Medtronic Files Opening Brief in Tax Court Transfer Pricing Case

MEDTRONIC, INC., & CONSOLIDATED SUBSIDIARIES,
Petitioner,
v.
COMMISSIONER OF INTERNAL REVENUE,
Respondent.

UNITED STATES TAX COURT

Judge Kerrigan

MEDTRONIC'S AMENDED POST-TRIAL BRIEF

---

TABLE OF CONTENTS

I. NATURE OF CONTROVERSY AND TAX INVOLVED

II. POINTS RELIED UPON

III. PROPOSED FINDINGS OF FACT

IV. ARGUMENT

TABLE OF AUTHORITIES

Cases

Alexander v. Jensen-Carter, 711 F.3d 905 (8th Cir. 2013)

Amazon.com, Inc. v. Commissioner, 148 T.C. 108 (2017)

Bass v. Stryker Corp., 669 F.3d 501 (5th Cir. 2012)

Bausch v. Stryker Corp., 630 F.3d 546 (7th Cir. 2010)

Bausch & LombInc. v. Commissioner, 92 T.C. 525 (1989)

Boltar, L.L.C. v. Commissioner, 136 T.C. 326 (2011)

Boyer v. Weyerhaeuser Co., 39 F. Supp. 3d 1036 (W.D. Wis. 2014)

Burkert v. Petrol Plus of Naugatuck, Inc., 579 A.2d 26 (Conn. 1990)

Centron DPL Co. v. Tilden Fin. Corp., 965 F.2d 673 (8th Cir. 1992)

Chay-Velasquez v. Ashcroft, 367 F.3d 751 (8th Cir. 2004)

Commissioner v. Duberstein, 363 U.S. 278 (1960)

Compaq Computer Corp. v. Commissioner, 78 T.C.M. (CCH) 20 (1999)

Cornett v. J&J, 998 A.2d 543 (N.J. Super. Ct. App. Div. 2010)

Georgia-Pacific v. U.S. Plywood, 318 F. Supp. 1116 (S.D.N.Y. 1970)

Hofts v. Howmedica Osteonics Corp., 597 F. Supp. 2d 830 (S.D. Ind. 2009)

Howard v. Sulzer Orthopedics, Inc., 382 F. App'x 436 (6th Cir. 2010)

LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51 (Fed. Cir. 2012)

Lear Eye Clinic Ltd. v. Commissioner, 106 T.C. 418 (1996)

Mandel v. Fischer, 205 A.D.2d 375 (N.Y. App. Div. 1994)

Medtronic, Inc. v. Commissioner, T.C. Memo. 2016-112

Medtronic, Inc. v. Commissioner, 900 F.3d 610 (8th Cir. 2018)

In re Medtronic, Inc. Sprint Fidelis Leads Prods. Liab. Litig., 592 F. Supp. 2d 1147 (D. Minn. 2009)

Miller v. DePuy Spine, Inc., 638 F. Supp. 2d 1226 (D. Nev. 2009)

Parker v. Stryker Corp., 584 F. Supp. 2d 1298 (D. Colo. 2008)

Pinsonneault v. St. Jude Med., Inc., 2014 WL 2879754 (D. Minn. June 24, 2014)

Purcel v. Adv. Bionics Corp., 2008 WL 3874713 (N.D. Tex. Aug. 13, 2008)

ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860 (Fed. Cir. 2010)

Riegelv. Medtronic, Inc., 552 U.S. 312 (2008)

Sundstrand Corp. v. Commissioner, 96 T.C. 226 (1991)

In re TMJ Implants Prods. Liab. Litig., 880 F. Supp. 1311 (D. Minn. 1995)

Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292 (Fed. Cir. 2011)

Veritas Software Corp. v. Commissioner, 133 T.C. 297 (2009)

Statutes & Regulations

21 U.S.C. § 360k

26 U.S.C. § 367

26 U.S.C. § 482

Treas. Reg. § 1.482-1

Treas. Reg. § 1.482-3

Treas. Reg. § 1.482-4

Treas. Reg. § 1.482-5

Other Authorities

Am. Bankr. Inst., Return on Assets So Useful . . . And So Misused (2001)

E. Minerd & R. Smith, Express and Implied Preemption for Premarket-Approved Medical Devices, Med Device Online (Apr. 9, 2018)

Fed. R. Evid. 702

IRS APA Study Guide (2013), https://www.irs.gov/pub/irs-apa/apa study guide.pdf

J. Beck, Riegel at 1 V2 : What Do We Know Now About Parallel Violation Claims?, Drug & Device Law Blog (July 30, 2009), https://www.druganddevicelawblog.com/2009/07/riegel-at-1-12-what-do-we-know-now.html

M. Herrmann et al., The Meaning of the Parallel Requirements Exception Under Lohr and Riegel, 65 NYU Ann. Surv. of Am. L. 545 (Mar. 18, 2010)

M. McLaughlin Kirmil, Transfer Pricing Answer Book, ch. 5 (2020 Ed.)

N.J. Royce, The Influence of Depreciation of Assets on Management Decisions, 27:2 J. Op. Res. Soc'y 471 (1976)

OECD, Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (1995)

S. Raymond, Judicial Politics and Medical Device Preemption After Riegel, 5 NYU J. L. & Liberty 745 (2010)

T. D. 8552, 1994-2 C.B. 93

---

I. NATURE OF CONTROVERSY AND TAX INVOLVED

This case concerns the 2005 and 2006 tax years of Medtronic, Inc., & Consolidated Subsidiaries (“Medtronic”).¹ For the one issue still in dispute — the arm's-length royalty rate for the “MPROC Licenses” — Respondent has determined deficiencies of $548,180,115 for tax year 2005 and $810,301,695 for 2006.²

II. POINTS RELIED UPON

The sole issue presented in this remand is: (i) the arm's-length royalty rate for technology licensed from Medtronic, Inc. (“Medtronic US”) to Medtronic's Puerto Rico-based operations (“Medtronic PR”) under the “MPROC Licenses” and (ii) as a function of that price, the return to Medtronic PR's Class III finished device manufacturing (and related activities). All other intercompany transactions that this Court addressed in Medtronic, Inc. v. Commissioner, T.C. Memo. 2016112 ("Medtronic J”) — i.e., component manufacturing, distribution and the Trademark License — have been priced at arm's length and are thus off-the-table.

For this narrow remaining issue, the Court reopened the record only for additional expert testimony, and did not reopen the factual record more broadly. May 13, 2019, Order at 2-4 (“Reopening Order”). And on appeal, the Eighth Circuit only requested supplemental findings to support the Court's original opinion. Medtronic, Inc. v. Commissioner, 900 F.3d 610, 617 (8th Cir. 2018) ("Medtronic IP"). In line with that mandate, the Court asked the parties to address six questions that relate to: (i) whether the “Pacesetter Agreement”³ is a valid CUT for the MPROC Licenses; (ii) the reliability of the adjustments made to the Pacesetter Agreement's royalty rate to establish the rate for the MPROC Licenses; and (iii) which party offers the more reliable transfer pricing method.

Medtronic has provided expert testimony on remand that, combined with the otherwise undisturbed factual record and the Court's findings in Medtronic I, overwhelmingly establishes that the Pacesetter Agreement is a CUT, and that Medtronic's expert, Dr. Jonathan Putnam, made reliable adjustments to the Pacesetter Agreement's royalty rate. Likewise, expert testimony further supports the Court's prior finding that Respondent's CPM method is wholly unreliable.

In addition, consistent with the Court's remarks at the remand proceeding, Medtronic presents an alternative unspecified method for pricing the MPROC Licenses. See Treas. Reg. § 1.482-4(d); see infra Part IV.H. This unspecified method applies a modified version of Medtronic's CUT method to allocate profit to Medtronic US's R&D activities and applies a modified version of Respondent's CPM to allocate profit to Medtronic PR. After the application of each party's method, the remaining profit is then allocated based on evidence that establishes how market participants would divide those profits. With this approach, the unspecified method accomplishes two things. First, it incorporates aspects of Medtronic's CUT and Respondent's CPM and addresses the Court's questions related to profitability. Second, the unspecified method and the primary CUT method mutually reinforce one another. Although the unspecified method results in a higher royalty rate for the MPROC Licenses (and thus less profit to Medtronic PR), these methods demonstrate that multiple approaches to adjusting for Medtronic's profitability converge on a range of reasonable results.

A. Medtronic's Experts on Remand

Medtronic offered experts with unique and highly relevant expertise who performed a “de novo” review of the Court's remand topics. Their conclusions are consistent with, and further support, the Court's findings in Medtronic I.

• Dr. Jonathan Putnam: An economist with significant experience in assessing the reliability of patent settlement agreements to benchmark patent damages. Based on his academic and professional background, Dr. Putnam is uniquely qualified to assess the comparability of the Pacesetter Agreement and to make appropriate adjustments to its royalty rate.

• Dr. Glenn Hubbard: An economist, Dean Emeritus at Columbia Business School, and public policy expert. Dr. Hubbard offered a broader economic perspective to assess both the Pacesetter Agreement and Respondent's CPM.

• Dr. Richard Cohen: A biomedical engineering professor at MIT and a faculty member at the MIT Institute for Medical Engineering and Science and the Harvard-MIT Health Sciences and Technology Program. As a scientist and career-long “student” of the CRDM industry, Dr. Cohen addressed various technical issues, including, for example, the relative importance of know-how to manufacturing Cardiac Devices and Leads.

• Mr. Fred McCoy: An industry expert and former CEO with 35 years of experience in the medical device industry. Significantly, as the former President of Guidant's CRDM Division, Mr. McCoy is uniquely positioned to offer market-grounded opinions on using the Pacesetter Agreement's royalty rate as a benchmark.

• Mr. Chris Spadea: An expert with over 25 years of experience valuing and managing IP. Mr. Spadea brought to bear that specialized expertise in showing the relative value of the IP here at issue over the life of the Pacesetter Agreement, as well as the market use of licenses that arise out of litigation in IP valuation and commercial negotiations.

B. These Experts (and the Evidence) Comprehensively Address the Court's Remand Topics

Medtronic's experts offered their own unique perspectives — grounded in objective evidence — that, together, comprehensively address the Court's questions.

1. Circumstantial Comparability

The evidence before the Court, including expert testimony, establishes that the circumstances surrounding the Pacesetter Agreement and the MPROC Licenses are comparable. Respondent raised two core challenges to this circumstantial comparability. First, Respondent argued that the settlement context of the Pacesetter Agreement precludes using it as a benchmark for the MPROC Licenses. Second, Respondent argued that the Pacesetter Agreement cannot be used as a benchmark because the parties to it — Siemens and Medtronic — were “horizontal” market competitors, whereas Medtronic PR and Medtronic US are “vertical” affiliates. Neither is true as a matter of fact, law, or economics.

(a) The Context of Litigation Bolsters the Pacesetter Agreement's Reliability

Expert Opinion of Both Parties: As Dr. Putnam, Dr. Hubbard and Mr. Spadea testified, the particular facts and circumstances of the Pacesetter Litigation bolster, rather than preclude, the use of the Pacesetter Agreement as a benchmark for the MPROC Licenses. Even Respondent's expert Dr. Michael Heimert testified that the litigation context of the Pacesetter Agreement does not, in and of itself, cause the Pacesetter Agreement to be an unreliable benchmark. And Respondent's expert Dr. Christine Meyer testified that she herself has used settlement licenses in the past as a tool to value patents.

St Jude's Acquisition: In addition, as Dr. Putnam and Dr. Hubbard testified, St. Jude's 1994 acquisition of Pacesetter establishes that the litigation context of the original Pacesetter Agreement did not distort its key terms. When St. Jude acquired Pacesetter, both St. Jude and Medtronic agreed to St. Jude's assumption of the rights and obligations under the Pacesetter Agreement. There was no pending patent litigation between Medtronic and Pacesetter or St. Jude, at the time of the acquisition. In addition, the acquisition caused the term of the Pacesetter Agreement to be extended by two years, resulting in the agreement remaining in effect through 2004 (i.e., Medtronic's tax year 2005). St. Jude could have structured the acquisition to avoid assuming the obligations under the Pacesetter Agreement, but St. Jude chose not to opt-out. Thus, St. Jude's independent decision to agree to pay a 7% royalty rate for access to Medtronic's CRDM patent portfolio for 10 years (1994 to 2004) eliminates any lingering concern that litigation distorted the terms of the Pacesetter Agreement in 1992.⁴

Maximum Rate Clause/Board Documents: Dr. Putnam and Dr. Hubbard also explained that litigation did not affect the Pacesetter Agreement's “Maximum Rate Clause.” This clause established each party's right to access the other's entire future CRDM portfolio for an aggregate rate that could not exceed 15%, which, for Pacesetter, was inclusive of the 7% royalty that it was already paying. The only exception to this 15% cap was narrow and short-term — for patents designated as “key.” This clause thus provides an arm's-length benchmark for the maximum rate that Medtronic US could charge Medtronic PR in 2004. And as Dr. Putnam explained, the Board Documents establish that litigation could not have distorted the Maximum Rate Clause. Moreover, the clause provided for licensing offuture patents — not in existence when the Pacesetter Agreement was executed — further establishing the clause's independence from any then-current litigation.

(b) That Medtronic/Siemens Were Competitors Enhances Comparability

Expert Opinion of Both Parties: As Dr. Hubbard and Dr. Putnam testified, that Medtronic and Siemens were competitors enhances — rather than undermines — comparability. As competitors, Medtronic and Siemens were each strongly motivated to negotiate the most favorable terms. This negotiating tension is the essence of an arm's-length result, as the section 482 regulations contemplate. See Treas. Reg. § 1.482-1 (d). Moreover, the structural comparison that Respondent finds objectionable — using an agreement between competitors as a benchmark to price an agreement between affiliates — is a feature, not a failure, of a good benchmark. By definition, CUTs use transactions between unrelated parties who are typically competitors. Dr. Heimert agreed that Pacesetter's status as a “horizontal” competitor does not, in and of itself, break comparability with the MPROC Licenses. Dr. Heimert also conceded that a negotiation between competitors creates sufficient negotiating tension for an arm's-length result.

2. Comparability of Intangibles

The key question for the comparability of intangibles is whether changes to Medtronic's CRDM patent portfolio between 1992 and 2004 undermine comparability between the patent portfolio licensed in the Pacesetter Agreement and in the MPROC Licenses. The evidence establishes that Medtronic satisfied the section 482's applicable comparability standards.

Market Evidence: Whether Medtronic PR is required to pay a higher royalty rate than the Pacesetter rate should be considered in light of how CRDM market participants actually behaved at arm's length. First, as Dr. Hubbard testified, the royalty rate that Medtronic US could charge Medtronic PR is influenced by the extent to which Medtronic's competitors also had access to Medtronic's CRDM patent portfolio. And, as Mr. McCoy and Dr. Cohen observed, Medtronic had licensed its CRDM patent portfolio to several competitors: CPI/Eli Lilly, Siemens and St. Jude. Medtronic US thus, consistent with the arm's-length standard, could not charge Medtronic PR an exorbitant rate for patents that were available to many of Medtronic's competitors. Moreover, the terms under which Medtronic licensed its CRDM patent portfolio to its competitors created a benchmark in the marketplace. The 7% rate that Siemens agreed to pay to Medtronic was known in the industry, including by Mr. McCoy as an executive at Guidant (another major industry player), and that rate would have been a benchmark in any subsequent negotiations. Mr. McCoy and Dr. Cohen testified that changes to Medtronic's CRDM patent portfolio between 1992 and 2004 did not alter that dynamic.

The testimony of Respondent's expert Dr. Iain Cockburn corroborates this point. He explained that, once a seminal cardiac patent had been licensed to multiple parties, the royalty rate became sticky, and subsequent licensees would pay the same rate. See Tr. (Cockburn) 2249-50. The “Mirowski” patents are a well-known example — seminal ICD patents licensed to several market players at a consistent 3% royalty rate for over 30 years. For Medtronic, the Pacesetter Agreement set the baseline market price for its CRDM patent portfolio at 7%.

Maximum Rate Clause: Further, the Maximum Rate Clause in the Pacesetter Agreement bridges the time period from 1992 to 2004 for two independent reasons. First, when this clause was negotiated, it established a 15% maximum rate that Pacesetter would pay for access to Medtronic's entire CRDM patent portfolio through 2002. Given the breadth and duration of that right, Medtronic executives negotiating the Pacesetter Agreement had to account both for the potential for new, high-value Medtronic patents, and that the CRDM market could expand and become more profitable over that ten-year period. They had to choose a sufficiently high rate to address those potential developments.

Second, and independent of the first point, the Pacesetter Agreement's operation in practice and the parties' conduct establish that, in 2004, Medtronic US could charge Medtronic PR no more than 15% for its CRDM patent portfolio. St. Jude's acquisition of Pacesetter and its assumption of the Pacesetter Agreement extended the term — including for the Maximum Rate Clause — for two years, into 2004. In addition, the evidence establishes that Medtronic never designated any patents as “key” between 1992 and 2004. Thus, when Medtronic US licensed its CRDM patent portfolio to Medtronic PR in 2004, Medtronic had an obligation to license that same patent portfolio to St. Jude at a rate of no more than 15%. This definitively establishes that an arm's-length, maximum rate for Medtronic US's CRDM patent portfolio in 2004 was 15%.

3. Comparability of Contractual Terms

The contractual terms of the two agreements are also comparable, and any differences are addressed by reliable adjustments, as Dr. Putnam concluded. Respondent has pointed to several differences in contractual terms — for example, the fact that the Pacesetter Agreement provided a cross-license, whereas the MPROC Licenses do not. None of these differences undermines comparability as demonstrated by, among other evidence, the Board Documents' record of Medtronic's contemporaneous and rigorous analysis of the Pacesetter Agreement's key terms. The Board Documents establish, for instance, that Medtronic did not assign any value to Pacesetter's patent portfolio, showing that the agreement effectively reflected a one-way license.

4. Dr. Putnam's Two Approaches to Adjusting the Royalty Rate

Dr. Putnam offered two approaches for adjusting the Pacesetter royalty rate for differences from the MPROC Licenses — one starting at the 7% rate and a second starting at the 15% rate from the Maximum Rate Clause. Both approaches reliably established an appropriate royalty rate for the MPROC Licenses. Yet Respondent's additional criticism of the CUT method is that Dr. Putnam made too many adjustments to the Pacesetter royalty rate, rendering his result unreliable.

Expert Opinion of Both Parties/Board Documents: For Dr. Putnam's first approach of starting with the 7% base royalty, the number of adjustments does not undermine reliability because each adjustment is grounded in the Pacesetter Agreement itself and in objectively observable evidence, including, in particular, the Board Documents. Dr. Putnam also relied on the expertise of scientific and industry experts — Dr. Cohen and Mr. McCoy — to support the adjustments. Finally, Dr. Hubbard concluded that Dr. Putnam's adjustments were reasonable.

Dr. Cockburn agreed that adjusting a royalty rate derived from a comparable agreement does not necessarily render the resulting rate unreliable. Dr. Cockburn can hardly be critical of Dr. Putnam's number or quantum of adjustments when Dr. Cockburn himself has, in other litigation, made adjustments to a comparable that increased the base amount by ten-fold. Tr. (Cockburn) 2273-75.

Maximum Rate Clause: Dr. Putnam's second approach of starting with the 15% maximum rate is not susceptible to Respondent's criticism on the number of adjustments. First, the “sub-license” adjustment to the 15% rate simply reflects rates that Medtronic US “passed-through” to Medtronic PR from third-party licenses. This method of pricing sub-licenses follows arm's-length market behavior. For example, the Pacesetter Agreement provides that each party will sublicense to the other third-party patents at a “pass-through” royalty rate. Ex. 2504-J at 4456-58. The only other adjustment to the 15% royalty rate is for know-how, and the evidence strongly supports Dr. Putnam's know-how adjustment.

5. Critique of Dr. Heimert's CPM

For the reasons that this Court aptly described in Medtronic I, Dr. Heimert's CPM is “arbitrary, capricious, and unreasonable.” Medtronic I at 118. Respondent did not appeal this finding. Yet Dr. Heimert's CPM has remained the same, even though the established factual record is contrary to core premises of his analysis. That Dr. Heimert's CPM conflicts directly with the Court's findings, in itself, establishes that his CPM cannot be the more reliable method in this case.

Further, Dr. Hubbard explained why Dr. Heimert's continued defense of these same 14 rejected comparables is unconvincing as an economic matter. Dr. Cohen and Mr. McCoy likewise concluded that Dr. Heimert's 14 companies fundamentally differ from Medtronic PR. And, as the Court already concluded, ROA is an inappropriate PLI where critical intangibles are not reflected on the balance sheet. Medtronic I at 113-14.

In attempting his own CPM, Dr. Hubbard found that the companies with potentially comparable functions were vertically integrated, like Medtronic PR, with no available segmented financial data. The relevant companies retain those functions in-house because finished device manufacturing is a core function in the Class III medical device industry generally (and the CRDM industry especially). The flaws in Dr. Heimert's data and analysis are thus interlinked: the lack of adequate data for his CPM stems, at least in part, from the central importance of Medtronic PR's function (which Dr. Heimert generally ignores).

III. PROPOSED FINDINGS OF FACT

A. The Evolution of the Pacesetter Agreement

1. The Pacesetter Litigation

1. In the late 1980s and early 1990s, Medtronic and Pacesetter litigated Medtronic's claim that Pacesetter's Cardiac Devices infringed on Medtronic CRDM patents — in particular, Medtronic's patent for a key technology known as “Activitrax.” The Activitrax technology established “rate-responsive” pacemakers that monitored and adapted to cardiac rhythm. Medtronic and Pacesetter had also filed lawsuits on other claims, such as unfair competition and anti-trust allegations (collectively, the “Pacesetter Litigation”). But Medtronic's patent infringement claims — most significantly for Activitrax — were the central issue in the Pacesetter Litigation. Medtronic I at 56; Ex. 6105-P (Putnam Opening) ¶¶ 344, 356-67; Ex. 6103-P (Cohen Opening) at 19-20; Ex. 6107-P (Spadea) ¶¶ 56-60.

2. The Pacesetter Litigation began in January 1988, when Medtronic sued Siemens in U.S. District Court for the Northern District of Illinois, alleging that certain of Pacesetter's CRDM products infringed on Medtronic's patents, including Activitrax (the “Chicago Litigation”). Ex. 6105-P (Putnam Opening) ¶ 357; Ex. 2505-J at 5041-42. Siemens owned Pacesetter at that time.

3. To prevail in the Chicago Litigation, Medtronic had to establish that its relevant CRDM patents were valid, and that Pacesetter had infringed on one or more of them. In late 1991 and early 1992, the district court ruled that: (i) Medtronic's Activitrax patent was valid; (ii) Pacesetter was infringing on it; and (iii) Pacesetter was permanently enjoined from selling three of its five rate-responsive CRDM products. Pacesetter had developed two generations of rate-responsive Devices, and the court enjoined all of its first-generation products and one of its three second-generation products.⁵ This was a near-total victory for Medtronic. Medtronic I at 56-57; Ex. 6105-P (Putnam Opening) ¶¶ 357-59.

4. For the two CRDM Devices not enjoined, Medtronic sued in U.S. District Court for the District of Minnesota in March 1992 claiming patent infringement for those Devices and other products (the “Minnesota Litigation”). Id. ¶¶ 358, 361.

2. The Settlement Negotiations

5. In late 1991, Siemens initiated settlement discussions with Medtronic that ultimately resulted in the Pacesetter Agreement in August 1992. A contemporaneous and particularly detailed account of the negotiation of the Pacesetter Agreement is in the Medtronic “Board Documents,”⁶ which include:

• Documents and presentations provided to Medtronic's Board of Directors (the “Board”) to approve proposed settlement terms. These documents “provide[ ] an unusually clear window into the Board's positions” and “help assuage concerns that the Pacesetter agreement is 'contaminated' in some unspecified way, relative to an agreement reached in the absence of litigation.” Id. 144; Tr. (Putnam) 780-85.

• Medtronic's exhaustive quantitative and qualitative assessment of the pros and cons of settlement and licensing as compared to litigating.

• Analysis of litigation issues, including: (i) the insignificance of the lawsuits other than the Chicago Litigation and Minnesota Litigation in the licensing negotiations and (ii) the relatively small costs of continued litigation. Ex. 6105-P (Putnam Opening) 139-59; Tr. (Putnam) 780-85.

6. During that negotiation period, Medtronic management analyzed potential settlement terms several times and presented that analysis to the Board on at least three occasions. First, in March 1992, Medtronic management presented to its Board settlement terms with an expected benefit to Medtronic in net present value terms (“NPV”)⁷ of $157 million. By May 1992, Medtronic management had sufficiently advanced the negotiations to recommend that the Board approve a resolution of all disputes with Pacesetter. On May 29, 1992, at a special Board meeting, Ron Lund, then Senior Vice President and General Counsel, presented settlement terms and a financial analysis showing that Medtronic was expected to realize a benefit of $200 million (NPV). As Mr. Lund noted, the revised settlement terms produced a projected financial benefit to Medtronic that was higher than the terms anticipated in March. Mr. Lund sought approval from the Board to settle on terms equal to or better than the terms presented (with their $200 million NPV). The Board agreed. Medtronic I at 57; Ex. 2524-J at 4414; Exs. 2522-J; 2525-J; 2526-J; 2527-J; 2528-J; 2529-J; 2532-J; 2533-J; 2534-J; MDT I Tr. (Mahle) 35253; MDTI Tr. (Ellwein) 3841-57, 3861.

7. After receiving that approval, Medtronic agreed with Siemens on a framework for a settlement, and the parties spent the summer of 1992 finalizing specific terms. On June 25, 1992, June Takafuji, Senior Counsel at Siemens, wrote to Michael Ellwein, then Medtronic's VP for Corporate Development, about the still-outstanding issues in their discussions. According to Ms. Takafuji, the “primary area” left to be negotiated was a term to allow Pacesetter to access Medtronic's future patents (i.e., patents that Medtronic developed post-settlement). As Ms. Takafuji wrote, Siemens “need[ed] from Medtronic more than a promise to license future patents at a reasonably royalty (whatever that is) and only after a period of time that could run from 3 to 8 years. We cannot predict the cost of unknowns such as this, and cannot safely proceed to rely upon its existence.” Ex. 2526-J at 4427; Ex. 6312-R at 9691-92.

8. Settlement negotiations continued and, on August 18, 1992, Mr. Lund informed the Board that the parties had reached agreement. The Board met on August 26, 1992, and ratified the negotiated terms. These terms were more favorable to Medtronic than those the Board had approved in May; the projected financial benefit to Medtronic increased from NPV $200 million in May to $205.1 million. Medtronic I at 57; Ex. 2527-J at 4430, 32; Ex. 2529-J at 4437; Tr. (Putnam) 767-73; Ex. 6105-P (Putnam Opening) ¶ 164.

9. While the parties were engaged in settlement negotiations — and after Medtronic had filed the Minnesota Litigation — Pacesetter filed a countersuit in April 1992 alleging patent infringement by two Medtronic pacemakers, “Elite” and “Legend.” This was the first time that Pacesetter claimed infringement by Medtronic. Ex. 6219-R; Ex. 6105-P (Putnam Opening) ¶ 366; Tr. (Cockburn) 2081-87. Pacesetter's lawsuit against Medtronic apparently did not improve Siemens' negotiating leverage, however: the projected value of the settlement to Medtronic increased after Pacesetter's lawsuit. PFF ¶ 7; see also Ex. 6105-P (Putnam Opening) ¶ 164(c).

3. Terms of the Pacesetter Agreement

10. The Pacesetter settlement comprised two documents: a patent license (the Pacesetter Agreement) and a settlement agreement that resolved the Pacesetter Litigation. Exs. 2504-J; 2505-J. By their terms, the two documents “comprise one agreement and the entire agreement of the parties.” Ex. 2505-J at 5046. In the Pacesetter Agreement, the parties cross-licensed their existing CRDM patents and agreed to a framework for licensing future patents. The major terms were:

11. Royalties: Pacesetter agreed to pay Medtronic:

• A 7% royalty on CRDM Devices and Leads sales in the US and Japan, and a 3.5% royalty on all other international sales. Medtronic I at 58; Ex. 2504-J.

• A $50 million upfront payment to compensate Medtronic for Pacesetter's past infringement. Medtronic I at 58; Ex. 2504-J at 4466.

• A $25 million royalty pre-payment credited against the 1.8% “portfolio access fee” added to the base rates. Medtronic I at 58; Ex. 2504-J at 4466.

Medtronic did not pay Pacesetter any royalty for the license of Pacesetter's patents. Ex. 2504-J; Ex. 2527-J at 4431.

12. Future Patents: The Maximum Rate Clause permitted each party to compel a license to any of the other party's CRDM patents developed during the agreement's term for an aggregate rate of no more than 15%. Thus, each party had access to both (i) the other party's existing CRDM patents, and (ii) all future patents issued thereafter. The royalty rate paid for all CRDM patents licensed could not be higher than 15%. For Siemens, the Maximum Rate Clause thus entitled it to license all of Medtronic's CRDM patents for an aggregate rate not higher than 15%, after taking into account the 7% royalty that Pacesetter was already paying for current patents.⁸ The one limited exception was set forth in the “Key Patent Clause,” under which each party could designate up to three patents per year as “key,” thus providing a roughly three-year period during which the other party could not compel a license of the key patent. If a party did not designate a patent issued in a given year as “key,” the party lost the right to designate it later. Thus, a party's decision not to designate patents as “key” subjected it to a potential license of all post-1992 patents. Ex. 2504-J.

13. Term & Assignment: The term was 10 years from 1992. The parties agreed that the term would reset if Siemens sold Pacesetter, to extend for ten years post-sale (but not more than 15 years total). Medtronic I at 58-59. If certain conditions were met, Siemens could assign its rights to the purchaser. Ex. 2504-J § 9.02(d).

4. St. Jude's Acquisition of Pacesetter

14. In 1994, Siemens sold Pacesetter to another medical device company, St. Jude Medical, Inc. (“St. Jude”). Medtronic I at 58; Ex. 6105-P (Putnam Opening) If ¶ 114; Ex. 2905-J (St. Jude 8-K, Sept. 30, 1994). When St. Jude acquired Pacesetter, it also acquired all of Siemens' rights and obligations under the Pacesetter Agreement. See, e.g., Medtronic I at 58-59; Fourth Stip. Fact ¶ 990 (Feb. 2, 2005); Ex. 2905-J (St. Jude 8-K, Sept. 30, 1994).

15. As a result of St. Jude's acquisition, the term of the Pacesetter Agreement reset, extending the termination date by two years. Medtronic I at 58-59; Ex. 2504J § 9.02(f); Ex. 6105-P (Putnam Opening) 315; Tr. (Putnam) 777:17-778:16. St. Jude thus paid royalties under the Pacesetter Agreement through September 2004 (i.e., into Medtronic's tax year 2005, which began in May 2004). Exs. 2509-J to 2519-J; MDT I Tr. (Ellwein) 3843; MDT I Tr. (Dennis) 3891-92.

