Mylan Argues Patent Litigation Costs Not Subject to Capitalization 

MYLAN, INC. & SUBSIDIARIES, ET AL.,
Petitioners,
v.
COMMISSIONER OF INTERNAL REVENUE,
Respondent

PETITIONERS' SERIATIM REPLY BRIEF

---

UNITED STATES TAX COURT

Judge Patrick J. Urda

PETITIONER'S REPLY BRIEF

---

TABLE OF CONTENTS

I. INTRODUCTION

II. LAW AND ARGUMENT

III. CONCLUSION

Appendix A — Petitioner's Objection to Respondent's Proposed Findings of Facts

TABLE OF AUTHORITIES

Cases

Eli Lilly & Co. v. Medtronic, Inc., 496 U.S. 661 (1990)

INDOPCO, Inc. v. Commissioner, 503 U.S. 79(1992)

Statutes & Regulations

21 U.S.C. § 355

26 U.S.C. § 162

26 U.S.C. § 263

35 U.S.C. § 2716,

Treas. Reg. § 1.263(a)-4

Treas. Reg. § 1.263(a)-5

T.D. 9107, 2004-1 C.B. 447

Drug Price Competition and Patent Term Restoration Act, Pub. L. No. 98-417, 98 Stat. 1585 (1984)

---

I. INTRODUCTION¹

As explained in Petitioner's Opening Brief, every court to consider the issue has held that patent infringement litigation costs are deductible. Respondent concedes that, in general, most patent infringement litigation costs are deductible, but purports to discern a special capitalization rule that applies only to generic drug companies. Respondent's Answering Brief fails to provide any coherent defense of its proposed rule.

The sole legal rationale cited by Respondent is that Petitioner's patent litigation defense costs “facilitate” the acquisition or creation of an intangible asset, namely, FDA approval of a generic drug. Under the regulations, an expenditure facilitates “the acquisition or creation of an intangible (the transaction) if the amount is paid in the process of investigating or otherwise pursuing the transaction.” Treas. Reg. § 1.263(a)-4(e)(1). Thus, the Court must make two determinations: (1) it must identify a “transaction” that results in the acquisition or creation of an intangible asset; and (2) it must determine whether Petitioner's patent litigation defense costs “facilitate” that transaction (and hence the acquisition of an intangible).

Patent litigation is not a “transaction.” The resolution of a patent infringement claim does not result in the acquisition or creation of any intangible asset. The focus of the patent suit is on the property rights of the patent holder. A successful defense against a patent infringement suit does not grant a generic drug company any affirmative property right, but merely confirms that the patent holder does not have the right to prevent it from making and selling a generic drug that has received FDA approval. Thus, the costs of defending against a patent infringement claim do not facilitate the acquisition of any intangible asset, including FDA approval.

Although Respondent contends that Petitioner's patent litigation defense costs facilitated a transaction in which Petitioner acquired an intangible asset, at trial Respondent failed to articulate a cogent and consistent definition of the relevant transaction. Respondent's Answering Brief continues to dance around the issue. At times, Respondent defines the transaction broadly to include all steps necessary to bring a new generic drug to market:

Petitioner's objective was to bring its generic drugs to market at the earliest possible date. Every step taken by petitioner in furtherance of that objective . . . was taken as part of a “transaction” within the meaning of the regulation. Because petitioner could not practically unilaterally abandon a Hatch-Waxman suit, the “transaction” continued until petitioner had both obtained FDA approval and resolved the Hatch-Waxman suit.

Respondent's Answering Brief (“Resp. Br.”) at 125 (citations omitted). At other times, Respondent purports to narrow his definition by characterizing the defense of patent infringement cases as “one step in a series of steps carried out as a single plan to obtain an FDA-approved ANDA.” Resp. Br. at 159.

None of Respondent's various formulations comports with the section 263 regulations. Respondent's suggestion that the transaction is getting a new drug to market ignores the fact that the regulations equate “transaction” with the acquisition or creation of an intangible asset and identify specific categories of intangible assets subject to capitalization. Getting a drug to market is not a separate and distinct intangible asset. Getting a drug to market may involve numerous other separate transactions and processes, such as research and development activities; acquisition or modification of manufacturing facilities; and marketing studies to ascertain the potential demand.

Getting a drug to market also requires a license from the FDA.² The license or permit received from the FDA is an intangible asset subject to capitalization. But that does not justify capitalizing other costs of getting a drug to market into the FDA license. Indeed, the section 263 regulations explicitly recognize that the costs of getting a drug to market are deductible unless incurred directly in the acquisition of an intangible asset identified in the section 263 regulations. See Treas. Reg. § 1.263(a)-4(l), Example 7.

Respondent's alternative formulation defines FDA approval as the relevant transaction, but then broadens the scope of the definition by labeling patent litigation as “one step in a series of steps carried out as a single plan” to get FDA approval. Resp. Br. at 159.

Whether defining the transaction broadly as bringing a generic drug to market or more narrowly as obtaining FDA approval of an ANDA, Respondent's fundamental error is the same. Respondent tries to meld two separate and discrete processes — patent infringement litigation and the FDA approval process — into a single transaction in order to trigger the capitalization rules of section 263(a). The separate processes move on separate tracks, have separate timetables, and involve different parties and different property rights. See Ptr. Op. Br. at 56-60. The FDA process culminates in the acquisition of an intangible: a permit or license (in the form of FDA approval of an ANDA) that permits the applicant to make and sell drugs without violating food and drug laws.

