The Medtronic transfer pricing case is the largest and longest-running intercompany pricing dispute in U.S. Tax Court history, a fight over roughly $1.4 billion in additional taxes tied to how much profit Medtronic could leave in its Puerto Rico manufacturing subsidiary for 2005 and 2006. After two Tax Court trials and two Eighth Circuit reversals, the case is heading back to the Tax Court for a third round, and this time the IRS’s preferred pricing method is the one under the microscope.
What the Dispute Is Actually About
Medtronic, the Minneapolis-based medical device manufacturer, ran a subsidiary in Puerto Rico called Medtronic Puerto Rico Operations Co., or MPROC. Puerto Rico offered substantial corporate tax incentives, so a dollar of profit booked there was worth more after tax than a dollar booked in the United States. MPROC manufactured Class III finished cardiac rhythm management devices and leads, among the most heavily regulated products in the industry.
The U.S. parent licensed its patents, manufacturing know-how, regulatory approvals, trade secrets, and other intellectual property to MPROC under agreements called the Technology Licenses. MPROC paid royalties back to the parent in return. The royalty rate is the entire ballgame: a low royalty leaves more profit in Puerto Rico, and a high royalty pulls it back to the United States where it is taxed at U.S. rates.
Earlier tax years had been partly resolved through a memorandum of understanding setting wholesale royalty rates of 44% for devices and 26% for leads, but Medtronic and the IRS could not agree on how those rates should apply to 2005 and 2006.1United States Court of Appeals for the Eighth Circuit. Medtronic Inc and Consolidated Subsidiaries v Commissioner of Internal Revenue (2025) The IRS audited the consolidated returns for those two years and issued a notice of deficiency totaling roughly $1.4 billion.
The Legal Rule at the Center
Section 482 of the Internal Revenue Code lets the IRS reallocate income between related entities when necessary to prevent tax evasion or to clearly reflect income, and for transfers of intangible property it requires the income to be “commensurate with the income attributable to the intangible.”2Office of the Law Revision Counsel. 26 USC 482 – Allocation of Income and Deductions Among Taxpayers Treasury regulations put this into practice through the arm’s length standard: the price between related parties has to match what unrelated parties would have charged in the same circumstances. The regulations then apply a “best method rule” — taxpayers must use whichever pricing method produces the most reliable measure of an arm’s length result, with no fixed hierarchy among methods.3GovInfo. Treasury Regulation 1.482-1 – In General
Medtronic and the IRS each argued a different method was best, and the gap between the two results was enormous.
The Two Methods That Divide the Case
Medtronic’s CUT Method
Medtronic used the Comparable Uncontrolled Transaction method, which prices a controlled deal by pointing to a similar deal between unrelated parties. When a close real-world comparable exists, this is generally considered the most direct approach because it rests on an actual market price.4GovInfo. Treasury Regulation 1.482-4 – Methods to Determine Taxable Income in Connection With a Transfer of Intangible Property
The comparable Medtronic put forward was the Pacesetter Agreement, a patent-licensing deal between Medtronic U.S. and Siemens Pacesetter, an unrelated third party, covering certain cardiac rhythm management patents.1United States Court of Appeals for the Eighth Circuit. Medtronic Inc and Consolidated Subsidiaries v Commissioner of Internal Revenue (2025) Medtronic’s experts adjusted for differences in geographic market, duration, and IP scope, and concluded the intercompany royalties were consistent with what unrelated parties would negotiate.
The comparability of that agreement became the single most contested factual question in the case. The Technology Licenses to MPROC covered patents, manufacturing know-how, regulatory approvals, secret processes, technical expertise, and copyrights. The Pacesetter Agreement licensed only patents. Profit potential also diverged sharply: Pacesetter’s product profit margin from the licensed IP averaged 29% from 1993 to 1995, while Medtronic’s average product profit margin was 54% during 2005 and 2006.1United States Court of Appeals for the Eighth Circuit. Medtronic Inc and Consolidated Subsidiaries v Commissioner of Internal Revenue (2025) Treasury regulations require comparable intangible property to have similar profit potential and do not allow adjustments for differences in the intangible property itself, only for differences in the circumstances of the transaction.4GovInfo. Treasury Regulation 1.482-4 – Methods to Determine Taxable Income in Connection With a Transfer of Intangible Property
The IRS’s CPM
The IRS rejected the CUT approach and proposed the Comparable Profits Method. Instead of finding a comparable licensing deal, the CPM benchmarks the operating profitability of the tested party — MPROC — against unrelated companies performing similar functions. The IRS treated MPROC as a routine contract manufacturer with limited risk, entitled to only a modest market-rate return on its manufacturing assets. Anything above that routine return belonged to the U.S. parent as owner of the IP, and would be reallocated back to the United States.1United States Court of Appeals for the Eighth Circuit. Medtronic Inc and Consolidated Subsidiaries v Commissioner of Internal Revenue (2025) The IRS identified five unrelated companies to benchmark MPROC’s routine profit. That framing generated the $1.4 billion deficiency.
Medtronic disputed the “routine manufacturer” label, arguing that MPROC produced Class III devices in a heavily regulated environment, carried meaningful product liability risk, and performed sophisticated quality control work well beyond simple assembly.
Twenty Years of Litigation, in Four Rulings
Tax Court, 2016
The Tax Court’s first opinion, T.C. Memo 2016-112, rejected both sides. It found the IRS’s Section 482 reallocations “arbitrary, capricious, or unreasonable,” criticizing the routine-manufacturer characterization and the selection of comparable companies. It also found Medtronic’s CUT analysis unreliable because the company had not made persuasive adjustments to the Pacesetter Agreement. Working from a modified CUT approach, the court set wholesale royalty rates of 44% for devices and 22% for leads. Both sides appealed.
