It is a truth universally acknowledged that a judge, in possession of a good transfer pricing case, must be in want of a compromise. Although it initially concluded that the merits were nearly all on the side of Medtronic's comparable uncontrolled transaction (CUT) method, on closer examination and a second look, the Tax Court was willing to find some faults with its CUT. However, rather than settling for the IRS's proposed comparable profits method (CPM), the Court instead found a perfect match for a compromise: the unspecified method.

Alright, this is not Pride and Prejudice, the unspecified method is not the elusive Mr Darcy, and the Court's conclusion was not based on romantic sensibility, but instead a re-evaluation of its prior decision in light of appellate criticism. Still, while the Court was wooed by the CPM in the recent Coca-Cola case, it rejected the CPM in Medtronic II in favour of another previously disfavoured method. But, alas, the story must be told properly: from the beginning.


Medtronic is a leading global medical technology company that manufactures and sells implantable medical devices and the leads that connect the devices to the human body. In particular, these devices and leads are Class III medical devices, which means they are life supporting or life sustaining.

Medtronic, Inc (Medtronic US), had several functions, including responsibility for research and development (R&D) and component manufacturing. Medtronic Puerto Rico (MPROC), created in 2001, was, in the 2005 and 2006 periods under audit, responsible for manufacturing and selling certain finished medical devices and leads.

In 2001, Medtronic US licensed to MPROC intangible property used in the manufacture of such devices and leads, and MPROC paid to Medtronic US a royalty of 29% of US net intercompany sales of devices and 15% of net intercompany sales of leads. Before the audit of its 2005 and 2006 years, Medtronic adjusted these rates to 44% and 26%, respectively, as agreed in the IRS's memorandum of understanding (MOU), which resolved Medtronic's 2002 audit.

In its final notice of deficiency for the audit years, however, the IRS made a further adjustment of almost $1.4 billion, primarily reflecting an increased royalty for the licensed intangibles. The IRS supported its adjustments using the CPM, arguing that MPROC's only role was to perform the final manufacturing steps for devices and leads, and that its profits were therefore vastly overstated.

In Medtronic I, the Tax Court rejected the IRS adjustments as arbitrary and capricious. It concluded that the IRS's application of the CPM downplayed the role of MPROC, particularly its role in ensuring the quality of finished devices and leads; used inappropriate comparables with different functions, capabilities, and assets from MPROC; utilised a return on assets profit level indicator that put too much emphasis on physical assets and ignored the value of licensed intangibles; and inappropriately aggregated multiple transactions.

Medtronic, for its part, argued that the pre-MOU rates in its agreements were arm's length, using a CUT analysis that relied heavily on an agreement that had resolved patent litigation between Medtronic US and Siemens Pacesetter, Inc (Pacesetter). In that agreement, Pacesetter agreed to pay a 7% royalty on US sales and a 3.5% royalty on rest of world sales of certain cardiac devices.

The Tax Court rejected Medtronic's proposed CUT analysis, determining that appropriate adjustments were required for differences between the Pacesetter agreement and the intangibles licence to MPROC, including a different array of devices, differences in the intangible property licensed and potential profit potential differences.

Instead, the Tax Court in Medtronic I, like it has done in many transfer pricing cases before, sought to find a middle ground. It started with the Pacesetter agreement as a CUT and made a series of adjustments to address the comparability differences, arriving at a royalty rate of 44% for devices and 22% for leads. This result was very close to the negotiated MOU rates, though the Court was careful to explain that this was not by design.

On appeal, the Eighth Circuit concluded that the Tax Court did not make sufficient factual findings to enable it to evaluate the determination and application of the best method. In particular, the appellate court found issues with the Tax Court's failure to determine whether the Pacesetter agreement was created in the ordinary course of business, to analyse the degree of comparability between the Pacesetter agreement and the MPROC licence, to evaluate the comparability of the additional intangibles covered by the MPROC licence, and to decide on the proper allocation of risk and product liability expense between Medtronic US and MPROC.