16. Notably, instead of assigning the agreement to St. Jude, Siemens could have “bought out” its remaining royalty obligations — based on a formula set forth in the Pacesetter Agreement — for approximately 2 years of royalties (around $53 million). Ex. 2504-J. That one-time payment would have been small (~10%) relative to the $550 million price that St. Jude paid to acquire Pacesetter. Ex. 6105P (Putnam Opening) ¶ 114. That buy-out provision could have been exercised if St. Jude decided that it wanted to extinguish both Siemens' and St. Jude's rights and obligations under the Pacesetter Agreement. Id. ¶ 350. But St. Jude did not seek a buy-out. Rather, St. Jude made an independent commercial decision to accept the Pacesetter Agreement. Consequently, St. Jude became the assignee of all of Siemens' rights and obligations under the Pacesetter Agreement. Medtronic likewise accepted the transfer. Medtronic I at 58-59; Ex. 6105-P (Putnam Opening) ¶¶ 114-16; Tr. (Putnam) 786-88; See Exs. 2504-J; 2509-J to 2520-J; MDT I Tr. (Ellwein) 3857-58; MDT I Tr. (Coyle) 1338-42.

17. The Pacesetter Agreement was assigned to St. Jude in its entirety, including the Maximum Rate Clause. A number of agreements entered by St. Jude, starting with the Pacesetter acquisition in 1994, recognized that the Pacesetter Agreement continued with full force and effect post-assignment. In 1996, St. Jude acquired Ventritex, a CRDM competitor. The agreement that memorialized that Ventritex acquisition affirmed that the “Medtronic Agreement” — defined as “[t]he License Agreement, dated August 26, 1992, between Medtronic Inc. and Siemens AG as assigned to [St. Jude] on August 23, 1994” — “is in full force and effect and will not by its terms terminate by reason of the [Ventritex] Merger.” Ex. 2892-J § 3.17 (emphasis added). In 2002, an amendment to the Pacesetter Agreement confirmed that St. Jude was the assignee of “all” of Siemens' right, title and interest, and that the Pacesetter Agreement remains in “full force and effect” other than the amended terms, none of which related to the IP license. See Ex. 2506-J at 1 & ¶ 7.

18. Mr. Ellwein, the “primary negotiator” of the Pacesetter Agreement for Medtronic, testified that Medtronic did not designate any Key Patents. MDT I Tr. (Ellwein) 3861:5-10. The categorical answer implies that Medtronic did not designate any patents during the entire tenure of the Pacesetter Agreement, including when St. Jude was the counterparty. Likewise, Michael Coyle, who worked for St. Jude on the Pacesetter acquisition, confirmed that access to Pacesetter's IP, including rights under the Pacesetter Agreement, was a crucial driver of the acquisition. MDT I Tr. (Coyle) at 1339:1-21.

19. Dr. Putnam testified that, from an economic perspective, it is logical that the Maximum Rate Clause transferred to St. Jude. “The purpose of this [provision] was to prevent the potentially catastrophic war and the near-death experience that Pacesetter” had vis-à-vis patent infringement litigation. Tr. (Putnam) 862:4-862:7. It would be irrational for Medtronic and Pacesetter to state in the Pacesetter Agreement that, “upon a change of control, then we're going to reopen that war, or the possibility of that war with respect to future patent.” Id. 862:8-862:11; see also id. 777-79, 851-62. Respondent's expert Dr. Cockburn also recognized, based on his initial review of the provision, that the Maximum Rate Clause transferred to St. Jude upon its acquisition of Pacesetter. Tr. (Cockburn) 2339-44.⁹

20. Further, Dr. Putnam reviewed the evidence regarding St. Jude's acquisition and found no evidence that St. Jude impaired or otherwise adjusted the value of the Pacesetter Agreement when it acquired Pacesetter.

B. The Pacesetter Agreement Is an Appropriate Benchmark

21. The Pacesetter Agreement is an arm's-length benchmark for pricing the IP transferred in the MPROC Licenses.

1. Patent Licenses Establish Benchmarks in the CRDM Industry

22. Patent licenses in the CRDM industry, such as the Pacesetter Agreement, regularly established benchmarks for other parties to use in subsequent license negotiations. Such benchmarks were available in the CRDM industry because market participants regularly licensed their patents to one another, and the key terms were often known to other industry participants. As Fred McCoy explained, “[o]ver time, the terms of the resulting settlement agreements and licenses took on a general pattern that provided a general set of benchmarks. Gradually, the competitors were able to roughly forecast where disputes might ultimately land.” Ex. 6101-P (McCoy Opening) at 8. Mr. McCoy, who himself directly participated in such license negotiations during his time at Guidant, confirmed that he would rely on prior CRDM licenses as benchmarks when negotiating a new license. Tr. (McCoy) 100:20-23, 215-16; Ex. 6102-P (McCoy Rebuttal) at 4.

23. Once a license set a royalty rate for a particular patent portfolio, that rate generally remained consistent. Dr. Cockburn agreed that it was “commonplace” for seminal patents to be licensed at a particular rate early in their lifecycle, and for that rate to persist over time. Tr. (Cockburn) 2249:22-2250:3; see also id. 2249-53. As Dr. Putnam explained, such a constant royalty rate for a given patent portfolio is expected. To justify a higher rate, one would expect an evolution in circumstances — in particular, a material change in the IP. Tr. (Putnam) 757.

24. The persistent nature of CRDM benchmark royalty rates is demonstrated by the Mirowski patent portfolio, “the most central license” in the CRDM industry. Tr. (Cohen) 577:2-6; see also Ex. 6101-P (McCoy Opening) at 6. This seminal patent portfolio was consistently licensed for over 30 years at a royalty rate of 3%. Ex. 6102-P (McCoy Rebuttal) at 6; Tr. (McCoy) 246-47; Ex. 6103-P (Cohen Opening) at 29; Tr. (Cohen) 577, 677-78; Ex. 6107-P (Spadea) ¶ 52.

25. Medtronic licensed its patents to competitors, both before and after the Pacesetter Agreement. Dr. Cockburn recognized that Medtronic had entered crosslicenses with multiple competitors before it entered the Pacesetter Agreement. Ex. 6238-R at 0048 (Q&A no. 9); Tr. (Cockburn) 2261-63 (discussing Ex. 6238-R); Ex. 6103-P (Cohen Opening) at 17-18; Ex. 6104-P (Cohen Rebuttal) at 17-18; Ex. 6154-P (Spadea Sealed) ¶¶ 44-54, 67-72; Tr. (McCoy) 107. For example, in 1991, Medtronic entered a cross-license with CPI/Eli Lilly that gave Medtronic access to the Mirowski patent portfolio in exchange for access to Medtronic's Activitrax patent. Thus, in a highly concentrated industry,¹⁰ Medtronic licensed the Activitrax patent to three key competitors: CPI/Eli Lilly, Siemens and St. Jude. Siemens and St. Jude affirmatively agreed to pay a 7% royalty for Medtronic's CRDM patent portfolio. Accord Medtronic I at 118 (“[Medtronic] had made transfers of similar intangibles to competitors, and the values of the intangibles were known.”).

2. The Pacesetter Royalty Rate Is a Valid Benchmark

26. In this context, several factors establish the reliability of using the Pacesetter royalty rate as a benchmark for the MPROC Licenses: the high baseline royalty rate; the fact that seminal patents were licensed; competitive tension between the parties; and market players' use of the rate as a benchmark.

(a) The Pacesetter Rate Was High In the Industry

27. First, the Pacesetter Agreement's 7% rate was one of the highest IP royalty rates in the CRDM industry, as established by this Court's findings and fact and expert witness testimony. See, e.g., Medtronic I at 59; MDT I Tr. (Ellwein) 3846:22-23; Ex. 6105-P (Putnam Opening) ¶ 164(a)-(b); Ex. 6101-P (McCoy Opening) at 9-11; Ex. 6102-P (McCoy Rebuttal) at 6; Tr. (McCoy) 100-02, 108, 239; 246-47; Ex. 6103-P (Cohen Opening) 28; Tr. (Cohen) 577; Ex. 6387-R. There is no contrary evidence on this point.

28. Second, royalty rates resulting from patent litigation settlements, like the Pacesetter Agreement's 7% rate, are often higher than rates resulting from purely commercial licenses. An empirical study of royalty rates — on which Respondent's expert Dr. Cockburn relied — demonstrated this trend. Ex. 6331-R at 6616 (T. R. Varner, Technology royalty rates in SEC filings, 45 les Nouvelles 120 (Sept. 2010) (the “Varner Study”)). The Varner Study shows that the average royalty rate in bare patent licenses (i.e., in the absence of litigation) is 3.7%, while in settlement agreements the average royalty rate is 5.9%, from which Dr. Varner infers a settlement royalty premium of 2.2%. Id. at 6616. That premium is consistent with the facts and circumstances of the Pacesetter Agreement. Ex. 6106-P (Putnam Rebuttal) ¶ 17; Tr. (Putnam) 1142-43. The Eighth Circuit referenced an article in Medtronic II that reaches the same conclusion. See Medtronic II (Shepherd, J., concurring) (citing Jonathan S. Masur, The Use and Misuse of Patent Licenses, 110 Nw. U. L. Rev. 115, 124 (2015)); Ex. 6116-P (Masur) at 4611-12 (explaining that settlement licenses “will be more accurate gauges of patent value than licenses negotiated outside of litigation” and that “[t]he closer the plaintiff is to being 100% certain of prevailing, the more accurate the value of the license.”).

(b) The Pacesetter Agreement Involved Seminal Patents

29. The Pacesetter Agreement licensed the patent for Medtronic's “Activitrax” technology. That technology “was an immediate market success.” Ex. 6103-P (Cohen Opening) at 19; see also id. at 19-20, 34; Ex. 6101-P (McCoy Opening) at 4; Tr. (McCoy) 94-95; Ex. 6104-P (Cohen Rebuttal) at 18; Tr. (Cohen) 686. Respondent's expert Mr. Crosby similarly credited Activitrax as seminal. Tr. (Crosby) 3020-22. Other players in the CRDM industry were eager to develop their own products that included rate-responsive pacing. Tr. (McCoy) 107:18-22; id. 308:11-310:7; Ex. 6101-P (McCoy Opening) at 4.

(c) The Pacesetter Agreement was Negotiated Between Competitors

30. The Pacesetter Agreement and its royalty rate were all driven by sharp competitive “tension” between Medtronic and Pacesetter. Medtronic and Pacesetter were “fierce competitors” in the CRDM industry. Through the 1990s and 2000s, the industry was dominated by 3-5 major players, including Medtronic and Pacesetter, and later, St. Jude. Medtronic I at 9; id. at 11 (listing “Medtronic, and then Guidant and St. Jude,” as the dominant competitors from the late 1990s through 2006); Ex. 6101-P (McCoy Opening) at 3-4, 9, 10; Ex. 5528-R (Kruger Opening) at 8, 16-18, 20-24. Leading up to the Pacesetter Agreement, Medtronic had the largest market share, followed by Pacesetter. Ex. 6103-P (Cohen Opening) at 21; Ex. 6101-P (McCoy Opening) at 4. In settling their litigation — in which Medtronic had obtained sweeping injunctive relief — Medtronic was strongly incentivized, and able, to obtain the highest possible royalty from Pacesetter, who naturally had the opposite incentive. Tr. (McCoy) at 106:22-107:10; Ex. 6101-P (McCoy Opening) at 9-10; Ex. 6102-P (McCoy Rebuttal) at 4; Tr. (Putnam) 110305; Ex. 6107-P (Spadea) ¶ 127.

31. The role of competition is also reflected in the Board Documents, which show that Medtronic weighed the exclusionary benefits of Medtronic's patents against the gains from licensing to Pacesetter as a key competitor. Those gains prominently included the competitive benefits of Medtronic's “tax” on its principal competitor, an “ongoing expense to Siemens.” Ex. 2525-J at 4424. This Board analysis demonstrates that the conditions of the Pacesetter Agreement were ideal for profit-maximizing on Medtronic's part. The resulting 7% royalty rate is likely conservative (i.e., high) relative to an agreement in which the licensee is not a competitor, and thus is not subject to the same zero-sum “tax.” Ex. 2525-J; Ex. 6106-P (Putnam Rebuttal) ¶ 35.

(d) Market Players Considered the Pacesetter Rate as a Benchmark

32. The Pacesetter Agreement's 7% royalty rate served as a commercial benchmark for other CRDM participants negotiating with Medtronic to license its patent portfolio. For example, Mr. McCoy testified that he would have considered the Pacesetter royalty rate as a benchmark during his time at Guidant in 2005 and 2006. Ex. 6101-P (McCoy Opening) at 10-11; Ex. 6102-P (McCoy Rebuttal) at 6; Tr. (McCoy) 69-70, 100-02, 239, 307. Respondent's expert Dr. Cockburn agreed that the Pacesetter Agreement “would be considered . . . by business people negotiating an agreement”; “[t]hey would take it into account.” Tr. (Cockburn) 2255:7-11; see also Tr. (Cockburn) 2254-56.

3. The Maximum Rate Clause Further Bridges Differences from 1992 to 2004

33. The Maximum Rate Clause further bridges any possible differences in the patent portfolios that Medtronic US licensed to Pacesetter in 1992 and to Medtronic PR in 2004. During the term of the Pacesetter Agreement, even if Medtronic developed a patent even more foundational than Activitrax or Mirowski, Pacesetter would still never pay more than 15% for Medtronic's entire CRDM portfolio under Maximum Rate Clause. This was true even taking into account the Key Patent Clause (discussed further below), which would have only delayed Pacesetter's access to any of Medtronic's designated patents.

(a) The Negotiation of the Maximum Rate Clause

34. Medtronic and Pacesetter negotiated the Maximum Rate Clause because they foresaw that one or both of them might develop then-unforeseen technologies that could become the subject of future dispute. Ex. 2526-J at 4427. The Maximum Rate Clause addresses this possibility by compelling the parties to negotiate licenses for any future patents, subject to the 15% cap. The Maximum Rate Clause thus balanced the parties' “need for flexibility in responding to future developments with their need for overall strategic predictability.” Ex. 6105-P (Putnam Opening) ¶ 276; see id. ¶¶ 28, 51, 83, 273-76; Tr. (Putnam) 760, 763-67.

35. For the Maximum Rate Clause, Medtronic and Pacesetter had to predict how high royalty rates could go for up to 15 years. As Mr. McCoy described, “Pacesetter and Medtronic, in 1992, must have judged market conditions to be sufficiently predictable such that each could reasonably establish an important agreement that could remain in force for fifteen years.” Ex. 6101-P (McCoy Opening) at 10; see Tr. (Spadea) 1213; Ex. 6105-P (Putnam Opening) ¶¶ 51, 277; Tr. (Putnam) 762-67; Tr. (Cockburn) 2159:6-2160:13; Tr. (Hubbard) 1434.

(b) The Prospective Nature of the Maximum Rate Clause

36. Even putting aside that the Maximum Rate Clause later transferred to St. Jude, the parties' agreement to this clause in 1992 for the agreement's 10-year term establishes the reliability and relevance of the 15% cap. The parties determined ex ante that Medtronic's entire patent portfolio (subject to temporary delays under the Key Patent Clause) would be licensed through 2002 for no more than a 15% royalty, regardless of any changes in the patents or the CRDM industry over that time. Further, had the parties actually invoked the Maximum Rate Clause in the late 1990s or early 2000s, any patent license granted by Medtronic thereunder — covering a period of up to the life of the patent(s) — could have been in effect well beyond the 10-year term of the Pacesetter Agreement. Ex. 6105-P (Putnam Opening) ¶¶ 13(b), 51-52, 81-83, 274-77; Tr. (Putnam) 758-59, 762-67, 774-77; Ex. 6153-PD (Putnam Demo.) at 9.

(c) The Parties' Conduct Reinforces the Relevance of the 15% Cap

37. The parties' actual conduct reinforces the 15% cap as a ceiling for pricing the MPROC Licenses — specifically, Medtronic's decision to not designate any patents as “key” under the Pacesetter Agreement (from 1992 to 2004). MDT I Tr. (Ellwein) 3857-58; Ex. 6105-P (Putnam Opening) ¶ 84. The Key Patent Clause required, in effect, that Medtronic annually assess the value of its CRDM patent portfolio. If Medtronic developed any patent(s) that allowed it to charge more than 15%, then Medtronic — as a rational, commercial actor — would have designated those patent(s) as “key” either to obtain exclusivity or charge a higher royalty rate for a period of time. Ex. 6105-P (Putnam Opening) ¶¶ 82-85, 274-77; Tr. (Putnam) at 760-62, 774-76; 953-54; Ex. 6108-P (Hubbard Opening) ¶ 149 & n.218.

(d) The Settlement Context Did Not Distort The Maximum Rate Clause

38. The Board Documents and other evidence demonstrate that the settlement context of the Pacesetter Agreement did not distort its terms. This is particularly true for the Maximum Rate Clause, which was developed independently of the other terms in the Pacesetter Agreement.

• The Board Documents establish that the Maximum Rate Clause was not contemplated when the parties discussed initial terms in May 1992. At that time, the term most similar to the Maximum Rate Clause merely provided that future patents “would be made available to the other party at reasonable royalty and other terms to be negotiated in the future.” Ex. 2524-J at 4413.

• This vague “reasonable royalty” provision was insufficient from Pacesetter's perspective. In June 1992, Ms. Takafuji insisted on a more robust provision to give Pacesetter access to Medtronic's future patents. Ex. 2526-J.

• Only after this letter — in August 1992, after the other major terms of the Pacesetter Agreement had generally been agreed — did revised settlement terms introduce the Maximum Rate Clause. Ex. 2527-J.

See also Tr. (Putnam) 767-73; Ex. 6105-P (Putnam Opening) ¶ 82.

(e) St. Jude's Acquisition of Pacesetter

39. St. Jude's assumption of the Pacesetter Agreement, including the Maximum Rate Clause, bolsters the reliability of the Maximum Rate Clause. PFF ¶¶ 14-20.

C. Comparability of Intangibles

40. The intangibles licensed in the Pacesetter Agreement and the MPROC Licenses are comparable, and differences are addressed by specific, targeted adjustments to the Pacesetter Agreement's base royalty rate.

1. Comparability of Medtronic's Patent Portfolio in 1992 and 2005

41. The patents licensed in the MPROC Licenses are comparable to the patents licensed in the Pacesetter Agreement.

42. Looking first at the CRDM industry broadly, the state of technology did not experience a “paradigm shift” between 1992 and 2004 that could disrupt technological comparability between the two time periods — for what is otherwise the same type of IP in the same industry for producing the same products. Dr. Cohen defined such a “paradigm shift” as a material change in technology that would materially alter the market and, as a result, put upward pressure on royalty rates. Based on his review of the technology in the CRDM industry, Dr. Cohen determined that no such shift occurred; no new key or foundational patents were introduced during the 1992-to-2004 timeframe. Tr. (Cohen) 472, 477; Ex. 6103-P (Cohen Opening) at 29-30; Ex. 6104-P (Cohen Rebuttal) at 10.

43. Instead, the industry experienced “incremental” or “iterative” improvements in technology between 1992 and 2004. The most important such advance was the development of clinical applications for CRT, or cardiac resynchronization therapy, which was based on technology patented prior to 1992. While these developments helped to sustain steady growth for the industry, they did not change the field's core technology. Ex. 6103-P (Cohen Opening) at 29-30; Tr. (Cohen) 472-73, 704:10-12; Ex. 6101-P (McCoy Opening) at 9; Tr. (McCoy) 94-98, 382. Respondent's expert in Medtronic I, Kurt Kruger, confirmed this. See Ex. 5528-R (Kruger Opening) at 8 (recognizing that profitability during this period was “maintained at above-average rates due to manufacturing efficiencies, evolutionary (not revolutionary) product changes, and developed marketing and selling channels” (emphasis added)). Indeed, Mr. McCoy described the CRDM products in 1992 and 2004 as “strikingly similar.” Tr. (McCoy) 387:9-14.

44. In line with these technological trends, Medtronic's patent portfolio was comparable as between 1992 and 2004. Mr. Spadea determined that the patent portfolios licensed under the Pacesetter Agreement and the MPROC Licenses were technologically comparable, covering similar medical technologies. Ex. 6107-P (Spadea) ¶ 14; Tr. (Spadea) 1205-06; see also Medtronic I at 121 (“The Pacesetter agreement covered some of the same intangibles that were licensed to MPROC.”).

45. Dr. Cohen likewise determined that Medtronic developed no new key patents between 1992 and 2004. As noted above, Medtronic did not identify any “key” patents during the Pacesetter Agreement, and Pacesetter/St. Jude did not request to license any of Medtronic's future patents. That further shows that Medtronic did not develop any critically important patents during the 1992-to-2004 timeframe. Ex. 6104-P (Cohen Rebuttal) at 10; Tr. (Cohen) 477-78.

46. Quantitative analysis of Medtronic's patents also demonstrate substantial similarity in 1992 and 2004. Mr. Spadea determined from a patent citation analysis that the Medtronic patents licensed to Medtronic PR were nearly “identical” to, and comparable with, the patents that were licensed to Pacesetter. Ex. 6107-P (Spadea) ¶¶ 74-79; Tr. (Spadea) 1226-33; Ex. 6155-P (Spadea Demo.) at 5.

47. Though there are differences in the patent portfolios in 1992 and 2004, these differences alone do not indicate that the value of the patent portfolio was greater in 2004 than in 1992. The value of a large patent portfolio is often concentrated in a handful of patents, as explained by Dr. Cohen. Ex. 6104-P (Cohen Rebuttal) at 10; Tr. (Cohen) 477; Ex. 6107-P (Spadea) ¶¶ 61-65; Tr. (Putnam) 870-71. Dr. Cockburn agreed that this principle is not disputed by economists who study patents and stated that “it's very likely that the [tax year] 2005 portfolio's value will — as distributed across the patents that make up that portfolio, will be highly skewed”; i.e., that the value of the patent portfolio licensed to Medtronic PR in 2004 would largely be attributable to a subset of Medtronic's patents. Tr. (Cockburn) 2239:15-2239:18; see also id. 2237-39.

48. Dr. Putnam conducted his own patent citation analysis to assess whether, and to what extent, the value of Medtronic's patent portfolio in 1992 survived into tax year 2005. Dr. Putnam concluded that there was “much greater economic overlap” of the two patent portfolios than suggested by Respondent's expert Dr. Cockburn. Ex. 6106-P (Putnam Rebuttal) ¶¶ 64-65; Tr. (Putnam) 870-71.

49. Dr. Cohen conducted a third patent citation analysis and concluded that Medtronic PR would not have paid a higher royalty because of differences in the patent portfolio in 1992 and 2004. Specifically, the proportion of Medtronic patents in the CRDM patent universe decreased from 45% of CRDM patents in 1992 to only 30% in 2004. Citations to Medtronic patents also decreased, from 44% of all CRDM citations in 1992 to only 37% in 2004. These declines occurred despite the fact that the total number of CRDM patents increased during this period. Thus, the value of Medtronic's patent portfolio was either comparable in 1992 and 2004, or actually weakened during that period. Ex. 6104-P (Cohen Rebuttal) at 5, 10-16; Tr. (Cohen) 479-81; Ex. 6150-PD (Cohen Demo.) at 15.

2. The Value of Patents Versus Know-How

50. The MPROC Licenses licensed IP in the form of patents and know-how, while the Pacesetter Agreement only licensed patents. Medtronic I at 135. The know-how conveyed to Medtronic PR provides significantly less value and protection than the licensed patents.

51. IP has value because it protects innovation. CRDM companies regularly endeavor to innovate and develop new IP to increase profits and market share. A critical element in that process is protecting the developed IP, so the company can exclusively exploit it for a period of time. Tr. (McCoy) 83-85; Medtronic I at 9.

52. Patents were the primary IP vehicle used in the CRDM industry for protecting innovations. Tr. (McCoy) 85:5-7; see also Tr. (McCoy) 83-85; Tr. (Spadea) 1221:24-25 (it was “a very patent-centric” industry); Ex. 6101-P (McCoy Opening) at 7. Innovations can also be embodied in “know-how,” or the knowledge developed by a company as part of the design and manufacture of a product. As relevant for this case, know-how includes three primary components:

53. Product-Specific Know-How: Dr. Cohen defines this concept as know-how that is practical to transfer in a license agreement in association with specific patents or other legally protected intangibles being transferred, such as product blueprints. Ex. 6103-P (Cohen Opening) at 6, 31. Both Medtronic US and Medtronic PR contribute to the development of product-specific know-how. As this Court found, Medtronic PR played a crucial role in design and product development for Devices and Leads in collaboration with Medtronic US. Medtronic PR personnel partnered with Medtronic US through every development phase of new products to ensure that newly developed products were manufacturable at commercial scale. In this regard, Medtronic PR's role in product development was critical. E.g., Medtronic I at 25-27, 29-34, 44-48, 54, 103-08. This type of know-how does not include materials with independent legal protections, such as copyrighted software and trademarks. Ex. 6104-P (Cohen Rebuttal) at 20-21; Tr. (Cohen) 456-57.

54. Quality-Manufacturing Know-How: Dr. Cohen defines this concept as know-how associated with high-volume, high-quality manufacturing capability. This type of know-how is organically embedded in an organization that performs such manufacturing and critically depends on the accumulated expertise and knowledge of its members; policies and processes integrated into the functioning of the organization; and the organization's culture. Ex. 6103-P (Cohen Opening) at 7, 32-33; Tr. (Cohen) 457. Dr. Cohen observed that the quality-manufacturing know-how that a manufacturer like Medtronic PR provides to a product developer like Medtronic US is highly valuable. Ex. 6103-P (Cohen Opening) at 37-42; see also Medtronic I at 40-45, 99-105, 117. Mr. McCoy similarly observed that this know-how “was vested primarily in people — their hard-won experiences applied to new facts and situations. It was also vested in systems and processes that came to comprise something akin to a unique dialect. The know-how was embedded, therefore, in the people and processes of a given industry participant. . . .” Ex. 6101-P (McCoy Opening) at 13. Of the two types of know-how, only product-specific know-how can be readily licensed to a third party. Tr. (Cohen) 456.

55. Regulatory/Clinical Background and Approvals: Under the MPROC Licenses, know-how also includes access to regulatory approvals for Devices and Leads. Both Medtronic US and Medtronic PR contributed to that FDA approval process. Ex. 6105-P (Putnam Opening) ¶ 234; Ex. 6103-P (Cohen Opening) at 13, 41; see also Medtronic I at 30-31, 44-45, 54. Both Medtronic US and Medtronic PR also contributed to clinical trials. Id. at 48. Clinical and regulatory functions may be relatively routine: every FDA-approved seller performs them (themselves or through third-parties), so they do not generally convey a competitive advantage. Ex. 6104-P (Cohen Rebuttal) at 7; Ex. 6106-P (Putnam Rebuttal) ¶ 47; Tr. (Cohen) 502-04; Ex. 6102-P (McCoy) at 7. The competitive advantage of clinical trials was further reduced to the extent that Medtronic made the results of its clinical trials available to competitors. Medtronic I at 15-17.

56. Patents provide significantly greater protection for CRDM innovations than product-specific know-how. Thus, the transfer of product-specific know-how from Medtronic US to Medtronic PR conveyed significantly less value than the transfer of patent rights. Ex. 6103-P (Cohen Opening) at 31-37; Tr. (Cohen) 463, 701-02.

57. Product-specific know-how provides relatively little protection in this industry because a competitor can reverse-engineer a company's product — vitiating the protection of product-specific know-how — in a relatively short period of time, such as a matter of months. A competitor “simply need[s] to know what the novel features of that product are in order to reverse engineer a competitive product. And in so doing, [the competitor] develop [s its] own product-specific know-how relevant to [its] own product.” Tr. (Cohen) 459:16-23, see also id. 45860, 701-02; Ex. 6103-P (Cohen Opening) at 33-34.

58. Because a product can be reverse-engineered so quickly, a license for product-specific know-how will only slightly accelerate a licensee's product development. Thus, a company that is considering whether to license product-specific know-how for Devices and Leads will place a relatively low value on it. Ex. 6103-P (Cohen Opening) at 33-34; Tr. (Cohen) 458-60, 465, 701-02.

59. In contrast, patents provide a long period of protection, generally either 17 or 20 years from the date that the patent was issued. Thus, patents generally provide substantially greater value than product-specific know-how in this industry. Ex. 6103-P (Cohen Opening) at 33-34; Tr. (Cohen) 460, 462, 702.

60. This relative weakness of product-specific know-how is demonstrated by facts in this case: Pacesetter's infringement of Medtronic's Activitrax patent. Medtronic received the patent for this technology in 1984, and in 1986 introduced its Activitrax product into the market. By September 1985, however, Pacesetter had already begun working on its own product using the same rate-responsive pacing technology. And by September 1986, Pacesetter implanted its first “Sensolog” pacemaker using that rate-responsive pacing technology. Because Pacesetter reverse-engineered Medtronic's Activitrax technology in less a year, Medtronic's product-specific know-how lasted for no more than 12 months. In contrast, because Pacesetter infringed the Activitrax patent, Medtronic was able to leverage its patent rights to ultimately receive over $500 million in royalties from Pacesetter. So while Medtronic's product-specific know-how lasted no more than one year, the patents provided substantial value to Medtronic for 17 years. Ex. 6103-P (Cohen Opening) at 34; Tr. (Cohen) 461-63.

61. The relative strength of patents is further demonstrated by CPI's success using its Mirowski patent rights to block Medtronic from the ICD market for 10 years. The Mirowski patent, issued in 1976, was foundational for ICD technology. Medtronic began working on its own ICD product in 1982 — i.e., developed its own product-specific know-how. CPI (the exclusive Mirowski licensee) sued Medtronic for infringement and effectively blocked Medtronic's entrance into the ICD market, notwithstanding that Medtronic had the necessary product-specific know-how. The parties settled in 1991, and CPI then licensed the Mirowski patent to Medtronic. With that license, Medtronic was finally able to release its ICD product in 1992. Thus, the Mirowski patent blocked Medtronic's entry into the ICD market for 10 years. Ex. 6103-P (Cohen Opening) at 34-35; Tr. (Cohen) 461-65.

62. The fact that the license of product-specific know-how to Medtronic PR was “exclusive” does not alter the relatively low value of product-specific know-how. The “exclusivity” of IP is relevant only to the extent that the IP is of value to a potential competitor. A competitor may be able to use another company's product-specific know-how to reverse-engineer a product; but, as described above, the competitor likely does not need that know-how to reverse-engineer successfully. Moreover, each competitor will generally develop its own product with its own product-specific know-how. Thus, the value of “exclusivity” is de minimis. Ex. 6104-P (Cohen Rebuttal) at 19-21; Tr. (Cohen) 465-66, 468.

63. Ultimately, license negotiations in the CRDM industry are focused on licensing patents, not know-how. Cf. Tr. (McCoy) 110-11. The license of know-how — if licensed at all — is a secondary consideration that is often handled in a separate agreement. Such agreements generally require the licensor to make its employees available to the licensee for the purpose of sharing know-how, and the licensee pays for that access on an hourly basis, rather than through a royalty rate on product sales. Ex. 6103-P (Cohen Opening) at 35-36; Tr. (Cohen) 460-61.

