to appeal to the Court of Customs Appeals on questions of law affecting the findings.[41]

Bakelite arose out of an administrative dispute after the Tariff Commission imposed a heavy tariff on the plaintiffs goods. But it is clear that the definition of public rights cited in Bakelite is far narrower than the one adopted in Justice Thomas’ opinion. The Bakelite Court said thatConspicuous among such matters  [public rights] are claims against the United States. These may arise in many ways and may be for money, lands or other things.[42] The patent dispute in Oil States is not a claim against the United States, but a property dispute between private parties; holding that the public rights doctrine extends to Oil States was a dramatic expansion of the doctrine, not a mere application of Bakelite’s principles.

D. Franchise, Properly Understood

After misconstruing Bakelite, Justice Thomas strings together a set of quotations that make it appear as though patent claims should be considered franchises,which he contrasts with other forms of property grants that must be adjudicated in Article III court. But the term franchiseis a synonym, not an antonym, of property. Thomas writes:

As this Court has long recognized, the grant of a patent is a matter between “‘the public, who are the grantors, and . . . the patentee.’” [United States v.] Duell, [172 U.S. 576,] 586 [1899] (quoting Butterworth v. United States ex rel. Hoe, 112 U.S. 50, 59 (1884)). By issuing patents,the PTO take[s] from the public rights of immense value, and bestow[s] them upon the patentee.United States v. American Bell Telephone Co., 128 U.S. 315, 370 (1888). Specifically, patents are public franchisesthat the Government grants to the inventors of new and useful improvements.Seymour v. Osborne, 11 Wall. 516, 533 (1871) . . . . The franchise gives the patent owner the right to exclude others from making, using, offering for sale, or selling the invention throughout the United States.” 35 U.S.C. § 154(a)(1). That right did not exist at common law.” Gayler v. Wilder, 10 How. 477, 494 (1851). Rather, it is a ‘creature of statute law.’ Crown Die & Tool Co. v. Nye Tool & Machine Works, 261 U.S. 24, 40 (1923).[43]

The quotations employed here are worth investigating further. Do the precedents quoted shed any light on how the Court has thought about patent rights?

1. Butterworth and American Bell Telephone Co.

The quotation from Butterworth is part of larger sentence that stands for the opposite proposition:

The legislation based on this provision regards the right of property in the inventor as the medium of the public advantage derived from his invention; so that in every grant of the limited monopoly two interests are involved, that of the public, who are the grantors, and that of the patentee.[44]

The italicized words make it clear that these patents, once granted, are rights of property, which then receive the same level of protection under the Due Process Clause as all other forms of property. The same point applies to the excerpt from United States v. American Bell Telephone Co., which reads with full context:

In this respect the government and its officers are acting as the agents of the people, and have, under the authority of law vested in them, taken from the people this valuable privilege and conferred it as an exclusive right upon the patentee. This is property, property of a value so large that nobody has been able to estimate it.[45]

2. Seymour

Similarly, the Court in Seymour v. Osborne declared that inventions secured by letters patent are property in the holder of the patent, and as such are as much entitled to protection as any other property, consisting of a franchise, during the term for which the franchise or the exclusive right is granted.[46] Qualifying “franchise” with the word publicdoes nothing to change the meaning. A public franchise is a form of private property. The full passage from which Justice Thomas extracted the Seymour quotation reads:

Letters patent are not to be regarded as monopolies, created by the executive authority at the expense and to the prejudice of all the community except the persons therein named as patentees, but as public franchises granted to the inventors of new and useful improvements for the purpose of securing to them, as such inventors, for the limited term therein mentioned, the exclusive right and liberty to make and use and vend to others to be used their own inventions, as tending to promote the progress of science and the useful arts, and as matter of compensation to the inventors for their labor, toil, and expense in making the inventions, and reducing the same to practice for the public benefit, as contemplated by the Constitution and sanctioned by the laws of Congress.[47]

Seymour rightly notes that a patent is not a monopoly but an exclusive grant that compensates for services rendered to the public at large. In attempting to rebrand patents as something less than property, Justice Thomas misreads the very precedents that prove him wrong.