Eighth Circuit, 2018
The Eighth Circuit vacated at 900 F.3d 610, holding that the Tax Court had not made enough factual findings to allow appellate review of whether the best method had actually been applied.5United States Court of Appeals for the Eighth Circuit. Medtronic Inc and Consolidated Subsidiaries v Commissioner of Internal Revenue (2018) The court sent the case back with instructions to explain why its modified approach was justified, particularly whether the Pacesetter Agreement was a valid comparable at all.
Tax Court, 2022
On remand, T.C. Memo 2022-84 again rejected both parties’ methods. This time the Tax Court adopted what it called an “unspecified method,” a hybrid Medtronic had proposed as an alternative, drawing on both CUT and CPM elements. The result: a profit split of roughly 69% to the Medtronic U.S. affiliates and 31% to MPROC, with an overall royalty rate of 48.8% for both devices and leads. The court found the Pacesetter Agreement did not qualify as a valid CUT because it did not involve intangible property with similar profit potential, yet still used Pacesetter data as one input in its three-step hybrid.
Eighth Circuit, 2025
On September 3, 2025, the Eighth Circuit vacated again.1United States Court of Appeals for the Eighth Circuit. Medtronic Inc and Consolidated Subsidiaries v Commissioner of Internal Revenue (2025) The appellate court affirmed that the Pacesetter Agreement was not a valid comparable uncontrolled transaction. The profit potential gap (29% versus 54%) and the difference between licensing only patents versus the full bundle of intangibles were too significant. Three of the five general comparability factors in the regulations were not satisfied: the functions performed were different, the economic conditions were not comparable, and the intangible property was not similar.
Because Pacesetter failed as a CUT, the court held it also tainted the Tax Court’s unspecified method, which had relied on Pacesetter data as a key input. The hybrid method was rejected as unreliable and not the best method for determining arm’s length pricing.
What the 2025 Remand Orders
The most consequential piece of the 2025 ruling is what it says about the CPM. The Eighth Circuit found the Tax Court had applied the wrong legal standard in rejecting the IRS’s method: the Tax Court had discarded the IRS’s proposed comparables because they “did not make solely Class III medical devices,” and the appellate court held this overemphasized product similarity, which matters less under the CPM than functional similarity.
The remand directs the Tax Court to conduct a fresh CPM analysis with specific tasks:
- Determine whether the IRS’s five proposed comparable companies were sufficiently similar to MPROC, and if not, whether reliable adjustments could correct any material differences.
- Make specific findings about how differences in asset composition between MPROC and the comparables affected profit allocation, and whether adjustments could account for those differences.
- Reconsider whether the functions performed by MPROC and the comparables were sufficiently alike under the regulatory framework.
- Quantify the product liability risk borne by MPROC versus the comparables and decide whether any difference was material enough to affect comparability of their profits.
The practical effect is striking. After nearly two decades of litigation in which the Tax Court twice rejected the CPM, the Eighth Circuit has told it to give the CPM a serious, properly conducted evaluation. The CUT method and the unspecified hybrid are both off the table. When the case returns to the Tax Court, the CPM will be the primary methodology under consideration.
What Else Is at Stake: Section 6662 Penalties
Large transfer pricing adjustments can trigger substantial accuracy-related penalties on top of the tax deficiency. Section 6662(e) imposes a 20% penalty for a substantial valuation misstatement — when the transfer price claimed is 200% or more (or 50% or less) of the correct arm’s length amount, or when the net Section 482 adjustment for the year exceeds the lesser of $5 million or 10% of gross receipts. A 40% penalty applies for a gross valuation misstatement, triggered when the price is 400% or more (or 25% or less) of the correct amount, or when the net Section 482 adjustment exceeds the lesser of $20 million or 20% of gross receipts.6Office of the Law Revision Counsel. 26 USC 6662 – Imposition of Accuracy-Related Penalty on Underpayments On an adjustment above $1 billion, these penalties can add hundreds of millions of dollars in exposure, and the main defense is contemporaneous transfer pricing documentation meeting Treasury Regulation 1.6662-6(d).7Internal Revenue Service. The Section 6662(e) Substantial and Gross Valuation Misstatement Penalty
What the Case Has Already Changed
Even without a final resolution, Medtronic has reshaped transfer pricing practice in concrete ways.
The CUT method took a hit. The 2025 opinion makes clear that comparability requirements under Treasury Regulation 1.482-4(c) are demanding, especially for intangibles. A comparable must involve property with similar profit potential, and a large profit-potential gap cannot be bridged through adjustments. The intangible itself has to be genuinely comparable before adjustments enter the picture. For companies leaning on internal license agreements with third parties as CUT comparables, a license that covers different IP or generates materially different margins will not survive scrutiny.
The CPM gained ground. The Eighth Circuit’s instruction to give the method a proper evaluation signals that profit-based methods remain viable in complex IP cases, provided the comparables are functionally similar. The court explicitly rejected the Tax Court’s insistence on exact product similarity and focused instead on similar functions and risks, which fits how the IRS typically builds its comparable sets.
The case also marks the limits of judicial creativity. Twice the Tax Court fashioned its own method rather than accept either party’s proposal, and twice the Eighth Circuit sent it back. The best method rule gives courts flexibility, but a court-invented approach has to stand on its own factual and regulatory foundation and cannot borrow data from a comparable the court itself found unreliable.3GovInfo. Treasury Regulation 1.482-1 – In General
The case now heads back to the Tax Court for a third trial-level proceeding. The outcome will affect not just Medtronic’s tax bill for 2005 and 2006 but the broader framework multinationals and the IRS use to negotiate, audit, and litigate intercompany pricing for high-value intangibles, and it will not produce a final answer soon.