Guest post by Professor Andrew C. Michaels (University of Houston Law Center)
With Arthrex certiorari petitions hitting the Supreme Court over the past weeks, one issue that seems to be underappreciated is the judicial retroactivity question raised by Judge Dyk in his Bedgear concurrence, which was joined by Judge Newman. It is fair to say that these two venerable judges don’t always agree, so on the rare occasion when they team up for a concurrence one might reasonably expect that they have a serious point. They do, for the dozens of remands now stayed before the USPTO pending the petitions were unnecessary and at least mostly imprudent under the law of retroactivity as properly understood. If the Supreme Court were to take up this issue, it would have an opportunity to eliminate this wasteful multiplication of administrative hearings, as well as to clarify retroactivity law.
The Supreme Court has stated: “The principle that statutes operate only prospectively, while judicial decisions operate retrospectively, is familiar to every law student.” Rivers v. Roadway Express, Inc., 511 U.S. 298, 311-12 (1994). But this fundamental principle was overlooked by the Federal Circuit in its Arthrex decision, where the court ruled that the relevant Administrative Patent Judges were “not constitutionally appointed at the time” they had previously issued final appealable decisions, despite the fact that the court at least purportedly cured the unconstitutionality with its as-applied severance of removal restrictions. See Arthrex, Inc. v. Smith & Nephew, Inc., 941 F.3d 1320, 1338-39 (Fed. Cir. Oct. 31 2019).
The Arthrex opinion’s notion that the APJs were not constitutional prior to the decision’s Halloween day release, but then spookily became constitutional as of that day, flies in the face of the Court’s basic retroactivity doctrine, under which the as-applied severance fix should have been considered retroactive. The court’s decision to treat Halloween 2019 as the “effective date” of its as-applied severance was inappropriately more legislative than judicial in character. See Paul J. Mishkin, Forward: The High Court, The Great Writ, and the Due Process of Time and Law, 79 Harv. L. Rev. 56, 65-66 (1965) (“the question of an effective date . . . smacks of the legislative process; for it is ordinarily taken for granted . . . that judicial decisions operate with inevitable retroactive effect”).
The Federal Circuit viewed remands as required by the Supreme Court decision in Lucia, despite the fact that in Lucia there was no judicial fix to make retroactive. If there was a fix in Lucia it came from the agency after the relevant ALJ had heard the case, and was thus properly considered prospective only because the other branches presumptively act prospectively whereas the judiciary at least presumptively acts retroactively. See Lucia v. SEC, 138 S. Ct. 2044, 2055 n.6 (2018); 783 Fed. Appx. 1029 (Fed. Cir. Nov. 7, 2019) (Dyk, J., concurring in the judgment, joined by Newman, J.).
The judicial retroactivity principal holds for statutory alterations and invalidations. When a court invalidates a statute, courts generally should treat the invalid statute as though it never existed in the first place. See Reynoldsville Casket Co. v. Hyde, 514 U.S. 749, 759-60 (1995) (Scalia, J., concurring) (“In fact, what a court does with regard to unconstitutional law is simply to ignore it. It decides the case ‘disregarding the unconstitutional law,’ . . . because a law repugnant to the Constitution ‘is void, and is as no law.’”) (quoting Marbury v. Madison, 5 U.S. 137 (1803), and Ex parte Siebold, 100 U.S. 371, 376 (1880)).
Similarly, when a court interprets a statute, the newly announced statutory construction is properly considered to have been the law all along. See, e.g., DIRECTV, Inc. v. Imburgia, 136 S. Ct. 463, 469 (2015) (“judicial construction of a statute ordinarily applies retroactively”); Rivers, 511 U.S. at 312-13 (“A judicial construction of a statute is an authoritative statement of what the statute meant before as well as after the decision of the case giving rise to that construction.”).