3. Gayler

Justice Thomas next quotes Gayler v. Wilder, which he claims stands for the proposition that “the right [to exclude others from making, using, offering for sale, or selling a patented invention in the U.S.] did not exist at common law.’[48] But again, context is everything, as the full passage from which this quotation is extracted makes clear:

Now the monopoly granted to the patentee is for one entire thing—it is the exclusive right of making, using, and vending to others to be used, the improvement he has invented, and for which the patent is granted. The monopoly did not exist at common law, and the rights, therefore, which may be exercised under it cannot be regulated by the rules of the common law. It is created by the act of Congress, and no rights can be acquired in it unless authorized by statute, and in the manner the statute prescribes. . . . the patentee may assign his whole interest or an undivided part of it. But if he assigns a part under this section, it must be an undivided portion of his entire interest under the patent, placing the assignee upon an equal footing with himself for the part assigned.[49]

The Gayler litigation took place in Circuit Court, and there is no hint that any administrative forum decided, or could have decided, the case. The decision assumes that the monopoly grantedwhen a patent issued is treated as a full property right. The fact that the attributes of the patent are regulated by statute instead of common law does not mean that patent rights are therefore weaker than those created at common law. Congress had required that only undivided interests could be assigned because it wanted to prevent complex interests in property rights from imposing extra burdens on licensees; Gayler merely recognizes that questions of assignability raised difficult issues in nineteenth century law generally.[50] Those difficulties carried over to the assignment of interests in patents, where additional complexities had to be addressed. None of this complexity calls into question the status of patent rights as property rights.

4. Crown Die

Justice Thomas similarly ignores context in Crown Die & Toll Co. v. Nye Tool & Machine Works, which explicitly relied on Gayler to understand the role of assignments in patent litigation.[51] The action involved a breach of contract dispute that placed at issue the validity of a patent assignment. The court held that even though the parties were not diverse, jurisdiction was proper in federal court because the patent was claimed to be valid under the statutes of the United States.[52] Justice Thomas quotes Crown Die to make the point that patents are creatures of statute, not common law. But he fails to consider that Crown Die treats the patent at issue as a property right that had been validly assigned under the applicable statutory rule. The case supports the proposition that a patent, once granted, is an ordinary property right entitled to full protection in an Article III court, and not a lesser right that can be administratively revoked.

5. Louisville Bridge

Unfortunately, the Oil States opinion also misreads other franchise cases. Justice Thomas writes:

Patent claims are granted subject to the qualification that the PTO has “the authority to reexamine—and perhaps cancel—a patent claim” in an inter partes review. . . . This Court has recognized that franchises can be qualified in this manner. For example, Congress can grant a franchise that permits a company to erect a toll bridge, but qualify the grant by reserving its authority to revoke or amend the franchise. See, e.g., Louisville Bridge Co. v. United States, 242 U.S. 409, 421 (1917) (collecting cases). Even after the bridge is built, the Government can exercise its reserved authority through legislation or an administrative proceeding.[53]

This passage makes it appear as though Louisville Bridge held that any dispute over a franchise can be determined by an administrative proceeding.

This reading is wrong. The case points to the exact opposite position, namely, that a judicial proceeding is required to enforce any conditions or limitations attached to a grant. The Ohio Falls Bridge was built in 1879 to cross the Ohio River at Louisville, Kentucky. The terms of the grant to the Louisville Bridge Company contained a proviso that said bridge and draws shall be so constructed as not to interrupt the navigation of the Ohio River.”[54] During World War I, the Secretary of War gave the bridge owner notice that the bridge was out of compliance because it did not allow for sufficient horizontal clearance over the river. After a number of administrative hearings failed to resolve the dispute, the Secretary of War ordered that the bridge be repaired within three years. When the bridge owner refused to comply:

[T]he Attorney General filed a bill for an injunction in the district court; appellant answered, setting up its claims as above indicated, and the case was brought to a hearing upon stipulated facts presenting, as the sole question to be determined, the legality of the order of the Secretary of War as applied to the bridge in question.[55]

The initial grant to the bridge owner by the government created certain property rights. No one questioned that any dispute over the scope of these rights had to take place in federal district court. The only point of contention was over the proper construction of the grant, on which Justice Mahlon Pitney (my favorite) wrote as follows:

[W]hen private rights of an indefeasible nature are sought to be derived from regulatory provisions established in the exercise of [the commerce] power, the case is peculiarly one for the application of the universal rule that grants of special franchises and privileges are to be strictly construed in favor of the public right, and nothing is to be taken as granted concerning which any reasonable doubt may be raised.[56]

Louisville Bridge does not even hint that the United States could resolve the case—which, like the dispute in Oil States, involved property rights obtained under a grant from the government—through an administrative proceeding.[57]

E. McCormick and Cases Cited

Whether patent disputes may be resolved in administrative proceedings is addressed in other cases, which Justice Thomas evaluates as follows:

To be sure, two of the cases make broad declarations that [t]he only authority competent to set a patent aside, or to annul it, or to correct it for any reason whatever, is vested in the courts of the United States, and not in the department which issued the patent.McCormick Harvesting Machine Co., supra, at 609; accord, American Bell Telephone Co., 128 U.S., at 364. But those cases were decided under the Patent Act of 1870. . . . That version of the Patent Act did not include any provision for post-issuance administrative review. Those precedents, then, are best read as a description of the statutory scheme that existed at that time. They do not resolve Congressauthority under the Constitution to establish a different scheme.[58]