The patent infringement litigation addresses the implications of what would occur following FDA approval. The court determines whether, if the generic drug company were to receive FDA approval, its manufacture and sale of the drug would infringe on a valid patent. The resolution of patent disputes occurs outside of, and independently of, the FDA approval process.³ A generic drug company may prevail in the patent infringement dispute but fail to receive FDA approval of its ANDA; conversely, it may lose the patent infringement litigation yet still receive FDA approval effective upon expiration of the patent. Any potential delay in the effective date of final FDA approval results from the court's determination that the patent holder has a separate property right to prevent the generic drug company from making and selling the product for a limited period of time.

Respondent tries to merge the FDA and patent processes by invoking provisions of the Hatch-Waxman Act. In particular, Respondent stresses the process by which a generic drug company files a paragraph IV certification, which results in an “artificial” act of infringement that the brand drug company or patent owner may use to file a patent infringement case against the generic drug company. Resp. Br. at 123, 140-41. According to Respondent, if a taxpayer makes a paragraph IV certification, any patent-related costs incurred must be capitalized as costs of the FDA approval. Resp. Br. at 216. In Respondent's eyes, the filing of a paragraph IV certification converts the FDA into a “patent enforcement agency.” Resp. Br. at 178-80, 187. Patent litigation ceases to be a stand-alone procedure that resolves disputes between private litigants over private rights to make, use, and sell a patented invention and, instead, becomes a mere “step” in “an integrated process” by which the generic drug company obtains a license from the FDA to make and sell the drug in the United States. See Resp. Br. at 120-21.

Respondent is incorrect. The Hatch-Waxman Act did not change the character of patent infringement litigation. The paragraph IV certification is purely procedural; its sole purpose is to accelerate the resolution of patent issues by permitting patent litigation to proceed prior to the manufacture and sale of the generic drug.⁴ Thus, the paragraph IV certification facilitates the resolution of patent issues; however, it does not facilitate the FDA process, does not make the FDA into a patent enforcement agency, and does not make patent adjudication a part of the FDA approval process. Ptr. Op. Br. at 56-60.

Having concocted an untenable interpretation of the Hatch-Waxman Act, Respondent proceeds to rely on that interpretation to try to justify a tax result that undermines the very objectives of that statute. Respondent would require generic companies to capitalize their patent infringement litigation costs and recover them over periods as long as 20-30 years (Resp. Br. at 210-13, 216), even though Respondent acknowledges that brand companies are entitled to an immediate deduction for their patent litigation costs. Resp. Br. at 133-34. Respondent's position tilts the playing field in favor of brand companies in a way that the drafters of the Hatch-Waxman Act could never have intended.

Petitioner's Reply Brief addresses three principal issues. First, Petitioner's Opening Brief identified multiple examples in the section 263 regulations that directly support the point that the Hatch-Waxman Act did not make patent litigation and FDA approval part of the same “transaction.” These examples demonstrate that costs incurred directly in one process do not facilitate an intangible acquired in another process, even where the two processes are so interrelated that neither would be undertaken without the other. In trying to explain away those examples, Respondent's Answering Brief misstates their facts and devises explanations that defy common sense.

Second, throughout this litigation Respondent has cited various procedural aspects of Hatch-Waxman as support for his “one transaction” theory. Respondent's Answering Brief demonstrates, however, that Respondent would treat FDA approval and patent litigation as part of the same transaction even when his perceived special rules and benefits attributed to Hatch-Waxman do not apply.

Finally, Respondent doubles down on support for his expert witnesses and lobs criticisms at Petitioner's experts. Respondent's criticisms employ misdirection to make up for the fact that Respondent failed to offer a single fact or expert witness with as much as one day of experience at the FDA.

II. LAW AND ARGUMENT

1. Respondent Mischaracterizes the Transaction.

a. Respondent conflates FDA approval with the larger transaction of bringing a pharmaceutical product to market.

The parties agree that FDA approval of an ANDA conveys a government permit or license described in Treas. Reg. § 1.263(a)-4(d)(5)(i), which requires capitalization of amounts paid to a governmental agency to obtain a license, permit, or similar right granted by the agency. Since Petitioner does not pay its patent litigation fees and costs to the FDA, such amounts are not subject to capitalization unless they “facilitate” the acquisition of a license or permit from the FDA. Treas. Reg. § 1.263(a)-4(b)(1)(v). For this purpose, a cost facilitates the creation or acquisition of an intangible if incurred “in the process of . . . pursuing the transaction.” Treas. Reg. § 1.263(a)-4(e)(1)(i).

The examples of legal fees that facilitate the acquisition of an intangible invariably describe costs incurred directly to acquire or create property. See Treas. Reg. § 1.263(a)-4(l), Example 1 (fees paid to counsel to prepare license application and represent the taxpayer in the licensing process), Example 2 (fees paid to counsel to negotiate franchise agreement), Example 3 (fees paid to counsel to draft covenant not to compete), and Example 9 (fees paid to counsel to obtain trademark and copyright protection). There is not a single example in which legal fees incurred in one process are required to be capitalized because they indirectly facilitate a separate process.

