On August 7, 2018, the Ninth Circuit panel vacated its earlier opinion in Altera v. Commissioner that, in a nutshell, had concluded that Treasury had complied with the Administrative Procedure Act (the “APA”) in issuing final regulations requiring that cost-sharing agreements take into account stock-based compensation costs such as employee stock options. That decision was 2-1, with Judge Reinhardt in the majority.[1] Judge Reinhardt’s death earlier this year precipitated a change in the composition of the deciding panel. Ninth Circuit rules required a replacement of Judge Reinhardt by lot, resulting in Judge Susan Graber as the new third panel member. The panel has scheduled a new oral argument for October 16.

On one level, the odds are 50-50 the court will affirm its former holding because the two holdover judges took opposite sides in the original decision. However, this would be a simplistic outlook. The “reconstituted panel” will now confer according to the August 7 order and also will hear oral argument. The panel will enter into a new dialogue that could result in a different opinion, which could be either a reversal or affirmance of the Tax Court. Recall the Ninth Circuit’s initial 2-1 decision in Xilinx v. Commissioner reversing the Tax Court and its subsequent withdrawal of that decision and issuance of a 2-1 decision affirming the Tax Court. We have discussed this previous history here and here.

Xilinx is noteworthy not only for its decisional history but also for its fundamental ruling. In Xilinx, the question was whether stock-based compensation costs were required to be shared by related parties pursuant to a regulation that required “all costs” to be shared. The issue arose because parties acting at arm’s length did not share stock-based compensation costs and a fundamental transfer pricing regulation enshrined the arm’s length standard as a measure for all transactions. Ultimately, the Ninth Circuit decided the arm’s length regulation must govern, but it did so in a way, in this author’s view, that answered the core question presented in Altera. As we previously observed:

The main conclusion [of Xilinx] is not that there is an irreconcilable conflict between the two regulations, but that their apparent conflict creates an ambiguity regarding what is the standard for determining which costs must be shared in an intangibles CSA. This is an interesting maneuver around what had been described as a “winner takes all” conflict between the two pertinent regulations. It allowed the majority to appeal to a reading of the regulations which comports with the purpose of the regulations versus a mechanical application of the specific versus general rule of construction. The majority reasoned that the purpose of both the regulations and the statute would be undermined by allowing the all costs regulation to trump the arm’s length standard. It then supported its view by referencing the U.S-Irish Tax Treaty enshrinement of the arm’s length standard and concluded that the standard of interpretation which applies is the arm’s length standard. Though the majority opinion did not specifically state, it suggested that the all costs regulation must be applied so as to allow for the operation of the arm’s length principle rather than holding that the all costs regulation is invalid.

In other words, the best reading of Xilinx is that the arm’s length principle acts as a kind of litmus test, rendering some expenses to be disregarded as “costs” for transfer pricing purposes. Thus, the “all costs” regulation must be understood to mean “all costs which are shared by parties at arm’s length.” Therefore, where a regulation, such as the one at issue in Altera, specifically requires related parties to share stock-based compensation costs, it is an empty rule to the extent that parties at arm’s length do not share the costs in question. If that is true, then Treasury’s exercise of regulatory power forcing taxpayers to share such costs cannot be anything other than arbitrary and capricious conduct, the very type of conduct Congress intended to prevent through enactment of the APA.

Certainly the question of what Xilinx held is one among several key questions now facing the panel. However, perhaps the more important and underlying decision is whether to affirm Treasury’s authority to decide important economic matters ipse dixit or to affirm the rule of law that measures all transfer pricing transactions according to the arm’s length principle.