Once again, a fuller examination shows that Justice Thomas has misconstrued this precedent. In McCormick Harvesting Machine Co. v. Aultman, the owner of a patent was frustrated in his effort to obtain a reissue of the patent on the ground that its subject matter was not novel in light of his previous patents. Ownership of the patent then passed by assignment to the plaintiff corporation, which abandoned the application for a reissue, and requested and obtained from the patent office the return of the original patent.[59] The full passage discussing this application reads:

It has been settled by repeated decisions of this court that when a patent has received the signature of the Secretary of the Interior, countersigned by the Commissioner of Patents, and has had affixed to it the seal of the Patent Office, it has passed beyond the control and jurisdiction of that office, and is not subject to be revoked or cancelled by the President or any other officer of the government. United States vSchurz, 102 U.S. 378United States vAm. Bell Telephone Co., 128 U.S. 315, 363. It has become the property of the patentee, and as such is entitled to the same legal protection as other property. Seymour vOsborne, 11 Wall. 516Cammeyer v. Newton, 94 U.S. 225United States v. Palmer, 128 U.S. 262, 271, citing James v. Campbell, 104 U.S. 356. The only authority competent to set a patent aside, or to annul it, or to correct it for any reason whatever, is vested in the courts of the United States, and not in the department which issued the patent. Moore v. Robbins, 96 U.S. 530, 533United States v. Am. Bell Telephone Co., 128 U.S. 315, 364Michigan Land and Lumber Co. v. Rust, 168 U.S. 589, 593. And in this respect a patent for an invention stands in the same position and is subject to the same limitations as a patent for a grant of lands.[60]

Each of these internal citations stands exactly for the proposition for which it is quoted in McCormick, and against the proposition for which McCormick is cited in Oil States.

United States v. Schurz involved the issuance of a land patent for which the rule on revocation was stated as follows: But we have also held that when, by the action of these officers . . . the title to the lands has passed from the government, the question as to the real ownership of them is open in the proper courts to all the considerations appropriate to the case.[61] Later, the opinion relies on Blackstone for the proposition that patents, once issued, “are then perfect grants, and no mention is made of delivery as a prerequisite to their validity. After this, they can only be revoked or annulled by scire facias or other judicial proceeding.[62] It was clear that the decision about land patents in Schurz carried over to patents for inventions, for in American Bell Telephone, the Court concluded that:

The patent, then, is not the exercise of any prerogative power or discretion by the President or by any other officer of the government, but it is the result of a course of proceeding, quasi judicial in its character, and is not subject to be repealed or revoked by the President, the Secretary of the Interior, or the Commissioner of Patents, when once issued.[63]

Other cases follow Schurz’s lead. In Moore v. Robbins, the Supreme Court wrote:

This decision [of the commissioner] is subject to an appeal to the secretary, if taken in time. But if no such appeal be taken, and the patent issued under the seal of the United States, and signed by the President, is delivered to and accepted by the party, the title of the government passes with this delivery. With the title passes away all authority or control of the Executive Department over the land, and over the title which it has conveyed.[64]

Michigan Land & Lumber Co. v. Rust says, to the same effect, After the issue of the patent the matter becomes subject to inquiry only in the courts and by judicial proceedings.[65]

The question in United States v. Stone was whether the United States could revoke a land patent because it had been issued without legal authority.[66] The Stone Court noted that the United States filed a bill in the Federal court of Kansas, against the Indian chiefs and Stone, to have them judicially decreed null, and the instruments themselves delivered up for cancellation.[67] Thereafter, the Court makes no bones about the legal rule:

A patent is the highest evidence of title, and is conclusive as against the government and all claiming under junior patents or titles until it is set aside or annulled by some judicial tribunal. In England this was originally done by scire facias, but a bill in chancery is found a more convenient remedy.[68]

Iron Silver Mining Co. v. Campbell, another land patent case, concluded that such patents are always and ultimately a question of judicial cognizance.[69] Likewise, Noble v. Union River Logging R.R. Co. held that [w]ith the title passes away all authority or control of the Executive Department over the land, and over the title which it has conveyed. . . . The functions of that department necessarily cease when the title has passed from the government.”[70] The same rule was adopted in United States v. Missouri, Kansas & Texas Railway Co., which involved a suit in equity for the cancellation of certain land patents. As in the other cases, it was well understood that once title had passed the matter could not be decided by the executive branch.[71]