Respondent's capitalization position thus requires him to characterize patent infringement litigation as merely “one step in a series of steps carried out as a single plan to obtain an FDA-approved ANDA.” Resp. Br. at 159. The scope of this “single plan” apparently depends on which of four certifications a generic drug company submits with its ANDA. If the generic drug company submits a paragraph I, paragraph II, or paragraph III certification, Respondent agrees that resolution of patent issues is not part of the plan for obtaining FDA approval.⁵ See Resp. Br. at 216. If, however, the generic drug company submits a paragraph IV certification to its ANDA, Respondent contends that all patent dispute costs are transformed into transaction costs incurred in the “process of . . . pursuing” FDA approval. Id. Respondent would require capitalization even when the entry of final judgment in the patent litigation occurs years after the FDA has approved the taxpayer's ANDA and even if the majority of the costs relate to patents that were not listed in the Orange Book and thus were not subject to any paragraph IV certification. Resp. Br. at 205-07.

Respondent fails to distinguish between the generic drug company's ultimate goal — getting its product to market — and the component transaction of obtaining FDA approval of its generic drug product. See Resp. Br. at 125, 157. Of course, there is a sense in which one might view all steps necessary to achieve an ultimate goal as being part of the same “transaction.” But the section 263 regulations bar such an expansive interpretation by defining the “transaction” as the acquisition or creation of an intangible asset. Although FDA approval is only one of many transactions and processes involved in getting the drug to market, for purposes of applying the section 263 regulations it is a stand-alone transaction. Cf. Treas. Reg. § 1.263(a)-4(l), Example 7 (confirming that certain costs incurred in bringing a new pharmaceutical product to market are deductible).

Although a taxpayer must capitalize the cost of obtaining FDA approval (e.g., payment of an application fee to the FDA), the capitalized cost of an FDA approval letter does not include costs that are not germane to the FDA approval process. The fact that the FDA approval costs and the costs of other transactions have a common purpose — getting a generic drug to market — does not mean those costs are incurred in the same transaction. And it does not warrant pulling costs from separate processes into the cost of FDA approval merely because FDA approval is one prerequisite to the ultimate goal.⁶

Petitioner's Opening Brief discussed several examples in the section 263 regulations that illustrate these basic principles. Respondent's efforts to explain away these examples highlight the inconsistency between his interpretation of “transaction” in this case and the interpretation reflected in the regulations.

b. Respondent's incoherent analysis of the regulatory examples addressed in Petitioner's Opening Brief illustrates the flaws in his position.

i. The lease termination example (Treas. Reg. § 1.263(a)-4(e)(5), Example 2).

In this example, the taxpayer (X) wants to enter into a lease of new equipment (the Z lease), but cannot do so without first terminating the existing Y lease. Treas. Reg. § 1.263(a)-4(e)(5), Example 2. The lease termination example concludes that X may deduct the cost of terminating the Y lease and need not capitalize those costs into the cost of the Z lease. Id.

This example rebuts Respondent's broad definitions of “facilitate” and “transaction.” Respondent tries to skirt around this fact by making two arguments. First, Respondent characterizes the example as illustrating a special rule for contract termination costs that has no broader implications beyond its narrow facts. Resp. Br. at 160-61. But the point of the example is not whether, in general, contract termination costs are deductible; rather, the point of the example is that termination costs are deductible even when termination of the contract is a prerequisite to obtaining another intangible (such as a new contract).

The facts are clear that, but for its plan to enter into the new Z lease, X would not terminate the Y lease. Treas. Reg. § 1.263(a)-4(e)(5), Example 2. The termination of the Y lease is, to paraphrase Respondent, “one step in a series of steps carried out as a single plan to obtain” the Z lease. See Resp. Br. at 159. Yet the termination of the Y lease and the entry into the Z lease are separate transactions, and the cost of terminating the Y lease is not capitalized into the tax basis of the Z lease. Treas. Reg. § 1.263(a)-4(e)(5), Example 2.

As the example demonstrates, achieving a taxpayer's objective often depends on interdependent steps or transactions, but the existence of the larger goal does not mean that costs incurred in one step of the process facilitate an intangible acquired in another step. The Y lease and the Z lease are separate transactions involving different counterparties, different assets, and discrete legal rights and obligations. Treas. Reg. § 1.263(a)-4(e)(5), Example 2. Similarly, patent litigation and the FDA approval process involve different parties, different legal rights, and different substantive issues. See Ptr. Op. Br. at 56-60. The lease termination example refutes Respondent's rationale for trying to convert patent litigation into a component step that facilitates FDA approval. Treas. Reg. § 1.263(a)-4(e)(5), Example 2.

Second, Respondent infers that the requirement to terminate the Y lease “must have been imposed by X.” Resp. Br. at 163. From this inference, Respondent reasons that the lease termination example is distinguishable because X made a voluntary decision to terminate the existing Y lease, whereas a generic drug company that submits a paragraph IV certification has no choice but to defend against any patent litigation that ensues. See Resp. Br. at 162-63.