"My good opinion once lost, is lost forever." – Mr Darcy

Akin to Mr Collins's repeated proposals to Lizzie, the IRS, despite a firm and clear rejection in Medtronic I, continued to argue on remand that the CPM was the best method and refused to provide any alternatives. The Tax Court again rejected the CPM and the IRS's application of it on fundamentally the same bases as in Medtronic I. Even with slight proposed modifications, the Court found the approach inappropriate and concluded that it was an abuse of discretion due to the flawed comparables used. Indeed, it stated: "The Court is not going to reverse its opinion that petitioner met its burden of showing that respondent's allocations were arbitrary and capricious." The Court further dismissed the IRS's invocation of Coca-Cola's acceptance of the CPM by noting the difference between the manufacturing of sweetened beverages and life-saving medical devices, particularly as to the role and importance of quality in the manufacturing processes.

In contrast, while Medtronic continued to argue that the CUT method was the best method and that the Pacesetter agreement was a reliable CUT, it also wisely offered the Court another option: the unspecified method. In Medtronic II, the Court concluded that the Pacesetter was not a CUT due to the lack of comparability in functions, economic conditions, and the property or services.

In particular, the Court found that as licensor to MPROC, Medtronic US performed significant R&D and marketing, but did not perform such functions for Pacesetter. Instead, Pacesetter performed such functions as licensee, while MPROC, whose functions were limited to manufacturing finished devices and leads, did not. Further, the Court concluded that that the difference in profit potential between the two licences, as demonstrated by the profit margin of the manufactured products, resulted in noncomparable economic conditions.

Finally, the Court found that the products and services were not comparable because the MPROC licence included devices related to neurological treatments in addition to cardiac devices, and because the number of patents available to MPROC was over five times higher than those available to Pacesetter. The Court did find that the Pacesetter agreement was reached in the ordinary course of business, but this was insufficient to revive its use as a CUT.

While the Court did not accept Medtronic's particular application of its proposed unspecified method because the resulting royalty rate was either below, or too close to, the rate from the Medtronic I decision, the Court did adopt Medtronic's three-step approach to bridging the gap between the parties. First, the Court found that there were sufficient similarities between the Pacesetter agreement and the MPROC licence that the former could be used as a starting point in step one of the proposed method, a modified CUT that allocated returns to Medtronic US. Step two allocated returns to MPROC using a modified CPM to adjust for MPROC's asset intensity. Although the Court found that Medtronic's asset intensity adjustment was too high, it decided to address this in step three, rather than making speculative adjustments in step two.

In step three, Medtronic proposed a profit split allocation to allocate the remaining profits based on the Pacesetter agreement, but the Tax Court rejected that approach, preferring to split profits with 80% Medtronic US and the remaining 20% to MPROC. The Court concluded that this approach both accounted for the imperfections of the CUT and CPM methods and counteracted the excess asset intensity adjustment in step two. But the Court did not explain in any amount of detail how it arrived at this 80/20 split, as compared to any other. To check its result, however, the Tax Court noted that its approach resulted in an overall profit split of 31.28% to MPROC, which in its view better reflected the role of MPROC in the Medtronic organisation than any other proposed result. As applied, the Tax Court's unspecified method resulted in a wholesale royalty rate of 48.8% applied consistently to both devices and leads.

"The distance is nothing when one has a motive." – Elizabeth Bennett

Unlike Elizabeth's motive for her three-mile rainy walk (to visit her ill sister), the Court's motive in applying Step 3 of its analysis is less clear. Although it detailed the evils that the adjusted allocation was intended to overcome, the Court failed to explain exactly how the 80/20 split had such an effect. Instead, the Court's allocation seemed aimed at reaching what it considered an appropriate overall profit split. Presumably, the Court did not consider a profit split as the appropriate method in the first place because neither party provided fact or expert evidence to support the application of such a method. Moreover, the profit split "check" used by the Court is based more on a general sense of each party's contributions rather than a mathematical analysis of the same. In short, and as seen in past cases, the Court's appears to simply to split of the difference between the parties.