64. Respondent's witness Dr. Cockburn argued that, contrary to this evidence, product-specific know-how has high value in this industry comparable to the value of patents. Tr. (Cockburn) 2052:1-18. Yet Dr. Cockburn's own articles undermine this argument. His article The Importance of Patents to Innovation: Updated Cross-Industry Comparisons with Biopharmaceuticals reported the results of a survey of companies in the biotech, pharmaceutical and medical-device sectors. The survey recipients were polled on the relative value of various IP protections. Respondents in all three sectors ranked patents as more important than know-how, trade secrets, trademarks and copyrights. Ex. 6124-P; Tr. (Cockburn) 2197-2204. In addition, the biggest gap between the relative importance of patents and know-how was in the pharmaceutical and healthcare industries, which includes medical devices. Specifically, 89% of respondents in the pharmaceutical and healthcare industries concluded that patents were “extremely important,” whereas only 47% of respondents identified know-how as “extremely important.” Ex. 6124-P at 4807.

65. Dr. Cockburn's position that know-how had a higher value is also based on flawed authorities. For example, Dr. Cockburn relied on a court opinion in an infringement lawsuit that St. Jude brought against both a former employee and a Chinese medical device company, Nervicon. The court in that case held that the defendants had infringed on St. Jude's non-patented IP (trade secrets) and awarded substantial damages to St. Jude. But that damages amount is, on its face, unusable as a relevant valuation. The defendants did not appear for any of the trial, much less to contest damages. St. Jude's “valuation” to support its damages claim was provided by an accountant — not an economist or valuation specialist — who primarily worked on divorces and general business valuation. The accountant based his valuation on public documents such as SEC filings, rather than documents typically used for such valuations, such as St. Jude cash-flow projections related to the non-patent IP. The accountant was not subject to cross-examination. Ultimately, the presiding judge questioned whether there was substantial evidence to support the jury's special verdict of $1.8 billion and reduced damages by over $1 billion. Tr. (Cockburn) 2222-35; Ex. 6317-R at 2.

66. Dr. Cockburn's report ignored these facts. Ex. 6202-R (Cockburn Rebuttal) ¶¶ 100-04. On cross-exam, when confronted with these failings in the Nervicon valuation, Dr. Cockburn testified: “I am in no way suggesting that the damages in [Nervicon] could somehow be used directly to value know-how to the valuation that MPROC might have placed on the know-how to which they gained access through the [MPROC Licenses].” Tr. (Cockburn) 2233:7-11; see also id. 2222-35.

3. The Value of Cardiac Versus Neuro

67. The Pacesetter Agreement licensed patents for use in cardiac Devices and Leads, while the MPROC Licenses licensed IP for use in Cardiac and Neuro Devices and Leads. Medtronic I at 136. This difference in scope of products does not undermine technological comparability because Cardiac and Neuro Devices and Leads are highly similar in their technology, manufacturing and profitability.

(a) Comparable Technology

68. First and foremost, Cardiac and Neuro Devices and Leads have substantial technological overlap. They are derived from the same base technology, with Neuro growing organically and logically out of developments in Cardiac. In Mr. McCoy's experience, “[t]he underlying technologies and product system elements of neurostimulators were so similar to CRM devices that virtually every CRM industry participant had ongoing development or a developed market presence in neurostimulation.” Ex. 6101-P (McCoy Opening) at 14; Tr. (McCoy) 98-99, 22122, 387:15-388:7; Ex. 6103-P (Cohen Opening) at 42-46; Tr. (Cohen) 706.

69. Specifically, both technologies involve implantable pulse generators and leads to deliver electrical energy to excitable tissue. Cardiac Devices were actually more sophisticated than Neuro Devices because they could sense electrical activity and other inputs that Neuro Devices could not. Medtronic I at 40-41, 51-52; Ex. 6103-P (Cohen Opening) at 8, 43; Tr. (Cohen) 706; Tr. (McCoy) 222.

70. The citation of neuro patents to cardiac patents likewise demonstrates the overlap of cardiac and neuro technology. Dr. Cohen analyzed how many patents with “neuro” in their title — i.e., patents for neuro technologies — cited Medtronic and Pacesetter cardiac patents licensed in the Pacesetter Agreement. He determined that, of the 559 cardiac patents in the Pacesetter Agreement, 121 cardiac patents were cited by a neuro patent. In total, 257 neuro patents cite at least one cardiac patent in the Pacesetter Agreement. Ex. 6103-P (Cohen Opening) at 43.

(b) Comparable Manufacturing

71. The manufacture of Cardiac and Neuro products was also substantially equivalent because they are both Class III implantable devices that require the highest level of quality manufacturing. Medtronic I at 52; Ex. 6103-P (Cohen Opening) at 8; Ex. 6149-P (Cohen Sealed Opening) at 43-44; Tr. (Cohen) 706.

(c) Cardiac Products Are More Profitable

72. Overall, Cardiac products are more profitable than Neuro products. The cardiac market is also larger. Dr. Cohen analyzed Medtronic's combined sales of Cardiac and Neuro products and found that Neuro was only a small fraction of Medtronic's overall profits. Ex. 6103-P (Cohen Opening) at 8, 45-46; Tr. (Cohen) 707; Tr. (Putnam) 800-01, 808; Tr. (McCoy) 387-88; Ex. 6101-P (McCoy Opening) at 14; Medtronic I at 137.

4. Devices and Leads

73. Devices and Leads are comparable for similar reasons to Cardiac and Neuro. Medtronic I at 29; Ex. 6101-P (McCoy Opening) at 12; Tr. (McCoy) 388.

D. Comparability of Circumstances

74. The circumstances surrounding the negotiation and terms of the Pacesetter Agreement were similar to those for the MPROC Licenses.

1. “Vertical” vs. “Horizontal” Relationships, Functions and Bargaining Power

75. The relative bargaining power of the respective licensors and licensees in the Pacesetter Agreement and the MPROC Licenses were similar, and the fact that Medtronic and Pacesetter were competitors in the CRDM industry further enhances the reliability of the Pacesetter Agreement's royalty rate as a benchmark. Respondent's experts challenge the circumstantial comparability of the Pacesetter Agreement and the MPROC Licenses — primarily based on the differing “vertical” and “horizontal” relationships and functions of the respective licensees in those two agreements (Pacesetter and Medtronic PR). These challenges are inconsistent with the factual record and applicable economic principles.

(a) Pacesetter's “Horizontal” Relationship Bolsters Comparability

76. Respondent's experts posited that the Pacesetter Agreement, as a license between “horizontal” competitors, cannot be a valid benchmark. Medtronic PR and Medtronic US were not “horizontal” competitors, but instead affiliates in a “vertical” relationship within one company. Ex. 6201-R (Cockburn Opening) ¶ 59; Ex. 6204-R (Heimert Rebuttal) ¶ 26; Tr. (Cockburn) 2048-50, 2285-87.

77. As discussed above, Pacesetter's status as a “horizontal” competitor in fact bolsters the reliability of the Pacesetter Agreement as benchmark. Dr. Heimert likewise testified that an agreement between competitors can be a valid CUT. Tr. (Heimert) 2420:10-2421:6. He agreed that a negotiation between competitors creates the negotiating tension for an arm's length rate because it “means that you have two parties that are independent from each other, and as a result of that . . . both party ha[ve] their own interests in getting the best possible outcome that they can from negotiating that agreement.” Id. 2422:10-22. Thus, Dr. Heimert acknowledged that Pacesetter's status as a “horizontal” competitor does not, in itself, break comparability with the MPROC Licenses.

(b) Functional Differences Do Not Impair Comparability

78. Respondent's experts also list functional differences between the licensees (Medtronic PR and Pacesetter), as well as differences in the functions of the licensor (Medtronic US) in the Pacesetter Agreement and MPROC Licenses. Ex. 6201-R (Cockburn Opening) ¶¶ 53-60; Ex. 6203-R (Heimert Opening) ¶¶ 27-31; Ex. 6205-R (Meyer Opening) ¶¶ 40-42. Respondent's experts argue the “more different the functions are, the less comparable the transactions are.” Ex. 6201-R (Cockburn Opening) ¶ 58. According to Dr. Cockburn, the key functions performed by Pacesetter, but not by Medtronic PR, are: R&D, quality control, regulatory and strategic business management. Id. ¶ 55; Tr. (Heimert) 2426-27.

79. This argument from Respondent's experts is at odds with applicable economic principles. As a threshold matter, there will often be functional differences between an intercompany transaction and a proposed comparable. That is because the taxpayer's functions will generally be divided among affiliates, which is what creates the need for a transfer pricing analysis in the first instance. By comparison, the overall operations of a third-party will frequently be more integrated. Thus, the relevant inquiry is not whether functions in controlled and uncontrolled transactions are similarly divided. Rather, the relevant question is whether any such functional differences render the uncontrolled transaction unreliable as a benchmark. Ex. 6106-P (Putnam Rebuttal) ¶¶ 39-42. As Dr. Putnam testified, the answer to that question in this case is “no” for several reasons.

(i) Other Functions Already Priced

80. The significant functions performed by Pacesetter, but not by Medtronic PR — component manufacturing, maintaining and enhancing the trademark and distribution — are already priced at arm's length. Medtronic I at 75, 129-30. As a result, Medtronic PR is in a similar position to Pacesetter: both have access to these functions (internally or through arm's-length pricing), but need to license technology IP from Medtronic US. Thus, on the facts here, the presence of these functions does not undermine the reliability of the Pacesetter Agreement. Ex. 6106-P (Putnam Rebuttal) ¶ 43; Tr. (Putnam) 990-91. Indeed, Dr. Heimert agreed that the fact that Pacesetter has these other functions does not impair the Pacesetter Agreement's validity as a CUT. Tr. (Heimert) 2428:4-22.

(ii) R&D Function

81. Certain of Respondent's experts cited R&D as another functional difference between Pacesetter and Medtronic PR. Ex. 6201-R (Cockburn Opening) ¶¶ 54, 56-57; Ex. 6203-R (Heimert Opening) ¶¶ 28-30; Ex. 6205-R (Meyer Opening) ¶¶ 40-41. Dr. Heimert agreed, however, that Pacesetter's R&D function does not, in and of itself, break comparability. Tr. (Heimert) 2433:15-19.

82. Differences in R&D activity are relevant to the extent that Medtronic PR accessed additional IP not licensed to Pacesetter. The MPROC Licenses did license know-how and the most current developments in Medtronic US' patent portfolio to Medtronic PR. Both of these items, however, are separately priced with the “portfolio access fee” and know-how adjustments. PFF ¶¶ 124-125, 136-144.

(iii) Other Know-How

83. Respondent's experts also cite quality control, clinical trials and regulatory approval as other functional differences. Ex. 6201-R (Cockburn Opening) ¶ 55; Ex. 6203-R (Heimert Opening) ¶ 31; Ex. 6205-R (Meyer Opening) ¶¶ 40-42. First, these are not categorical differences because Medtronic PR also contributed to these activities. Medtronic PR played a crucial role in quality control, clinical trials and regulatory approval compliance and procedure. Medtronic I at 25-27, 29-34, 44-48, 54, 103-08; Ex. 6105-P (Putnam Opening) ¶ 234; Ex. 6103-P (Cohen Opening) at 41. Second, these items are included in “know-how,” and thus are covered by Dr. Putnam's know-how adjustment. PFF ¶¶ 136-144. Third, clinical and regulatory operations may be relatively routine and typically do not convey a meaningful competitive advantage. PFF ¶ 55.

(iv) Strategic Business Management

84. Finally, Respondent's experts cite “strategic business management” as another functional difference. Ex. 6201-R (Cockburn Opening) ¶ 54; Ex. 6203-R (Heimert Opening) ¶ 31. To the extent not covered by the retail markup over the wholesale price that Medtronic USA, Inc. (“Med USA”) pays to Medtronic PR — which markup covers sales, general and administrative expenditures — this function is compensated through the uncontested, arm's-length pricing for the trademark royalty, component manufacturing and distribution, as well as the proposed royalty rate under the MPROC Licenses. Ex. 6106-P (Putnam Rebuttal) ¶ 49.

(c) Pacesetter/Medtronic PR's Similar Bargaining Power

85. Respondent's experts also argue that Pacesetter had greater bargaining power vis-à-vis Medtronic US than Medtronic PR did. Ex. 6206-R (Meyer Rebuttal) ¶¶ 32, 36; Tr. (Meyer) 1756-58. This claim is contrary to the evidence that Medtronic US had strong bargaining power relative to Pacesetter, similar to Medtronic US's bargaining power with Medtronic PR. PFF ¶¶ 3, 6, 8-9, 27-28, 30.

2. Settlement of Litigation

86. That the Pacesetter Agreement was part of a settlement of litigation, given the circumstances of that settlement, only enhanced the reliability of its terms and, therefore, the use of its royalty rate as a benchmark.

(a) Settlement Licenses in the CRDM Industry

87. Patent litigation and settlement licenses are common in the CRDM industry, particularly for broader patent licenses like the Pacesetter Agreement. Thus, that this commercial license arose out of settlement was not unusual for Medtronic or the CRDM industry generally. It was business as usual. Ex. 6101-P (McCoy Opening) at 6-8; Ex. 6103-P (Cohen Opening) at 19; Ex. 6107-P (Spadea) ¶¶ 21, 103-104; Tr. (Cohen) 700; Tr. (McCoy) 83-86; Tr. (Spadea) 1207-09.

88. Dr. Meyer relied in part on Medtronic's 2006 10-K to opine that “settlement does not appear to be an ordinary business activity for Medtronic US.” Ex. 6205 (Meyer Opening) at ¶ 48. Yet that same 10-K states, “[w]e operate in an industry characterized by extensive patent litigation[,] . . . [and] [a]t any given time, we are generally involved as both a plaintiff and a defendant in a number of patent infringement actions.” Ex. 31-J at 1528. Dr. Meyer did not address this language in reaching her conclusion. Tr. (Meyer) 1843:1-1844:15.

(b) Settlement Licenses More Generally

89. Virtually all patent licenses are negotiated in the shadow of litigation, even in absence of an active lawsuit. No licensee would pay a penny in royalties if the licensor lacked the means to enforce its rights via litigation. Royalty negotiations are thus necessarily based on the outcome the parties would expect in litigation. Ex. 6105-P (Putnam Opening) ¶ 130; Ex. 6116-P (Masur) at 4586; Ex. 6107-P (Spadea) ¶ 21; Tr. (Spadea) 1210-11.

90. This principle is reinforced by Medtronic's own commercial philosophy for licensing its patent portfolio, which was to consider the same factors whether the license arrangement was established in a litigation or non-litigation context. This overlap in philosophy was presented to Medtronic's Board when management sought approval of the terms of the Pacesetter Agreement. Ex. 2525-J at 4419-20; Ex. 6105-P (Putnam Opening) ¶¶ 147-48, 150-52; Ex. 6106-P (Putnam Rebuttal) ¶ 30; Tr. (Spadea) 1208.

91. In many cases, the terms of license agreements negotiated in settlement of litigation more accurately reflect the value of the licensed IP than licenses negotiated outside of direct litigation. Litigation can make the parties better informed as to their respective rights and obligations. In this regard, litigation can resolve, rather than cast, any “cloud” on a patent's value. Thus, litigation can move the bargain toward its true outcome — rather than distort the bargain — by resolving uncertainty and giving legal force to each party's rights. A license negotiated in the course of litigation that has progressed substantially (and with the patent owner was winning) will often be the most reliable indicator of IP value. The greater the probability that the patent owner would prevail at trial, the closer the value of the license to the true value of the patent. Tr. (Putnam) 764-66; Tr. (Spadea) 1211-12; Ex. 6105-P (Putnam Opening) ¶¶ 131-33, 165-70; Ex. 6107-P (Spadea) ¶¶ 15-17; Ex. 6116-P (Masur) at 4611-12; Ex. 6331-R (Varner).

92. Dr. Cockburn agreed that patent settlement licenses are not categorically unreliable. He agreed that in some cases, licenses negotiated as litigation settlements are highly reliable gauges of patent value — more reliable than licenses negotiated outside of litigation — where the plaintiff is winning handily. Dr. Cockburn considered a 95% likelihood of a court decision in favor of the plaintiff as a circumstance in which the plaintiff would be winning handily. Tr. (Cockburn) 2176-80. Dr. Heimert acknowledged that the context of settlement does not, in itself, cause the Pacesetter Agreement to be an unreliable benchmark. Tr. (Heimert) 2440, 2443. Dr. Meyer also agreed that patent settlement licenses can be probative of a reasonable royalty. Tr. (Meyer) 1922. Dr. Meyer has used settlement licenses as a tool to value patents. Id. 1929-33.

(c) Settlement Context of the Pacesetter Agreement

93. The litigation-related circumstances surrounding the Pacesetter Agreement enhanced the reliability of the Pacesetter royalty rate as a benchmark in this case. The Pacesetter Agreement was a paradigmatic case of litigation moving the bargain toward its true outcome.

(i) Procedural Posture

94. The procedural posture of the Pacesetter Litigation eliminates concerns about distortion of the royalty rate. Medtronic I at 56-57; Ex. 6105-P (Putnam Opening) ¶¶ 161-70; Tr. (Putnam) 764, 997-1001; Ex. 6107-P (Spadea) ¶¶ 19, 57. Before the negotiation of the Pacesetter Agreement began in earnest, the district court in the Chicago Litigation had ruled that Medtronic's Activitrax patent was valid and infringed upon by three of five Pacesetter products. Medtronic I at 56-57; Ex. 6105-P (Putnam Opening) ¶¶ 162, 357-67. Thus, at the time of the licensing negotiations with Pacesetter, Medtronic had succeeded resoundingly, such that Pacesetter had “an injunction gun to their heads.” MDT I Tr. (Dennis) 3896. And for Pacesetter's other two products on the market, Medtronic filed the Minnesota Litigation, and Medtronic assessed its likelihood of prevailing in that lawsuit at 60%. Ex. 6105-P (Putnam Opening) ¶¶ 162, 357-67.

95. Siemens recognized it had limited negotiating leverage. PFF ¶ 139. Indeed, Medtronic ultimately obtained more favorable terms from Pacesetter than expected during the earlier stages of the negotiation, notwithstanding Pacesetter had filed an intervening countersuit. PFF ¶¶ 3, 6, 8-9, 27-28.

96. Thus, the settlement context, if anything, created upward pressure on the Pacesetter royalty rate and drove it to the true value of Medtronic's patents. Ex. 6105-P (Putnam Opening) ¶¶ 161-70; Tr. (Putnam) 1008 (“[A]s a result of the litigation, Medtronic learned valuable things. It learned that it had valid patents and it learned that Pacesetter infringed them. That drove up the rate. It learned that it had an injunction at least against some Pacesetter products. That also drove up the rate because it increased Medtronic's bargaining power.”); Tr. (McCoy) 250:8-11 (“Pacesetter was losing in the courts. Medtronic had them over a barrel.”); Ex. 6107-P (Spadea) ¶ 18; Tr. (Spadea) 1211; Tr. (Cohen) 723; Ex. 6101-P (McCoy Opening) at 10; Ex. 6103-P (Cohen Opening) at 22; Ex. 6331-R (Varner).

(ii) Other Litigation Factors

97. It is true that, in certain circumstances, the litigation context may distort settlement terms because the parties may settle based on avoided litigation costs rather than the value of the underlying IP. Ex. 6105-P (Putnam Opening) ¶¶ 133-35, 166. In addition, in a litigation involving multiple claims, settlement of one claim may be affected by resolution of other, unrelated claims. None of these factors distorted the terms of the Pacesetter Agreement for a number of reasons.

98. First, one does not need to “speculate” as to whether factors such as litigation costs or the resolution of other litigation claims distorted the Pacesetter royalty rate; the Board Documents provide a contemporaneous record of Medtronic's assessment. These Board Documents establish that considerations such as litigation costs and the resolution of other litigation claims did not affect the Pacesetter royalty rate or other key terms (such as the Maximum Rate Clause). The Board and management exhaustively considered multiple business objectives and constraints for the Pacesetter Agreement, as it would for any large commercial contract. Any rational licensor would evaluate the same strategic goals, knowing that litigation was an alternative. Tr. (Putnam) 780-85; Tr. (Spadea) 1208; Ex. 6106-P (Putnam Rebuttal) ¶¶ 30; Ex. 6105-P (Putnam Opening) ¶¶ 144, 150-52.

99. In addition to its detailed qualitative assessment, the Board undertook an even more detailed quantitative assessment of its “settle vs. litigate” decision. This quantitative analysis of the license terms demonstrates that Medtronic solely factored in the value of Medtronic's patent infringement claims. Ex. 2522-J. None of the other claims were factored into that quantitative analysis. Elsewhere, the Board Documents acknowledged the mutual value of resolving other pending cases between Medtronic and Pacesetter; but that mutual value did not affect the terms of the Pacesetter Agreement's patent license. Ex. 6105-P (Putnam Opening) ¶¶ 153-59, App. I.C; Ex. 6106-P (Putnam Rebuttal) at 27; Tr. (Putnam) 784.

100. The Board Documents show that Medtronic expected to pay an additional $17 million (NPV) for litigating the Pacesetter case to conclusion. Ex. 2524-J at 4414. This is a reliable projection of Medtronic's litigation costs because it was a necessary input for Medtronic to assess whether to continue to pursue litigation or settle. Tr. (Putnam) 783-84; see also Tr. (Cockburn) 2097 (acknowledging that “it's reasonable to infer that that $5 million a year is connected to these patent cases in which the damages are being forecast”).

101. Evidence and expert testimony establish that this expected litigation cost did not distort the terms of the Pacesetter Agreement. First, the $17 million of avoided litigation costs is insignificant as compared to the expected future cash flows from the license itself, which ultimately had an NPV of over $205 million. Second, these litigation costs were symmetrical. Both Medtronic and Pacesetter were sophisticated market competitors that faced similar litigation costs, and both companies were mutually relieved of those costs through settlement. On these facts, litigation costs did not influence the settlement terms. Ex. 2527-J; Ex. 6105P (Putnam Opening) ¶¶ 134-35; Tr. (Putnam) 783-84, 1006-09; Tr. (Spadea) 126566, 1383-84. Dr. Meyer also conceded, while not accepting the term symmetry, that Siemens would have had litigation costs too, and Siemens could withstand litigation costs associated with prolonged litigation. Tr. (Meyer) 1870-71.

102. Moreover, Dr. Putnam and Mr. Spadea concluded that the avoided litigation costs were not factored into calculation of the expected royalty rate, based on Medtronic's quantitative assessment. Ex. 6105-P (Putnam Opening) ¶¶ 153-59; Tr. (Putnam) 782-84; Tr. (Spadea) 1259-60. Dr. Meyer agreed that the Board Documents showed no reduction in the royalty rate or other offset to account for reduced litigation costs. Tr. (Meyer) 1876.

103. Despite the Board Documents, Dr. Meyer argued that litigation costs distorted the royalty rates, citing Mr. Ellwein's statement that such litigation costs were “nearly crippling” and that the Pacesetter license was “financially attractive.” Ex. 6205-R (Meyer Opening) ¶¶ 30-31. Over the same 5-year period over which the $17 million NPV of litigation costs were forecast ($5 million per year), Medtronic's financial statements show over $10 billion in revenue. In this context, however Medtronic may have characterized its past litigation costs, the expected future litigation costs were not “crippling.” Ex. 6106-P (Putnam Rebuttal) ¶ 25.

104. The fact that the Pacesetter Agreement is predominantly a running royalty also strongly indicates that it is, fundamentally, a commercial license not distorted by litigation costs. Ex. 6105-P (Putnam Opening) ¶ 134. The Pacesetter Agreement did include two lump-sum payments along with the larger running royalty; one was for past damages, however, while the other was a prepaid credit against the running royalty and is reflected in Dr. Putnam's adjustments. PFF ¶¶ 106-108.

(iii) St. Jude Acquisition

105. Finally the St. Jude acquisition affirmed the arm's-length nature of the Pacesetter Agreement and demonstrates that litigation-related factors did not distort the royalty rate. PFF ¶¶ 14-20.

3. The Contractual Terms of the Pacesetter Agreement Support Its Use as a Benchmark

(a) $50 Million Fixed Payment

106. The $50 million upfront payment under the Pacesetter Agreement was for past infringement and therefore does not affect the go-forward royalty rate for the MPROC Licenses. Ex. 6105-P (Putnam Opening) ¶¶ 309-13; Ex. 6108-P (Hubbard Opening) ¶¶ 131, 156; Ex. 6107-P (Spadea) ¶ 63.

107. Dr. Meyer cites Ex. 6312-R (Response to IDR 83) to argue that the $50 million is part of the running royalty, but ignores the context in that document and other evidence that the $50 million was for royalties on past sales. Ex. 6205-R (Meyer Opening) 44; see Ex. 6202-R (Cockburn Rebuttal) 115. In particular:

• Plain Terms: The Pacesetter Agreement states that “$50,000,000 shall constitute royalties paid by Siemens as partial consideration for settlement of the Chicago Case, Chicago Federal Circuit Case and the Minnesota Patent Case.” Ex. 2504-J, § 3.02(a).

• Analysis Relative to Past Sales: The relationship of $50 million to past sales matches a 7% base royalty rate, showing that the $50 million did not affect the 7% base royalty rate for future sales. Ex. 6105-P (Putnam Opening) ¶¶ 309-11 & Ex. 6.1; Tr. (Putnam) 1019, 1021-22.

• IDR Response: The language that Dr. Meyer cited — that payments were “split into a $50 million upfront lump-sum royalty and a[n] ongoing royalty, there [was] no difference in substance between the two types of payments: they all constitute[d] royalties” — supports that the $50 million was for the past. Ex. 6205-R (Meyer Opening) ¶ 44. That language does not characterize the $50 million as related to future sales. Rather, it explains the $50 million was a royalty on Pacesetter's past sales. Ex. 6312-R (Response to IDR 83); Ex. 6106-P (Putnam Rebuttal) ¶ 72; Tr. (Putnam) 1019, 1023-30.

• Medtronic Accounting: A 1992 document showing Medtronic's financial accounting analysis of the settlement repeatedly stated that the $50 million was for “past royalties.” Ex. 6161-P; see Tr. (Meyer) 1902-03.

• Medtronic Public Messaging: An internal Medtronic memo prepared for the announcement of the Pacesetter Agreement states, “Siemens makes a payment of $50 million for access to Medtronic patents and for past royalties.” — contrasting with the immediately subsequent statement that “Siemens also prepays $25 million for future royalties.” Ex. 6238-R at 0049.

• Expert Analysis: Medtronic's experts opined that the $50 million was for past infringement. See Ex. 6108-P (Hubbard Opening) ¶¶ 131, 156; Ex. 6105-P (Putnam Opening) ¶¶ 309-13; Ex. 6107-P (Spadea) ¶ 63.

(b) $25 Million Fixed Payment

108. The fixed payment of $25 million was not a lump sum royalty, but rather a pre-paid credit against a portion of the future running royalty. Medtronic I at 58, 121. Dr. Putnam addressed this aspect of the future running royalty by making a “portfolio access fee” adjustment. PFF 124-125.

(c) The Cross-License

109. The cross-license provision in the Pacesetter Agreement does not undermine the reliability of the 7% rate as a benchmark. The Pacesetter Agreement essentially reflects a one-way license from Medtronic to Pacesetter, and Dr. Putnam in any event adjusted for any benefit that Medtronic potentially received from access to Pacesetter's patent portfolio. PFF ¶¶ 126-130.

(d) The Term of the Agreement

110. The initial term of Pacesetter Agreement was 10 years. Ex. 2504-J. The MPROC Licenses had an initial term of 3 years (Sept 2001-April 2003), renewable at both parties' option. Exs. 7-J; 11-J. The MPROC Licenses were renewed through tax year 2005, and the tax year 2006 amended agreement was then renewed for an additional three years. Exs. 8-J; 9-J; 10-J; 12-J; 13-J; 14-J § 7.1; Ex. 6105-P (Putnam Opening) ¶ 316. The difference in term length does not render the two agreements dissimilar. Nor does it require any adjustment to the base royalty rate beyond those already made to account for the differences in periods and circumstances (such as the 1.8% portfolio access fee adjustment and profitability adjustment). Ex. 6105-P (Putnam Opening) ¶¶ 314-17; PFF ¶¶ 124-125.

4. Medtronic PR and Pacesetter Bore Product Liability Risk in the Same Manner

111. Relative to Medtronic US, each of Pacesetter and Medtronic PR bore, under their respective agreements, the same amount of product liability risks over the products manufactured using the licensed Medtronic IP. Both licensees — Pacesetter and Medtronic PR — also bore catastrophic product liability risk associated with manufacturing Class III implantable CRDM devices. Accordingly, the allocation of product liability risk as between the licensor and the licensee, and the type of risk borne by the licensee, were the same under both agreements.

(a) Under the MPROC Licenses, Medtronic PR Bore All Product Liability Risk

112. The factual record from Medtronic I establishes that Medtronic PR — and not Medtronic US — bore all product liability risk relating to finished Devices and Leads under the MPROC Licenses. In particular, the MPROC Licenses provided that “Licensee [i.e., Medtronic PR] is liable for all costs and damages arising from recalls and product defects.” Exs. 7-J at 4177; 11-J at 4354 (emphasis added).¹¹

(b) Under the Pacesetter Agreement, Pacesetter Bore All Product Liability Risk

113. Under the Pacesetter Agreement, Pacesetter — not Medtronic — bore all product liability risk relating to the Class III CRDM implantable products manufactured using the licensed patents. Ex. 2504-J. While the Pacesetter Agreement did not specifically allocate the product liability costs to Pacesetter as the licensee or otherwise provide for indemnification of the licensor (Medtronic), express allocation or indemnity was unnecessary — Pacesetter controlled and performed all the activities with respect to producing the finished devices and leads (whereas Medtronic was merely the licensor). If Medtronic, as the licensor in a third-party context, had been concerned about any potential product liability risk relating to products manufactured by the licensee (Pacesetter), Medtronic would have insisted that Pacesetter provide an indemnity. No such indemnity was provided, and there is no evidence indicating that Medtronic requested such indemnity during their negotiations.

(c) Medtronic PR's Assumption of Product Liability Risk

114. The allocation of product liability risks and costs to Medtronic PR pursuant to the MPROC Licenses was fully consistent with:

• The parties' course of conduct.

• Medtronic PR's financial capacity to bear the product liability costs.

• Medtronic PR's control over the activities that directly influenced the magnitude of potential product liability costs.