Respondent cites no authority in the text of the section 263 regulations or elsewhere to support the existence of such a distinction. Nor does Respondent make any effort to explain why such a rale would make any sense.

Moreover, Respondent's newly concocted distinction provides no basis for distinguishing the lease termination example. X may have made a voluntary decision to terminate the Y lease, but it could not terminate the Y lease without paying termination costs to Y. Treas. Reg. § 1.263(a)-4(e)(5), Example 2. A decision by Petitioner to submit a paragraph IV certification with an ANDA is every bit as voluntary as X's decision to terminate the Y lease. And Petitioner's decision to contest any ensuing patent infringement litigation is no less voluntary than X's decision to cut a check to Y. Costs incurred in terminating an existing lease are not capitalized into the basis of a new lease, and costs incurred in resolving patent disputes should not be capitalized into the basis of an approval received from the FDA.

ii. The loan example (Treas. Reg. § 1.263(a)-5(l), Example 2).

The loan example demonstrates that the costs of incurring a loan that is necessary to finance a stock purchase do not facilitate the stock purchase. Treas. Reg. § 1.263(a)-5(l), Example 2.

Respondent first asserts that the loan example is not relevant because it arises under Treas. Reg. § 1.263(a)-5, rather than Treas. Reg. § 1.263(a)-4. Resp. Br. at 164. The two provisions, however, employ virtually identical terminology, and they were promulgated on the same day pursuant to the same Treasury Decision. See Ptr. Op. Br. at 60-61; T.D. 9107, 2004-1 C.B. 447. Respondent cites no authority for giving “facilitate” and “transaction” different meanings in Treas. Reg. § 1.263(a)-4 than in Treas. Reg. § 1.263(a)-5, or for asserting that the examples illustrating the latter can shed no light on the former.

Second, Respondent tries to rewrite the facts of the loan example. He labels the loan and stock purchase as separate “transactions” because “Q may . . . use, or not use” the loan proceeds to close on the stock purchase. Resp. Br. at 165. But the text of the example states that “Q must issue new debt” to finance the purchase. Treas. Reg. § 1.263(a)-5(l), Example 2 (emphasis added). Even if Q might have used the loan proceeds for some other purpose, from Q's perspective the loan was “one step in a series of steps carried out as a single plan” to acquire stock. Respondent's rewrite of the loan example undermines his position in this case.

If anything, Respondent's rationalization of the loan example should apply with even greater force to Petitioner's patent litigation expenses. There may be, as Respondent maintains, some theoretical possibility that Q might close on the loan but not use the proceeds to purchase stock. Resp. Br. at 165. But the possibility that Petitioner may prevail in patent litigation and yet, for reasons beyond its control, fail to obtain FDA approval of its ANDA is more than theoretical. Ptr. Op. Br. at 57-58; PFF ¶¶ 22, 51. And Petitioner can lose a patent litigation suit and still receive FDA approval of an ANDA. Ptr. Op. Br. at 67; PFF ¶ 100; see also Resp. Br. at 118, 211-13. Even under Respondent's rewrite of the facts, there is a much stronger case for treating the loan as facilitating the stock purchase than there is for treating patent litigation as facilitating FDA approval.

iii. The tort litigation example (Treas. Reg. § 1.263(a)-5(l), Example 18).

The tort litigation example describes a case in which a company subject to extensive tort litigation claims files a bankruptcy petition. Treas. Reg. § 1.263(a)-5(l), Example 18. Confirmation of the bankruptcy plan requires resolution of the tort litigation claims. The example concludes that the cost of resolving the tort claims does not “facilitate” the approval of the bankruptcy reorganization. Id.

Respondent observes that the tort litigation example illustrates a specific rule. Resp. Br. at 168. That explanation is consistent with Respondent's continued insinuations that the specific rules and examples set forth in the regulations are arbitrary, ad hoc determinations that are untethered from the meaning of the terms they illustrate. That is incorrect. The lessons these rules and examples teach regarding the meanings of the terms “facilitate” and “transaction” are essential to the proper understanding of the regulations.

c. The only example cited by Respondent supports Petitioner's interpretation of the regulations.

None of the regulatory examples that illustrate the scope of “facilitate” and “transaction” support capitalizing patent litigation defense costs into the cost of obtaining FDA approval of a generic drug. It is not surprising, therefore, that Respondent dismisses the relevance of the regulatory examples. Resp. Br. at 158-59.

There is one example that Respondent embraces: Example 4 of Treas. Reg. § 1.263(a)-4(e)(5). Respondent criticizes Petitioner for failing to cite it. See Resp. Br. at 169. Yet this example rebuts Respondent's assertion that patent litigation costs are part of the FDA approval transaction.

In the example, U owns a majority of the stock of T. Treas. Reg. § 1.263(a)-4(e)(5), Example 4. U votes in favor of a perpetual extension of T's charter. M opposes the extension, and under state law, because the charter extension occurs over M's objection U must purchase M's stock. Litigation ensues over the valuation of M's shares. The example concludes that the costs incurred by U in the litigation are part of the cost of the shares that U must purchase from M. Id.