Concurring in the court’s denial of rehearing en banc, Judge O’Malley appeared to suggest that judicial severance is an exception to these general retroactivity principals, claiming that the Court’s decision in Booker makes clear that judicial severance is “necessarily a prospective act.” See Arthrex, 953 F.3d at 768 (citing United States v. Booker, 543 U.S. 220, 268 (2005)). This assertion is contrary not only to retroactivity doctrine in general but also even to Booker itself, which in fact held that its judicial severance did have to be considered retroactive. See Booker, 543 U.S. at 268 (“we must apply . . . our remedial interpretation of the Sentencing Act – to all cases on direct review”) (citing Reynoldsville, 514 U.S. at 752; Harper v. Virginia Dep’t of Taxation, 509 U.S. 86 (1993)). Nothing in Booker carves out a judicial severance exception to foundational principles of judicial retroactivity.
The reason that a remand for rehearing was appropriate in Booker (despite the retroactivity of the severance) was that the prior statutory misrepresentation of law clearly made a difference, in that it led to Mr. Booker receiving a longer criminal sentence than he properly could have under the corrected statute. See Booker, 543 U.S. at 227, 245-46. Similarly in Harper, taxes had been collected under an invalid tax statute, so the Court remanded for state courts to consider refunding the taxes. See Harper, 509 U.S. at 102. In both of these cases, governmental action had been taken that adversely affected a party and could not have been taken under the law as correctly understood. Retroactivity doctrine provides for discretion to remedy these situations when equities so dictate. See, e.g., James B. Beam Distilling Co. v. Georgia, 501 U.S. 529, 543-44 (1992) (“nothing we say here precludes consideration of individual equities when deciding remedial issues in particular cases”).
The key difference though is that in Arthrex, the prior misrepresentation of law had no such effect. There is no apparent reason to think that any of the remanded cases would have been decided differently if the relevant APJs had known that they were in fact removable at will, and any argument that actual harm is present would seem to be a tenuous. Absent actual harm, a discretionary remand was at most prudent in the Arthrex case itself as an incentive creating reward for first winning the Appointments Clause challenge (and even that is questionable), but not in the dozens of other remanded matters.
Concurring in the denial of rehearing en banc, both Judge O’Malley and Judge Moore pointed to a government brief as having “rejected” Judge Dyk’s retroactivity argument. See Arthrex, 953 F.3d at 764 and n.3, 767 (citing Supp. Br. of United States, Polaris v. Kingston, Nos. 2018-1768, -1831, at 13-14). That brief asserts that the court’s as-applied severance was not “sufficient to eliminate the impact of the asserted constitutional violation on the original agency decision,” but tellingly provides no suggestion of what that impact might have been. In any event, the requirement of at least presumptive judicial retroactivity is rooted in Article III and the Court’s jurisprudence, and cannot be overridden or waived by an executive branch brief. See James B. Beam, 501 U.S. at 549 (Scalia, J., concurring in the judgment) (“‘the judicial Power of the United States’ . . . Art. III, § 1, must be deemed to be . . . the power to say what the law is . . . not the power to change it”).
Apart from the lack of even an assertion of any actual harm caused by the prior statutory mirage of removal restrictions, such harm is unlikely also because the Supreme Court has applied standing (specifically traceability) requirements rather loosely in the Appointments Clause context. See Free Enterprise Fund v. Pub. Co. Accounting Oversight Bd., 561 U.S. 477, 511-12 n.12 (2010). It is one thing to allow litigants to raise Appointments Clause challenges that make no likely difference to their case, but it is another to retrospectively vacate prior agency actions that were almost certainly unaffected by those issues, especially where doing so is not required by and in fact runs counter to the Court’s retroactivity jurisprudence.
Finally, concurring in the denial of rehearing en banc, Judge Moore downplayed the disruption of the unnecessary rehearings, stating that the Arthrex decision would result in at most eighty-one remands. See Arthrex, 953 F.3d at 764 n.4. Squaring this statement with the PTAB general order issued on the first of May staying over one-hundred remanded matters (and expecting more to come) pending certiorari petitions would seem to be more than trivial. Regardless, even if the PTAB chooses not to reopen briefing or the record, a new hearing before a new panel of APJs plus a new final written decision subject to a new appeal, in each of the dozens of remanded matters, is not without significant expense, disruption, and waste.
More to the point though, the disruption is legally improper, as explained further in my forthcoming article, Retroactivity and Appointments, available here: http://ssrn.com/abstract=3650166