(i) Parties' Course of Conduct

115. Consistent with the requirements in the section 482 regulations, Medtronic PR paid all costs and damages arising from recalls and product defects, in accordance with its obligations under the MPROC Licenses. During Medtronic's 2005 and 2006 tax years, Medtronic PR in fact paid all product liability costs, except for the amount that it reallocated to Medtronic US solely to comply with the terms of the Memorandum of Understanding (“MOU”) entered into with Respondent. After Respondent abandoned the MOU, Medtronic submitted a refund claim to Respondent (which was denied) to, among other things, reallocate all of the product liability costs back to Medtronic PR in accordance with the MPROC Licenses and Medtronic PR's obligations. The MOU did not in any way alter the terms of the MPROC Licenses or the parties' responsibilities and obligations regarding recalls and product defects; it merely revised the pricing of the transactions between them. Once clear that Respondent would no longer respect the terms of the MOU, Medtronic reverted to having Medtronic PR pay for all product liability costs that it otherwise bore, in accordance with the terms of the MPROC Licenses and how the parties have viewed their responsibilities for product liability before, during, and after the MOU was in effect. Medtronic I at 61; MDT I Tr. (Albert) 3546-67; Exs. 2726-J; 2727-J; 2728-J; 2730-J.

(ii) Medtronic PR's Financial Capacity

116. Medtronic PR had the financial wherewithal to bear catastrophic product liability risk. Medtronic PR had the financial wherewithal to bear multi-billion-dollar product liability claims based on its cash flows from 2002 forward. Ex. 6109-P (Hubbard Rebuttal) ¶¶ 71-78. Dr. Hubbard's analysis demonstrated how Medtronic PR would be able to issue debt to satisfy such a product liability claims and then repay that debt using future cash flows. Dr. Hubbard's conservative approach severely discounted those future cash flows to account for a presumptive sharp sales decrease due to the product liability issue. Id. ¶ 77.

117. With respect to past cash flows, Respondent's critique of Medtronic PR's financial wherewithal focuses on Medtronic PR's current net worth. Ex. 6109-P (Hubbard Rebuttal) ¶ 74. This argument ignores the cash distributions that Medtronic PR made to its parent, Medtronic Holdings Switzerland GmbH (“Swiss HoldCo”), for efficient cash management (i.e., non-tax business reasons). Medtronic I at 23-24; Ex. 2249-P at 6155-56 & n.4; Ex. 118-J at 5217. Swiss HoldCo was an international treasury center for Medtronic and collected and deployed cash from different foreign operations, including Medtronic PR. Medtronic PR thus distributed cash to Swiss HoldCo as part of Medtronic's broader internal cash management and treasury functions, undertaken by any multinational business in the ordinary course. Yet, Respondent excludes Medtronic PR's distributions to Swiss HoldCo from Medtronic PR's current net worth.

118. Those prior cash flows are properly included in determining Medtronic PR's financial wherewithal. With the benefit of this distributed cash — which Medtronic PR would, of course, not distribute as an arm's-length party if it saw major product liability claims on the horizon — Medtronic PR would clearly have financial wherewithal to withstand a catastrophic product liability event.

(iii) Medtronic PR's Control Over Activities that Influenced Product Liability Risk

119. Medtronic PR had full control over the business activities that would directly, and most significantly, influence the amount of loss from product defects.

120. The stages of development and business activities involved in the production of Devices and Leads included design and product development, manufacturing and regulatory compliance. Medtronic I at 7-9, 16-17, 25-27, 29-34, 44-48, 54, 103-08; Ex. 5485-P at 31-33, 39, 41-43; MDT I Tr. (Schultz) 4372-77, 4390, 4425; Ex. 5486-P (Chappell Opening) at 43-44, 46-62; MDT I Tr. (Chappell) 4529-30, 4661-62, 4666-75, 4679-84, 4695-97, 4710-11. Both Medtronic US and Medtronic PR performed the activities for these operations to varying degrees, with the exception of manufacturing. PFF ¶ 53-55.

121. Medtronic PR had full control over, and was solely responsible for, the manufacturing operations for the finished Devices and Leads — from beginning to end. Medtronic PR sourced and purchased all the necessary components and materials, managed the supplier relationships, inspected the quality of components and materials, assembled those raw materials and built them into Class III finished Devices and Leads, tested and sterilized the products, packaged the products and, finally, Medtronic PR employees were the last individuals to inspect and handle the finished Devices and Leads prior to their sale. Neither Medtronic US nor Med USA (nor any of their employees or divisions) performed any of these functions. Medtronic I at 28-29, 31-33, 42-43, 45-46, 52-54, 108, 117.

122. The manufacturing process for Devices and Leads is extremely complicated and labor intensive. Id. at 28-30, 53-54. Moreover, Medtronic PR's reliable manufacture of the products at scale was critical to ensuring product quality and, ultimately, Medtronic's success as a Class III Device and Leads manufacturer. Id. at 10, 29, 31-34, 44-45, 99-108. Because Devices and Leads must last for many years in the human body, it is and was critical that Medtronic PR meet its quality obligations 100% of the time. Id. at 14, 29, 99-100, 102, 105, 106; MDT I Tr. (Steinhaus) 1046-50, 1066-67; MDT I Tr. (Mahle) 460. Manufacturing and the resulting quality of the final Device or Lead is and was the most critical function that would determine the success or failure of a Device or Lead. Medtronic I at 10, 29-30, 99-105.

E. Potential Adjustments to the Pacesetter Agreement

123. The factual record and expert testimony presented on remand support certain targeted adjustments to the Pacesetter Agreement's 7% base rate to determine the arm's-length royalty rate for the MPROC Licenses.

1. “Portfolio Access” Fee

124. The Pacesetter Agreement included an 1.8% upward adjustment to the royalty rate on sales into the United States — the “portfolio access” fee — resulting in an 8.8% royalty rate. Ex. 2504-J at 4466. During negotiation of the Pacesetter Agreement, Medtronic requested, and Pacesetter agreed, to this higher 8.8% royalty rate during the initial period of the Pacesetter Agreement, when Medtronic's patents subject to the license were more current. Pacesetter thus paid this additional 1.8% “portfolio access” fee to compensate Medtronic for the greater value of Medtronic's patents during the initial license period, when those patents were most current and thus incrementally more valuable. This “portfolio access” fee was prepaid via Pacesetter's $25 million payment, which was credited against the 1.8% fee. Ex. 6105-P (Putnam Opening) ¶¶ 30, 99-100, 102-08, 312-13; Ex. 6106-P (Putnam Rebuttal) ¶¶ 44-46; Tr. (Putnam) 790:1-8, 991-96. Once Pacesetter had accumulated $25 million in credits against this prepayment, the royalty rate reverted to the 7% base rate for the term of the agreement.

125. In the MPROC Licenses, Medtronic US provided Medtronic PR with ongoing access to the current patents and other IP in Medtronic US's portfolio. E.g., Exs. 7-J to 14-J. Thus, in terms of access to the most-current patents, Medtronic PR in the MPROC Licenses was in a comparable position to Pacesetter during the earlier period of the Pacesetter Agreement (i.e., when the Medtronic patents available to Pacesetter were most current). The 1.8% “portfolio access” fee is therefore properly applied to Medtronic PR through the term of the MPROC Licenses. Ex. 6105-P (Putnam Opening) ¶ 108; Tr. (Putnam) 790:9-13 (“Since MPROC gets Medtronic's most recent technology in perpetuity, then we adapt that to the MPROC Agreement and permanently add [the “portfolio access” fee].”).

2. Cross-License

126. The Pacesetter Agreement is nominally a cross-license of the parties' respective patent portfolios. Ex. 2504-J at 4458-60. Thus, in theory, the Pacesetter Agreement's 7% royalty rate could have been a “net” rate. In other words, the 7% rate on the face of the Pacesetter Agreement could have been “net” of an implied royalty paid by Medtronic to Pacesetter for the value of Pacesetter's patents to Medtronic. But this was not the case. As found in the first trial, and further supported by the expert testimony, Pacesetter's patents cross-licensed to Medtronic under the Pacesetter Agreement provided no value to Medtronic. See Medtronic I at 58; Ex. 6105-P (Putnam Opening) ¶¶ 173-94; Tr. (Putnam) 790:15-17 (“The evidence is overwhelming that those [Pacesetter] patents were not worth anything. . . .”).

127. The Pacesetter Agreement only meters royalty payments based on Pacesetter's sales, with no offset based on a percentage of Medtronic's sales. Rather, Pacesetter's license to Medtronic was “irrevocable” and “fully paid up” from the beginning, Ex. 2504-J at 4459, while Medtronic could terminate Pacesetter's license upon non-payment. Also, the Board's sophisticated modeling did not reflect any netting of Medtronic's and Pacesetter's sales. Moreover, the structure of the Maximum Rate Clause demonstrates that the 7% rate is not a “net” rate. Tr. (Putnam) at 1014:22-25 (“[T]he characterization by both Medtronic and Siemens was that this was a payment from Siemens to Medtronic with nothing coming back the other way.”); id. at 1015-18; Ex. 6105-P (Putnam Opening) ¶ 186.

128. In addition, Analysis of the $50 million fixed payment for past infringement also illustrates that the 7% base rate was the “gross” rate for Medtronic's license to Pacesetter. Tr. (Putnam) at 1021:19-23; Ex. 6105-P (Putnam Opening) ¶ 187.

129. Mr. Ellwein also testified that the value of Pacesetter's patents to Medtronic was “very small, none”; Medtronic “didn't infringe any of those patents” and Siemens' patents “didn't involve any seminal patents on new technology.” MDT I Tr. (Ellwein) 3844:1-8; see MDT I Tr. (Coyle) 1340:2-4 (testifying to the “imbalance in the patents being offered by the two companies”). Board Documents likewise show that Medtronic's management did not assign any value to a license to Pacesetter's portfolio. Ex. 2524-J at 4412-15. Even Siemens viewed the arrangement as Pacesetter paying for Medtronic's patents, not vice versa. Ex. 2526-J at 4427; see also Ex. 6312-R at 9686 (expressing Siemens' recognition to Medtronic that “our future is in your hands”). Expert testimony likewise concluded that Medtronic did not materially value the Pacesetter patent portfolio. Ex. 6103-P (Cohen Opening) at 21-22; Ex. 6101-P (McCoy Opening) at 12; Tr. (McCoy) 28891; Ex. 6107-P (Spadea) ¶ 55. In contrast, Pacesetter needed Medtronic's foundational rate-responsive technology. Ex. 6101-P (McCoy Opening) at 9-10.

130. Given the clearly minimal value to Medtronic of the cross-license, the 7% base rate was not a “net” rate, as Respondent has suggested. There was some theoretical value to Medtronic for access to Pacesetter's current patents if they could unexpectedly later became relevant to future product development.¹² Tr. (Putnam) 790:17-20. But this theoretical value to Medtronic was minimal for several reasons, and so is more-than-adequately addressed by Dr. Putnam's 0.5% to 1.0% adjustment for it. Ex. 6105-P (Putnam Opening) ¶¶ 195-200.

3. Profitability

131. Dr. Putnam calculated a profitability adjustment based on the relative profit margins of Pacesetter in its 1992-to-1994 fiscal years and the global Medtronic CRDM business in its 2003-to-2005 fiscal years. The Pacesetter 1992 to 1994 financial data was from St. Jude's public financials as part of its acquisition of Pacesetter. The Medtronic financial data was in Medtronic's “Annual Operating Plans,” which are contemporaneous Medtronic financial records used for internal business decision-making and that provide measures of profitability for various Medtronic divisions and geographies. Exs. 8-J; 6273-R; Ex. 6105-P (Putnam Opening) ¶¶ 256-93; Ex. 6106-P (Putnam Rebuttal) ¶ 142.

132. Dr. Putnam originally believed that the Pacesetter profit margins were for Pacesetter's global business, so he compared the Pacesetter margins to Medtronic's margins for its global CRDM business. Subsequently, he learned that the Pacesetter margins in St. Jude's public filings were limited to Pacesetter's North American business. Ultimately, however, Dr. Putnam determined that his original profitability adjustment remained correct (and conservative) because comparisons against Medtronic's US/North America margins resulted in adjustments equal to or below his original adjustment range. Tr. (Putnam) 1161:12-1164:21.

133. There is, economically speaking, no magic formula for relating profitability and royalty rate. Ex. 6105-P (Putnam Opening) ¶¶ 265-68. Rather, as the Federal Circuit stated in Uniloc USA, Inc. v. Microsoft Corp., in applying the Georgia-Pacific factors (including “factor 12” relating to profit),¹³ the analysis “must be tied to the relevant facts and circumstances of the particular case at issue.” 632 F.3d 1292, 1317-18 (Fed. Cir. 2011) (emphasis added).

134. Dr. Putnam's analysis did precisely that, looking at the economic facts and circumstances of the Pacesetter Agreement and the MPROC Licenses to determine: (1) whether an adjustment is necessary; and (2) how such an adjustment should be calculated. On whether to make a profitability adjustment, Dr. Putnam concluded that there are “cognizable arguments for and against further adjustment to the base rate to reflect differences in . . . profitability.” Ex. 6105-P (Putnam Opening) ¶ 258. For this judgment call, Dr. Putnam followed the “conservative]” approach of making an adjustment. Ex. 6105-P (Putnam Opening) If ¶ 293. On how to adjust, Dr. Putnam concluded that an economically appropriate profitability adjustment should be “proportional,” adjusting the base royalty rate in proportion to the relative difference in profit between Pacesetter and Medtronic PR. Ex. 6105-P (Putnam Opening) ¶¶ 294-95.

135. Within this “facts and circumstances” framework, Dr. Putnam's conclusions, including the use of a proportional adjustment, are supported by:

• The Maximum Rate Clause: As Dr. Putnam states, “transformational patents that could generate significantly higher profitability for the licensee” were subject to that provision. Ex. 6105-P (Putnam Opening) ¶ 277. In other words, the royalty “cap” in the Maximum Rate Clause cabins how much Dr. Putnam expected the royalty rate to increase with profit, supporting a proportional adjustment in this case.

• “[T]he significant overlap in the patents licensed to Pacesetter and MPROC,” id. ¶ 292, which Dr. Putnam analyzed extensively in his report. Id. ¶¶ 78-86. This overlap in patents cabins the profitability adjustment in a manner similar to the Maximum Rate Clause.

• Other Pacesetter Agreement Terms: Particularly that “the Pacesetter Agreement provides some evidence that royalty rates may vary with profitability: Medtronic charged Pacesetter higher rates in regions where profits were higher (the US and Japan) than where they were lower (the rest of the world).” Id. ¶ 39; Tr. (Putnam) 794-795.

4. Know-How

136. The Pacesetter Agreement did not license know-how, while the MPROC Licenses did. Dr. Putnam concluded that an adjustment is necessary to account for know-how. Dr. Putnam concluded that an adjustment of between 1% and 3% adequately compensated Medtronic US for the annual “net” know-how transferred to Medtronic PR. Ex. 6105-P (Putnam Opening) ¶¶ 235-46.

137. The record demonstrates that the 1% to 3% rate is appropriate:

138. Collaboration on Know-How: While Medtronic US licenses know-how to Medtronic PR, the Court recognized in Medtronic I that Medtronic PR also shares, and collaborates on, know-how with Medtronic US. The value of the quality manufacturing know-how that Medtronic PR provides to a product developer like Medtronic US is highly valuable. Both Medtronic US and Medtronic PR had know-how, which flowed generally in both directions. PFF ¶ 84; Tr. (McCoy) at 110. Given the ongoing sharing of know-how between the two entities and the nature of know-how, Dr. Putnam recognized that it would be difficult to precisely identify the know-how possessed by Medtronic PR that was solely attributable to what Medtronic US shared. Ex. 6105-P (Putnam Opening) ¶ 241.

139. Annual Payment: Medtronic PR should only have to compensate Medtronic US for the know-how created and transferred each year. Since its creation in Medtronic's 2002 tax year, and through the 2004 tax year, Medtronic PR compensated Medtronic US for all the intangible property it licensed, including know-how, through royalty payments under earlier iterations of the MPROC Licenses. And historic manufacturing know-how that existed prior to the creation of Medtronic PR was also compensated for as part of the 2001 restructuring. Respondent has agreed the royalties paid by Medtronic PR from tax years 2002 through 2004, including for know-how, were arm's length. Medtronic I at 66-67.¹⁴

140. Short Useful Life: Know-how has a short useful life, and product-specific know-how for CRDM devices has significantly less value than the patents. PFF ¶¶ 50-56; Ex. 6105-P (Putnam Opening) ¶ 243(a). Thus, Dr. Cohen testified that a rate of “far less” than 3% would be reasonable for product-specific know-how, and 1% would be “more than enough.” Tr. (Cohen) 703:9-15.

141. Comparison to Foundational Patents: Mirowski licensed his foundational CRDM patents to several licensees for a 3% royalty. As described by Dr. Cohen, a know-how license likely would not exceed that 3% Mirowski rate. Ex. 6103-P (Cohen Opening) at 36; Tr. (Cohen) 469-70, 702-03.

142. Other Testimony: Mr. Spadea agreed that a 1% to 3% rate was typical for the transfer of CRDM know-how. Tr. (Spadea) 1222. And licensing expert Mr. Louis Berneman testified that a 2% to 3% adjustment for know-how was appropriate. MDT I Tr. (Berneman) 5010:10-12. Mr. Charles Dennis, VP of open innovation and IP for Medtronic's CRDM division, likewise testified that the typical know-how license has a 1% to 3% rate. MDT I Tr. (Dennis) 3955:3-4.

143. Based on these observations, Dr. Putnam correctly concluded that an adjustment of 1% to 3% — substantially less than the 7% base rate for patents — compensated for the know-how. Ex. 6105-P (Putnam Opening) ¶¶ 238-46.

144. Dr. Putnam's conclusions were reinforced by the Varner Study cited by Respondent's expert, Dr. Cockburn, as well as a paper that Dr. Cockburn authored. The Varner Study surveyed nearly 3,000 license agreements to quantify a “know-how premium” for IP licenses that license know-how in addition to patents. Ex. 6201-R (Cockburn Opening) at 17 n.39 (citing Varner). Dr. Putnam read Dr. Varner's study as establishing a “know-how premium” of 1.5% to 3%. Ex. 6106-P (Putnam Rebuttal) ¶¶ 96, 124-27; Tr. (Putnam) 956-64. Dr. Hubbard similarly reviewed Dr. Varner's study and concluded that 3% is the upper limit for know-how. Tr. (Hubbard) 1449:24-1450:6. Furthermore, Dr. Cockburn's article The Importance of Patents to Innovation concluded that survey respondents in the biotech, pharmaceutical and medical-device sectors ranked patents as more important than know-how. PFF ¶ 64; Ex. 6124-P at 4804-08.

5. Sub-Licenses

145. Another difference between the rights in the Pacesetter Agreement and the MPROC Licenses relates to the third-party patent rights. Medtronic US periodically licenses patent rights from third parties, such as the Mirowski patents. These third-party patent rights relating to Cardiac and Neuro products were made available to Medtronic PR. Although the MPROC Licenses did not formally include a sub-license provision, to the extent Medtronic US needed access to other patents (e.g., the Mirowski patents or Guidant's CRDM patents), so did Medtronic PR. Thus, their licensing arrangement included Medtronic PR's ability to sublicense from Medtronic US. Ex. 6105-P (Putnam Opening) ¶ 207.

146. The Pacesetter Agreement also included a sub-license provision, whereby each party that had the right to sub-license patent rights was obligated to do so, if requested by the other party. Ex. 2504-J at 4461-62. The sub-licensee was obligated to pay the same royalty rate for sub-licensed patent rights as the licensee (i.e., a “pass-through” rate). This “pass-through” pricing is typical in arm's-length patent licensing. Ex. 6105-P (Putnam Opening) ¶¶ 32, 172(b)-(c), 203-05, 208. Thus, to put Medtronic PR and Pacesetter in the same position, Medtronic PR must separately pay for third-party licenses — with “pass-through” pricing in accordance with the Pacesetter Agreement's terms and general market practice. Id. ¶¶ 209-10.

147. In accordance with this arm's-length approach, Dr. Putnam calculated a “sub-license” adjustment of 4% to 5% for cardiac Devices and Leads (and 1% for Neuro) by reference to three Medtronic commercial licenses:

• Mirowski: Medtronic's license of the Mirowski patents, which were generally recognized as necessary to the manufacture and sale of certain CRDM devices, and were widely licensed in the CRDM industry at a royalty rate of 3% of net sales. Id. ¶ 219.

• Guidant: Medtronic's cross-license with Guidant for a small subset of CRDM patents, in 2004. Id. ¶¶ 220-224.

• NeuroPace: Medtronic's license of certain patents related to neurostimulation devices from NeuroPace, for which Medtronic paid a royalty rate of 1% of net sales. Id. ¶ 219.

148. If anything, these amounts overstate the royalties that Medtronic US paid (and create a conservatively larger adjustment) because these third-party royalties were not paid on all sales subject to the MPROC Licenses, and the sub-licenses also expired at various points. To check the reasonableness of his sub-license adjustment, Dr. Putnam reviewed the royalties that Medtronic paid to third parties in the United States. He determined that his adjustment was substantially higher than the third-party royalties actually paid by Medtronic. Id. ¶¶ 213, 215, 218, 225.

6. Other Adjustments from Medtronic I

149. In Medtronic I, this Court considered certain other adjustments to the Pacesetter Agreement's base rate that, in light of the relevant expert testimony on remand and other adjustments, are not necessary.

(a) Past Know-How

150. The Medtronic I opinion made a 4% to 5% adjustment for past know-how, as part of an overall 7% adjustment for know-how. Medtronic I at 135. Based on the expert testimony on remand, it is clear that such an adjustment for past knowhow is not needed. Any such “past” know-how — i.e., know-how created and conveyed to Medtronic PR in tax years prior to those at issue in this case — was compensated for by the royalties that Medtronic PR paid to Medtronic US under prior iterations of the MPROC Licenses, which had been in place since tax year 2002. Tr. (Putnam) 803:11-20 In addition, as described above, the relevant knowhow conveyed to Medtronic PR here is, as a general matter, relatively short-lived. PFF ¶¶ 57-62. For these reasons, know-how transferred/licensed in the past should not require additional compensation under the MPROC Licenses in the tax years at issue. Ex. 6105-P (Putnam Opening) ¶ 230.

(b) Profit Potential

151. The Medtronic I opinion made a 3.5% adjustment for profit potential. Medtronic I at 135-36. Based on the expert testimony on remand (and the adjustments provided for in such expert testimony), no further adjustment is needed for profit potential. Dr. Putnam's “profitability” adjustment already computes an adjustment based on the relative profit of Pacesetter at the time of the Pacesetter Agreement and Medtronic at the time of the MPROC Licenses. PFF ¶¶ 131-135. And moreover, the Maximum Rate Clause “caps” the effect of differences in profit potential given that the 15% cap applied regardless of the profit potential of the future patents. Tr. (Putnam) 803-04.

(c) Exclusivity

152. The Medtronic I opinion made a 7% adjustment for exclusivity because the MPROC Licenses were styled as an “exclusive” license, while the Pacesetter Agreement was non-exclusive by its terms. Medtronic I at 124, 134. Based on the expert testimony on remand (and the adjustments provided for in such expert testimony), no adjustment is needed for exclusivity. First, the MPROC Licenses, although styled as exclusive, did not convey to Medtronic PR, as an economic matter, “market exclusivity” to the licensed IP (i.e., Medtronic PR was not the exclusive licensee of the licensed IP). Ex. 6105-P (Putnam Opening) ¶¶ 248-52. In particular, Medtronic's competitors such as Pacesetter and Guidant licensed Medtronic CRDM patents at the same time as Medtronic PR. Ex. 6105-P (Putnam Opening) 251-52; Ex. 6107-P (Spadea) ¶¶ 33, 38, 44. As explained by Medtronic's experts on remand, the MPROC Licenses referred to licensing the IP on an “exclusive” basis, but then provided an exception to that exclusivity in certain circumstances, including “settlement of litigation.” Ex. 6105-P (Putnam Opening) ¶¶ 248-50; e.g., Ex. 7-J at 4175; Ex. 6107-P (Spadea) ¶ 113. This exception swallowed the rule. Actual exclusivity, in the sense of market exclusivity, could warrant a higher royalty rate. Ex. 6105-P (Putnam Opening) ¶ 249. But that was not what was conveyed here. Tr. (Putnam) 804-07.

153. An exclusivity adjustment could also be theoretically appropriate to address that Medtronic PR, as the finished manufacturer for the largest and most profitable firm in the industry, was selling into a more profitable market at the time of the MPROC Licenses than Pacesetter was at the time of the Pacesetter Agreement. But this circumstance is separately addressed by Dr. Putnam's profitability adjustment. That profitability adjustment addresses the same issues of profitability and market share that underlie this element of a potential exclusivity adjustment. Tr. (Putnam) 806:10-13 (“[T]he adjustment for exclusivity is really embodied within the profitability adjustment because MPROC participates in Medtronic's success.”). Further, to the extent that the value-driver at issue is Medtronic's exclusive right to sell Medtronic-branded Devices and Leads in the United States, that is already priced at arm's length through the Trademark License. Tr. (Putnam) 806:14-807:9.

(d) Scope of Products: Neurostimulation Devices

154. The Medtronic I opinion made a 2.5% adjustment for scope of products — i. e., because the Pacesetter Agreement only licensed patents for Cardiac products, while the MPROC Licenses covered both Cardiac and Neuro products. Medtronic I at 136-37. Cardiac and Neuro Devices and Leads are highly comparable. PFF ¶¶ 67-72. It is thus unnecessary to adjust the royalty rate for Neuro/scope of products. Medtronic US is appropriately compensated for the broader scope of products because Medtronic PR pays the royalty rate on a correspondingly broader set of sales (i.e., Cardiac and Neuro sales). And more importantly, Neuro Devices and Leads are less profitable than Cardiac Devices and Leads.¹⁵ Indeed, Dr. Heimert apparently agreed that the Cardiac and Neuro rates should be the same. Tr. (Heimert) 2570:6-11. Thus, the sole difference between Cardiac/Neuro royalty rates is the difference in relevant sub-licenses as described above.

7. Leads

155. The Medtronic I opinion applied a royalty rate to Leads that was half of the royalty rate applied to Devices. Medtronic I at 138. The Pacesetter Agreement, however, applied the same rate for Devices and Leads. Tr. (Putnam) 809:1-8. Dr. Heimert and Medtronic's experts agree that the royalty rate for Devices and Leads should be the same. Tr. (Heimert) 2571:9-13; Tr. (McCoy) 388:11-16.

F. Respondent's Transfer Pricing Method

1. Dr. Heimert's CPM

156. In this remand proceeding, Dr. Heimert presented the same CPM analysis that he presented (and that this Court rejected) in Medtronic I. Dr. Heimert again presented Medtronic PR as the “tested party”; used the same 14 companies to benchmark the return to Medtronic PR; and used the same profit level indicator (“PLI”) — Return on Assets (“ROA”). Dr. Heimert made no real changes to his value-chain CPM or to any of its individual elements in response to Medtronic I. For example, Dr. Heimert did not identify any new comparable companies beyond his 14 from Medtronic I. Compare 5543-R (MDTI Heimert Opening) at 126-27, with Ex. 6203-R (Heimert Opening) at Tbl. 9. Even the points in his Medtronic III report that Dr. Heimert characterized as “clarifications” of his prior analysis simply restated points that he already presented in Medtronic I. Ex. 6203-R (Heimert Opening) at ¶¶ 68-79; Tr. (Heimert) at 2448-49, 2469-74, 2477-83, 2486-91. As Dr. Heimert testified, his view of Medtronic PR and its capabilities remained the same before the opinion in Medtronic I and now on remand. Tr. (Hemiert) 2491:11-14 (“I didn't feel, you know, less or better about MPROC [after the Medtronic I opinion]. It was pretty much, you know, a similar perception that I had about MPROC's capabilities”).

157. The Court concluded in Medtronic I that Dr. Heimert's CPM was arbitrary, capricious and unreasonable based on several key findings of fact, including the resolution of a range of issues relevant to Dr. Heimert's recycled CPM analysis:

• The Companies Are Not Comparable. Several specific findings that Dr. Heimert's 14 companies fundamentally differ from Medtronic PR in product offerings; their role in product quality; the product liability risk that they bear and manage; and their scale, asset composition and geographic markets. See Medtronic I at 109-11; Ex. 6109-P (Hubbard Rebuttal) at 9-19; Ex. 6101-P (McCoy Opening) at 14-17; Ex. 6102-P (McCoy Rebuttal) at 9-10; Tr. (Crosby) 2955:25-2961:6.

• ROA Is the Wrong PLI Dr. Heimert's use of ROA as the PLI “ignores valuable intangible assets that were obtained through the devices and leads licenses because these assets are not recorded on petitioner's balance sheet.” Medtronic I at 114; see also id. at 112-14.

• Dr. Heimert Is Wrong About Realistic Alternatives: The Court rejected Dr. Heimert's “realistic alternative” analysis, and found that Medtronic PR's role could not have been replicated “without substantial time and costs. MPROC workers were highly trained, and its workforce could not be replaced overnight. The Swiss facility only made devices and could not make enough devices for both Europe and the United States. Petitioner was so concerned about quality that it never considered outsourcing the activities of MPROC. It did not want to rely on external partners for such a vital role.” Id. at 107-08; see also id. at 34, 113.

158. Dr. Heimert also failed in Medtronic III to take into account another key finding from Medtronic I — that it was improper to aggregate functions performed by various Medtronic entities in the United States. Id. at 114-16. Dr. Heimert explained in Medtronic I that the first step of his value-chain CPM analysis involved “aggregating the transactions,” which he did “because by aggregating the transactions[,] I felt that it was a more reliable measure of an arm's length result.” MDTI Tr. (Heimert) 7282:9-11. As Dr. Heimert stated in his affirmative report in Medtronic I, “[t]hus, while the royalty rates [combined for both the MPROC Licenses and the Trademark License] may seem high in isolation, they are reasonable when considered in the context of the covered transactions in the aggregate.” Ex. 5543-R (Heimert Opening) at 152 (emphasis added).

159. In Medtronic I, the Court rejected Dr. Heimert's “aggregation” of the four separate intercompany transactions as a matter of fact. Medtronic I at 116. On remand, Dr. Heimert testified that his analysis no longer relies on the presence of the Trademark License, sales and distribution and component manufacturing. He agreed that Medtronic's intercompany pricing for those three functions was at arm's length, and so those transactions are “off the table” in the transfer pricing analysis for the MPROC Licenses. Tr. (Heimert) 2417:24-2418:12.

2. Dr. Hubbard's CPM Analysis

160. In contrast to Dr. Heimert, Dr. Hubbard attempted to construct a reliable CPM from the ground-up. Dr. Hubbard reexamined the broader universe of available financial data and potential comparables, and performed his analysis based on the factual record as it stands in this remand proceeding. Notably, he started from — rather than ignored — the fact that only one intercompany transaction remains to be priced (the MPROC Licenses). Ex. 6108-P (Hubbard Opening) ¶ 13.

161. In Dr. Hubbard's attempt to construct a CPM, he genuinely reexamined the choice of tested party, id. ¶¶ 58-84; the potentially comparable companies for either Medtronic PR or Medtronic US as the tested party, id. ¶¶ 85-96; and the choice of PLI, id. ¶¶ 122-24. Ultimately, Dr. Hubbard concluded that a reliable CPM could not be performed. Tr. (Hubbard) at 1405, 1436-39.