In trumpeting this example, Respondent overlooks how the example defines the transaction. U's purchase of the M shares results from, and is a requirement of, the decision to amend T's corporate charter. Treas. Reg. § 1.263(a)-4(e)(5), Example 4. Amending a corporate charter to make it perpetual fits within the classic definition of a reorganization described in section 368(a)(1)(F). Treating the costs as facilitating the reorganization of T would subject them to Treas. Reg. § 1.263(a)-5 rather than Treas. Reg. § 1.263(a)-4. The result would be to recast U's costs as capital contributions that facilitate the reorganization of T and would increase U's basis in all of its T shares, including those held prior to the reorganization. See Treas. Reg. § 1.263(a)-4(e)(5), Example 4.

Yet the example does not treat U's costs as incurred in the reorganization of T. Id. Instead, the example treats U's purchase of M's shares as a separate transaction and capitalizes U's costs into the basis of the shares purchased from M. Id. Thus, the stock appraisal example is in complete accord with the lease termination example, the loan example, and the tort litigation example.

The stock appraisal example contradicts Respondent's definition of the relevant transaction here. Treas. Reg. § 1.263(a)-4(e)(5), Example 4. The extension of T's corporate charter creates a mandatory obligation for U to buy M's shares, yet the example treats the mandatory stock purchase as a separate transaction from the charter extension.⁷ Id. Similarly, the resolution of the brand drug company's patent rights is a separate event from the FDA approval, even if the FDA's acceptance of an ANDA with a paragraph IV certification obligates the applicant to issue a notice letter to a patent owner that may result in patent litigation.

2. Respondent's Answers to Questions Posed in Petitioner's Opening Brief Demonstrate the Flaws and Inconsistencies in Respondent's Position.

Throughout the audit and litigation process, Respondent has continued to adjust his position while dodging questions about the implications of shoe-homing patent litigation into the FDA approval process. See, e.g., Tr. 616:15-617:6. Petitioner's Opening Brief posed some questions Respondent had avoided; Respondent's answers to those questions reveal the internal inconsistencies and flaws in his interpretation of the “facilitate” standard.

a. Respondent's position requiring capitalization of costs incurred in unsuccessful patent litigation produces bizarre and unfair results that are inconsistent with Respondent's stated rationale for requiring capitalization.

Respondent has sought to justify capitalization of the patent litigation costs at issue in this case by alleging that the receipt of FDA approval of an ANDA with a paragraph IV certification provides a generic drug company with benefits that it could not otherwise obtain. See, e.g., Resp. Br. at 155 (ascribing “immense value” to securing FDA approval of an ANDA subject to a paragraph IV certification). According to Respondent, obtaining these benefits is why generic drug companies file paragraph IV certifications and why they defend themselves against infringement litigation brought by brand companies under 35 U.S.C. § 271(e)(2). Resp. Br. at 190. And, according to Respondent, these timing benefits justify treating the defense of patent infringement suits as a step in the FDA process. Id.

A generic drug company that loses a patent infringement case receives no such benefits. Instead, it is in the same position that it would have occupied had it submitted a paragraph III certification rather than a paragraph IV certification. The only difference is the litigation costs it has incurred. Yet Respondent contends that even if a generic drug company loses a patent infringement case, it must capitalize all of its patent infringement defense costs into the basis of the FDA approval that it eventually will receive upon expiration of the disputed patents. Resp. Br. at 118, 211-13.

Respondent's position produces results inconsistent with the Supreme Court's decision in INDOPCO, Inc. v. Commissioner, 503 U.S. 79 (1992). Respondent's justification for these results confirms Petitioner's interpretation of the regulations.

i. Because costs incurred in unsuccessful patent infringement litigation produce no future benefits, Respondent's position produces unfair results that violate INDOPCO.

As explained in Petitioner's Opening Brief, prior to INDOPCO, many taxpayers had assumed that capitalization was limited to expenditures that create or enhance a separate and distinct asset. INDOPCO rejected that interpretation, holding that capitalization may be appropriate if an expenditure produces a benefit beyond the current tax year, even if not traceable to a specific asset. 503 U.S. at 88-89. The Court recognized that a long-term benefit will not result in capitalization in all cases, but it confirmed that such a benefit is essential to the capitalization requirement. Id. at 87. Cf. 26 U.S.C. § 263(a)(1) (basing the capitalization requirement on “permanent improvements or betterments made to increase the value of any property”).

Thus, even under the most expansive interpretation of INDOPCO, an expenditure that does not produce a benefit beyond the current year is not subject to capitalization. In promulgating the section 263 regulations, the Treasury Department sought to narrow and clarify INDOPCO by identifying cases in which the presence of long-term benefits will not automatically result in capitalization. Ptr. Op. Br. at 48-50. Yet Respondent invokes those very regulations to justify a result that expands the capitalization requirement beyond the limits contemplated by INDOPCO or section 263(a)(1).