Despite the nebulous basis for the result, the Tax Court's use and approval of the unspecified method, and the three steps applied in Medtronic II in particular, may create a guidepost for Exam or Appeals in transfer pricing cases. The method that the Court applied is arguably a variation of the residual profit split, whereby the non-US party and the US party first receive their baseline return for routine or benchmarkable contributions, and then have the remaining profit split in, ideally, some principled manner. So is that really an unspecified method? Maybe more a method with some unspecified aspects? Regardless, Exam or Appeals may be more likely to consider this three-step approach in a narrow set of circumstances in which the following are all true:

  • Two separate methods are proposed by the taxpayer and by Exam, each of which is rejected by the other party.
  • The difference between the two rejected methods is sufficiently far apart that neither method could be reliably adjusted to bridge the difference.
  • Both of these methods are methods that address the benchmarkable contributions of the parties to the transaction.

However, applying the three-part approach from Medtronic II could lead to an unusual result by making compromise less likely, given the outsized impact of step three. In this step, the Court applied a judgement-based split of remaining profit, allocating 80% to Medtronic US and 20% to MPROC, with little explanation of its reasoning. Generally, neither taxpayers nor Exam will proactively bridge the typically wide divide between their positions in a dispute, and the third-step split could be manipulated to produce desired results. Taxpayers and Exam could be incentivised to pick one-sided methods in the tug of war for allocating profit. Awkwardly, the Court's goal of achieving a compromise resolution in this case could actually widen the distance between taxpayer and Exam positions when applying one-sided methods, thereby resulting in more future tax disputes.

This three-step approach can only be reliably applied when the taxpayer and Exam select methods that each reflect the contributions of just one party. If either applies a method that accounts for the contributions of both parties, the Medtronic II method may be more challenging to apply. Indeed, if a selected method appropriately reflects the contributions of all of the parties to the relevant transaction, then the Court's method should not be applicable. In the aftermath of this case, taxpayers, Exam teams and Appeals may want to think carefully about proactively laying out "two-sided" methods, or alternative allocations to split the remaining profit.

Future of CUT
"There are few people whom I really love, and still fewer of whom I think well." – Elizabeth Bennett

No one method is perfect in every circumstance, though the CUT has, for the obvious reason of reliability, reigned for some time as the gold standard in transfer pricing cases. With the recent decisions in Coca-Cola (adopting the CPM) and Medtronic II (applying an unspecified method), the question arises whether the CUT still makes the cut as a transfer pricing method. Can poor Mr Wickham be redeemed? In short, yes.

Medtronic II may have created a new path forward, albeit a narrow one, but it has also lit a candle in the darkness on how to utilise CUTs more effectively. In rejecting Medtronic's CUT, the Tax Court provided guidance on its analytical approach to CUTs, especially as the Pacesetter agreement was both accepted (in Medtronic I) and rejected (in Medtronic II) by the Court. Most companies that have tested IP transactions similar to those in Medtronic II likely have third-party transactions that can be considered as potential CUTs. Until now, taxpayers have generally thought about using these from a defence or reserve perspective, but have not been able to weigh the reliability of their potential comparables. Taxpayers can take comfort that other third-party transactions may be disregarded when they align with the rejection criteria laid out in these decisions. Moreover, a principled approach in applying the CUT, using the Court's guidance, may also increase the comfort of a taxpayer's auditor from a reserve perspective.

The history of transfer pricing case law shows that the CUT has been the predominant means to achieve resolution in the Tax Court. Now, taxpayers can begin to further appreciate some of their beloved CUTs going forward. Perhaps, the George Wickham stand-in, the CUT, will make a comeback in the courts. Only time will reveal its fate, or perhaps another revision of this tale by the Eighth Circuit.

"I declare after all there is no enjoyment like reading!" – Caroline Bingley

For further information on this topic please contact Katie Marcusse, Gene Tien or Jason Law at Baker McKenzie by telephone (+1 415 576 3000) or email ([email protected] or [email protected] or [email protected]). The Baker McKenzie website can be accessed at