F. Patent Rights Are Property Rights

Justice Thomas simply brushes off this uniform line of authority when he says that these precedents are best read as a description of the statutory scheme that existed at the time.[72] Unfortunately, he gets the causal inference exactly backwards. As Justice Gorsuch notes, Justice Thomas’ view ignores the palpable constitutional dimension in the case: “Allowing the Executive to withdraw a patent, McCormick said, ‘would be to deprive the applicant of his property without due process of law, and would be in fact an invasion of the judicial branch of the government by the executive.’[73] These principles apply with equal force to patents for land and patents for inventions. There was no statutory provision for IPR in the 1870 Patent Act because it was universally understood that such a provision would amount to an impermissible encroachment on the jurisdiction of Article III courts.[74] Instead of confessing error on any of these issues, Justice Thomas vainly tries to sidestep Justice Gorsuch’s straightforward constitutional argument by insisting that the constitutional deficit could be cured by amending the unique and anachronistic features of the 1870 Patent Act. But this just repeats his earlier mistakes and fails to explain how the entire body of nineteenth century constitutional law can be erased by a simple statutory fix that no one has ever suggested, let alone attempted.

II. Is Inter Partes Review Good for the Patent System?

A. Pre-Grant v. Post-Grant Review

Justice Thomas second strand of argument is a theoretical justification for the AIA procedures governing IPR:

The primary distinction between inter partes review and the initial grant of a patent is that inter partes review occurs after the patent has issued. But that distinction does not make a difference here. Patent claims are granted subject to the qualification that the PTO has the authority to reexamine—and perhaps cancel—a patent claimin an inter partes review.[75]

Justice Thomas offers no explanation as to why the standards for review should be the same both before and after the patent has issued. But there are strong reasons for the earlier practice that required higher post-grant standards. Before a patent is granted, it is certainly appropriate for the relevant patent authority to take steps to see that it is valid. There is no reason to allow flawed deeds to take effect and forestall other salutary innovations, especially if a patent is likely to be subject to challenges after it is issued. But once that examination is made and the patent has issued, the landscape has changed radically. The initial review process should give rise to some confidence that the patent has been rightly issued so that the odds that a post-issue review will correct some previously unknown error are lower than they would have been if the patent had initially been issued without any review at all. This argument renders problematic all iterations of post-issuance IPR.[76]

B. Patent Quality and Increased Innovation

One might argue that this departure from constitutional text and practice is justified by positive empirical results showing IPR improves patent quality and spurs innovation. But the patent quality issue was nowhere mentioned in Oil States, and the empirical evidence on point does not show any systematic weakness in the patent system—certainly none that could not be cured by beefing up the pre-grant patent examination process.[77] Indeed, adding further layers of review carries its own serious systemic risks. One such risk is that the level of scrutiny during initial review will be reduced as the likelihood of a second review increases. As Professor Dolin observes:

Congress has adopted an overly simplistic approach that can be described as “one set of eyes is good, two is better, three is better still, etc.” But as it turns out, the relationship between patent quality (however defined), certainty of patent rights, and the number of levels of review is not linear. Importantly, more opportunities to challenge issued patents also means more opportunities to engage in abusive practices to undermine legitimate patent rights.[78]

This conclusion should not come as a surprise. The argument for limiting post-grant review of patents is analogous to the argument for adhering to a principle of res judicata, under which a claim that has once been resolved should not be subject to a second review given the additional time, expense, and uncertainty that would necessarily result.[79] That is all the more true given the dicey form of review before the PTAB, with its serious due process weaknesses.

C. Due Process

1. Reliance

In addition, the patentees reliance interest increases after the issuance of a patent. Prior to issuance, the patentee knows that he proceeds to practice or license the patent at his own risk. Hence, the patent applicant has to be cautious in deciding how much to invest in developing the new technology. But once the patent has been issued, the patentee rightly receives a presumption of patent validity, having gone through an exhaustive administrative process.[80] Enabling an infringement defendant to kick a case back to the PTAB at any time will undermine the patentees ability to exploit the patent through use or license. For a patentee who has successfully asserted his patent multiple times in court, going back to PTAB could undercut these victories and resource expenditures. This risk is especially great if the courts are prepared after Cuozzo to defer to the PTABs rules on claim construction; PTAB has adopted a rule that gives claims their broadest reasonable interpretation,which increases the likelihood that patents will be found invalid because they cover material that is already covered by previous patents.[81]

2. Due Process and Separation of Powers

At this point, the question of separation of powers tends to merge with the concerns over procedural due process.[82] The amalgamation of separate powers in one set of hands raises the odds of abuse by undercutting the constitutional system of checks and balances. This theme was evident in the nineteenth century cases dealing with the separation of powers; references to Blackstones Commentaries were not just idle embellishment. But Justice Thomas makes light of the distinctive nature of the American patent experience when he observes:

Based on the practice of the Privy Council, it was well understood at the founding that a patent system could include a practice of granting patents subject to potential cancellation in the executive proceeding of the Privy Council. The parties have cited nothing in the text or history of the Patent Clause or Article III to suggest that the Framers were not aware of this common practice.[83]

He is indeed correct that the Framers were aware of these common practices. But he did not look hard enough to find the abundant evidence of the differences between English and American practice that undercut his argument. As Justice Gorsuch notes in dissent:

While the Court is correct that the Constitution’s Patent Clause was written against the backdrop of English practice, it’s also true that the Clause sought to reject some of early English practice. Reflecting the growing sentiment that patents shouldn’t be used for anticompetitive monopolies over goods or businesses which had long before been enjoyed by the public, the framers wrote the Clause to protect only procompetitive invention patents that are the product of hard work and insight and add to the sum of useful knowledge.[84]

As Professor Adam Mossoff—a leading expert on early American patent law—has shown, the American patent system was organized by Thomas Jefferson in accordance with democratic principles at odds with the highly restrictive rules of English patents.[85] Indeed, it was precisely this rejection of older English practices that led the Framers to insist on the principle of separation of powers so that the president could never claim the powers of a King. This point was made with great force in Federalist No. 69, where Hamilton writes of the president as commander-in-chief:

The President is to be commander-in-chief of the army and navy of the United States. In this respect his authority would be nominally the same with that of the king of Great Britain, but in substance much inferior to it. It would amount to nothing more than the supreme command and direction of the military and naval forces, as first General and admiral of the Confederacy; while that of the British king extends to the DECLARING of war and to the RAISING and REGULATING of fleets and armies, all which, by the Constitution under consideration, would appertain to the legislature.[86]

In exactly the same fashion, the King was never regarded as a role model for the United States on matters related to patents. Indeed, in one striking passage in American Bell Telephone, the Court went out of its way to point to the critical differences between the English monarch and the American president:

But whatever may have been the course of procedure usual or requisite in the English jurisprudence, to enable the king to repeal, revoke or nullify his own patents, issued under his prerogative right, it can have but little force in limiting or restricting the measures by which the government of the United States shall have a remedy for an imposition upon it or its officers in the procurement or issue of a patent. We have no king in this country; we have here no prerogative right of the crown; and letters patent, whether for inventions or for grants of land, issue not from the President but from the United States. The President has no prerogative in the matter. He has no right to issue a patent, and, though it is the custom for patents for lands to be signed by him, they are of no avail until the proper seal of the government is affixed to them.[87]

The passage should stand as a warning against the concentration of power inside the executive branch, because any breakdown in the principle of separation of powers will lead to loss of the liberty which has long been part and parcel of the American constitutional tradition. It is worth repeating that the origin of the doctrine of unconstitutional conditions came in connection with the role of federal courts and by way of early decisions that no state could condition a corporate license to do business within the state on the willingness to sacrifice diversity jurisdiction.[88] The same principle limits the ability of Congress by statute to announce a rule that says henceforth all patents are issued on the condition that they will be subject to IPR. The whole point of the unconstitutional conditions doctrine is to make sure that no branch of government is able to use its unilateral powers to upset the balance of power among the several branches. There is no reason to exempt PTAB from that rule. The failure to respect separation of powers principles is an open invitation to flout the elementary conditions of procedural due process, which the PTAB does in its standard mode of operation.[89]

D. PTAB v. Other Article I Courts

In this regard, it is important to contrast the serious institutional abuses inside the PTAB with the very different situation in the well-established Article I courts that deal with such matters as taxation and bankruptcy. Indeed, as a matter of first principles, I think that the independence of the judiciary would be better protected by giving judges long terms on the bench rather than life tenure, at which point the risks of entrenchment and incompetence become too large.[90] But there is no reason ever to tolerate a system in which so-called judges are appointed to administrative tribunals on a case-by-case basis, or by a head judge who is empowered to tip the balance of a case if any panel previously chosen strays from the PTAB leadership’s preferred vision.

III. Conclusion

If Oil States had been faithful to the precedents that cut overwhelmingly against its holding, the issue of bias in PTAB proceedings would not have to be raised separately under the Due Process Clause. Yet those challenges surely will arise, and sooner rather than later, at which time the Court will have to decide whether to turn a blind eye to the problem, or whether to address the internal abuse of the PTAB on a retail or a wholesale basis. The separation of powers framework should be understood as a consistent effort to nip due process violations in the bud before they infect actual cases, where, once embedded, they will be difficult to detect and to root out. But make no mistake: this issue is percolating even now, and it will present first the lower courts and then the Supreme Court with a set of challenges that they should never have to face in the first instance. But there is a high price to be paid when a clear majority of the Supreme Court signs on to an opinion that goes so wrong in its treatment of legal precedent and constitutional theory on key issues of separation of powers, due process, and unconstitutional conditions. These mistakes assume dramatic proportions in an age where the administrative state has grown too big for its breeches.


