Interview: Transfer Pricing Litigation Update II

David D. Stewart: Welcome to the podcast. I'm David Stewart, editor in chief of Tax Notes Today International. This week: trials and tabulations.

Periodically, we like to take a look at the state of transfer pricing litigation here in the United States, and since our previous episode on the subject last year, there have been two major decisions.

Eaton and Medtronic cover two vastly different areas, but continue to develop the new landscape, where, rather than routine IRS losses, we're seeing much more mixed results.

I'll be joined by Tax Notes contributing editor Ryan Finley to talk more about this. Ryan, welcome back to the podcast.

Ryan Finley: Thanks for having me.

David D. Stewart: Since the last time you were here, I understand there have been two major decisions. Could you tell us about them?

Ryan Finley: The first is the Medtronic II opinion, which the Tax Court issued in August. It's the decision that comes after the case was remanded by the Eighth Circuit in 2018, which came after the Tax Court decided the case largely in Medtronic's favor in 2016.

The other one has to do with the Eaton case, which has to do with advance pricing agreements and cancellation. That case was decided by the Sixth Circuit, and it basically affirmed the Tax Court opinion from 2017.

David D. Stewart: Why don't we go in reverse chronological order and dive into the Eaton case first. Could you give us some background on the company?

Ryan Finley: Sure. Eaton Corp. is a U.S.-based multinational that manufactures electrical equipment — devices that have to do with power management — and they manufactured some of these devices at offshore plants in Puerto Rico and Dominican Republic. It was those transactions that were the subject of the advance pricing agreement at issue in the case.

David D. Stewart: What was the main issue going to trial?

Ryan Finley: The main issue was whether the IRS was within its rights to cancel a pair of advance pricing agreements that it had entered into with Eaton Corp.

Basically, Eaton made a number of acknowledged and self-reported, but major, errors in its compliance with the terms of the advance pricing agreement, and in its annual compliance reports that it had to release under the advance pricing agreement.

On that basis, the IRS said that these were under the relevant cancellation standards, which allow the IRS to cancel for material admissions effect and material misrepresentations. Basically, whether the IRS had the right to cancel these advance pricing agreements on the basis of these inadvertent, but major, errors.

David D. Stewart: Taking one quick step back for anybody that may not be familiar, what are these advance pricing agreements?

Ryan Finley: An advance pricing agreement is basically an audit and settlement before the fact in transfer pricing. Companies that want to reach agreement on the transfer pricing — usually for complicated transactions, to avoid an audit, an exam, and potentially significant exposure down the line — enter negotiations with the IRS to form a contract that stipulates, "Here's how our transfer pricing is going to work."

There are compliance obligations, but generally speaking, that advance pricing agreement is binding on both parties. The taxpayer has to follow it, and the IRS has to respect what it agreed to, subject to these exceptions that allow cancellation or, in even more egregious situations, revocation.

David D. Stewart: What happened at the Tax Court when this case came up for trial?

Ryan Finley: In 2017 the Tax Court held essentially that the errors could not satisfy the materiality standard laid out in these revenue procedures.

There are two revenue procedures that detail the whole advance pricing agreement process, compliance, negotiation, et cetera, and they set out the standard for cancellation. Applying those standards, Judge Kathleen Kerrigan held that this kind of situation, where you have a self-reported, self-corrected, but again, still major error in compliance, those sorts of errors cannot satisfy the specific materiality standard that allows the IRS to cancel an advance pricing agreement. It was a win for Eaton.

Importantly, the judge did side with the IRS on the standard of review. On the one hand, the IRS said its decision to cancel the advance pricing agreement should be reviewed, whereas Eaton wanted it to be looked at more as a contract between equal parties, equally binding on both sides, and cancellation would only be permitted to the extent that the contract allows it.

David D. Stewart: Then this case went up on appeal, and what happened at the appeals court?

Ryan Finley: Technically, this was a cross-appeal, because there were some secondary issues involved regarding the assessment of section 66.62 penalties.

But the main issue was this cancellation decision, and it was a clear-cut win for Eaton. The Sixth Circuit held that based on parsing the language in the revenue procedures regarding cancellation, the kind of noncompliance that the parties agreed had taken place would not fall within that standard of materiality.

The Sixth Circuit affirmed the outcome but rejected the Tax Court's holding that the standard of review was for abuse of discretion. The Sixth Circuit said, "No, this is just a contract," and it was up to the IRS to approve that under the contract; it had the right to cancel it.