162. In his ground-up analysis, Dr. Hubbard explained the key facts that drove his conclusion. Most fundamentally, Dr. Hubbard found that there were inadequate financial data for a reliable CPM for either Medtronic PR's finished device manufacturing or Medtronic US's R&D function. In his comprehensive search for comparable companies, Dr. Hubbard determined that there were no comparable companies — in terms of functions, risks and products — to Medtronic PR or to Medtronic US's R&D function. Ex. 6108-P (Hubbard Opening) ¶¶ 85-96. And within the Class III medical device industry (including the CRDM industry), Dr. Hubbard found that the companies that had potentially comparable functions were vertically integrated, with no segmented financial data that could be used to perform a reliable CPM. Id. ¶¶ 118-21. As Dr. Hubbard explained, this vertical integration of finished manufacturing and R&D was itself instructive:

In the industry, and for Medtronic in particular, the vertical integration, both manufacturing and R&D are core functions. So with that finished device manufacturing as a core competency, the data are not going to be adequate for the task. . . .

Tr. (Hubbard) 1435:7-12. In other words, because finished device manufacturing and R&D are both core functions in the Class III medical device industry generally (and the CRDM industry in particular), relevant companies retain those functions in-house. Id. 1435; Ex. 6108-P (Hubbard Opening) ¶¶ 111-17. As a result, the financial data for those functions are not available on a standalone or segmented basis, generally precluding a reliable CPM for them.

3. Analysis of Dr. Heimert's 14 Companies

163. Medtronic's experts provided further testimony on the 14 companies used in Dr. Heimert's CPM from economic, industry and scientific perspectives. Economically, Dr. Hubbard explained why Dr. Heimert's arguments defending these same rejected comparables are unconvincing. In particular, Dr. Heimert “paints in unreasonably broad strokes,” reducing Medtronic PR's functions and attributes to meaningless generalities in an attempt to establish comparability with his 14 companies. Ex. 6109-P (Hubbard Rebuttal) ¶ 39; id. ¶¶ 19-42.

164. For example, Dr. Heimert, in defending the functional comparability of his 14 companies to Medtronic PR on remand, claimed that his companies are similar to Medtronic PR in terms of the importance of product quality (and their role in ensuring that quality). Ex. 6203-R (Heimert Opening) ¶ 85. As Dr. Hubbard explained, “it appears that the only support that [Dr. Heimert] provides for his assertion is that: 'quality is also a necessary condition for market success for the Comparables,' as indicated by the fact that: 'all of the Comparables' 10-Ks discuss quality as a risk factor.” Ex. 6109-P (Hubbard Rebuttal) ¶ 34.

165. As Dr. Hubbard noted, such 10-K disclosures say nothing about the degree of importance of product quality.¹⁶ Indeed, as Dr. Hubbard explains, even the maker of Swingline staplers discloses product quality as a risk factor in the same manner as Dr. Heimert's 14 companies. Id. ¶¶ 33-38. Dr. Heimert's analysis thus fails because, “[w]hile product quality is an important consideration with respect to both pacemakers and staplers, it is of course far more important with respect to the former.” Id. ¶ 38. The same is true as between Medtronic PR and Dr. Heimert's 14 companies. Accord Medtronic I at 109 (“Many of the products made by the comparable companies do not require the same level of FDA scrutiny as implantable class III medical devices.”).

166. While Dr. Heimert's analysis was too general and generic, Dr. Hubbard relied on Medtronic's other experts who provided nuanced scientific and industry perspectives. Both Dr. Cohen and Mr. McCoy concluded that Dr. Heimert's 14 companies are fundamentally different form Medtronic PR in terms of products and risks assumed. As Mr. McCoy concluded, “any comparison of the activities or the profits of these listed companies to [Medtronic PR] would be such a stretch as to be meaningless.” Ex. 6101-P (McCoy Opening) at 17. Mr. McCoy viewed the risks assumed and related functions performed to be too different to allow for an economically reasonable comparison.

167. As Dr. Cohen explained, the 14 companies are, in whole or predominantly, manufacturers of Class I and Class II medical products, and the functions and risks associated with Class I and Class II are different in kind from those of a finished manufacturer of Class III CRDM devices.

• As. Dr. Cohen explained, “[t]he demand for quality manufacturing is of course highest for Class III devices.” Ex. 6104-P (Cohen Rebuttal) at 22; see Ex. 6103-P (Cohen Opening) at 37-38; Medtronic I at 8 (“Class III medical devices are higher risk and more novel than are those of classes I and II.”).

• Beyond this distinction among classes, CRDM devices are unique within the Class III category because “they pose among the highest potential risks in the category of Class III devices.” Ex. 6104-P (Cohen Rebuttal) at 22. Failure of a Class III CRDM device can lead to death of a patient within minutes. With other Class III devices, adverse effects generally occur over a much longer period of time, providing opportunity for medical intervention to prevent patient death or other serious adverse consequences. Id.

These fundamental technological differences informed Dr. Hubbard's conclusion that a reliable CPM could not be performed for the MPROC Licenses.

4. The “Implantables” Are Not Implantables

168. Dr. Heimert indicated during his testimony that a subset of his 14 companies — the so-called “implantables” — could perhaps be used to perform a CPM for Medtronic PR. Tr. (Heimert) 2545:22-2546:3.

169. Dr. Heimert himself concluded, however, that there is not enough data to create an “implantable” set within his 14 companies for purposes of adjusting his CPM. As Dr. Heimert stated in his opening remand report, he originally considered adjusting his set of companies to “eliminate any non-implantable products within each Comparable's manufacturing function.” Ex. 6203-R (Heimert Opening) 102. He concluded, that such an adjustment could not be done, explaining: “given public filings do not contain such segregated data, I chose not to perform any such adjustments because the assumptions that would need to be made and the lack of reliable information would not increase the reliability of my results.” Id. (emphasis added). Dr. Heimert further explained in his testimony that SEC filings and other publicly available information is insufficient to determine “the mix of Class III versus Class II products that these companies are manufacturing.” Tr. (Heimert) 2550:22-2551:10; see also id. 2480:9-2481:12, 2483:12-17. At most, this subset of Dr. Heimert's companies is “implantables” in the sense that they produce at least one implantable product. Ex. 6203-R (Heimert Opening) ¶ 84.

170. Dr. Hubbard was also clear that the “implantable” subset of Dr. Heimert's comparables presented “inappropriate benchmarks for the profitability of MPROC.” Ex. 6109-P (Hubbard Rebuttal) ¶ 27; Tr. (Hubbard) 1445-46. This subset is not actually “implantables” because all five companies manufactured Class I and/or Class II medical devices, not Class III devices like Medtronic PR. Ex. 6109-P (Hubbard Rebuttal) ¶ 27. Furthermore, the five “implantables” did not manufacture finished CRDM products, which are unique even among Class III devices. Id. ¶¶ 28-29; see also Ex. 6104-P (Cohen Rebuttal) at 23. The “implantables” also still presented “blended profitability measures across the several functions of each of Dr. Heimert's comparables” when one of those functions “(i.e., manufacturing of finished medical devices) is relevant for the CPM in this matter.” Ex. 6109-P (Hubbard Rebuttal) ¶ 27.

171. Respondent's experts, in their direct testimony, stated incorrectly that Dr. Hubbard had somehow endorsed the “implantables” as valid comparables and manufacturers of Class III medical devices like Medtronic PR. Tr. (Heimert) 237879, 2385-88; Tr. (Becker) 2747:13-23. That is simply not true. And on cross-exam, these experts admitted that their claims about Dr. Hubbard and the implantables were incorrect. Tr. (Heimert) 2526; Tr. (Becker) 2801-02.

5. ROA Is Not an Appropriate PLI

172. Dr. Hubbard also concluded that Dr. Heimert's use of the ROA as his PLI does not capture the value that Medtronic PR contributes through its workforce skills and quality/institutional know-how accumulated over three decades:

[W]hile ROA can be a suitable PLI for PP&E-intensive activities such as industrial manufacturing, it is less suitable for Class III medical device manufacturing functions such as those performed by MPROC, the returns of which may be driven to a lesser extent by PP&E than by labor and institutional know-how. The same observation can be made for ROA as a PLI for MED Inc.'s R&D function.

Ex. 6108-P (Hubbard Opening) ¶ 123; accord Medtronic I at 113-14. In other words, ROA can be an appropriate profit benchmark when returns are attributable to the assets being measured. Accord Medtronic I at 113. Here, the asset base used for Medtronic PR in Dr. Heimert's CPM “do[es] not reflect the value of this technical know-how, process improvements, and MPROC's workforce in place.” Ex. 6109-P (Hubbard Rebuttal) ¶ 53. As a result, the use of the ROA is fundamentally flawed.

173. Mr. McCoy reinforced that ROA was an inappropriate PLI at least in part because he — a former CEO and CFO of one of Medtronic's direct competitors — never used ROA to analyze the financial performance of the finished device manufacturing function at his CRDM companies. Tr. (McCoy) 405:15-406:11. ROA was not a financially relevant metric for that function. In other words, Mr. McCoy's experience proves Dr. Hubbard's theoretical economic point.

174. Dr. Hubbard addressed other critical failures in Dr. Heimert's use and defense of his ROA calculation. Ex. 6109-P (Hubbard Rebuttal) ¶¶ 46-58; Tr. (Hubbard) 1440:4-18; cf. Ex. 6108-P (Hubbard Opening) ¶¶49-57. For example, Dr. Cockburn and Dr. Heimert defended Dr. Heimert's use of the ROA on the basis that Dr. Heimert uniformly excluded intangible assets from the asset base for each of Medtronic PR and the 14 companies. Ex. 6203-R (Heimert Opening) ¶77; Ex. 6201-R (Cockburn Opening) ¶99. Dr. Hubbard explained why excluding intangible assets in this manner does not at all solve for them. Ex. 6109-P (Hubbard Rebuttal) ¶ 56-57; see also Ex. 5550-P (Pindyck) ¶ 15.3. As recognized by this Court, Dr. Cohen and Mr. McCoy, Dr. Heimert's 14 companies did not possess workforce skills or quality of institutional know-how comparable to Medtronic PR, which means that they understate the appropriate returns to Medtronic PR in any event. Second, intangible assets of the 14 companies would presumably be spread across their functions.

175. In addition, Dr. Heimert and Dr. Cockburn purportedly performed “crosschecks” on the ROA that Dr. Heimert assigned to Medtronic PR. Dr. Heimert compared his ROA assigned to Medtronic PR for the at-issue years to the industry-average ROA in SIC Code 384, and found that his assigned ROA is “within the range of ROAs experienced by other companies in SIC code 384.” Ex. 6203-R (Heimert Opening) ¶ 105. Dr. Cockburn compared Dr. Heimert's calculation of Medtronic PR's ROA in different scenarios to those of: (i) Dr. Heimert's comparables, (ii) all companies in SIC code 384, and (iii) companies in manufacturing sectors that, according to Dr. Cockburn, “require precision manufacturing and are subject to significant (even fatal) consequences in the case of a product failure.” Ex. 6201-R (Cockburn Opening) ¶ 107; see also id. ¶¶ 102. However, as Dr. Hubbard demonstrated, these conclusions are circular. The CPM benchmarks Medtronic PR's ROA to the selected comparables, and then derives royalty rates for Medtronic PR from that benchmarking. Conducting the same analysis in reverse — using Dr. Heimert's assigned royalty rates to calculate Medtronic PR's ROA, and then comparing that back against the ROAs for the original comparables (or a more general set of similar comparables) — will always “support” Dr. Heimert's CPM. Tr. (Hubbard) 1440-41.

IV. ARGUMENT

A. The Scope of Remand & Standard of Review

1. The Remaining Issue on Remand Is Narrow

The scope of the remand proceeding is narrow because of the limited scope of: (i) the Eighth Circuit's opinion, (ii) this Court's reopening on remand and (iii) Respondent's appeal of the Medtronic I decision.

First, the Eighth Circuit requested additional findings limited to two of the Court's conclusions from Medtronic I: (1) that the Pacesetter Agreement is a CUT for the MPROC Licenses, and (2) the amount of product liability risk that should be allocated between Medtronic US and Medtronic PR. For these topics, the Eighth Circuit asked for “additional findings” and did not disturb the Court's original findings of fact. Medtronic II at 615 (emphasis added).¹⁷ In accordance with that direction, the Eighth Circuit vacated this Court's judgment in Medtronic I and not the underlying opinion. Id. at 615 (“Accordingly, we vacate the tax court's January 25, 2017, order and remand the case for further consideration in light of the views set forth in this opinion. See Centron DPL Co. v. Tilden Fin. Corp., 965 F.2d 673, 676 (8th Cir. 1992) ('In the absence of . . . critical factual determinations, we must vacate the judgment of the district court and remand for reconsideration with instructions to make more complete factual findings.').” (alteration in original) (emphasis added)).

The Eighth Circuit requested additional findings in three specific areas:

• Comparability of Circumstances: “[W]hether the circumstances of the settlement between Pacesetter and Medtronic US were comparable to the licensing agreement between Medtronic and Medtronic Puerto Rico.” Medtronic II at 614.

• Comparability of Contractual Terms: “[T]he degree of comparability of the Pacesetter agreement's contractual terms and those of the Medtronic Puerto Rico licensing agreement.” Id.

• Comparability of Intangibles: “[H]ow the different treatment of intangibles affected the comparability of the Pacesetter agreement and the Medtronic Puerto Rico licensing agreement.” Id. at 614-15.

Second, in its May 3, 2019 Reopening Order, the Court “reopen[ed] the record to a limited degree to permit expert testimony” on six specific questions (emphasis added), which elaborate on the Eighth Circuit's topics:

• Whether the Pacesetter Agreement is a CUT;

• Whether this Court made appropriate adjustments to the Pacesetter Agreement as a CUT;

• Whether the circumstances between Pacesetter and Medtronic US were comparable to the licensing agreement between Medtronic US and Medtronic PR, and whether the Pacesetter Agreement was an agreement created in the ordinary course of business;

• An analysis of the degree of comparability of the Pacesetter Agreement's contractual terms and those of the Medtronic PR licensing agreement;

• An evaluation of how the different intangibles affected the comparability of the Pacesetter Agreement and the Medtronic PR licensing agreement; and

• An analysis that contrasts and compares the CUT method using the Pacesetter agreement with or without adjustments and the CPM, including which method is the best method.

Reopening Order at 2-3 (emphasis added).

This Court further explained the scope of this remand by stating that this expert testimony “should not be duplicative of the existing record's expert testimony.” Id. at 3 (emphasis added). The Reopening Order also instructed the parties to address the product liability issue on brief. Id. In addition, consistent with the Court's decision not to reopen the factual record, the Court has indicated that it will not make additional findings of fact, other than findings based on expert testimony that the parties offered on remand.

Third, the scope of issues in the remand proceeding is also constrained by Respondent's narrow appeal of the Court's Medtronic I decision, limited to legal aspects of this Court's best method analysis for the MPROC Licenses. This remand can be no broader than the appeal that precipitated it. Thus, the points that Respondent did not appeal from Medtronic I still stand.

In Medtronic I, this Court addressed numerous issues based on a trial record that included the testimony of nearly 70 witnesses and more than 5,000 exhibits. Respondent appealed only certain legal elements of this Court's best method analysis of the arm's-length royalties under the MPROC Licenses. MDT II Resp't Opening Brief at 2, 29-30. Respondent contended in in his opening appellate brief that this Court “erred as a matter of law” and remand was required for two purposes. Id. at i. First, so the Court “may reevaluate the best transfer-pricing method” for the MPROC Licenses. Id. And alternatively, so that “certain adjustments may be made to the transfer price adopted by the Tax Court” for the MPROC Licenses. Id.

Respondent did not appeal, and never disputed before the Eighth Circuit, this Court's factual findings in Medtronic I. In particular, Respondent never argued that any Medtronic I factual finding was “clearly erroneous,”¹⁸ instead claiming on appeal that this Court failed to make certain findings required as a matter of law to sustain its holding on the best method for the MPROC Licenses. MDT II Resp't Reply Brief at 6.¹⁹ It is axiomatic that Respondent waived any challenge to points that he did not appeal. E.g., Chay-Velasquez v. Ashcroft, 367 F.3d 751, 756 (8th Cir. 2004); Alexander v. Jensen-Carter, 711 F.3d 905, 909 (8th Cir. 2013). Specifically, he has waived any challenge to this Court's existing findings of fact and this Court's other conclusions in Medtronic I other than the aspects of this Court's best method analysis described above. Respondent has essentially agreed that this Court's determinations from Medtronic I are final to the extent that he did not appeal them. In his pre-trial memorandum, Respondent specifically invoked the “cross-appeal rule” — a sub-genre of the broader “waiver” doctrine — to argue that Medtronic cannot receive a more favorable judgment on remand than the judgment in Medtronic I. Resp't Pretrial Mem. at 8-10. Medtronic agrees with Respondent. Of course, broader waiver principles apply to Respondent in equal measure. And as a result, as Respondent has recognized on remand, he waived issues that he did not appeal — specifically, as Respondent conceded, he waived further challenge to the Court's section 367(d) ruling and to the arm's-length pricing of the intercompany transactions for components, distribution and sales and the Trademark License. Resp't Mot. Leave to File Out of Time Second Amend. Answer ¶ 6; Resp't Pretrial Mem. at 1 n.1.

2. The Court Should Incorporate by Reference Its Prior Findings and Un-appealed Conclusions

For procedural clarity, Medtronic respectfully submits that it would be appropriate for this Court, in its supplemental opinion on remand, to incorporate by reference its existing findings of fact and un-appealed conclusions from Medtronic I and to reaffirm their continuing validity. This Court is empowered to incorporate by reference elements of its prior opinion following a remand. Lear Eye Clinic Ltd. v. Commissioner, 106 T.C. 418, 419 (1996).

B. Standard of Review

Because the sole remaining issue on remand is the last step in the section 482 analysis for the last transaction still at issue (the MPROC Licenses), Respondent's notice of deficiency is not entitled to any deference. It is settled that Respondent abused his discretion because his CPM and resulting allocations were arbitrary, capricious and unreasonable, Medtronic I at 118. The Court's finding on this score eliminates any deference to Respondent's section 482 determination, his method and his supporting analysis. Id. at 85-87.

C. The Pacesetter Agreement Is a Valid CUT, and Dr. Putnam's Methodology is the Best Method

1. Section 482 Legal Standard

Treas. Reg. § 1.482-4(c)(1) provides that the CUT method “evaluates whether the amount charged for a controlled transfer of intangible property was arm's length by reference to the amount charged in a comparable uncontrolled transaction.” The CUT method has been considered to be the best measure of arm's-length prices for intercompany transfers of intangibles. See Treas. Reg. § 1.482-4(c)(2)(ii); Veritas Software Corp. v. Commissioner, 133 T.C. 297 (2009) (relying on actual market transactions to establish the appropriate royalty rates for technology intangibles); see also OECD, Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations ch. 2.7 (1995) (“Where it is possible to locate comparable uncontrolled transactions, the [comparable uncontrolled price method] is the most direct and reliable way to apply the arm's length principle.”).

The CUT method is predicated on finding “comparable” uncontrolled transactions. Comparability is assessed under two provisions: (i) the general comparability factors set forth in Treas. Reg. § 1.482-1(d)(3), which are applicable to all methods; and (ii) specific comparability considerations for the CUT method set forth in Treas. Reg. § 1.482-4(c). Although “all of the factors described in § 1.482-1(d)(3) must be considered” for a CUT, “specific factors [described in Treas. Reg. § 1.482-4(c)] may be particularly relevant to” the CUT method. Treas. Reg. § 1.482-4(c)(2)(iii)(A); see T.D. 8552, 1994-2 C.B. 93.

(a) Section 482 Requires Comparable, Not Identical, Transactions

(i) Specific Comparability Factors

For the intangible property in an uncontrolled transaction to be comparable to the intangible property in a controlled transaction, both sets of intangibles must:

(i) Be used in connection with similar products or processes within the same general industry or market; and (ii) [h]ave similar profit potential. The profit potential of an intangible is most reliably measured by directly calculating the net present value of the benefits to be realized (based on prospective profits to be realized or costs to be saved) through the use or subsequent transfer of the intangible, considering the capital investment and start-up expenses required, the risks to be assumed, and other relevant considerations.

Treas. Reg. § 1.482-4(c)(2)(iii)(B)(1).

In addition, Treas. Reg. § 1.482-4(c)(2)(iii)(B)(2) provides that particular aspects of the comparability factors for contractual terms and economic conditions are specifically relevant to the CUT method. Under these CUT-specific comparability considerations, the CUT method is valid, and even preferred, even if identical comparable transactions do not exist. As this Court has held:

• Sundstrand Corp. v. Commissioner, 96 T.C. 226 (1991):²⁰ Imperfect comparables serve “as a base from which to determine the arm's-length consideration for the intangible property involved in this case.” Id. at 393.

• Veritas, 133 T.C. at 331, 335: Comparability standards are satisfied for the use of the CUT method when they were “sufficiently comparable to the controlled transaction” even though they were “certainly not identical to the controlled transaction.”

• In Veritas and Sundstrand, the controlled transaction involved IP not transferred in the uncontrolled transaction — trademarks in the former; marketing intangibles in the latter — but those differences did not impede threshold comparability. See also Compaq Computer Corp. v. Commissioner, 78 T.C.M. (CCH) 20, 31 (1999).

(ii) General Comparability Factors

Treas. Reg. § 1.482-1(d)(3) describes the five general comparability factors that are considered for all methods, namely: (i) functions, (ii) contractual terms, (iii) risks, (iv) economic conditions and (v) property or services. Treas. Reg. § 1.482-4(c) emphasizes three of those five general factors for CUTs (contractual terms, economic conditions, and property or services).

2. The Pacesetter Agreement Satisfies the Section 482 Requirements for Comparability

The Pacesetter Agreement satisfies both sets of comparability criteria and is a valid CUT for establishing the royalty rate under the MPROC Licenses. As with most comparable transactions, the intangible property is not identical, and the contract terms differ. In addition, the relevant market conditions changed between 1992 when the Pacesetter Agreement was originally executed and 2004 when the MPROC Licenses were executed. However, there is a sufficient degree of comparability on all three of these dimensions to satisfy the standards in Treas. Reg. § 1.482-4(c)(2). In addition, based on the opinions of Medtronic's experts in Medtronic III, the Court has the analytical support necessary to make reliable adjustments to account for any differences.

(a) The Two Sets of Intangible Property Are Comparable

One of the key questions for comparability under Treas. Reg. § 1.482-4(c) is whether the intangible property transferred in the Pacesetter Agreement and the MPROC Licenses was sufficiently comparable. Intangible property is considered comparable if it is used in connection with similar products or processes within the same general industry or market and has similar profit potential.

The intangible property transferred in the MPROC Licenses can be divided into three categories: (i) patents relating to cardiac Devices and Leads; (ii) patents relating to Neuro Devices and Leads; and (iii) know-how. As discussed further below, all three categories of intangibles licensed in the MPROC Licenses were used to manufacture and sell similar products within the same medical device industry and market, and have similar profit potential as the patents licensed under the Pacesetter Agreement.

(i) CRDM Patents Were Highly Valuable, Seminal Patents, and Pacesetter Royalty Rate Created a Benchmark

Three key elements establish the comparability of the CRDM patents licensed under the two sets of agreements. First, the value of a patent portfolio is often concentrated in a small subset of the portfolio. Second, both agreements licensed the most valuable patents. Third, Medtronic's repeated licensing of those patents — including pursuant to the Pacesetter Agreement — created a benchmark.

Specifically, as Dr. Cohen explained and Dr. Cockburn agreed, the value of a patent portfolio is often concentrated in a small subset of the portfolio. PFF ¶ 47. Medtronic's CRDM patents involved groundbreaking rate-responsive pacing technology that Medtronic introduced to the market in its Activitrax pacemaker. PFF If ¶ 29. Activitrax was an immediate market success and permanently changed the pacing industry. Years later, Medtronic's CRDM patents were still the most valuable of the three categories of intangibles licensed to Medtronic PR under the MPROC Licenses. ¶¶ 29, 42-72.

Patent licenses in the CRDM industry often establish benchmark royalty rates that other market players use in their subsequent license negotiations. PFF 22-24. Dr. Cockburn agreed that seminal patents — such as the Mirowski patents — establish a benchmark royalty rate early on in the life of the patent that is difficult to shake. A constant royalty rate for a given patent portfolio is to be expected, absent an evolution in the circumstances or licensed technology. PFF ¶ 23.

In fact, the Pacesetter royalty rate established such a benchmark in the CRDM industry that competitors would use to negotiate a license for Medtronic's patent portfolio. Medtronic repeatedly licensed its patents to competitors, which created a market price. For example, as Dr. Cockburn recognized, Medtronic had licensed its patents to CPI/Eli Lilly before entering into the Pacesetter Agreement. By 1994, Medtronic had licensed those patents to three key competitors — CPI/Eli Lilly, Siemens and St. Jude — two of which agreed to pay a 7% royalty for Medtronic's CRDM patent portfolio. PFF ¶ 25. That 7% royalty rate thus created a market price for Medtronic's CRDM patent portfolio. That was particularly the case in light of the fact that the CRDM industry was dominated by 3-5 major players (including Medtronic and Pacesetter, and later, St. Jude) throughout the 1990s and 2000s, meaning that those players paid close attention to seminal agreements such as the Pacesetter Agreement. Indeed, Mr. McCoy testified that he would have considered the Pacesetter royalty rate as a benchmark during his time at Guidant in 2005 and 2006 when negotiating CRDM patent licenses. PFF ¶ 32.

Despite recognizing the significance of benchmarks across the CRDM industry, Dr. Cockburn claimed during trial that there is no objective, arm's-length price for patents, but rather, the price depends on the parties' own subjective valuation of the patents. Tr. (Cockburn) 2258-2261, 2263-2266. That position is fundamentally at odds with section 482. If adopted, it would generally preclude the use of the CUT for pricing IP licenses.

(ii) The Term of the Pacesetter Agreement and the MPROC Licenses Directly Overlapped

The time gap in the execution of the two sets of agreements does not undermine comparability. While the two agreements were executed twelve years apart, the terms of the Pacesetter Agreement were in effect when the MPROC Licenses were entered (i.e., May 2004) because the Pacesetter Agreement's term was automatically extended when St. Jude acquired Pacesetter. PFF ¶ 15. As a result of the Pacesetter Agreement's ten-year term, which could be extended for a maximum of five years if assigned, the parties anticipated, in 1992, that the agreement could extend up through 2007. Ex. 2504-J § 9.02(d).

Medtronic and Pacesetter's willingness to license Medtronic's CRDM patents for 7% through 2007 establishes the arm's-length price for the license of Medtronic's key CRDM patents through 2007. This is because the parties necessarily assessed, at the execution of the Pacesetter Agreement, whether its terms (including the 7% rate) would reasonably withstand changes in the CRDM business such as changes in the technology, market conditions, profitability, etc., over 15 years. That market and technological assessment was performed again, two years later, by St. Jude — a different commercial actor, who was not a party to the Pacesetter Litigation and who sought to become a major player in the CRDM industry. Both St. Jude and Medtronic reached the same conclusion two years later: 7% was a commercially acceptable rate for Medtronic's CRDM patent portfolio. PFF ¶ 16.

(iii) The Pacesetter Agreement's High Baseline Royalty Rate Bridges the Gap, and There Was No Paradigm Shift

Critically, the fact that the Pacesetter royalty rate is high makes it a particularly reliable benchmark for pricing the MPROC Licenses because it narrows the question to whether the Medtronic 2004 patent portfolio justifies an even higher rate. The 7% royalty rate Siemens paid Medtronic was among the highest rates paid for CRDM patents. There is no evidence to the contrary. PFF ¶¶ 27-28, 29-31, 30-35. Having started with a high baseline rate, in order for Medtronic US to justify charging Medtronic PR an even higher rate, either Medtronic's CRDM patents or market conditions, or both, would have had to have undergone a significant transformation between 1992 and 2004.

There was no such transformation during that period. PFF ¶¶ 40-49. First, there was no transformation in CRDM technology during that intervening period. Dr. Cohen concluded that, between 1992 and 2004, there was no technological “paradigm shift” in the CRDM industry; rather, the industry experienced “incremental” or “iterative” technological improvements during that period. PFF ¶¶ 42-43. Second, Medtronic's patent portfolio itself was comparable in 1992 and 2004, both qualitatively and as further supported by quantitative expert patent citation analyses. PFF ¶¶ 44-49. Certain differences between the patent portfolios in 1992 and 2004 do not undermine their comparability because the value was highly concentrated in the subset of Medtronic's CRDM patents licensed under both sets of agreements — a key principle that Dr. Cockburn recognized. PFF ¶ 47. Thus, the core Medtronic CRDM portfolio remained technologically comparable between the two sets of agreements.

Starting with a highly reliable 7% base rate, Dr. Putnam was able to make adjustments to that base rate to account for the differences between the two sets of agreements. Several of those adjustments are addressed if the Court uses the other available benchmark rate of 15%. PFF ¶¶ 123-155; Part IV.D.

(iv) The Maximum Rate Clause Further Bridges the Gap Between 1992 and 2004

(1) 15% Was the Maximum Rate Medtronic US Could Charge Medtronic PR

The Pacesetter Agreement's Maximum Rate Clause provides definitive evidence of the maximum arm's-length price that Medtronic US could charge Medtronic PR in 2004 for CRDM patents. PFF ¶¶ 33-39. The Maximum Rate Clause allowed Pacesetter to license all of Medtronic's CRDM patents during the term of the Pacesetter Agreement for an aggregate royalty rate not higher than 15%, after taking into account the royalty that Pacesetter was paying Medtronic for the patents Medtronic had already licensed. PFF ¶ 12. This provision required the parties to predict — over a possible 15-year period — the highest possible royalty rate that they would be willing to accept for each other's current and future CRDM patent portfolios, including any transformational patents that could generate significantly higher profitability for the licensee. PFF ¶ 35.

Importantly, Medtronic's and Pacesetter's prospective agreement to this provision in 1992 for the Pacesetter Agreement's initial 10-year term definitively establishes the reliability and relevance of the 15% cap for pricing the MPROC Licenses in 2004. By agreeing to a royalty rate through the early 2000s, the parties identified a rate that would account for substantial changes in the patent portfolio and market. Moreover, if Pacesetter had invoked the Maximum Rate Clause in the late 1990s or early 2000s, any patent license granted by Medtronic under that provision could have been in effect well beyond the Pacesetter Agreement's initial 10-year term — a time period that would directly overlap with the MPROC Licenses even today. That is because any such license for newly issued future patents could have been spanned over a period up to the life of those patents. Thus, the parties determined ex ante that Medtronic's entire patent portfolio (subject to any temporary delays under the Key Patent Clause) would be licensed for no more than an aggregate 15% royalty, regardless of any changes in patent portfolios or the CRDM industry. Under these circumstances, the Maximum Rate Clause bridges any differences between when the Pacesetter Agreement and the MPROC Licenses were each entered, regardless of the Court's conclusion on whether the Maximum Rate Clause transferred to St. Jude. PFF ¶ 36.

Further, the evidentiary record, including the Board Documents, establishes the reliability of the Maximum Rate Clause. Not only did the parties give special attention to this provision, but it was also targeted to address future patents — it was not impacted by the parties' then-existing patent disputes. PFF ¶ 38.