[1]      584 U.S. ___, 138 S. Ct. 1365 (Apr. 24, 2018).


[2]      For my commentary on the case after certiorari was granted and after oral argument, see Richard A. Epstein, The Supreme Court Tackles Patent Reform: Why the Supreme Court Should End Inter Partes Review in Oil States, 18 Fed. Soc’y Rev. 116 (2018) [hereinafter OS I] and Richard A. Epstein, Further Reflections on the Oil States Case after Oral Argument Before the Supreme Court, 18 Fed. Soc’y Rev. 124 (2018) [hereinafter OS II].


[3]     Oil States, 138 S. Ct. at 1380-86 (Gorsuch, J., dissenting).


[4]     This is a combination to which I have, it is no secret, been long opposed. See Richard A. Epstein & Kayvan B. Noroozi, Why Incentives for Patent HoldoutThreaten to Dismantle FRAND, and Why It Matters, 32 Berkeley Tech. L.J. 1381 (2017); Richard A. Epstein, The Disintegration of Intellectual Property? A Classical Liberal Response to Premature Obituary, 62 Stan. L. Rev. 455 (2010).


[5]      See eBay Inc. v. MercExchange, LLC, 547 U.S. 388 (2006) (limiting availability of injunctive relief).


[6]      See Quanta Computer, Inc. v. LG Elecs., Inc., 553 U.S. 617 (2008) (narrow reading of exhaustion doctrine that limits the persons subject to suit by patentee).


[7]      See Chevron U.S.A. Inc. v. NRDC, 467 U.S. 837 (1984).


[8]      See Cuozzo Speed Technologies, LLC v. Lee, 579 U.S. ___, 136 S. Ct. 2131 (2016). See also OS II, supra note 2, at 130 (criticizing Cuozzo).


[9]      See 37 C.F.R. §42.108(c) (2017).


[10]     Leahy-Smith America Invents Act, 35 U.S.C. § 66 et seq. (2012). For a critique of the AIA, see OS I, supra note 2, at 116-17.


[11]    OS I, supra note 2, at 116-18. The USPTO defines inter partes review as “a trial proceeding conducted at the [PTAB] to review the patentability of one or more claims in a patent only on a ground that could be raised under §§ 102 [dealing with novelty and prior art] or 103 [dealing with non-obvious subject matter], and only on the basis of prior art consisting of patents or printed publications. United States Patent and Trademark Office, Inter Partes Review, Patent Trial an Appeal Board (May 9, 2017), The relevant passages from that definition are neither quoted nor analyzed.


[12]     Oil States, 138 S. Ct. at 1379.


[13]     Gregory Dolin, Dubious Patent Reform, 56 B.C. L. Rev. 881, 883-85, 896-97 (2015).


[14]     See infra at sections I.E. & I.F.


[15]     Act of December 12, 1980 (Bayh-Dole Act), Pub. L. No. 96-517, ch. 30, § 302, 94 Stat. 3015, 3015 (codified at 35 U.S.C. § 302 (2012)).


[16]     American Inventors Protection Act of 1999, Pub. L. No. 106-113, 113 Stat. 1501 (codified in relevant part at 35 U.S.C. §§ 311–318 (2006)) (repealed 2012).


[17]     See OS I, supra note 2, at 117-18.


[18]     Oil States, 138 S. Ct. at 1373.


[19]     Id. at 1372 (citing Stern v. Marshall, 564 U.S. 462, 488 (2011) and Northern Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 69 (1982)).


[20]     Statute of Monopolies, 1624, 21 Jac. I, c. 3, discussed in OS I, supra note 2, at 118.


[21]     Stern, 564 U.S. at 484 (citing Murray’s Lessee v. Hoboken Land & Improvement Co., 59 U.S. 272 (1856)).


[22]     Northern Pipeline, 458 U.S. at 60.


[23]     Id. at 58.


[24]     See 35 U.S.C. § 6(c) (“Each . . . inter partes review shall be heard by at least three members of the Board, who shall be designated by the Director.”).