David D. Stewart: What sort of ramifications does this decision have going forward?

Ryan Finley: It is kind of an isolated case, in that if you look at advance pricing agreement statistics, there are very few. It's a very small percentage that the IRS tries to cancel or revoke, so it probably does not affect a ton of taxpayers.

But an appellate court's holding essentially that the IRS is not going to be given much leeway in its decision to cancel or revoke an advance pricing agreement could potentially make the IRS a little more wary of entering in advance pricing agreements in the first place.

I would note that the IRS is now working on a new revenue procedure governing advance pricing agreements, and it's possible that its experience from this case will be reflected in the cancellation standards in the new revenue procedure.

On the other hand, it does solidify the binding force of an advance pricing agreement for taxpayers. In that sense, it could be pro-taxpayer. They could be more confident that courts will really strictly scrutinize any IRS decision to try to cancel the advance pricing agreement.

David D. Stewart: Let's turn to the Medtronic decision. What happened?

Ryan Finley: This is obviously Medtronic II, so the case has quite a history. I believe the petition was filed in 2011, so this case has been going on for quite a while.

It's really representative of this recurring battle between the IRS and taxpayers over the selection of a transfer pricing method — especially transactions that involve the transfer of intangibles, usually developed in the United States, by a U.S. parent, to what's usually a low-taxed, offshore subsidiary.

Taxpayers in Medtronic, and cases like it, tend to favor transactional methods like this method called a comparable uncontrolled transaction method, or CUT method. For various reasons, this often yields a lower royalty rate for the licensor than what the IRS tends to prefer, which is the comparable profits method. The comparable profits method leaves the licensee in the offshore jurisdiction with less of the total profit attributable to the controlled transaction.

David D. Stewart: Taking a step back just real quickly, could you tell us about Medtronic as a company and what this transaction is about?

Ryan Finley: Sure. Medtronic's a very prominent, U.S.-based medical device manufacturer, specifically implantable medical devices. They're best known for their cardiovascular implantable devices — pacemakers, things like that. They also have a significant implantable neurological device business.

Those two business lines were what were issue in the case. The specific transaction that the case centered around was Medtronic U.S.'s license of basically all the IP necessary to manufacture these cardiological and neurological devices to a subsidiary that manufactured them in Puerto Rico.

The license covered patents. It covered related kind of know-how and technology that would be necessary to successfully manufacture these devices according to the standards that they have to meet for implantable medical devices.

The case is about what should the royalty rate be for MPROC [Medtronic Puerto Rico Operations Co.], this Puerto Rican manufacturing subsidiary. What royalty should it pay Medtronic U.S. for the right to use these intangibles to manufacture these devices?

David D. Stewart: You mentioned that this has been going on since 2011. What has happened in the case over the last 11 years?

Ryan Finley: Right. The first court opinion was in 2016 by the Tax Court. It was an opinion also written by Kerrigan. It did not entirely accept Medtronic's position, but on this core question of whether the CUT method or the comparable profits method was the best method, the Court squarely sided with Medtronic. Kerrigan, in her 2016 Medtronic I opinion, found that additional royalty rate adjustments to raise that royalty were necessary, but substantially, she held in favor of Medtronic.

The IRS subsequently appealed the case, basically arguing that the transaction that Medtronic had used as a comparable in its CUT method analysis was not a reliable comparable. There are too many differences. Too many adjustments were necessary to try to bring it in line with the license to this MPROC subsidiary, and the result was not reliable.

Specifically, the IRS claimed that the Tax Court had failed to really consider and apply the comparability standards that the CUT regulations stipulate.

The Eighth Circuit agreed, and they vacated the Tax Court's decision and remanded the case, basically saying, "We need more factual development to review your decision that this comparable..."

And the comparable was a license. It was actually a litigation settlement agreement with Siemens Pacesetter Inc., from a decade before the controlled license at issue.

But there are numerous differences between that and the MPROC license. Essentially, the Eighth Circuit directed the Tax Court to make the factual findings necessary to determine whether this Pacesetter license was really a reliable comparable, and to use it as the reference point to price this MPROC license.

After that, in 2021, the Tax Court held its Medtronic II trial, and that trial was based on what the Eighth Circuit opinion said. It was focused on essentially a reassessment of which method was more reliable.