Respondent's Exhibit 100-J illustrates the real-world implications of Respondent's position. In June 2015, Petitioner received FDA approval of acetylcysteine, but lost its patent case in September 2015. The patent will not expire until May 2026, which is the earliest Petitioner can obtain FDA approval of its ANDA. See Ex. 100-J, at LIT-0005250. Respondent's position would delay Petitioner's recovery of the patent litigation costs incurred in 2012-14 until May 2026, at which point Petitioner could begin deducting those costs over the ensuing 15 years. The earliest point at which Petitioner could obtain full recovery of these costs would be May 2041, a delay of almost 30 years in the recovery of expenditures that, by definition, produced no economic benefit to Petitioner. This result demonstrates the absurdity and unfairness of Respondent's position.

ii. Respondent's rationale for requiring capitalization of costs incurred in unsuccessful patent infringement litigation underscores the fundamental flaw in his litigating position.

Given the distortions created by requiring capitalization of costs incurred in unsuccessful patent litigation, Respondent's defense of his position is striking. Respondent contends that capitalization is justified because, whether a generic drug company wins or loses its infringement litigation, it eventually will receive FDA approval of an ANDA. Resp. Br. at 211-212.

That explanation is ironic, to say the least. Respondent repeatedly describes the “transaction” as the receipt of FDA approval prior to the expiration of potentially applicable patents. See, e.g., Resp. Br. at 120, 147, 212-13. A defeat in a patent infringement case means that, at least with respect to Orange Book-listed patents, the transaction as defined by Respondent has ended. Yet Respondent continues to insist on capitalization.

More importantly, Respondent's explanation confirms the central point of Petitioner's case: patent litigation defense costs do not facilitate FDA approval. The outcome of patent litigation does not affect whether a generic drug company will obtain FDA approval. At most, it affects when that approval will be effective. See Ptr. Op. Br. at 64. The FDA process resolves whether a generic product satisfies the FDA Standards for manufacture and sale within the United States. The patent litigation process determines whether the manufacturer of the brand-name drug has the right to prevent the generic drug company from manufacturing and selling a generic product that otherwise meets the FDA Standards.

As Respondent concedes, the patent infringement litigation affects when a company may begin to make and sell an FDA-approved generic drug; it has no impact on whether the company will receive FDA approval. Resp. Br. at 118.

b. Respondent's views on the relevance of the notice requirement, 30-month stay, and related procedures undermine his position that patent litigation is part of the FDA process.

Respondent invokes various procedural and timing rules imposed by Hatch-Waxman — the maintenance of the Orange Book, the notice requirement that follows from the filing of an ANDA with a paragraph IV certification, the 30-month stay, and the possibility of a 180-day exclusivity period — to assert that Hatch-Waxman merged patent litigation and FDA approval into a single unitary process. See, e.g., Resp. Br. at 122-25, 139-41.

The myriad exceptions to these procedural rules demonstrate the opposite. A generic drug company may incur patent litigation defense costs outside of the procedures described in Hatch-Waxman. If a patent owner or NDA holder fails to list an eligible patent in the Orange Book, there is no paragraph IV certification, no notice requirement, and no potential for a 30-month stay. If a patent owner or NDA holder lists a patent in the Orange Book, but fails to file an infringement suit within 45 days of receiving notice of a paragraph IV certification, there is no 30-month stay. If an NDA holder has patent rights that are ineligible for inclusion in the Orange Book (such as patents on a process), it may bring infringement litigation relating to those patents independently of any claims relating to Orange Book-listed patents.

In all of these circumstances, Respondent contends that generic drug companies must capitalize any patent litigation costs. See Resp. Br. at 204-09. For example, if a patent owner or NDA holder lists a patent in the Orange Book after a generic drug company has filed an ANDA and thereafter initiates infringement litigation, Respondent would require the generic drug company to capitalize its defense costs into the tax basis of its final FDA approval. See Resp. Br. at 206. Respondent also would require capitalization of costs attributable to the disputed patents that were ineligible for inclusion in the Orange Book and thus outside the scope of the Hatch-Waxman procedures. See Resp. Br. at 205.

In other words, while Respondent purports to rely on the Hatch-Waxman Act, his position is much broader.

c. Respondent's position regarding the deductibility of costs incurred after a generic drug company has received FDA approval illustrates the unfairness and inconsistency of his position.

From the inception of this case, Respondent has based his position on the assertion that resolution of any disputed patent issues is a prerequisite to the receipt of FDA approval of an ANDA with a paragraph IV certification. See Respondent's Pretrial Memorandum at 27, 30. Respondent continues to assert that the FDA “will not approve an ANDA that infringes on a valid patent.” Resp. Br. at 198; RPFF ¶ 239.

That assertion is untrue. The FDA may approve a drug that infringes on patents. This may occur, for example, if the drug purportedly infringes on a process patent that is not eligible for inclusion in the Orange Book. Even if a patent is listed in the Orange Book, the FDA may approve a generic drug that is later determined to have infringed on the listed patent. PFF ¶¶ 72, 96-97. Approval may occur because the brand drug company failed to file an infringement suit within 45 days of receiving notice of the paragraph IV certification, because the brand drug company added the patent to the Orange Book after the generic drug company filed its ANDA with the FDA, or because the 30-month stay has expired. See PFF ¶¶ 50, 91, 96. If any of these circumstances apply, the FDA may grant a final approval of an ANDA with a paragraph IV certification, even though the patent issues remain unresolved. Id. That has happened with respect to several generic drugs at issue in this case, including acetylcysteine and certain strengths of doxycycline hyclate. See Ex. 100-J, at LIT-0005250-53.