[25]     The U.S. Tax Court consists of 19 members appointed by the president for 15-year terms. 26 U.S.C. § 7443 (2018). Assignment of judges follows from their geographic assignment by the chief judge for each trial sessioneach judge is responsible for a portion of the Courts national jurisdiction. Nicholas R. Metcalf & Mary W. Prosser, Litigating a Case Before the U.S. Tax Court, The Federal Lawyer, 34-35 (Aug. 2014). Special trial judges may be assigned by the chief judge of the Tax Court to preside over small tax matters subject to special constraints and the ability of parties to seek review. 26 U.S.C. § 7443A; U.S. Tax Ct. R. 180-183 (July 2012).

Bankruptcy judges are considered judicial officers of the federal district courts and are appointed by the Circuit Courts of Appeals in which they serve for 14-year terms. 28 U.S.C. § 152(a)(1) (2018). These judges follow local rules for case assignment which, as a general matter, mandate random drawings and strict recusal procedures to avoid the appearance of impropriety. See United States Courts, FAQs: Filing a Case,

[26]     285 U.S. 22 (1932).


[27]     Act of March 4, 1927, ch. 509, 44 Stat. 1424; 33 U.S.C. §§ 901-950 (2018).


[28]     Crowell, 285 U.S. at 65.


[29]     Oil States, 138 S. Ct. at 1373 (citing Crowell, 285 U.S. at 50).


[30]     Crowell, 285 U.S. at 50 (emphasis added).


[31]     Id. at 45-46.


[32]     See OS II, supra note 2, 129-30 (citing Cuozzo, 136 S. Ct. at 2131). In several recent decisions, the Federal Circuit has given immediate issue-preclusive effect to its affirmance of PTAB findings of unpatentability. See, e.g., XY, LLC v. Trans Ova Genetics, L.C., 2018 U.S. App. LEXIS 13530, 2018 WL 2324460 (May 23, 2018).


[33]     See, e.g., Freytag v. Commissioner, 501 U.S. 868 (1991) (special trial judges of the Tax Court are inferior officers even though they do not have final authority); Lucia v. SEC, No. 17-130 (U.S. June 21, 2018) (same for SEC administrative judges).


[34]     Crowell, 285 U.S. at 50 (citing Murray’s Lessee, 59 U.S. at 284).


[35]     Id. at 51.


[36]     Id.


[37]     Id. at 52-53.


[38]     59 U.S. 272.


[39]     For a more complete discussion, see OS I, supra note 2, at 118-19.


[40]     Oil States, 138 S. Ct. at 1373 (citing Crowell, 285 U.S. at 50-51 (quoting Bakelite, 279 U.S. 438, 451 (1929)) (internal quotation marks omitted)).


[41]     Bakelite, 279 U.S. at 446-47 (citation omitted) (emphasis added).


[42]     Id. at 452.


[43]     Oil States, 138 S. Ct. at 1373-74.


[44]     Butterworth v. United States ex rel. Hoe, 112 U.S. 50, 59 (1884) (emphasis added).


[45]     United States v. American Bell Telephone Co., 128 U.S. 315, 370 (1888).


[46]     78 U.S. 516, 533 (1870).


[47]     Id. at 533-544.


[48]    Oil States, 138 S. Ct. at 1374 (quoting Gayler v. Wilder, 51 U.S. 477, 494 (1851)) (internal citations omitted). See also 35 U.S.C. § 154(a)(1) (listing the rights of patentees). 


[49]     Oil States, 138 S. Ct. at 1374.


[50]     For an exhaustive discussion of the many layers of this historical difficulty in connection with personal property, see 1 Grant Gilmore, Security Interests in Personal Property ch. 7 (1965). Patents of course were frequently analogized to real property as well. See Horne v. Dep’t of Agric., 576 U.S. ___, 135 S. Ct. 2419, 2425-28 (2015).


[51]     261 U.S. 24, 40 (1923).


[52]     Id. at 33.


[53]     Oil States, 138 S. Ct. at 1375.


[54]     Louisville Bridge Co. v. United States, 242 U.S. 409, 414 (1917).


[55]     Id. at 416.


[56]     Id. at 417.


[57]     It is also worth noting that the rule of construction Justice Pitney gave favoring the public right for government land grants was not followed in patent cases. Historically, the government retained conditions over the subject matter of land grants but not grants of patents. See, e.g., Whitney v. Emmett, 29 F. Cas. 1074, 1080 (CCED Pa. 1831) (No. 17,585); see generally Adam Mossoff, Who Cares What Thomas Jefferson Thought About Patents? Reevaluating the Patent “Privilege” in Historical Context, 92 Cornell L. Rev. 953, 998-1001 (2007).


[58]     Oil States, 138 S. Ct. at 1376.


[59]     McCormick Harvesting Machine Co. v. Aultman, 169 U.S. 606, 608 (1898).


[60]     Id. at 608-09.