But in particular, Kerrigan was especially interested in the possibility that comparability adjustments could be tacked onto one of the party's methods, to sort of "bridge the gap between the results of the two parties' methods," as she often said.

That was followed by about a year of post-trial briefing. In general, the parties stuck to their original positions. For Medtronic it was that the CUT method was the best method. For the IRS it was that the comparable profits method was.

But they based their positions on Kerrigan's invitation at the end of the Medtronic II trial to propose an unspecified method, which is something that in some circumstances the regulations allow. It's basically a method that the regulations do not expressly acknowledge or describe, but they offer that possibility that in some cases maybe a method that the regulations don't identify could be the best method.

David D. Stewart: All right, so the rules allow for unspecified methods, and Medtronic is arguing for them. What did the court do with that?

Ryan Finley: The court accepted the unspecified method proposed by Medtronic as the framework for what it ultimately found to be the most reliable method, but sort of tweaked the quantitative parameters in a way that materially affected the outcome.

Kerrigan did accept that method conceptually as the foundation for how she decided Medtronic II.

It's an interesting method. It sort of resembles the residual profit split method, in that it uses traditional comparables-based transfer pricing methods in the first couple steps, to give the parties a certain share of the "system profit," as they call it, and then basically splits up what remains based on some allocation percentage. But that's where the similarities end with the profit split method.

As part of those steps, step 1 was basically to give to Medtronic U.S. a royalty rate, after adjustments, that was drawn from this Pacesetter agreement. It was the same license that was used as the basis for the CUT method, but in this case, it was nominally the first step in the unspecified method.

Medtronic gets the royalties that were determined using the royalty rates for the Pacesetter agreement, subject to upward adjustments, but not including any adjustment for profit potential, because that was addressed in a later step.

The second step gave MPROC a return based on the comparable profits method, a nod to what the IRS thought ought to be done. Medtronic proposed splitting that remaining pot in more favorable percentages for MPROC. Kerrigan held that the most reasonable split of that remaining profit pool was 80 percent to Medtronic U.S. and 20 percent to MPROC.

It's important to not get lost in the steps here. This really is, in essence, the CUT method, because it starts with the royalty rate drawn from a comparable uncontrolled transaction. It's just that this more byzantine and indirect way of calculating adjustments is not something that would ever be permitted under the actual CUT method regulations.

It raises the issue of whether this was just a noncompliant application of the CUT method or if it really was in fact an unspecified method.

David D. Stewart: I'm curious about this issue of unspecified methods. The regulations have set out certain ways that you're supposed to divvy up profits between entities, but then they have this one odd section of make something up. How is that supposed to work?

Ryan Finley: Yeah, it's a good question, and it's something that really hasn't been tested a whole lot in litigation under the current regulatory scheme. There's a subsection of the regulations dealing with controlled and tangible transfers that says that taxpayers can apply methods other than the specified methods.

They can apply an unspecified method subject to two conditions. The first is that it really is the best method under the general kind of principles for identifying the best method. The second, which is sort of built into the first, is that it complies with the realistic alternatives principle, which was something that was actually added to the statute by the Tax Cuts and Jobs Act.

But the realistic alternatives principle essentially says that for an unspecified method to be the best method, it can't leave one of the parties worse off than it would have been had the party engaged in a realistically available alternative transaction.

Essentially, the forgone profit, or maybe the forgone sales revenue that flowed from entering the controlled transaction, serves as the baseline to measure the outcome of the unspecified method.

There's an example in the regulations that the situation deals with, a U.S. parent. They license the IP necessary to manufacture an industrial adhesive to a foreign subsidiary, and they charge X as a royalty for the European subsidiary to make that stuff and sell it in Europe.

However, under the assumed facts, the U.S. parent would have earned far more than that royalty income had it just sold directly into the European market itself. That was its realistic alternative, and that was the forgone profit, and because that number is much greater than the amount of royalty income that it was getting under the arrangement that it did enter, then that cannot have been an arm's-length transfer price.

An unspecified method basically has to respect this sort of principle for evaluating the arm's-length nature of a transfer price.

David D. Stewart: The court accepted Medtronic's basic method, so was this an unqualified win for Medtronic?

Ryan Finley: Well, no. It was not an unqualified win.