Respondent would require capitalization even in cases in which a generic drug company has launched “at risk,” reasoning that the FDA's final approval remains contingent until all patent litigation has ended. Resp. Br. at 146.⁸

The possibility that the outcome of a patent infringement suit may lead to the FDA's revocation of an approval does not mean that the FDA process is not complete when it issues final approval. The FDA may revoke its approval for a brand-name or generic drug based on a variety of reasons, including new concerns about safety or efficacy of the drug itself. See 21 U.S.C. § 355(e); Tr. (Talton) at 293:21-294:9. That possibility does not mean that the FDA approval process is ongoing; otherwise, a variety of costs incurred by the drug company — for example, costs incurred to maintain its manufacturing plant — could be subject to capitalization as costs incurred in obtaining an FDA approval received decades earlier.⁹

Requiring a generic drug company to capitalize costs incurred even after it has launched at risk highlights the uneven playing field that Respondent's position creates between generic companies and brand companies. Respondent tries to justify this uneven playing field by observing that a brand drug company is protecting a current income stream. Resp. Br. at 133-34. But the same is true for generic drug companies. When a generic drug company launches a product, it begins generating income from that product, and its patent litigation defense costs are protecting that income stream. Respondent ignores the obvious disparity that its position creates.

Finally, Respondent ignores the fact that commercial launch changes the remedies available to the brand drug company. Respondent's brief bases its proposed capitalization requirement on the filing of an infringement action under 35 U.S.C. § 271(e)(2). Resp. Br. at 203-04, 216. Respondent's position appears to be that costs incurred in defending infringement cases brought under other provisions of the patent laws are not subject to capitalization. Once a company launches a generic product, however, a brand drug company that believes the product infringes on its patents may bring a patent infringement suit under 35 U.S.C. § 271(a), (b), or (c). See Ex. 35-J, at LIT-0001481-82. According to Respondent, if the brand drug company files an action under 35 U.S.C. § 271(e)(2), the generic drug company must capitalize all defense costs, whenever incurred; if, however, the brand drug company files its action under 35 U.S.C. § 271(a), immediate deductibility appears to be the order of the day. Respondent's position would enable a brand drug company to control the generic drug company's deduction.

3. Only Petitioner Offered Credible Expert Testimony on the Role of Patent Litigation on the FDA Approval Process.

a. Respondent's brief offers no plausible argument for challenging the testimony of Petitioner's experts.

At trial, Petitioner introduced testimony from three experts, Sheldon Bradshaw, Tony Figg, and Dr. Nicholas Fleischer, two of whom worked at the FDA. The Court had the opportunity to hear their testimony and assess their credibility. Respondent offers no meaningful challenges to the substance of their testimony. Respondent makes three points that warrant a reply.

First, Respondent suggests that Mr. Bradshaw's testimony that the FDA does not engage in patent enforcement or patent regulation somehow lacks credibility because his term as Chief Counsel at the FDA did not coincide with the tax years at issue in this case. Resp. Br. at 179. Respondent's critique borders on the frivolous. Under Respondent's theory of the case, the enactment of Hatch-Waxman made the FDA into a “patent enforcement agency.” That alleged metamorphosis occurred in 1984, two decades before Mr. Bradshaw became Chief Counsel of the FDA. Respondent does not explain how the purported conversion of the FDA into a patent enforcer could have escaped the notice of the Chief Counsel of the FDA. Nor does Respondent cite any evidence that the FDA's role changed between Mr. Bradshaw's tenure at the FDA and the years at issue in this case.

Respondent's attempt at making something out of the timing of Mr. Bradshaw's tenure at the FDA is the direct result of his inability to rebut Mr. Bradshaw's testimony. Respondent had every opportunity to rebut that testimony by offering a fact or expert witness with actual FDA experience. He failed to do so.

Second, Respondent suggests that Mr. Figg testified on the ultimate issue in the case when he opined that patent litigation does not “facilitate” FDA approval. Resp. Br. at 191-92. A review of the context of that statement makes it clear, however, that Mr. Figg was using “facilitate” in its ordinary, everyday meaning, and not rendering any opinion on the definition of that term for purposes of the section 263 regulations. His simple point was that patent disputes may delay, but never expedite, receipt of FDA approval.

Third, Respondent alleges that prior work by Messrs. Bradshaw and Figg for Petitioner undermines the credibility of their testimony. Resp. Br. at 177-78. The suggestion hardly warrants a response. The Court had the opportunity to see Mr. Bradshaw and Mr. Figg and can make its own judgments as to their credibility.¹⁰

b. Respondent failed to offer any credible testimony to support his contentions regarding the role of patent litigation in the FDA process.

In his Answering Brief, Respondent attempts to rehabilitate Professor Thomas. Respondent asserts that Petitioner “appears to start and stop with Professor Thomas's choice of descriptive language” and “makes no effort to comprehend Professor Thomas's actual meaning.” Resp. Br. at 178. Later, Respondent reprimands Petitioner for “focus[ing] on Professor Thomas's terminology.” Id. at 179. In essence, Respondent tries to defend his own expert by contending Professor Thomas's testimony is not to be taken literally, and that the fault lies with Petitioner for focusing on his choice of words rather than divining his “actual meaning.”

