[8/7/2018 Update: In an order issued today by a Ninth Circuit three-judge panel where Judge Graber has replaced Judge Reinhardt, the newly constituted panel states that “[t]he Opinions filed July 24, 2018, are hereby withdrawn to allow time for the reconstituted panel to confer on this appeal.”]

Today the Ninth Circuit issued a 2-1 decision in a very important tax and administrative law case, Altera v. Commissioner. My co-blogger Andy Grewal had a short post about one aspect of the case last year: the amici law professors’ suggestion to remand without vacatur. And two groups of tax professors filed amicus briefs in support of the IRS, and those briefs are definitely worth reading (here and here). Importantly, both briefs reject tax exceptionalism, as the Ninth Circuit majority also purports to do here, and instead argue that the IRS action under review is substantially and procedurally proper under the Administrative Procedure Act and Internal Revenue Code.

Despite my scholarly interest in tax exceptionalism, I have generally not covered the substance of the case on the blog, as I represented the U.S. Chamber of Commerce as amicus curiae in this case. That brief is available here. I try to avoid mixing my academic and of counsel roles when blogging or otherwise engaged in academic work. But I did contribute a short post with some reflections on the administrative law aspects of the case to The Surly Subgroup as part of the blog’s 2017 Mini-Symposium on Tax Enforcement and Administration. Here’s the key part of the Chamber’s amicus brief, which I also reproduced for the online symposium:

This case is illustrative. Before the Tax Court, the IRS sought Chevron deference for its statutory interpretation, but disagreed that the regulation at issue is “a legislative rule,” which would subject it to all of the APA’s notice-and-comment rulemaking requirements. ER49.

Similarly, the IRS argued that administrative law’s reasoned decisionmaking requirement, as articulated by the Supreme Court in Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Auto Insurance Company, 463 U.S. 29 (1983), did not apply to this rulemaking, even though the new Treasury regulation departs from the IRS’s prior position without reasoned explanation. ER51. In arguing why the reasoned decisionmaking standard that applies to the rest of the regulatory state should not apply to the IRS, the IRS claimed that “the Supreme Court has never, and [the Tax] Court has rarely, reviewed Treasury regulations under State Farm.” ER55.

Moreover, as detailed at length in Altera’s opening brief (at 38-70), now that the IRS is in court, it has essentially abandoned the stated rationale for the regulation (the arm’s-length standard) and, instead, attempted to defend its rule on a new argument (the commensurate-with-income standard). Apparently the IRS does not feel it should be bound by the Chenery doctrine, which holds that an agency action “can-not be upheld unless the grounds upon which the agency acted in exercising its powers were those upon which its action can be sustained.” Chenery I, 318 U.S. at 95. Nor does the IRS seem to consider itself bound by Supreme Court precedent that an agency must provide reasons in the rule itself for changing its position.

The dangers inherent in the IRS’s tactics should be plain: the IRS wants to take advantage of the agency discretion afforded by judicial deference doctrines that apply to administrative interpretations of law without also being bound by the constraints administrative law imposes on federal agency action in order to ensure an agency’s discretion is not exercised in an arbitrary and capricious manner. The Supreme Court, the D.C. Circuit, and now the Tax Court have rejected any such claims of tax exceptionalism. This Court should send a clear message to the IRS that it must play by the same rules of the road that govern the rest of the federal regulatory state.

On the merits, the dissent filed by Federal Circuit Judge O’Malley, who was sitting by designation on the Ninth Circuit, largely tracks this argument regarding the IRS’s failure to engage in reasoned decisionmaking.

But as the title of this post suggests, I’m very uncomfortable with the judicial decisionmaking process here. This is a 2-1 decision with Judge Reinhardt casting the deciding vote in the majority. Judge Reinhardt passed away on March 29, 2018, nearly four months before the opinion was published. As the opinion’s author is Chief Judge Thomas, who has always struck me as a careful and deliberate judge, I’m sure he conferred with all of his colleagues on the Ninth Circuit before deciding to publish this decision. The case was argued way back in October 2017, so the panel no doubt weighed the costs of additional delay and reargument.

And, to be sure, the opinion itself notes that “Judge Reinhardt fully participated in this case and formally concurred in the majority opinion prior to his death.” In other words, I infer that Chief Judge Thomas had likely circulated his opinion prior to Judge Reinhardt’s death, and Judge Reinhardt had reviewed it and sent the formal “I’m pleased to join” notification. But Judge O’Malley likely hadn’t drafted, or at least circulated, her dissent prior to Judge Reinhardt’s death, hence the four-month delay in publication.

As Howard Bashman has noted, #appellatetwitter folks disagree about what to do with judges’ votes in opinions issued after they die. Consider me firmly among those who find this practice of dead-hand voting quite unsettling, especially to cast the deciding vote and even more so if the deceased judge did not have an opportunity to review and respond to the dissent. On the substance of the administrative law issues implicated by this case, I wouldn’t be surprised to see the Ninth Circuit reconsider the case en banc and/or the Supreme Court deciding to review the case. But I wonder how this procedural oddity will factor, if at all, into the decision to grant further review.