As I said before, Kerrigan repeatedly expressed an intention, or at least an aspiration, to find a middle ground between the parties' positions. The 80-20 split that she found to be appropriate led to a royalty rate that fell very near the halfway point between the royalty rates that each party said were correct.

David D. Stewart: Eleven years of litigation, and we wound up just the average between the two?

Ryan Finley: It's interesting. There's language in the opinion preemptively denying that that's what was going on. Kerrigan said that she's not simply taking the average of two methods. In her opinion, she wrote that she thought that this approach was sound, and overall led to a reasonable outcome.

Now, the changes she made, the dollar amount implications, would significantly raise Medtronic's tax bill by hundreds of millions or billions of dollars, depending on how many tax years you're considering, relative to the original 2016 opinion.

On the other hand, it was still considerably less than what the IRS said it ought to be, and it introduces this whole new kind of wrinkle in transfer pricing litigation, of whether you can use this catch-all unspecified method to make up what you think gets you to your preconceived idea of an arm's-length or reasonable outcome.

Under the regulations, an unspecified method can't just be whatever gets you to the halfway point. It has to conform to these reliability end principles and the realistic alternatives principle, and it's really not clear from the opinion whether that requirement was seriously taken into account when Kerrigan found that this was the right method.

It's a loss for both parties, in some sense. But I would caution that it is very possible that there will be a second appeal in this case. If I had to guess, I'd say there probably will be, just because of how important the legal issues are and how much money is at stake in this case and in related cases that are ongoing.

But for now, we can say that this case is not necessarily over. We don't really know who won yet, because it could still be appealed.

David D. Stewart: What sort of danger is lurking out there for the IRS if it doesn't win on appeal?

Ryan Finley: I think the greatest danger is that the IRS and Treasury wrote the CUT method regulations in such a way with very strict comparability and reliability standards.

If you read through it, the standards are far more explicit and prescriptive than what you'll find usually for other transfer pricing methods. They did that on purpose, because there's a ton of money at stake in these outbound IP transfers, and there's a high risk that really valuable, unique IP could be transferred for less than what it's worth by using unreliable transactional comparables.

This raises the risk that taxpayers can pick and choose which of those CUT method requirements they would like to follow, disregard others, and simply call it an unspecified method, and in so doing, circumvent the requirements they don't like.

As I said before, the later steps in the unspecified method accepted by the court really function as a profit potential adjustment, and in my view, the best reading of the regulations is that a significant difference in profit potential outright disqualifies an uncontrolled transaction as a comparable. It's simply not something that could be adjusted for.

If you look at this method as the CUT method simply trying to go under another name, then it's something that's pretty clearly at odds with the regs. The risk is that taxpayers would be able to sidestep those by just renaming the method.

David D. Stewart: Does this case have any direct implications on other cases in the pipeline?

Ryan Finley: There are two trends and issues that I would highlight, and both of them are actually related to Medtronic.

The first is the Amgen case, which is in early stages of litigation. Based on what appears in the Tax Court petition, and then Amgen's securities filings describing the situation, it's eerily similar to Medtronic.

There's a Puerto Rican manufacturing facility. They seem to be, from the filings, hinting at the selection of method. It really sounds like Medtronic III, only the dollar amounts at stake are an order of magnitude greater. It raises the stakes for any possible Medtronic appeal.

The other case I'd mention is Coca-Cola. This sort of came up in the post-trial briefing, whether the result in Coca-Cola requires the Tax Court's acceptance of the IRS's favored comparable profits method analysis in this case.

In Coca-Cola, it was widely regarded as a major win for the IRS. They succeeded in getting the Tax Court to agree that the comparable profits method, and not other methods, including the CUT method, were the best method.

David D. Stewart: Are there any other cases you're hoping to see a decision on in the near future?

Ryan Finley: At one point, I was hoping to see an opinion in the 3M case. I don't know whether hopes are warranted at this point, given that it's been fully briefed for six years, but you know, you never know.

Maybe at one point, we'll get a 3M opinion, and if that were to happen, it would be a very big deal. It's actually what's holding up Coca-Cola from proceeding into the appeal process, because the Coca-Cola opinion stayed that issue pending 3M.

There's also the Perrigo case. It's been fully briefed, and we're waiting for an opinion in that case at some point. There are a handful of others as well. A lot to look forward to.

David D. Stewart: As we see these opinions, we'll have you come back and explain what they all mean. Ryan, thank you for being here.

Ryan Finley: Thanks for having me.