Yet whatever concerns Respondent harbors over Professor Thomas's choice of words does not prevent him from doubling down on Professor Thomas's terminology. For example, Respondent disputes Petitioner's statement that Professor Thomas characterized the FDA as a patent enforcement agency. Resp. Br. at 178. Yet Respondent proceeds to explain why he believes that the FDA is a patent enforcement agency, and then dismisses testimony from a former FDA Chief Counsel that the FDA does not view itself as a patent enforcement agency. See Resp. Br. at 179.

Respondent employs a similar approach to Professor Thomas's description of the FDA as an agent for patent owners. Respondent criticizes Petitioner for focusing on that description, then suggests the issue that Respondent and his own expert raised is irrelevant, and finally concludes that, after all, Hatch-Waxman gave the FDA “a type of agency role.” Resp. Br. at 179-80. Once again, Respondent ignores Mr. Bradshaw's testimony that the FDA has never viewed itself as an agent for patent owners or other persons.

Respondent attributes Petitioner's criticisms to frustration at Professor Thomas's refusal to let himself be “corral[led]” in cross-examination over the meaning of “ministerial.” Resp. Br. at 183. But evasiveness is hardly a praiseworthy attribute of a witness. And Professor Thomas's unwillingness to explain the meaning of the term contrasts with Mr. Bradshaw's forthright and succinct testimony regarding the FDA's understanding of “ministerial.” Tr. (Bradshaw) at 354:21-356:4.

c. Respondent offers no rationale for the relevance of Dr. Mortimer's testimony.

At trial, Petitioner objected to the relevance of Dr. Mortimer's testimony. Tr. at 537:13-19. And Petitioner's Opening Brief noted that Respondent had failed to explain the relevance of Dr. Mortimer's testimony. Ptr. Op. Br. at 77-81.

Respondent's Answering Brief tries to demonstrate relevance. See Resp. Br. at 188-90. According to Respondent, Dr. Mortimer's testimony shows that the potential rewards of filing a paragraph IV certification exceed the applicant's patent litigation defense costs and therefore demonstrates that patent litigation defense is part of the transaction in which a generic drug company obtains FDA approval of an ANDA. Resp. Br. at 190.

That explanation is nonsense. Respondent cites nothing in the section 263 regulations that even remotely suggests that cost/benefit analysis plays a role in defining the scope of the transaction. And with good reason: the existence of such a belief is irrelevant to the definition of transaction. A rational taxpayer will never incur an expenditure unless it believes that the “potential reward . . . is well worth the expenditure.” Resp. Br. at 190. That does not mean that the expenditure and reward are part of the same transaction.

Once again, the examples contained in the section 263 regulations refute Respondent's reasoning. In the lease termination example, the taxpayer (X) would never enter into the new Z lease unless the benefits of the Z lease outweighed the cost of terminating the old Y lease. Treas. Reg. § 1.263(a)-4(e)(5), Example 2. Yet the termination of the Y lease and the Z lease are two different transactions. Id. Similarly, in the loan example, Q would never have bought the stock unless it expected the benefits of stock ownership to exceed the interest expense and other costs incurred in raising the necessary financing. Treas. Reg. § 1.263(a)-5(l), Example 2. Yet the loan and stock purchase remain two transactions. Id.

The examples demonstrate that Dr. Mortimer's cost/benefit comparisons are irrelevant to whether patent litigation and FDA approval are part of the same transaction. Respondent has failed to offer any coherent justification for the relevance of Dr. Mortimer's testimony.¹¹

III. CONCLUSION

As explained in Petitioner's Opening Brief, the costs of defending against patent infringement suits are properly deductible under longstanding tax principles. Respondent has sought to justify his departure from those principles by trying to shoe-horn patent litigation into the FDA approval process. As Petitioner has demonstrated, Respondent's position is incompatible with the consistent teaching of the section 263 regulations. Nothing in Hatch-Waxman justifies a different result here.

Petitioner respectfully requests that the Court conclude that the costs at issue are properly deductible under section 162.

Respectfully submitted,

PETITIONER

David J. Curtin
Morgan, Lewis & Bockius LLP
1111 Pennsylvania Avenue NW
Washington, D.C. 20004
Telephone No: (202) 373-6669
Tax Court Bar No: CD0395
david.curtin@morganlewis.com

James D. Bridgeman
Morgan, Lewis & Bockius LLP
1111 Pennsylvania Avenue NW
Washington, D.C. 20004
Telephone No: (202) 373-6628
Tax Court Bar No: BJ1764
james.bridgeman@morganlewis.com

William F. Nelson
Morgan, Lewis & Bockius LLP
1111 Pennsylvania Avenue NW
Washington, D.C. 20004
Telephone No: (202) 373-6782
Tax Court Bar No: NW0106
william.nelson@morganlewis.com

James G. Steele, III
Morgan, Lewis & Bockius
1111 Pennsylvania Avenue NW
Washington, D.C. 20004
Telephone No: (202) 739-5790
Tax Court Bar No: SJ2284
james.steele@morganlewis.com

Dated: June 27, 2019