(2) The Parties' Actual Subsequent Conduct Confirms 15% Was Still the Maximum Rate in 2004

There is additional evidence in the record — namely, the parties' actual conduct through 2004 — that reinforces the conclusion that Medtronic US could charge Medtronic PR no more than 15% for the CRDM patent portfolio. Specifically, Medtronic did not designate any of the CRDM patents as “key” throughout the term of the Pacesetter Agreement, from 1992 to 2004. PFF ¶ 37.

The Key Patent Clause required, in effect, that Medtronic perform an annual assessment of the value of its CRDM patent portfolio, taking into account the quality of Medtronic's patents and the market conditions, including the profit potential of those patents. If Medtronic had believed that certain of its patents could be licensed at a rate higher than 15% or that Medtronic could gain a market advantage by having exclusive rights to exploit the patent(s) for a period of time, it would have designated up to three patents per year as “key.” Medtronic's assessment of whether to designate one or more patents as “key” would have necessarily considered (i) the quality of Medtronic's patents and (ii) the market conditions, including profit potential, of those patents. PFF ¶ 37.

Thus, Medtronic's assessment of the value of its CRDM patent portfolio extended well beyond 1992 when it executed the Pacesetter Agreement. In fact, having decided to never designate any patents as key, Medtronic effectively offered — every single year — all of its patent portfolio for a maximum 15% royalty rate. That annual offer extended through May 2004, when the MPROC Licenses were entered. Medtronic US, therefore, effectively offered its entire CRDM patent portfolio to St. Jude in 2004 for no more than 15%, at the same time it offered that same CRDM patent portfolio to Medtronic PR.

The combination of the Maximum Rate Clause and Medtronic's decision to not designate any patents as “key” addresses the key factors in Treas. Reg. § 1.482-4(c)(2)(iii)(B)(1) relevant to evaluating a CUT. Medtronic was willing to license its entire CRDM patent portfolio to a third-party for no more than 15% taking into account the specific circumstances present in 2004, including: (i) the specific characteristics of Medtronic's CRDM patent portfolio; (ii) the potential profitability of the patent portfolio; and (iii) the relevant market conditions.

(v) Medtronic US's License of Neuro and Know-How to Medtronic PR Does Not Undermine Comparability

The intangible property transferred in the MPROC Licenses also included patents relating to Neuro Devices and Leads and know-how. These differences do not undermine the comparability of the Pacesetter Agreement, and Dr. Putnam otherwise made reliable adjustments to account for the transfer of know-how.

Neuro. Neuro Devices and Leads are sufficiently comparable to cardiac Devices and Leads — both technologically and from a profitability perspective. And, on that basis, Medtronic's experts concluded that it is appropriate to use the Pacesetter royalty rate for CRDM patents, as adjusted by Dr. Putnam, for pricing the Neuro patents licensed to Medtronic PR. PFF ¶¶ 67-72, 154.

Know-How. The Pacesetter Agreement did not license know-how like the MPROC Licenses did. A reliable adjustment can be made for know-how conveyed to Medtronic PR, however, and the MPROC Licenses' transfer of these additional rights therefore does not undermine comparability. Importantly, for comparability purposes, all of the intangibles in the Pacesetter Agreement and the MPROC Licenses, including know-how, are being used “in connection with similar products” (i.e., cardiac and Neuro Devices and Leads) and “within the same general industry or market” (i.e., the U.S. retail market for those products). Treas. Reg. § 1.482-4(c)(2)(iii)(B)(1)(i). PFF ¶¶ 50-66, 136-144.

(b) The Circumstantial Comparability Factors Are Satisfied

Treas. Reg. § 1.482-4(c)(2)(iii)(B)(2) sets forth specific economic and contractual factors that are particularly relevant to assessing the comparability between a controlled and uncontrolled transaction for the CUT Method. Each of these factors is addressed in various sections of this Memorandum. For ease of reference, the chart below identifies each factor and the section in this Memorandum where it is addressed:

Circumstantial Factor	Section
The terms of the transfer, including the exploitation rights granted in the intangible, the exclusive or nonexclusive character of any rights granted, any restrictions on use, or any limitations on the geographic area in which the rights may be exploited.	IV.C.2.(a) IV.C.2.(b).(i) IV.D & PFF ¶¶ 123-155
The stage of development of the intangible (including, where appropriate, necessary governmental approvals, authorizations, or licenses) in the market in which the intangible is to be used.	IV.C.2.(a)
Rights to receive updates, revisions, or modifications of the intangible.	IV.D & PFF ¶¶ 124-125
The uniqueness of the property and the period for which it remains unique, including the degree and duration of protection afforded to the property under the laws of the relevant countries.	IV.C.2.(a) IV.D & PFF ¶¶ 123-155
The duration of the license, contract, or other agreement, and any termination or renegotiation rights.	IV.C.2.(b).(i) IV.C.2.(b).(i).(2) IV.C.2.(a).(ii) IV.C.2.(a).(iii) IV.C.2.(a).(iv)
Any economic and product liability risks to be assumed by the transferee.	IV.C.2.(b).(ii)
The existence and extent of any collateral transactions or ongoing business relationships between the transferee and transferor.	IV.E.1
The functions to be performed by the transferor and transferee, including any ancillary or subsidiary services.	IV.E.2

(i) The Contractual Terms of the Pacesetter Agreement and MPROC Licenses Are Comparable

The contractual terms of the Pacesetter Agreement and MPROC Licenses are comparable. Specifically, the payment terms are comparable, as is the scope of each licensee's right to use the licensed intangible property. Moreover, while there are certain differences in terms, they either do not undermine comparability or can be adjusted for, as reflected in Dr. Putnam's adjustments.

(1) The Payment Terms Are Comparable

The terms of payment under the two agreements are comparable. Both agreements structured a running royalty based on sales of Devices and Leads. The fact that the Pacesetter Agreement also required two lump sum payments ($50 million and $25 million) does not undermine comparability. First, the $50 million upfront payment was for past infringement and therefore does not affect the royalty rate for the MPROC Licenses. PFF ¶¶ 106-107. Second, the fixed payment of $25 million was not a lump sum royalty, but rather a pre-paid credit against a portion of the future running royalty. Dr. Putnam addressed this factor through his 1.8% upward adjustment. PFF ¶¶ 108, 124-125.

(2) Other Differences Do Not Undermine Comparability

The fact that the Pacesetter Agreement had a cross-license provision, whereas MPROC Licenses do not, does not undermine comparability because the Pacesetter Agreement effectively reflects a one-way license from Medtronic to Pacesetter, and Dr. Putnam made an adjustment in any event to account for any potential differences. PFF ¶¶ 126-130. Moreover, while the duration of the two agreements were different, that difference likewise does not undermine comparability, nor does it require any adjustments in addition to the ones Dr. Putnam already made. PFF ¶ 110.

(ii) The Licensee's Product Liability Risk Assumption Under the Two Agreements Are Comparable

The contractual allocation of product liability risk and the resulting risk in fact borne by each licensee under the two sets of agreements are comparable under Treas. Reg. § 1.482-4(c)(2)(iii)(B)(2)(vi). Both Pacesetter and Medtronic PR, as the licensees, assumed all of the product liability risk relative to Medtronic US. PFF ¶¶ 111-122.

(1) Risk Allocation Under Section 482

Under section 482, the starting point for determining which controlled taxpayer bears a particular risk — including product liability risk — is the relevant contractual terms. Treas. Reg. § 1.482-1(d)(3)(ii)(B), (d)(3)(iii)(B). The regulations provide that the contractual terms, including the parties' contractual allocation of risk, are respected if such terms are consistent with the economic substance of the underlying transactions. Id. § 1.482-1(d)(3)(ii)(B)(1).

(2) Both Medtronic PR and Pacesetter Bore All the Product Liability Risk

Pursuant to the MPROC Licenses, all the product liability risk was allocated to, and borne by, Medtronic PR. PFF ¶ 112. As explained further below, that allocation reflected Medtronic PR's product liability risk under common law. By contrast, Medtronic US had no product liability risk under common law in its capacity as licensor. Likewise, under the Pacesetter Agreement, Pacesetter — not Medtronic — bore all product liability risk relating to the products manufactured using the licensed patents. PFF ¶ 113. That allocation also reflected the parties' underlying legal obligations — Pacesetter bore primary product liability under common law as the finished products manufacturer, and Medtronic had no liability in its capacity as the licensor.

(A) Medtronic PR's Product Liability Under Common Law Generally

Medtronic PR's assumption of product liability costs under the MPROC Licenses reflected its product liability risk under common law. Medtronic's expert from Medtronic I, Maya Eaton, and Respondent's expert from Medtronic I, Michael Green, agreed that Medtronic PR always bore primary liability for defects with respect to the Class III products as the finished products manufacturer. Although other entities within the chain of distribution may have product liability exposure, the existence of those parties does not eliminate the intrinsic risk to Medtronic PR. Mr. Green agreed that Medtronic PR is the one entity that can always be sued by putative injured plaintiffs and, thus, is always subject to product liability risk. Ex. 5487-P (Eaton Opening) ¶¶ 15-17, 29-55; MDT I Tr. (Eaton) 4733-42, 4748-51, 4757-58, 4767-72, 4777-78; MDT I Tr. (Green) 7096, 7107-08, 7139-42, 7150. The MPROC Licenses memorialized Medtronic PR's common law tort liability for recalls and defects. Exs. 7-J at 4175, 4177; 11-J at 4352, 4354.

(B) Medtronic US's Product Liability In Its Capacity as Licensor Was Limited

Medtronic US, by contrast, had limited product liability exposure under common law in its capacity as the licensor. A licensor of IP generally is not subject to product liability risk for products derived from that IP by a licensee. An IP license does not alter the fundamental proposition that the manufacturer of the licensed product remains primarily responsible for ensuring the safety of the product. See In re TMJ Implants Prods. Liab. Litig., 880 F. Supp. 1311, 1321 (D. Minn. 1995) (a trademark licensor is not responsible in tort simply because of a licensing agreement); Burkert v. Petrol Plus of Naugatuck, Inc., 579 A.2d 26, 35 (Conn. 1990); Boyer v. Weyerhaeuser Co., 39 F. Supp. 3d 1036, 1045 (W.D. Wis. 2014) (“[a]s a matter of law . . . plaintiffs cannot state a negligence claim based only on a party's licensing of a technology, even if used by the licensee to create a product containing asbestos.”). See also MDT I Tr. (Eaton) 4735:1-13.

Moreover, that Medtronic US could have theoretically been liable for a pure design defect risk under common law — if it were somehow separable from the manufacturing risk²¹ — does not undermine the fact that Medtronic PR in fact bore all the product liability risk relating to the Devices and Leads. First, product liability risks arising from design or manufacturing failures were contractually allocated to Medtronic PR under the MPROC Licenses, and Medtronic PR in fact paid for all such expenses. Second, as discussed below and as experts testified in Medtronic I, product liability risk solely with respect to the design of Class III medical devices (such as the Devices and Leads, the design of which both Medtronic US and Medtronic PR created) — is relatively insignificant in any event because of federal preemption.

(3) MPROC Licenses' Product Liability Risk Allocation Was Consistent with Economic Substance

Under section 482, as long as the contractual allocation under the MPROC Licenses was consistent with the economic substance, that allocation is to be respected. The economic substance inquiry considers all relevant facts and circumstances, including the following set forth in the Treasury Regulations:

• Course of Conduct: Whether the pattern of the controlled taxpayer's conduct over time is consistent with the purported allocation of risk between the controlled taxpayers; or where the pattern is changed, whether the relevant contractual arrangements have been modified accordingly. Treas. Reg. § 1.482-1(d)(3)(iii)(B)(1).

• Financial Wherewithal: Whether a controlled taxpayer has the financial capacity to fund losses that might be expected to occur as the result of the assumption of a risk, or whether, at arm's length, another party to the controlled transaction would ultimately suffer the consequences of such losses. Id. § 1.482-1(d)(3)(iii)(B)(2).

• Managerial/Operational Control: The extent to which each controlled taxpayer exercises managerial or operational control over the business activities that directly influence the amount of income or loss realized. In arm's-length dealings, parties ordinarily bear a greater share of those risks over which they have relatively more control. Id. § 1.482-1(d)(3)(iii)(B)(3).

In evaluating economic substance, the greatest weight is given to the actual conduct of the parties and their respective legal rights. Id. § 1.482-1(d)(3)(ii)(B).

The evidentiary record from Medtronic I demonstrates that the MPROC Licenses' allocation of product liability risk was fully consistent with the economic substance. Four factors support this. First, the parties' course of conduct was consistent with the allocation because Medtronic PR actually paid all the costs and damages arising from recalls and product defects, other than when Medtronic PR shared a portion of those expenses with Medtronic US to stay in compliance with the MOU. PFF ¶ 115. Medtronic's temporary reallocation of these costs to cater to Respondent's MOU has no effect on economic substance. Second, Medtronic PR had the financial wherewithal to bear catastrophic product liability risk based on its past and future cash flows. PFF ¶¶ 116-118. Third, Medtronic PR — and not Medtronic US — had sole managerial and operational control over the manufacturing operations for the finished Devices and Leads, which implicated the significant product liability risk. PFF ¶¶ 119-122; see Part IV.C.2.(b).(ii).(3).(A)-(C). Fourth, the product liability risk allocation reflected the parties' legal obligations under applicable state and federal law. See Part IV.C.2.(b).(ii).(3).(A)-(C).

Under these circumstances, the section 482 regulations require Respondent to respect the MPROC Licenses' product liability risk allocation.

Regarding the fourth factor, as both experts of Respondent and Petitioner explained in Medtronic I and as discussed in greater detail below, federal preemption under the Medical Device Amendments (“MDA”) to the Food, Drug & Cosmetic Act significantly limited product liability claims related to design and warning defects, but did not limit manufacturing defect claims with respect to Class III medical devices. Manufacturing defect risk with respect to Class III medical devices therefore reflected the significant product liability risk during tax years 2005 and 2006, and all other product liability risks were immaterial. Thus, any potential product liability exposure that Medtronic US may have faced under common law for its involvement in the design and product development of the Devices and Leads (i.e., beyond Medtronic US's capacity as the IP licensor) was immaterial at best, while Medtronic PR's potential liability for manufacturing defects was significant.

(A) Federal Preemption under the MDA

The MDA preempts state law causes of action against Class III medical devices that received a Premarket Approval (“PMA”) to the extent that any state “requirement . . . is different from, or in addition to” any federal requirement under the MDA. 21 U.S.C. § 360k(a); Ex. 5487-P (Eaton Opening) ¶ 21.

As explained by Ms. Eaton, in 2008, the Supreme Court clarified the scope of this preemption in Riegel v. Medtronic, Inc., 552 U.S. 312 (2008). The Supreme Court held that the PMA process established federal requirements; therefore, for a PMA-approved device, state law claims of negligence, strict liability and breach of implied warranty were preempted because they imposed additional or different requirements than federal law. Following Riegel, claims that survive preemption are generally referred to as “parallel claims” because the state-law duty must “parallel” a federal requirement to avoid preemption. Parallel claims include those: that allege a manufacturer deviated from federal requirements in manufacturing a PMA device; that allege a failure to comply with FDA requirements other than a PMA, such as the Good Manufacturing Practices or reporting requirements; or that allege a device was negligently manufactured in violation of PMA requirements. As a result of parallel claims, state law product liability claims continue to present significant product liability risk for manufacturers of Class III medical devices such as Medtronic PR. Ex. 5487-P (Eaton Opening) ¶¶ 39, 40-48, 51-55.

Respondent's expert Mr. Green agreed on the effect of federal preemption on product liability claims with respect to Class III devices and the limited scope of parallel claims that survive preemption:

Riegel, in my opinion, eliminated all of the standard products liability claims that exist under state law. Strict products liability, negligence, and breach of warranty, particularly breach of the implied warranty of merchantability, have been and remain the most frequently asserted products liability theories by plaintiffs. And within those three theories, negligent or defective design, warnings, and manufacture are the three bases on which product liability can be found.

Ex. 5536-R (Green Opening) ¶ 65 (emphasis added). Specifically, Mr. Green agreed that only a limited number of parallel claims survive preemption as described above. Ex. 5536-R (Green Opening) ¶¶ 69-70.

(B) Manufacturing Defect Risk is the Most Significant Risk

While it is theoretically possible for a plaintiff to bring a claim against a company that did not follow FDA-approved design or labeling (or failed to report a significant adverse event), such violations rarely occur in practice. Companies typically use the design and labeling that were approved by the FDA, and regulatory reporting is ordinary-course for Class III device manufacturers. By contrast, the manufacture of a PMA-approved device is an extremely complicated and intricate process that is subject to human error and various risks of malfunction (e.g., defects in machinery or components, incorrect assembly). The risk is much higher that the manufacture of a device would deviate from the PMA-approved specifications, giving rise to potential parallel claims for manufacturing defects. In addition to greater susceptibility to error, the consequences of those manufacturing errors are particularly severe. Because Devices and Leads are inserted into the human body, it is and was critical that Medtronic PR meet its quality obligations 100% of the time, and the resulting quality is and was the most critical function that would determine the success or failure of — and relatedly, any potential product liability losses with respect to — a Device or Lead. PFF ¶ 122.

Accordingly, as relevant to Medtronic's Devices and Leads, the types of design or labeling defects that remained viable (i.e., non-preempted) were limited. In effect, Medtronic US could only face design or labeling defect claims — the types of claims influenced by Medtronic US's activities — in the narrow circumstances where Medtronic failed to follow its own PMA-approved specifications. Such failures were extremely unlikely, and in any event, Medtronic PR would have been subject to liability given its role in designing the products. By contrast, a manufacturing defect claim remained much more likely to occur, given that Medtronic PR constantly engaged in activities that could give rise to a non-preempted claim. In other words, following PMA-approved specifications was easy; complete perfection in Medtronic PR's multi-faceted manufacturing processes on every single Device and Lead manufactured was far more difficult.

(C) Empirical Evidence Post-Riegel

Empirical evidence confirms that manufacturing defect risk is the most significant type of non-preempted parallel claim, and warning or designed-based claims are relatively uncommon and usually unsuccessful.

One study analyzed cases decided shortly after Riegel and found that the majority of the claims that survived preemption were manufacturing defect claims. S. Raymond, Judicial Politics and Medical Device Preemption After Riegel, 5 NYU J. L. & Liberty 745, 757-69 (2010) (Raymond). This article examined 75 cases decided between February 2008 — when Riegel was decided — and July 2010, which ruled on product liability claims for PMA-approved devices subject to the MDA. Of those 75 cases, only 17 survived preemption — i.e., over 77% of cases had preempted claims. Id. at 760. The study found that “[t]he broadest claim some courts have found not to be preempted after Riegel involve injuries arising from manufacturing defect.” Id. at 768 (emphasis added). Such manufacturing defect claims generally asserted that the device at issue was not manufactured in accordance with the FDA-approved standards. The author concluded that “[t]hese manufacturing claims provide[d] the largest number of unpreempted parallel cases” at the time. Id. at 768-69. The first federal court of appeals to rule on this question agreed that manufacturing defects were not preempted by Riegel, and several other courts of appeal have since decided the same.²²

By contrast, for design defects, the study found that, while claims based on a company's deviation from an approved design were possible, such claims were “an infrequent factual occurrence.” Raymond at 767. In fact, such a claim was allowed to proceed only once, and no other plaintiff had successfully alleged such facts post-Riegel during the examined period. See id.; Purcel v. Adv. Bionics Corp., 2008 WL 3874713 (N.D. Tex. Aug. 13, 2008).

As for labeling-based claims, while some lower courts have allowed claims based on violations of express warranties to move forward (i.e., on the basis that a company made statements not authorized by the FDA),²³ other courts have rejected such claims on preemption grounds.²⁴

In sum, both Medtronic PR and Pacesetter bore 100% of product liability risk, which further bolsters the comparability of the two agreements.

(c) The Context of Settlement Only Enhances the Pacesetter Agreement's Comparability

The fact that the Pacesetter Agreement was originally entered into as part of a litigation settlement does not affect its comparability to the MPROC Licenses. Based on the facts and circumstances of Medtronic's settlement with Siemens, the Pacesetter Agreement was executed in the “ordinary course of business” within the meaning of the section 482 regulations. And, in fact, the specific litigation-related circumstances that led to the Pacesetter Agreement only enhance the reliability of using its royalty rate as a benchmark for the MPROC Licenses.

(i) The “Ordinary Course” Standard

Treas. Reg. § 1.482-1(d)(4)(iii)(A)(1) provides that transactions “ordinarily will not constitute reliable measures of an arm's length result” if they are “not made in the ordinary course of business.” The Treasury Regulations do not expressly define “ordinary course of business,” but they do provide an example of what is not an “ordinary course” transaction.

Specifically, Treas. Reg. § 1.482-1(d)(4)(iii)(B), Example 1 provides:

USP, a United States manufacturer of computer software, sells its products to FSub, its foreign distributor in country X. Compco, a United States competitor of USP, also sells its products in X through unrelated distributors. However, in the year under review, Compco is forced into bankruptcy, and Compco liquidates its inventory by selling all of its products to unrelated distributors in X for a liquidation price. Because the sale of its entire inventory was not a sale in the ordinary course of business, Compco's sale cannot be used as an uncontrolled comparable to determine USP's arm's length result from its controlled transaction.

As Example 1 demonstrates, the “ordinary course of business” concept is intended to ensure that a potentially comparable uncontrolled transaction was not materially distorted by unusual circumstances. In Example 1, Compco's unusual circumstances were that it was selling inventory “for a liquidation price” because it was in bankruptcy. The example implies that Compco's bankruptcy forced it into a “fire sale” and that therefore the sale price was likely depressed. The potential distortion of the sales price caused by the distressed sale of the inventory results in an unreliable uncontrolled transaction.²⁵

The standards articulated by the federal courts considering whether to use a litigation settlement agreement as a comparable for computing patent infringement damages are similar. Specifically, as articulated by the Federal Circuit, when calculating patent damages based on a settlement agreement, the trial court must “consider the license in its proper context within the hypothetical negotiation frame-work to ensure that the reasonable royalty rate reflects 'the economic demand for the claimed technology.'” LaserDynamics, Inc. v. Quanta Computer, Inc., 694 F.3d 51, 77 (Fed. Cir. 2012) (citation omitted).²⁶

(ii) The Pacesetter Agreement Was in the“Ordinary Course of Business”

Under these standards, the Pacesetter Agreement — agreed to in the context of resolving litigation — was entered in the ordinary course of business. Given the concessions by Respondent's experts, the claim that the litigation context of the Pacesetter Agreement undermines its reliability as a CUT has been virtually eliminated. Dr. Heimert, as a transfer pricing economist, acknowledged that the context of settlement does not, in itself, cause the Pacesetter Agreement to be an unreliable benchmark. PFF ¶ 92. Dr. Cockburn likewise agreed that patent settlement licenses are not categorically unreliable. Dr. Cockburn “absolutely agree[d] that the probative value of the evidentiary value of the usefulness of a royalty rate which emerges in the course of a litigation can be high.” Tr. (Cockburn) 2177:14-17. He agreed that in some cases, licenses negotiated as litigation settlements are highly reliable gauges of patent value — more reliable than licenses negotiated outside of litigation — where the plaintiff is winning handily. The evidence in record, overwhelmingly establishes that Medtronic was winning “handily” and that Pacesetter's “future was in Medtronic's hands.” PFF ¶¶ 92, 94-96.

(1) Patent Licenses to Resolve Litigation Were Common and Can More Accurately Reflect IP Value

Respondent's experts' concessions follow the analysis of Medtronic's experts. As explained by Dr. Cohen, Mr. Spadea and Mr. McCoy, patent litigation and settlement licenses were and are common in the CRDM industry, particularly for broader patent licenses like the Pacesetter Agreement. PFF ¶¶ 87-88. Even in the absence of an active lawsuit, virtually all patent licenses — settlement or not — are negotiated in the shadow of litigation. Royalty negotiations are thus necessarily based on the outcome the parties would expect in litigation. PFF ¶ 89. This is reinforced by Medtronic's own commercial philosophy for patent licensing, which was to employ the same fundamental considerations whether the license arrangement was established in a litigation or non-litigation context. PFF ¶ 90.

Importantly, as explained by Dr. Putnam as well as an academic paper (Masur) cited by the Eighth Circuit in this case, in many cases, the terms of license agreements negotiated in settlement of litigation can more accurately reflect the value of the licensed IP than licenses negotiated outside of direct litigation. Litigation can make the parties better informed as to their respective legal rights and obligations, moving the bargain toward its true outcome. As Masur explains, the most reliable indicator of IP value will often be a license negotiated in the course of litigation that has progressed substantially, where the patent owner was winning. The greater the probability that the patent owner would win, the closer the value of the license to its true value. PFF ¶ 91.

(2) The Pacesetter Royalty Rate Was Not Distorted

As Medtronic's experts explained, the specific litigation-related circumstances of the Pacesetter Agreement in fact produced a royalty rate that was both more reliable for purposes of comparability and likely significantly higher than if Medtronic and Pacesetter had negotiated a license before Medtronic had initiated litigation. The Pacesetter Agreement presented a paradigmatic case of litigation moving the bargain toward its true outcome. PFF ¶¶ 93-96.

At the time of the settlement negotiations with Pacesetter, Medtronic had succeeded resoundingly in the Chicago Litigation and had obtained a permanent injunction. The Pacesetter Litigation thus substantially clarified the parties' rights and obligations over Medtronic's patents. Pacesetter needed access to those patents to stay in business. Medtronic had no such need for Pacesetter's patents. PFF ¶¶ 94-95, ¶¶ 126-129. Thus, the settlement context of the Pacesetter Agreement, to the extent that it had an effect on the terms of the Pacesetter Agreement, created upward pressure on the Pacesetter royalty rate and drove it to the true value of Medtronic's patents. PFF ¶ 96. This is consistent with empirical studies (including the Varner Study, which Dr. Cockburn relied on) showing that the average royalty rates for patent licenses entered as part of litigation settlement are higher than the average royalty rates for patent licenses absent active litigation. PFF ¶ 28.

(3) The Board Documents Reinforce that Litigation Did Not Distort the Terms

The litigation context may, in certain circumstances, distort settlement terms because, for example, the parties may settle based on avoided litigation costs rather than the value of the underlying IP. In addition, in a litigation involving multiple claims, settlement of one claim may be affected by resolution of other, unrelated claims. None of these factors distorted the terms of the Pacesetter Agreement. PFF ¶¶ 97-104.

First, one does not need to “speculate” as to whether factors such as litigation costs and resolution of other litigation claims distorted the Pacesetter royalty rate. The Board Documents establish that such considerations did not affect the Pacesetter royalty rate or other key terms. PFF ¶¶ 98-102.

With respect to litigation costs, the Board Documents show that Medtronic expected to pay an additional $17 million (NPV) for litigating the Pacesetter case to conclusion. This is a reliable projection of Medtronic's litigation costs because it was a necessary input for Medtronic to assess whether to continue to pursue litigation or settle. PFF 100. This expected litigation cost did not distort the terms of the Pacesetter Agreement because $17 million of avoided litigation costs is insignificant relative to the expected future cash flows from the license itself. Moreover, the benefit of avoiding litigation costs was symmetrical as both Medtronic and Pacesetter — then the number one and number two companies in the industry — were relieved of those costs. On these facts, litigation costs did not influence the settlement terms. PFF ¶ 101.

The parties' resolution of other litigation claims also did not distort the key settlement terms. The Board Documents' quantitative analysis of the license terms demonstrates that Medtronic solely factored in the value of Medtronic's patent infringement claims, and no other claims. PFF ¶¶ 99, 102.

(4) St. Jude Acquisition Validates Arm's-Length Nature

Importantly, St. Jude's acquisition of Pacesetter eliminates any questions about distortive effects of the Pacesetter Litigation. When St. Jude acquired Pacesetter in 1994, the sole question for St. Jude and Medtronic was whether to accept and continue the terms of the Pacesetter Agreement, including the 7% base rate and the Maximum Rate Clause, for 10 years. The parties evaluated that question in a purely commercial setting, apart from any active, ongoing litigation, and concluded that the terms of the agreement were commercially acceptable. In fact, the parties could have caused Siemens to exercise the “buy out” provision as part of the acquisition, but St. Jude instead agreed to accept the Pacesetter Agreement. PFF ¶ 16. Whatever litigation factors may have existed at the time of the Pacesetter Agreement, those factors were absent for St. Jude's 1994 acquisition, and thus, by definition, could not have influenced the decision-making of St. Jude and Medtronic. Therefore, St. Jude's assumption of the Pacesetter Agreement validates its arm's-length nature and propriety as a CUT.

Consistent with this Court's Medtronic I findings, there is also no dispute that the Pacesetter Agreement was assigned to, and assumed by, St. Jude (the “Assignment”). PFF ¶¶ 14-19; Medtronic I at 58-59. Respondent has claimed on remand, however, that while the Pacesetter Agreement transferred, the Maximum Rate Clause — and only the Maximum Rate Clause — did not. Although Respondent has not yet clearly articulated his reasoning, it appears to be that: (1) there is no assignment agreement in the record; and (2) Section 2.04 of the Pacesetter Agreement, which contains the Maximum Rate Clause, specifically terminated upon St. Jude's assumption of the agreement. Ex. 2504-J; Resp't Motion to Show Cause (Feb. 12, 2021). That position is contrary to both the record and the Pacesetter Agreement's express terms.²⁷

The factual record shows that the Maximum Rate Clause transferred to St. Jude, along with the rest of the Pacesetter Agreement. First, a number of agreements that St. Jude subsequently executed — including an amendment to the Pacesetter Agreement itself — confirm this. Second, Mr. Ellwein — the primary negotiator of the Pacesetter Agreement — implicitly affirmed that the Maximum Rate Clause transferred and was effective post-assignment to St. Jude when he testified that Medtronic did not designate any Key Patents during the life of the Pacesetter Agreement. Had the Maximum Rate and Key Patent Clauses terminated in 1994, as Respondent claims, Mr. Ellwein's answer, when asked whether any Key Patents were ever designated, would instead have been that that question was irrelevant for almost the entire term of the agreement (i.e., 10 out of 12 years). Third, Michael Coyle, who worked on the Pacesetter acquisition for St. Jude, confirmed that getting access to Pacesetter's IP, including Pacesetter's rights under the Pacesetter Agreement, was a crucial driver of the acquisition. Thus, accessing Medtronic's patents — including future patents — through the Pacesetter Agreement was part of St. Jude's commercial goals and strategy for the transaction. PFF ¶¶ 17-18.

Moreover, the Maximum Rate Clause was transferable as a contractual matter. Pursuant to the assignment provisions of the Pacesetter Agreement, Siemens was entitled to assign all of its rights and obligations — including those under the Maximum Rate Clause — to St. Jude:

• Section 9.02(a): “The Agreement shall be binding upon and inure to the benefit of the parties hereto, their Affiliates, and their permitted successors and assigns. . . .” Ex. 2504-J (emphasis added).