[61]     103 U.S. 378, 396 (1880).


[62]     Id. at 398. Scire facias was “a judicial writ requiring a defendant to appear in court and prove why an existing judgment should not be executed against him or her.” The Free Dictionary (accessed June 19, 2018),


[63]     American Bell Telephone, 128 U.S. at 363.


[64]     Moore v. Robbins, 96 U.S. 530, 533 (1878).


[65]     168 U.S. 589 (1897).


[66]     69 U.S. 525 (1864).


[67]     Id. at 528. For a detailed account of Stone, see Gary Lawson, Appallingly Illegal Adjudication: The AIA Through a Constitutional Lens, 41 Geo. Mason L. Rev. ___ (forthcoming 2018). Lawson develops a powerful case that the AIA’s IPR processes do not satisfy the requisites of the appointments clause insofar as it deals with inferior officers. The relevant text of the Constitution reads:

[The President] shall nominate, and by and with the Advice and Consent of the Senate, shall appoint Ambassadors, other public Ministers and Consuls, Judges of the supreme Court, and all other Officers of the United States, whose Appointments are not herein otherwise provided for, and which shall be established by Law: but the Congress may by Law vest the Appointment of such inferior Officers, as they think proper, in the President alone, in the Courts of Law, or in the Heads of Departments.

U.S. Const. art. II, § 2, cl. 2.


[68]     Stone, 69 U.S. at 535.


[69]     135 U.S. 286, 293 (1890).


[70]     147 U.S. 165, 176 (1893).


[71]     Missouri, Kansas & Texas Railway Co., 141 U.S. 358 (1891).


[72]     Oil States, 138 S. Ct. at 1369.


[73]     Id. at 1385 (Gorsuch, J., dissenting) (quoting McCormick, 169 U.S. at 612).


[74]     This is the principal lesson gleaned from review of the cases cited by Justice Thomas for the opposite proposition. See supra sections I.A.-I.E.


[75]     Oil States, 138 S. Ct. at 1375 (citing Cuozzo, 136 S. Ct. at 2137).


[76]     Furthermore, although the point was not raised in Oil States, it seems clear that no breach in the wall between the executive and judicial branches should be tolerated. The 1981 and 1999 statutes that introduced more limited versions of IPR should be considered just as unconstitutional as the more invasive provisions of the AIA.


[77]     See Dolin, supra note 13; Neal Solomon, The Myth of Patent Quality (Sept. 14, 2017),, discussed in OS I, supra note 2, at 116-17.


[78]     Dolin, supra note 13, at 883. For an empirical investigation, see id. at Part IV, 923–31 (noting multiple challenges to key patents, and the costs of defense and delay even when the patents are found valid by IPR).


[79]     See Allen v. McCurry, 449 U.S. 90, 94 (1980) (Stewart, J.) (“As this court and other courts have often recognized, res judicata and collateral estoppel relieve parties of the costs and vexation of multiple lawsuits, conserve judicial resources, and by preventing inconsistent decisions, encourage reliance on adjudication.”).


[80]     35 U.S.C. § 282(a) (2018).


[81]     See OS II, supra note 2, at 130 (discussing Cuozzo, 136 S. Ct. at 2136, 2142-44).


[82]     For an argument that they are explicitly connected, see Nathan S. Chapman & Michael W. McConnell, Due Process as Separation of Powers, 121 Yale L.J. 1672, 1681-86 (2012).


[83]     Oil States, 138 S. Ct. at 1377.


[84]     Id. at 1385 (Gorsuch, J., dissenting) (internal quotations and citations omitted).


[85]     Justice Gorsuch quotes Prof. Mossoff’s work: “[A]n American patent in the late eighteenth century was radically different from the royal monopoly privilege dispensed by Queen Elizabeth or King James in the early seventeenth century. Patents no longer created, and sheltered from competition, manufacturing monopolies—they secured the exclusive control of an inventor over his novel and useful scientific or mechanical invention.” Id. at 1382 n.3 (citing Mossoff, supra note 57, at 967–68).


[86]     The Federalist No. 69 (Alexander Hamilton), available at


[87]    American Bell Telephone, 128 U.S. at 362-363.


[88]     See OS II, supra note 2, at 127; Insurance Co. v. Morse, 87 U.S. 445 (1874); Terral v. Burke Construction Co., 257 U.S. 529 (1922).


[89]     See discussion of Stern and Northern Pipeline, supra at section I.A.; OS I, supra note 2, at 122.


[90]     See Richard A. Epstein, Mandatory Retirement for Supreme Court Justices, in Reforming the Court: Term Limits for Supreme Court Justices 435 (Roger C. Cramton & Paul D. Carrington, eds., 2006).