• Section 9.02(d): “Subject to Section 9.02(c) hereof and the final sentence of this subsection (d), The Agreement may be assigned or otherwise transferred by Siemens to a purchaser of substantially all of the assets and liabilities of Siemens and its Affiliates relating to the research, development, manufacture and sale of cardiac Stimulation Devices or of substantially all of Siemens and its Affiliates voting common stock or other voting equity ownership interest in all Affiliates of Siemens which are engaged in the business of research, development, manufacture and sale of Cardiac Stimulation Devices. The foregoing transactions in this section 9.02(d) are referred to herein as a 'Sale of the Business.'. . . .” Id.

Under Section 9.02(d), Siemens was entitled to assign “The Agreement” — i.e., the entire Pacesetter Agreement — to St. Jude. That is because: (1) Siemens' sale of its entire CRDM business was a “Sale of Business,” as defined in Section 9.02(d); and (2) St. Jude was a permissible assignee under Section 9.02(c)-(d). Id.

Importantly, nothing in Section 9.02 or any other provision of the Pacesetter Agreement carved out the Maximum Rate Clause from the assignable “Agreement” or otherwise limited Siemens' assignment of the Maximum Rate Clause to St. Jude. Thus, post-Assignment, the Maximum Rate Clause became fully effective between Medtronic and St. Jude. Section 9.02(d) included language (the “Termination Language”) to expressly terminate and novate the rights and obligations between Medtronic and Siemens for future patents issued after the date of the Assignment (the “Termination”):

Subject to Section 9.02(c) hereof and the final sentence of this subsection (d), The Agreement may be assigned or otherwise transferred by Siemens to a purchaser. . . . In the event of an assignment or transfer of The Agreement as permitted in this subsection (d), the rights granted by Medtronic and its Affiliates to Siemens and its Affiliates or Siemens and its Affiliates to Medtronic and its Affiliates under this [Pacesetter] Agreement shall terminate with respect to Future Patents issued after the date of such assignment or transfer and Sublicensable Patents which have not yet been licensed to, respectively Medtronic or Siemens at the time of such assignment or transfer.

Ex. 2504-J § 9.02(d) (emphasis added).

Respondent has generally pointed to the Termination Language to argue that the Maximum Rate Clause terminated upon the Assignment and, therefore, that the provision was not effective as between Medtronic and St. Jude. Respondent's interpretation assumes that the Termination Language terminated the Maximum Rate Clause for all current and future parties to the Pacesetter Agreement after the initial assignment by Siemens. That interpretation is contrary to both the plain language of Section 9.02(d) and the overall framework of the Pacesetter Agreement. As for plain text, the Termination applies to future patent rights granted by “Medtronic” and “Siemens” (and their respective Affiliates) as between each other — it does not apply to the “purchaser” or “assignee” (i.e., St. Jude).

In stark contrast to the Termination Language, the provisions in Section 9.02 that do define the assignee's or purchaser's rights and obligations upon assignment use the term “purchaser” or “assignee.” See, e.g., Ex. 2504-J § 9.02(a) (“The Agreement shall be binding upon and inure to the benefit of the parties hereto . . . and their permitted successors and assigns.” (emphasis added)); § 9.02(d) (“. . . . The Agreement may be assigned . . . to a purchaser.” (emphasis added)); § 9.02(e) (“In the event of a Sale of the Business. . . . Siemens may elect to not transfer its rights under The Agreement to the purchaser. . . .” (emphasis added)); § 9.02(f) (“In the event of a Sale of the Business . . . such assignment and transfer shall only be effective at such time as the purchaser . . . has agreed . . . to fully assume and perform all obligations of Siemens under The Agreement.” (emphasis added)).

The contract drafters intentionally used “Siemens” in the Termination Language to limit the rights of Siemens only, and not those of the assignee. At least four independent points support this. First, the terms “Siemens” and “Affiliates” are defined terms under the Pacesetter Agreement that do not include Siemens' assignee or successor. Thus, the Termination, by definition, did not apply to St. Jude. Second, the reference to “Siemens” appears in the same paragraph that provides for the assignability of the Pacesetter Agreement to a “purchaser,” yet the clause terminating the future patent rights references only Siemens and its Affiliates. That is, Section 9.02(d) frames Siemens' rights as separate from the assignee's rights, specifically in the context of assignment. Third, that the rights and obligations of “purchaser” are distinct from the rights and obligations of Siemens is reinforced by Section 9.02(f), which imposes an obligation on the “purchaser” to execute a separate agreement with Medtronic to “fully assume and perform all obligations of Siemens under The Agreement.” In the context of assignment-related provisions, the contract drafters were precise in their use of the term purchaser/assignee as compared to Siemens. Fourth, Section 9.02(a) refers to the parties “and their permitted successors and assigns,” which term could have been added to the Termination Language if the intention were to limit the rights of both Siemens and its assignee pursuant to the Termination.

Critically, if the parties had intended to terminate the Maximum Rate Clause altogether upon the first assignment of the agreement (as implied by Respondent's position), there would have been a much more direct and obvious way to express that intent — for example: “The parties shall in no event assign the rights and obligations under Section 2.04 [the Maximum Rate Clause] with respect to Future Patents issued after assignment.” Indeed, the contract drafters knew how to limit assignment in precisely this manner — the preceding provisions of Section 9.02(d) and Section 9.02(c) use exactly this kind of language to make clear when assignment is and is not allowed. See Ex. 2504-J § 9.02(c) (“The Agreement shall not, in any event, be assigned or otherwise transferred . . . to, or inure to the benefit of, any person or entity” in certain categories). Respondent assumes that the parties to the Pacesetter Agreement chose to terminate the Maximum Rate Clause obliquely, in language entirely unlike all other provisions that actually and expressly limit the scope of assignment or otherwise actually “turn off' specific terms. E.g., id. § 2.05 (“turning off' a provision in the event of either party's material breach).

The reason the Termination Language was drafted with such precision — focusing on future patents issued after the assignment of the Pacesetter Agreement — is apparent from the potential operation of the Maximum Rate Clause in the event Pacesetter were sold to another party (as it was to St. Jude). Specifically, when Pacesetter was sold to St. Jude, any CRDM patents owned by Pacesetter prior to the sale transferred to St. Jude, which then assumed all obligations with respect to those patents under the assigned Pacesetter Agreement. There was a possibility, however, that Siemens could continue to create and acquire its own future CRDM patents post-assignment. Without the Termination Language, therefore, a question could arise as to whether Medtronic is entitled to request a license from Siemens for such future patents under the Maximum Rate Clause (and vice versa), notwithstanding that the Pacesetter Agreement was otherwise assigned to St. Jude. Without express novation, under New York law, Medtronic potentially would have had continuing rights against Siemens with respect to those post-assignment patents. Mandel v. Fischer, 205 A.D.2d 375, 376 (N.Y. App. Div. 1994). Indeed, Siemens had other continuing obligations under the Pacesetter Agreement that would survive any assignment, absent an express novation. See, e.g., Ex. 2505-J § 3. Thus, the Termination Language was included to prevent Medtronic and Siemens from being obligated to provide its future patents to each other, after any assignment of the Pacesetter Agreement.

Thus, neither the Termination Language nor any other provision of the Pacesetter Agreement (nor anything else in the record) supports Respondent's position that the Maximum Rate Clause did not transfer to St. Jude along with all other provisions of the Pacesetter Agreement.

D. The Pacesetter Agreement Can be Adjusted to Enhance its Comparability to the MPROC Licenses

Dr. Putnam made adjustments to account for differences between the Pacesetter Agreement and the MPROC Licenses. PFF ¶¶ 123-155. Consistent with Treas. Reg. §§ 1.482-1(c)(2), (d)(2), -4(c)(2), one core rule guided the analysis that Dr. Putnam performed to ensure that the adjustments he has determined (and areas in which he did not make adjustments) are reliable: every adjustment, or lack thereof, is grounded in the Pacesetter Agreement itself and in objectively observable evidence, whether it be the Board documents, financial data or public filings. As each part of his analysis shows, Dr. Putnam's views are grounded in facts and evidence. This is shown by two core principles underlying each of his decisions on whether and to what extent to adjust:

The Maximum Rate Clause. The Maximum Rate Clause provides guideposts for the maximum adjustments that could be made for the difference in Medtronic's CRDM patent portfolio in 1992 and 2004. The Maximum Rate Clause provides an objective parameter that definitively “caps” the reliable adjustments that an economist can make to the base royalty rate to account for differences in the CRDM patent portfolios. PFF ¶¶ 33, 135, 151. In this regard, the 15% cap offers a cross-check as to the reliability of Dr. Putnam's adjustments.

Maintain consistency with the CUT. For each of his adjustments, Dr. Putnam considered whether the Pacesetter Agreement itself or the parties' negotiating behavior offered assistance in identifying whether an adjustment is necessary, and, if so, in determining the size of the adjustment. By taking this approach, Dr. Putnam's adjustments are intended to closely match how the parties at arm's length — Medtronic and Siemens (and St. Jude) — would have acted.

For the reasons above, Petitioner's CUT is the best method under Treas. Reg. § 1.482-1(c).

E. The Commissioner's Other Criticisms of the Pacesetter Agreement as a CUT Are Invalid

1. The Criticism that Medtronic PR Was Not Competing with Medtronic US Is Wrong as Matter of Law and Economics

Respondent's experts posited that, because Medtronic PR had a “vertical” relationship with its affiliate Medtronic US, the Pacesetter Agreement, as a license between “horizontal” competitors, cannot be a valid CUT for the MPROC Licenses. PFF ¶ 49. However, this would in effect put the end to any CUT. Indeed, Dr. Heimert conceded during trial that an agreement between competitors can be a valid CUT. He also agreed that a negotiation between competitors creates sufficient negotiating tension for an arm's-length result. PFF ¶ 51.

To the extent that Respondent continues to challenge the validity of the Pacesetter Agreement as a CUT based on “horizontal” vs. “vertical” differences, that argument has no basis in law or economics. It is fundamentally inconsistent with the arm's-length standard in section 482 and the section 482 regulations on the CUT Method. By definition, a controlled transaction between related parties is not — and will never be — between competitors. The arm's-length standard in section 482 is premised on the principle that the intangible property transferred between affiliates should be valued based on how uncontrolled parties in the relevant market would price the intangibles. See Treas. Reg. § 1.482-1(b).

The CUT method is, of course, the most direct way of establishing how uncontrolled parties would price intangibles because it relies upon actual market transactions. Within the category of CUTs, those executed between competitors would, in fact, be the most reliable transactions for establishing an arm's-length price, because each party has the heightened incentive to both optimize their own returns and to maximize the detriment to their competitor.

Consistent with these principles, both the section 482 regulations and decisions by this Court confirm the validity of a CUT in analogous circumstances. Two examples in the Treasury Regulations addressing the CUT method: (a) involve a controlled subsidiary (i.e., a vertical relationship) and (b) involve uncontrolled transactions with counterparties that are or could be competitors. Treas. Reg. § 1.482-4(c)(4), Examples 1 & 3 (Example 1 accepts a license agreement with a competitor in the pharmaceutical industry as a CUT, and the third-party licensee's status as a competitor in the industry is not presented, in any way, as an impediment to comparability).²⁸ If the uncontrolled counterparty's status as a competitor were harmful to the CUT method, then these examples would expressly address that point, given the apparent likelihood that the uncontrolled counterparty is a competitor. The absence of any such discussion in the examples demonstrates the legal irrelevance of Respondent's arguments.

Moreover, this Court has also agreed that the CUT method was appropriate and that a transaction with a competitor was comparable with the intercompany transaction between vertically integrated affiliates. See, e.g., Amazon.com, Inc. v. Commissioner, 148 T.C. 108 (2017) (both the taxpayer and the IRS agreed that a CUT between Amazon and Target, direct competitors, was appropriate to price an intercompany buy-in payment for certain intangibles).

Thus, the position of the Commissioner's experts that Pacesetter is not a valid CUT because Medtronic PR is not a competitor of Medtronic US cannot be reconciled with the arm's-length standard generally or the CUT method specifically. And, it is otherwise inconsistent with the factual record showing that the competitive dynamic between Medtronic US and Pacesetter bolstered the reliability of the Pacesetter royalty rate. PFF ¶¶ 50-52.

2. The Cited Functional Differences Are Irrelevant

Respondent's experts' arguments about the functions performed between the licensees (Pacesetter and Medtronic PR) and the licensor (Medtronic US) under the two agreements are likewise contrary to the law and applicable valuation principles. Dr. Cockburn pointed to several functions performed by Pacesetter but not Medtronic PR: R&D, quality control, regulatory and strategic business management. PFF ¶ 53.

To illustrate the flaw in the Commissioner's position, consider a simple hypothetical. Party A sells a building in New York City to Party B. Two weeks later, Party C makes an offer for a second building in New York City owned by Party A and identifies the sale to Party B as a comparable transaction. Whether Party C is in the same business or performs the same functions as Party B will not be a basis upon which the comparable transaction will be distinguished. The market transaction sets the price for the property, and the price of that property does not change based on the characteristics of the buyer.

The same principle applies for valuing intangibles under the CUT method. There almost always will be functional differences between an intercompany transaction and a proposed CUT: the taxpayer's functions will generally be divided among affiliates, which is what creates the need for the application of the transfer pricing rules in the first place. By comparison, the overall operations of a third party will frequently have more integrated functions. See Treas. Reg. § 1.482-4(c).

Consistent with these principles, Treas. Reg. § 1.482-4(c) does not require a comparison of the controlled and uncontrolled parties to determine whether the two parties are sufficiently comparable. The rules relating to the CUT method do not require tallying up the functions performed by Pacesetter and Medtronic PR. That type of comparison is explicitly required for other transfer pricing methods — such as for the cost-plus method and the CPM — that establish an arm's-length return based more on a comparison of the functions performed by the controlled and uncontrolled parties. Treas. Reg. §§ 1.482-3(d)(3)(ii)(A) (cost-plus method), -5(c)(2)(1) (CPM).

For the CUT method, the differences in functions performed are relevant only to the extent that Medtronic PR is realizing additional benefits from Medtronic US that were not conveyed by Medtronic to Pacesetter. This is confirmed by both the language of the Treasury Regulations and one of the examples demonstrating the application of the CUT method. See Treas. Reg. § 1.482-4(c)(4), Example 3. This Court has taken a similar approach in response to the Commissioner's prior attempts at this argument. See Bausch & Lomb Inc. v. Commissioner, 92 T.C. 525, 589 (1989) (“We have found as fact that B&L functioned as a distributor with respect to lenses it purchased from B&L Ireland. We fail to see the significance of the fact that B&L engaged in other functions in addition to distribution with respect to soft contact lenses.”).

Here, the significant functions performed by Pacesetter, but not by Medtronic PR — that is, component manufacturing, maintaining and enhancing the trademark and distribution — have already been priced at arm's length. All the activities already encompassed within such arm's-length pricing cannot undermine comparability. Indeed, Dr. Heimert agreed during trial that the fact that Pacesetter has these other, already-priced functions does not, in itself, impair the Pacesetter Agreement's validity as a CUT. PFF ¶ 55.

As for intangible rights that Medtronic US conveyed to the licensees, Dr. Heimert agreed that Pacesetter's R&D function does not, in itself, break comparability. PFF ¶ 56. Instead, differences in R&D activity are relevant to the extent that Medtronic PR accessed additional IP not licensed to Pacesetter. Medtronic US did provide Medtronic PR with additional intangible rights not conveyed to Siemens. Specifically, Medtronic US provides the most current developments in its patent portfolio and know-how. Both of these items, however, were separately priced by Dr. Putnam. PFF ¶ 57. Moreover, the other functions (such as quality control, clinical trials, regulatory approval and strategic business management) have likewise already been compensated under the other intercompany agreements or through Dr. Putnam's adjustments, PFF ¶¶ 58-59, or are not compensable under section 482. MDT I Pet'r Post-Trial Reply Brief at 140-43.

F. The Results Under Dr. Putnam's Royalty Rates Are Reasonable

1. The Resulting Profit Split Is Reasonable

To analyze the reasonableness of the CUT method's results, Dr. Hubbard reviewed the resulting relative split of profits — as between Medtronic US's R&D function and Medtronic PR's manufacturing function. Because the only remaining issue in this case is the pricing of the MPROC Licenses as between these two functions, this profit split is the more relevant one for testing reasonableness. Tr. (Hubbard) 1451. In other words, the more relevant profit split for this analysis is the profit split between Medtronic US's R&D function and Medtronic PR's manufacturing, not between Medtronic US and Medtronic PR more broadly. The Court has already determined that the other non-R&D-related functions that Medtronic US and Med USA perform — component manufacturing and distribution — are appropriately compensated.

Applying the royalty rates at the high-end of Dr. Putnam's ranges, 62% to 64% of allocable profit goes to Medtronic PR's manufacturing function, and 36% to 38% goes to Medtronic US's R&D function. Ex. 6109-P (Hubbard Rebuttal) ¶¶ 60-67. This split includes trademark income and costs in Medtronic US's return because the trademark value is a function of Medtronic US's R&D and Medtronic PR's high-quality finished device manufacturing. As Dr. Putnam explained:

MPROC also paid an 8% royalty to MED Inc. for rights to Medtronic's trademarks, and a portion of that royalty payment can also be viewed economically as, in part, a payment for MPROC's rights to R&D. As Medtronic's expert Mr. McCoy opined, the value of Medtronic's trademarks and brand is significantly influenced by product reliability and performance. While Mr. McCoy was focused on MPROC's manufacturing capabilities, R&D also plays an important role in product performance.

Ex. 6106-P (Putnam Rebuttal) ¶ 44 n.45 (citation omitted); see also Ex. 6102-P (McCoy Rebuttal) at 8 (“In the CRM industry, Quality in manufacturing was viewed to be closely correlated with high product reliability. Therefore, the work of MPROC to deliver Quality had direct bearing on the customer-decision-maker's perception of the Medtronic brand. Thus, MPROC had a direct and important role as-to the brand promise represented by the Medtronic trademark.”). Accordingly, to the extent that Medtronic US (as opposed to Medtronic PR) is being compensated for trademark value, that is properly viewed as a return to Medtronic US's R&D function. This differentiates the Trademark License from component manufacturing and distribution.

This relative split of profit is consistent with the allocation of profit between Medtronic and Pacesetter in the Pacesetter Agreement, as well as other indicators of a reasonable profit split between an IP licensor and licensee in the CRDM industry. Tr. (Hubbard) 1451:7-1454:32; Ex. 6158-PD at 9. This relative split is above the 29% share of profits that Professor Pindyck calculated, in Medtronic I, to have been retained by Medtronic as the licensor to the Pacesetter Agreement. Ex. 5550-P (Pindyck) ¶¶ 38-40. Medtronic's profit share under the MPROC Licenses is also consistent with the 35% share of operating margin that Mr. McCoy explained is the “practical upper limit” to what a licensor could retain in the industry. Ex. 6101-P (McCoy Opening) at 11. This split is also consistent with Dr. Hubbard's economic evaluation of the relative significance of patents and technology IP for CRDM Devices and Leads. See infra Part IV.G.2.(c). And lastly, this split is reasonable in light of this Court's findings on Medtronic PR's critical role in finished device manufacturing and beyond; Medtronic PR's bearing of product liability risk; and this Court's “reasonable alternatives” analysis for the types of and volume of CRDM Devices and Leads that Medtronic PR manufactured. As Dr. Hubbard explained, these data points on profit splits between market participants is instructive as to how parties behave at arm's length. Tr. (Hubbard) 1451-55.

2. Respondent's Criticisms of this Profit Split Are Meritless

Respondent's experts present various criticisms of the result under Dr. Putnam's royalty rates. These criticisms are meritless. In particular, Dr. Heimert presents, as he did in Medtronic I, comparisons of “value added” costs relative to profits. Ex. 6204-R (Heimert Rebuttal) ¶¶ 115-16. First, this analysis aggregates all of the intercompany transactions when it should focus on the transactions relevant to Medtronic US's return to R&D. In particular, distribution and component manufacturing already have been compensated at arm's length and have no relevance to Medtronic US's return on its R&D activities.

As agreed to by Dr. Heimert, because component manufacturing and distribution costs have been priced at arm's length, they are, by definition, equivalent to third-party expenses. Tr. (Heimert) 2540:5-11. Medtronic PR thus bears these costs in the same manner as it would if it purchased components from, or sold products through, a third party. Thus, Dr. Heimert's concept of “value-added” costs is particularly inapt given this established transfer pricing. Tr. (Hubbard) 1730:1-1732:4. Once costs are properly allocated and analyzed, the reality is that Medtronic PR incurred more costs than Medtronic US — over three-quarters of the costs as between Medtronic US R&D and Medtronic PR finished device manufacturing under Dr. Putnam's highest rates. Ex. 6109-P (Hubbard Rebuttal) ¶¶ 66-70.

G. The CPM Remains Arbitrary and Capricious and Is Not the Best Method

The transfer pricing analysis in Dr. Heimert's CPM — including Dr. Heimert's assessment of Medtronic PR and its capabilities — remains unchanged from Medtronic I, as Dr. Heimert himself testified. Even the points that Dr. Heimert characterized in his Medtronic III report as “clarifications” of his prior analysis simply restated points that he presented in Medtronic I. PFF ¶ 156.

For the reasons that this Court aptly described in Medtronic I, Dr. Heimert's CPM is arbitrary, capricious and unreasonable. Medtronic I at 118. Since then, Dr. Heimert's CPM has remained the same, even now that the established factual record is contrary to core premises of his analysis. Thus, the situation has only gotten worse since Medtronic I, which is all the more reason that Dr. Heimert's arbitrary and capricious method cannot be the best method.

In contrast, Medtronic's experts provided new, economically and scientifically-informed analysis, including Dr. Hubbard's attempt to construct a CPM from the ground-up. As Medtronic's experts demonstrate, the economics, the science and the factual record are all at odds with Dr. Heimert's CPM.

1. Dr. Heimert's 14 Companies Are Not Comparable

Dr. Heimert's 14 companies — as well as any subset thereof — still are not comparable to Medtronic PR, per this Court's prior findings. PFF ¶¶ 156-157; see Treas. Reg. § 1.482-5(c) (describing comparability between controlled and uncontrolled parties in the CPM). In Medtronic I, as now, Dr. Heimert's analysis miscomprehends the nature of Medtronic PR's functions, resources and risks. Dr. Heimert's continued misapplication of these key comparability factors remains fatal to his CPM's reliability. See Treas. Reg. § 1.482-5(c)(2)(ii).

As this Court found in Medtronic I, Respondent and Dr. Heimert analogized Medtronic PR to a components manufacturer — indeed, using the same 14 companies for both functions — and argued that Medtronic PR had no particular responsibility for quality relative to the overall product “value chain.” Medtronic I at 106, 112; Tr. (Heimert) 2469:25-2470:8. Even after Medtronic I, Dr. Heimert did not analyze or account for the difference in quality manufacturing operations and expectations between component manufacturing and Medtronic PR's finished device manufacturing.²⁹

This characterization of Medtronic PR is not correct. It improperly downplays both the importance of quality to Medtronic's CRDM business, Medtronic I at 97-102 (“Respondent does not place enough emphasis on the importance of quality in the industry.”), and Medtronic PR's central role in ensuring that quality, id. at 102-05. Wholly unlike a component manufacturer, Medtronic PR was “an FDA-registered facility responsible for putting together sophisticated medical devices that would remain in the human body for years.” Id at 107. Medtronic PR “leveraged its systems engineering expertise to make the manufacturing process design improvements to Class III finished medical devices, enabling a safe product to be made.” Id. at 106-07. In addition, Medtronic PR “contributed throughout the design process and had a role in product development.” Id. at 108.

Dr. Heimert's 14 companies thus fundamentally differ from Medtronic PR in product offerings, their role in product quality, the product liability risk that they bear and manage, scale, asset composition and geographic markets. Id. at 109-10; PFF ¶ 166; see Tr. (Heimert) 2535:22-2536:4.

(a) Expert Testimony on Remand Further Demonstrates that Dr. Heimert's Companies are Not Comparable

On remand, Medtronic's experts further demonstrated that Dr. Heimert's 14 companies are not comparable through credible economic, scientific and industry perspectives. PFF ¶¶ 163-167.

Economically, Dr. Heimert's recycled arguments defending the same rejected companies are unconvincing. As Dr. Hubbard explained, Dr. Heimert's generic arguments “paint[ ] in unreasonably broad strokes,” reducing Medtronic PR's functions and attributes to meaningless generalities in order to attempt to establish comparability with his 14 companies. Likewise, Dr. Heimert argued that Medtronic PR and his 14 companies are comparable in terms of the importance of product quality (and their role in ensuring that quality) because his companies' 10-Ks disclose quality as a risk factor. This shared, generic element among financial statements proves nothing about the degree of importance of product quality.³⁰ As Dr. Hubbard points out, even the maker of Swingline staplers discloses quality as a risk factor. Any argument suggesting that a Class III implantable medical device is akin to a stapler proves too much. PFF ¶¶ 163-165.

Dr. Heimert's 14 companies are also fundamentally different in terms of products and risks assumed. Mr. McCoy concluded that any comparison of the activities or the profits of the 14 companies with Medtronic PR “would be such a stretch as to be meaningless[,]” because the risks assumed and related functions performed to be too different to allow for an economically reasonable comparison. PFF ¶ 166. And, as Dr. Cohen explained, the 14 companies are, in whole or predominantly, manufacturers of Class I and Class II medical products, and the functions and risks associated with Class I and Class II are substantially different from those of a finished manufacturer of Class III CRDM devices. PFF ¶ 167; see also Medtronic I at 8-9, 109-12.

(b) The So-Called “Implantables” Are as Flawed as the Rest of Dr. Heimert's 14 Companies

Dr. Heimert testified that a subset of his 14 companies — the so-called “implantables” — could perhaps be used to perform a CPM for Medtronic PR. But this subset is the fruit of a poisoned tree — they suffer the same flaws as the rest of Dr. Heimert's 14 companies. Indeed, this Court specifically rejected companies within this subset in Medtronic I. Id. at 109 (rejecting Orthofix); PFF ¶ 168.

But even more fundamentally, this subset of “implantable” companies fails because they are not even really “implantables.” Dr. Heimert himself concluded that there is not enough data to create an “implantable” set within his 14 companies for purposes of making a CPM adjustment. As Dr. Heimert stated in his opening remand report, he originally considered adjusting his set of companies to “eliminate any non-implantable products within each Comparable's manufacturing function.” Ex. 6203-R (Heimert Opening) ¶ 102. He concluded, however, that such an adjustment could not be done. PFF ¶ 169.

In other words, Dr. Heimert himself concluded that there is no real “implantable” set of companies because there is no “reliable” data on the relative product mix within Dr. Heimert's 14 companies, a point that Dr. Heimert conceded in his testimony. At most, this subset of Dr. Heimert's companies is “implantables” in the sense that they produce at least one implantable product. Dr. Heimert used no information about relative product mix beyond that. PFF ¶ 169.

This is why Dr. Heimert originally presented his subset of “implantables,” not as a CPM adjustment, but instead to demonstrate that his CPM was not sensitive to whether the comparables produced any implantable products. But Dr. Hubbard demonstrated that Dr. Heimert's claim was wrong: the output of Dr. Heimert's CPM was sensitive to which companies produced at least one implantable product. Dr. Hubbard was very clear that, in all events, the “implantable” subset of Dr. Heimert's comparables still presented inappropriate benchmarks for the profitability of MPROC. PFF ¶ 170. Respondent's experts stated incorrectly that Dr. Hubbard had somehow endorsed the use of the “implantables” as valid comparables and manufacturers of Class III medical devices like Medtronic PR, which they admitted on cross-examination. PFF ¶ 171.

2. Dr. Hubbard, Unlike Dr. Heimert, Credibly Attempted a CPM

(a) Dr. Hubbard Found a Reliable CPM to Be Impossible

These fundamental differences between Medtronic PR and Dr. Heimert's 14 companies informed Dr. Hubbard's conclusion that a reliable CPM could not be performed for the MPROC Licenses. Dr. Hubbard attempted to construct a reliable CPM from the ground-up, based on the broader universe of available financial data, potential comparables, and the factual record as it stands in this remand proceeding. Notably, Dr. Hubbard started from — rather than ignored — the fact that only one intercompany transaction remains to be priced, while the three others previously at issue have been priced at arm's length. PFF ¶¶ 160-162.

Dr. Hubbard also genuinely reexamined the choice of tested party; the potentially comparable companies for either Medtronic PR or Medtronic US as the tested party; and the choice of PLI. PFF ¶ 161. By contrast, Dr. Heimert failed to reassess these key choices. PFF ¶ 156; see Treas. Reg. § 1.482-5(b)(2), (4) (describing the selection of the tested party and PLI). It undercuts the credibility of Dr. Heimert's analysis that he did not reconsider any of these key premises after this Court's prior, un-appealed factual determinations related to his CPM.

For the reasons described above, Dr. Hubbard credibly concluded that a reliable CPM could not be performed to price the MPROC Licenses. PFF ¶ 161.

(b) Dr. Hubbard's Analysis Further Illustrates Why ROA Is Not an Appropriate PLI

In particular, Dr. Heimert failed to contend with this Court's prior finding that ROA fails to capture the value that Medtronic PR contributes through its workforce skills and quality/institutional know-how accumulated over three decades. Medtronic I at 112-14; PFF ¶ 157. Dr. Hubbard likewise explained that ROA is not suitable for Class III medical device manufacturing functions such as those performed by Medtronic PR. Here, Dr. Heimert's CPM used an asset base for Medtronic PR that does not reflect the value of technical know-how, process improvements, and MPROC's workforce in place. PFF ¶ 172. The Treasury Regulations caution against this exact problem in using ROA as the PLI — i.e., situations where the measured operating asset base fails to capture the true value of the enterprise for which profitability is being benchmarked. Treas. Reg. § 1.482-5(b)(4)(i) (“The reliability of this profit level indicator [ROA] increases as operating assets play a greater role in generating operating profits for both the tested party and the uncontrolled comparable. . . . [D]ifficulties in properly valuing operating assets will diminish the reliability of this profit level indicator.”).

Mr. McCoy's experience proves, in application, this theoretical economic point. In Mr. McCoy's view, ROA is not an appropriate PLI at least in part because, as a practical matter, he never used ROA to analyze the financial performance of the finished device manufacturing function at his CRDM companies. It is not a financially relevant metric for that function, both in Mr. McCoy's experience and for Medtronic PR. PFF ¶ 173.

In addition to the operating asset base failing to capture Medtronic PR's true value, the book values that Dr. Heimert uses for those operating assets are depressed. As described below, the book values that Dr. Heimert used for Medtronic PR are low relative to his purported comparables due to financial accounting conventions. See infra Part IV.H.2.b.

Dr. Heimert failed to address these fundamental issues with the ROA and instead presented unconvincing defenses of it. Dr. Heimert and Dr. Cockburn argued that ROA is acceptable because Dr. Heimert uniformly excluded intangible assets across Medtronic PR and his 14 comparables. As described above, this argument misses the mark. PFF ¶ 174.

In addition, Dr. Heimert's and Dr. Cockburn's purported “cross-checks” on the ROA that Dr. Heimert assigned to Medtronic PR are circular and unconvincing. The CPM benchmarks Medtronic PR's ROA to the selected comparables, and then derives royalty rates for Medtronic PR from that benchmarking. So of course, doing the same thing in reverse — using Dr. Heimert's assigned royalty rates to calculate Medtronic PR's ROA, and then comparing that back against the ROAs for the original comparables (or a more general set of similar comparables) — will always “support” Dr. Heimert's CPM. PFF ¶ 175. There is no magic to putting the rabbit in the hat.

Related arguments from Respondent's experts — to the effect that Dr. Putnam's royalty rates result in an insupportably high ROA for Medtronic PR compared to Dr. Heimert's companies and other firms — founder for similar reasons. E.g., Ex. 6203-R (Heimert Opening) ¶ 105. ROA, as applied by Dr. Heimert, is a mismeasurement in various ways. E.g., Ex. 6108-P (Hubbard Opening) ¶¶ 49-57; Tr. (Hubbard) 1440:4-18. In addition, a similarly skewed result arises from applying this type of cross-check to Medtronic US's profitability using Dr. Heimert's rates. Ex. 6108-P (Hubbard Opening) ¶ 58; Tr. (Hubbard) 1441:221443:15; Ex. 6158-PD at 6. That fact, along with other flaws in Dr. Heimert's CPM, renders these arguments unconvincing. See Tr. (Heimert) 2462:17-24.

(c) Dr. Hubbard's Analysis Demonstrated Why There is Relatively Poor Data in the Industry for a CPM

Another key insight from Dr. Hubbard's ground-up attempt at CPM is why there were inadequate financial data for the task. Dr. Hubbard found that the companies that had potentially comparable functions were vertically integrated, like Medtronic PR itself, with no segmented financial data for a reliable CPM. As Dr. Hubbard concluded, relevant companies retain finished device manufacturing in-house because it is a core function in the Class III medical device industry generally (and the CRDM industry in particular). The flaws in Dr. Heimert's data and analysis are thus interlinked: the lack of adequate data for a CPM stems, at least in part, from the importance of Medtronic PR's function (which Dr. Heimert generally ignores). As a result, the financial data for those functions is not available on a standalone or segmented basis, generally precluding a reliable CPM for them. PFF ¶¶ 162, 169; see Treas. Reg. § 1.482-5(c)(3)(i) (“The reliability of the results derived from the comparable profits method is affected by the quality of the data and assumptions used to apply this method.”).

By contrast, Dr. Hubbard found sufficient transactional data for a CUT at least in part because of how technology IP is shared/cross-licensed in the CRDM industry. Tr. (Hubbard) 1433:21-1434:1; see also Tr. (Putnam) 1090:15-1091:1.

3. Dr. Heimert's Analysis Ignores the Facts by Aggregating

Relative to Dr. Heimert's analysis in Medtronic I, the most disconcerting aspect of his CPM on remand is his position on aggregation. As Dr. Heimert testified at the first trial, aggregation was the starting premise of his CPM. And as his reports stated, his results were, in his view, reasonable because of aggregation. This Court rejected aggregation for the intercompany transactions involving Medtronic PR, Medtronic I at 114-16, a finding that Respondent did not appeal. Similarly, three of the four intercompany transactions that Dr. Heimert aggregated have now been priced at arm's length and are “off the table” in this remand proceeding. As Dr. Heimert agreed in his testimony, the royalty for the MPROC Licenses must be evaluated and established independently from the other three intercompany transactions. PFF ¶¶ 158-159.

Yet Dr. Heimert could not explain — in either his Medtronic III reports or at trial — why eliminating aggregation and taking these three functions “off the table” did not affect his CPM in any way whatsoever. That Dr. Heimert's CPM analysis is impervious to change — even when his core premises like aggregation are eliminated — critically undermines his method's credibility. His economic output bears no real relationship to changed factual inputs. Cf. Treas. Reg. 1.482-5(b)(2)(H) (“The tested party's operating profit must first be adjusted to reflect all other allocations under section 482, other than adjustments pursuant to this section” (emphasis added)).

In truth, Dr. Heimert reaches the same results because he is still applying aggregation. In his testimony on remand, Dr. Heimert avoided calling his approach “aggregation,” but then repeatedly justified his analysis based on the concept of aggregation by other names.³¹ This approach is particularly unsettling given the factual and legal failings in Dr. Heimert's Medtronic I aggregation arguments. MDT I Pet'r Post-Trial Reply Brief at 129-34.

But more fundamentally, an expert must opine based on the facts of the case. Fed. R. Evid. 702(d). The established facts in this case include that aggregation is inappropriate, and that all transactions other than the MPROC Licenses are off the table. Dr. Heimert attempts to honor those facts in the breach, by playing a game of Taboo with the word “aggregation” while still clearly relying on the concept. That is both deeply unconvincing and legally impermissible. Cf. Boltar, L.L.C. v. Commissioner, 136 T.C. 326, 335 (2011) (“Expert opinions that disregard relevant facts affecting valuation or exaggerate value to incredible levels are rejected.”). Thus, Dr. Heimert's CPM not only retains the analytical flaws that caused this Court to reject it in Medtronic I, but also, on remand, fails further as a matter of evidence and basic logic. It now contradicts the undisputed factual record.

H. Alternatively, the Court Could Consider an Unspecified Method

1. Overview of Medtronic's Unspecified Method

Medtronic offers, in the alternative, an unspecified method for pricing the MPROC Licenses (the “Unspecified Method”) in response to the Court's remarks at the remand proceeding.³² See Treas. Reg. § 1.482-4(d)(1). This Unspecified Method results in a royalty rate for the MPROC Licenses that is two-to-five percentage points higher than under Medtronic's primary CUT method (24.3%-27.2% as compared to 22.2%).³³

This Unspecified Method addresses two issues that the Court identified:

• It incorporates aspects of both Respondent's CPM and Medtronic's CUT.

• It adjusts for the profitability of Medtronic's U.S. CRDM business in tax years 2005 and 2006 (relative to the lower profitability of Pacesetter in 1992 and of Dr. Heimert's companies). See Tr. (Judge Kerrigan) 3056:6-10.

At a high-level, step one of the Unspecified Method applies a modified version of Medtronic's CUT and also the arm's-length rate for the Trademark License to allocate profit to Medtronic US's R&D activities. Step two applies a modified version of Respondent's CPM to allocate profit to Medtronic PR's activities. Step two also reduces this return to Medtronic PR for profits allocated to Medtronic US and Med USA based on the arm's-length prices for component manufacturing and distribution. (The order of steps one and two do not change the results of the Unspecified Method). After allocating profit for tax years 2005 and 2006 based on these modified methods and established arm's-length prices, a portion of the Device and Leads “system” profit remains unallocated. Step three of the Unspecified Method allocates this remaining profit between Medtronic US and Medtronic PR based on relevant commercial and economic evidence in the record.

This Unspecified Method differs from the methods that each party has advanced so far. The parties' existing CUT and CPM methods each price a key activity (R&D or finished device manufacturing respectively) and then allocate the remaining profit to the other activity. For example, Medtronic's CUT establishes a royalty rate for the MPROC Licenses; allocates profit to Medtronic US based on that royalty rate; and then allocates the remaining profit to Medtronic PR. Similarly, Respondent's CPM prices Medtronic PR's finished device manufacturing activities using Dr. Heimert's ROA; allocates profit to Medtronic PR on that basis; and then allocates the remaining profit to Medtronic US.

By contrast, the Unspecified Method uses versions of both the CUT and CPM as starting points to price each of Medtronic PR's and Medtronic US's respective activities and then divides the remaining profit between the two entities using commercial and economic evidence. See M. McLaughlin Kirmil, Transfer Pricing Answer Book, eh. 5 at 5-43 (2020 Ed.) (“In practice, unspecified methods are often combinations of two or more specified methods, or modified versions of specified methods.”).

To illustrate the Unspecified Method with a simplified example, assume that the tax year 2005-06 system profit for Devices and Leads is $1,000. In step one, the royalty rate under the modified CUT method, along with the arm's-length payment under the Trademark License, allocates $350 of profit to Medtronic US. In step two, the modified CPM allocates $450 to Medtronic PR, which return is then reduced by $100 for the arm's-length return to components and distribution paid to Medtronic US and Med USA. These calculations thus allocate $450 to Medtronic's US entities, and $350 to Medtronic PR. That leaves $200 of unallocated system profit ($1,000 - $450 - $350 = $200), which is intended to be a proxy for Medtronic's higher profitability in tax years 2005-06. The $200 is then allocated between Medtronic US and Medtronic PR using evidence of how commercial parties transact at arm's length.

A key aspect of the Unspecified Method is addressing the higher profitability of Medtronic's Devices and Leads business in tax years 2005 and 2006, relative to the lower profitability of both Pacesetter in 1992 and of Dr. Heimert's companies in tax years 2005-06. Accordingly, the first two steps apply the modified CUT and CPM in a manner that intentionally does not address this profitability. By excluding any “profitability” or “profit potential” adjustment at steps one and two, the remaining unallocated profit at step three is, in effect, a proxy for Medtronic's relatively higher profitability.

The data necessary for the Unspecified Method are already in the record.³⁴

The remainder of this Part walks through the Unspecified Method. Medtronic first explains why and how each existing method is adjusted in steps one and two, and then Medtronic presents the allocation of remaining profit in step three. Finally, Medtronic analyzes the resulting profit splits between Medtronic PR and Medtronic US/Med USA.

2. Unspecified Method: Steps One and Two

(a) The Modified CUT Is the Starting Point to Allocate Profit to Medtronic US

As the starting point to price Medtronic US's R&D activities, the Unspecified Method applies Dr. Putnam's CUT analysis, with its 7% base royalty rate and all of Dr. Putnam's adjustments except for his profitability adjustment. Ex. 6105-P (Putnam Opening), Tbl. 3. Many of Dr. Putnam's adjustments are a range (e.g., 1% to 3% for know-how). The Unspecified Method conservatively adopts the high end of each range. Further, in this modified CUT, the Maximum Rate Clause is not used because it (and its 15% max rate) addresses profitability differences that are instead addressed by step three in the Unspecified Method. Thus, the royalty rate for step one's modified CUT is 17.3%:

Royalty Rate for Modified CUT

Base rate	7.0%
Portfolio access fee	1.8%
Cross license	1.0%
Know-how	3.0%
Cardiac/Neuro avg. sub-license	4.5%
Total Royalty Rate	17.3%

Note that Dr. Putnam's sub-license adjustment is different for Cardiac (5%) and Neuro (1%) because of the differing third-party patent licenses relevant to those areas. Ex. 6105-P (Putnam Opening), Tbl. 3. The 4.5% adjustment above is the revenue-weighted average of his Cardiac and Neuro adjustments.³⁵

This modified CUT royalty — plus the established royalty rate of 5.4% for the Trademark License — allocates $674,352,148 of profit to Medtronic US for the 2005 and 2006 tax years, as shown below.

Unspecified Method	Med PR Profit for TY 2005-06	Med US Profit for TY 2005-06
Device and Lead System Profit To Allocate	$3,333,823,544
Step 1: Modified CUT + Trademark License allocates returns to Med US	—	$674,352,148

The Pacesetter Agreement with adjustments (other than for profitability) provides a highly reliable starting point to allocate profit to Medtronic US's R&D activities for the reasons articulated above. See supra Part IV.C-F. It is Medtronic's position that Dr. Putnam's profitability adjustment to the base rate is also highly reliable, and that the Maximum Rate Clause also separately caps the maximum adjustment for profitability. See supra Part IV.D. Medtronic understands, however, the Court's questions about profitability, and the Unspecified Method is intended to address those questions with an alternative approach.

(b) The Modified CPM Is the Starting Point to Allocate Profit to Medtronic PR

As the starting point to allocate profit to Medtronic PR, the Unspecified Method applies a modified version of Dr. Heimert's CPM. The Unspecified Method modifies the CPM to address one of its many fundamental issues: that the book values that Dr. Heimert uses for Medtronic PR's operating assets — i.e., the “Assets” in Dr. Heimert's “Return on Assets” — are low.

These low operating asset values are one reason why Dr. Heimert's CPM is unreliable. Properly allocating profits using an ROA depends, among other things, on using appropriate values for Medtronic PR's operating assets. Treas. Reg. § 1.482-5(c)(2)(i) (“[T]he degree of comparability between the tested party and the uncontrolled taxpayer depends upon all the relevant facts and circumstances, including . . . the asset composition employed (including the nature and quantity of tangible assets, intangible assets and working capital).” (emphasis added)). The more depressed the operating asset values on Medtronic PR's balance sheet, the lower the return that the CPM allocates to Medtronic PR.

As explained below, Medtronic PR's operating asset values for tax years 2005 and 2006 are understated in Dr. Heimert's CPM. To address this flaw, the Unspecified Method modifies the CPM by making an upward adjustment to the operating asset values. This upward adjustment is informed by financial metrics for Dr. Heimert's own purportedly comparable companies. The following subsections (i): explain how one can reliably compare operating asset values of medical device companies using the metric of “asset intensity”; (ii) compare Medtronic PR's “asset intensity” to other medical device companies; (iii) explain why Medtronic PR's asset intensity (and, thus, the Medtronic PR book asset values used in Dr. Heimert's CPM) are so low relative to various benchmarks; (iv) adjust Medtronic PR's book asset values based on this “asset intensity” comparison; and (v) allocate profits to Medtronic PR, under step two of the Unspecified Method, using this adjusted asset base.

(i) “Asset Intensity” Allows for More Reliable Comparison of Asset Values

Absolute comparisons of the book value of operating assets may be unreliable when comparing companies with different revenue, volume of production, etc., because each company's scale of production generally affects the absolute quantity of operating assets on its balance sheet. For example, a medical device company with $250 million in revenue (Company A) will almost certainly have less in operating assets than a medical device company with $5 billion in revenue (Company B). To control for these revenue differences, a ratio — “asset intensity” — can measure the asset values on a firm's balance sheet relative to the firm 's revenue. By controlling for revenue differences, “asset intensity” ratio allows for a more “apples-to-apples” comparison between firms.

For example, if Company A with $250 million of revenue had $125 million of operating assets, then Company A's “asset intensity” would be 50% ($125 million of operating assets divided by $250 million of revenue). Similarly, if Company B with $5 billion of revenue had $2 billion of operating assets, then that firm's “asset intensity” would be 40% ($2 billion/$5 billion):

Financial Metric	Company A	Company B
Operating Assets	$125,000,000	$2,000,000,000
Revenue	$250,000,000	$5,000,000,000
Asset Intensity (Operating Assets/Revenue)	50%	40%

For these two firms, Company A, with fewer operating assets on an absolute basis, actually has a higher asset value compared to Company B, relative to revenue.

(ii) Medtronic PR's Asset Intensity Is Low as Compared to Dr. Heimert's Companies

Asset intensity is an important metric for comparing Medtronic PR's book asset values to those of other medical device companies. Comparable companies are likely to have similar ratios of assets to revenue (i.e., similar asset intensities); dissimilar companies are more likely to have dissimilar asset intensities. With this principle in mind, a review of the asset intensity ratios for Dr. Heimert's 14 companies and Medtronic PR establishes two key points:

• The asset intensity ratios for Dr. Heimert's 14 companies converge within a relatively narrow band, consonant with the idea that more comparable companies have similar asset intensities.

• Medtronic PR's asset intensity — as calculated using the figures from Dr. Heimert's CPM — is dramatically lower than for these other companies. This demonstrates that Medtronic PR's operating asset values, as used in Dr. Heimert's CPM, are grossly understated.

For this review, Medtronic first analyzes the asset intensity of the five (among Dr. Heimert's 14) companies that manufacture at least one implantable device. Ex. 6410-RD at 10. Dr. Heimert testified that he believed these five companies could be more comparable to Medtronic PR. Tr. (Heimert) 2469. The chart below presents asset intensities for these five companies and Medtronic PR.³⁶

Company	Asset Intensity TY 2005-06
Bard (C.R.) Inc.	46.1%
Orthofix International NV	52.3%
Stryker Corp.	47.4%
Wright Medical	80.2%
Zimmer Holdings Inc.	66.2%
Medtronic PR	13.3%

The median asset intensity for Dr. Heimert's five companies is 52.3%, compared to only 13.3% for Medtronic PR.

Similarly, Medtronic PR's asset intensity is not close to any of Dr. Heimert's 14 companies. Below, Medtronic PR's asset intensity is on the far left at 13.3%, while the 14 companies are on the far right.

As a cross-check, one can also consider the asset intensities for Medtronic's close competitors — Boston Scientific, Guidant and St. Jude:

Company	Asset Intensity TY 2005-06
Boston Scientific	42.1%
Guidant	61.4%
St. Jude	61.2%

These asset intensity figures reinforce that Medtronic PR's asset intensity is depressed. The lowest asset intensity among these companies — 42.1% — is more than three times Medtronic PR's 13.3%.

Finally, another cross-check is Pacesetter's asset intensity in 1994, prior to St. Jude's acquisition — 45.6% — which is within the same range.

(iii) Medtronic PR's Asset Intensity Is Low Because Dr. Heimert's Book Values for Medtronic PR Are Low

The book value of operating assets from Medtronic PR's balance sheet is significantly lower (relative to revenue) than these other medical device companies due to financial accounting conventions. The value of operating assets that Medtronic PR carries on its balance sheet has been depreciated over time and, thus, that balance-sheet value does not reflect fair market value of the assets. An asset on a balance sheet that depreciates for financial statement purposes on a “straight line” basis (i.e., in equal increments) over 20 years will be carried at half of its value by year 10 and have zero value by year 20. This remains true even if the asset retains economic and commercial value. Thus, the longer that Medtronic PR owned the depreciating assets on its balance sheet as of 2005, the lower the carrying value of those assets due to deprecation. See Ex. 6109-P (Hubbard Rebuttal) ¶ 56 (“The market value of a company often far exceeds its book value because the book value typically reflects the original cost of the company's assets, which may have been purchased many years in the past.”).

The IRS has recognized this depreciation effect as a potential CPM issue:

[T]he reliability [of PLIs like the ROA in the CPM] can be diminished if there are problems in using book values as a proxy for the fair market values of tangible assets. For example, a company may have facilities that show a very low book value because of depreciation but in fact are still substantially productive.

IRS APA Study Guide, 13 (2013), https://www.irs.gov/pub/irs-apa/apa study guide.pdf. See also N. J. Royce, The Influence of Depreciation of Assets on Management Decisions, 2T.2 J. Op. Res. Soc'y 471, 472 (1976) (Depreciation “affects the timing of profit . . . and affects both the numerator and denominator of the return on assets.”).

This accounting-driven phenomenon becomes especially apparent when one compares the asset intensity of firms before and after an acquisition. Before an acquisition, the asset values on the acquired company's balance sheet will likely be lower because the acquired company has been depreciating those assets, perhaps for a significant period of time, for financial statement purposes. Upon an acquisition, the acquirer generally must mark the acquired company's balance-sheet asset values to their current fair market value as part of a “purchase price allocation.” Tr. (Heimert) 2487:11-15 (“an acquisition company when they make that acquisition will be required to perform a purchase price allocation. So you break out the purchase price into various buckets. . . .”). Thus, because the book value of the acquired company's assets will generally increase due to the purchase price allocation, the acquired company's asset intensity will also increase (because those assets are the numerator in the asset intensity ratio). This is true even though nothing has changed the “true” value of the acquired assets.

The experts in this case referenced three acquisitions — Pacesetter, Enpath, and Guidant — that provide concrete examples of this effect.³⁷

• Pacesetter: Siemens sold Pacesetter to St. Jude in 1994. Pacesetter's pre-acquisition asset intensity before St. Jude's acquisition was 46%, while its post-acquisition asset intensity was 65%.

• Guidant: Boston Scientific acquired Guidant in 2006. Tr. (McCoy) 54:1420. Guidant's pre-acquisition asset intensity was 61% and post-acquisition asset intensity was 259%.

• Enpath: Greatbatch acquired Enpath in 2007, as discussed with Dr. Heimert on cross-exam. Tr. (Heimert) 2500:16-22. Enpath's pre-acquisition asset intensity was 41% and post-acquisition asset intensity was 120%.

The pre- and post-acquisition asset intensities of these companies differ so significantly because the book asset values were significantly depreciated preacquisition. Nothing else about the assets changed in the acquisition — the balance sheet reflected the same pool of operating assets before and after.

Dr. Hubbard recognized that this accounting phenomenon could distort the CPM — describing it as part of an example using the CPM to value hotels:

[The] CPM-like approach could introduce distortions to the estimate of the arm's length price due to variations in the accounting practices of the comparable hotels. For instance, the historical financial data of comparable hotels that were acquired recently would likely reflect the amortization of their respective purchase prices. In contrast, the historical financial data of comparable hotels that were acquired long ago likely would not reflect that amortization, if the purchase prices of the hotels acquired long ago have been fully amortized in prior years.

Ex. 6108-P (Hubbard Opening) ¶ 29. See Am. Bankr. Inst., Return on Assets So Useful . . . And So Misused (2001) (detailing similar accounting issues in the bankruptcy context).

As of 2005, Medtronic PR's accounting records reflected the historical cost of its assets, less accumulated depreciation. This significantly depressed the asset values on Medtronic PR's balance sheet as compared to Dr. Heimert's companies, which represented broader consolidated groups in which acquisitions would have occurred and balance sheets would have been adjusted for the fair market value of assets. The proof — of this accounting distortion as between Medtronic PR and Dr. Heimert's companies — is in the pudding: their asset intensity ratios diverge dramatically as shown above.

(iv) Medtronic PR's Asset Intensity Should Be Adjusted to Be Similar to Dr. Heimert's Companies

To address Medtronic PR's undervalued operating assets in Dr. Heimert's CPM, the Unspecified Method adjusts Medtronic PR's asset intensity ratio upward from 13.3% to 52.3%, matching the median asset intensity for Dr. Heimert's five-company subset. Cf. Treas. Reg. § 1.482-5(c)(2)(iv) (“In some cases, the assets of an uncontrolled comparable may need to be adjusted to achieve greater comparability between the tested party and the uncontrolled comparable.” (emphasis added)); Treas. Reg. § 1.482-1(d)(2). If the CPM treats Medtronic PR as comparable to these companies, then an asset intensity adjustment is appropriate to incrementally improve the comparability of their operating asset values.

By adjusting Medtronic PR's asset intensity to 52.3%, the value of Medtronic PR's operating assets increases as shown below.

Year	Medtronic PR Average Operating Assets in Dr. Heimert's CPM	Adjusted Averaoe Operating Assets With 52.3% Asset Intensity
TY 2005	$393,029,644	$1,401,712,258
TY 2006	$424,192,500	$1,853,316,656

For the myriad reasons discussed at length above and in Medtronic I, none of Dr. Heimert's 14 companies, including these five, are sufficiently comparable to Medtronic PR to use for benchmarking Medtronic PR's returns. See supra Part IV.G. 1. This remains true after this “asset intensity” adjustment, which only addresses a deficiency in how Medtronic PR's operating assets shown on its balance sheet are valued. For example, this adjustment does not address the fact — found by this Court in Medtronic I — that Medtronic PR's balance sheet also utterly fails to capture Medtronic PR's intangible value contributions. See Medtronic I at 114; infra Part IV.G.2.(b). Thus, this modified CPM is — like the modified CUT — only the starting point for allocating profit.

In addition, because the modified CPM uses the asset intensity and ROA for Dr. Heimert's five-company subset to allocate returns to Medtronic PR, Medtronic PR's returns are consistent with these five companies, each of which is materially less profitable than Medtronic's Device and Leads business, as shown below:

Company	Operating Profit Margin (Avg. TY 2003-05)
Bard (C.R.) Inc.	20.2%
Orthofix International NV	21.6%
Stryker Corp.	19.3%
Wright Medical	11.7%
Zimmer Holdings Inc.	29.7%
Medtronic Devices/Leads	51%

Thus, in even the modified CPM, the companies used to benchmark Medtronic PR's profit allocation are still fundamentally different from, and materially less profitable than, Medtronic PR and its broader business unit.

(v) The Modified CPM Is the Starting Point to Allocate Profit to Medtronic PR

After making this “asset intensity” adjustment, the Unspecified Method's modified CPM uses Dr. Heimert's ROA analysis for the five-company subset. Thus, Medtronic PR is allocated profit using a 41.3% ROA — i.e., the average of Dr. Heimert's 2005-06 ROAs for his five-company subset, Ex. 6410-RD at 10 — as applied to Medtronic PR's adjusted asset base. This results in the allocation of $1,344,326,942 in profit to Medtronic PR for tax years 2005 and 2006:

Unspecified Method	Med PR Profit for TV 2005-06	Med US Profit for TY 2005-06
Device and Lead System Profit To Allocate	$3,333,323,544
Step 1: Modified CUT + Trademark License allocates returns to Med US	—	$674,352,148
Step 2(a): Modified CPM allocates returns to Med PR	$1,344,326,942	—

Medtronic PR also must compensate Medtronic US for components and Med USA for distribution using the arm's-length prices for those functions. PFF ¶ 80. Thus, the returns to those functions are subtracted from Medtronic PR's return and added to Medtronic US/Med USA’s returns, as shown below.

The modified CPM, in providing sufficient return to allow Medtronic PR to compensate for components and distribution, demonstrates an additional flaw in Dr. Heimert's CPM: his CPM does not allocate sufficient returns to Medtronic PR to allow it to compensate for those functions. Economically, the returns allocated to Medtronic PR should be sufficient to allow Medtronic PR to pay for these functions at the arm's-length prices, the same way that Medtronic would pay for them from a third party. Cf. Treas. Reg. § 1.482-5(b)(2)(ii) (“The tested party's operating profit must first be adjusted to reflect all other allocations under section 482, other than adjustments pursuant to this section.”).

Medtronic PR's obligation to pay for these functions from its own returns is also consistent with Dr. Heimert's five companies. They incur expenses and realize revenue attributable to component and distribution functions, which they perform themselves or pay for third parties to perform. Ex. 6108 (Hubbard Opening), Ex. 3. Their overall returns also thus include embedded revenue and expenses for component manufacturing and distribution. Thus, paying the resulting returns to component/distribution from Medtronic PR overall return creates an apples-to-apples comparison between Medtronic PR and Dr. Heimert's companies.³⁸ By contrast, these companies do not possess trademarks as valuable as those held by Medtronic US, as reflected by their relatively low profitability. See Tr. (Putnam) 807:6-9 (“Medtronic's enhanced profitability and higher sales are the source of additional compensation from MPROC through a trademark license”); Tr. (Heimert) 2532:6-10 (agreeing that Medtronic PR is paying for Medtronic's brand through the Trademark License). Cf. MDT I Tr. (Heimert) 7300. Thus, returns to such valuable trademarks are not embedded in the returns for these five companies, so an ROA derived from these companies would not include returns for the Trademark License. Accordingly, the trademark return is not subtracted from Medtronic PR's benchmarked CPM return in the same manner. Instead, the return to the Trademark License is added to Medtronic US's return in step one above.

3. Unspecified Method: Step Three

The profit not allocated in steps one and two is shown below. This residual profit proxies for Medtronic's relatively higher profitability in tax years 2005-06.

Step three allocates the remaining profit to each of Medtronic US and Medtronic PR. Based upon the evidence in the record, including the testimony of Mr. McCoy, Dr. Pindyck's analysis, and Dr. Hubbard's analysis, licensors and licensees allocate no more than 35% of the returns to CRDM technology IP (including as licensed in the MPROC Licenses) to the licensor. See supra Part IV.F.1. Thus, given that evidence, step three allocates 65% of the remaining profit to Medtronic PR as licensee and 35% to Medtronic US as licensor.

In evaluating this division of the profits between Medtronic US and Medtronic PR, several other points are relevant:

• Technology IP is generally less of a unique value driver to Medtronic and its competitors, given IP cross-licensing in the industry. See supra Part IV.G.2.(c).

• Medtronic PR's activities are not generally outsourced by Medtronic or its competitors, demonstrating those activities' core importance. See supra Part IV.G.2.(c).

• The factual record extensively demonstrates Medtronic PR's fundamental importance to Medtronic's CRDM & Neuro businesses.

• Medtronic US was already compensated for higher profitability through the Trademark License.

Given these considerations — as well as the other evidence described above that demonstrates a 65/35 licensee/licensor split is already generally an upper limit on the licensor's return — the remaining profits could not, in any event, conceivably be divided more in the licensor's favor than a 50/50 split between Medtronic PR as licensee and Medtronic US as licensor.

4. Summary of the Unspecified Method's Results

The profit splits and royalty rates under the Unspecified Method with either the 65/35 or 50/50 allocation at step three are presented below.

(a) Resulting Profit Splits

To summarize the Unspecified Method with a 65/35 allocation at step three:

Under this version, approximately 51% of system profit is allocated to Medtronic US and Med USA, while 49% is allocated to Medtronic PR.

Then, with a 50/50 allocation of remaining profit at step three:

Under this version, approximately 57% of system profit is allocated to Medtronic US and Med USA, while 43% is allocated to Medtronic PR.

These profit splits — which allocate greater profit to Medtronic US relative to the primary CUT method — could be viewed as reasonable for the same reasons as the profit split under the primary CUT method. See supra Part IV.F.l.

(b) Resulting Royalty Rate for the MPROC Licenses

The royalty rates under the Unspecified Method and the primary CUT method are summarized below.

Method	Rate for MPROC Licenses
Primary CUT Method	22.2%
Unspecified Method: 65/35 Allocation	24.3%
Unspecified Method: 50/50 Allocation	27.2%

These results under the Unspecified Method both provide an alternative transfer pricing method as the Court requested, and, separately, provide further support for Medtronic's primary CUT method. The Unspecified Method isolates a proxy for additional profit and reasonably allocates it based on evidence. That allocation results in a royalty rate within range of the rate calculated under the primary CUT method. This demonstrates that the primary CUT method sufficiently addresses Medtronic's profitability in tax years 2005 and 2006. Whether the primary CUT method or either version of the Unspecified Method is applied, that the methods converge on royalty rates within a range should give comfort that these two methods result in an arm's-length allocation of profits.

Should the Court adopt either version of the Unspecified Method, it would need to conclude that: the Unspecified Method is the best method; the method is applied in accordance with Treas. Reg. § 1.482-1; and it takes into account “reasonable alternatives.” See Treas. Reg. § 1.482-4(d)(1). This best method analysis could follow the Court's best-method and “realistic alternatives” findings from Medtronic I. The Court would presumably select the Unspecified Method over the primary CUT method based on the profit-related questions that the Court raised at the remand proceeding.

Respectfully submitted,

Thomas V. Linguanti
Tax Court Bar No. LT0209
Morgan, Lewis & Bockius LLP
110 N. Wacker Drive
Chicago, IL 60606
Tel: (312) 324-1486
thomas.linguanti@morganlewis.com

Raj Madan
Tax Court Bar No. MR1190
Skadden, Arps, Slate, Meagher & Flom LLP
1440 New York Avenue, N.W.
Washington, D.C. 20005
Tel: (202) 371-7020
raj.madan@skadden.com

Date: November 19, 2021