United Therapeutics Corporation

• Court: United States Tax Court
• Decided: May 17, 2023
• Docket: 10210-21
• Status: Published

Opinion

United States Tax Court

160 T.C. No. 12

UNITED THERAPEUTICS CORPORATION, Petitioner

v.

COMMISSIONER OF INTERNAL REVENUE, Respondent

—————

Docket No. 10210-21. Filed May 17, 2023.

P is a biotechnology company. For each of the tax years 2011 through 2014, P claimed both the research credit under I.R.C. § 41 and the orphan drug credit under I.R.C. § 45C. Some of P’s expenses during those years qualified as both qualified clinical testing expenses under I.R.C. § 45C and qualified research expenses under I.R.C. § 41. For those expenses, P elected to claim the orphan drug credit under I.R.C. § 45C.

In determining the research credit for 2014, P elected to use the alternative simplified credit calculation under I.R.C. § 41(c)(5) and the reduced credit under I.R.C. § 280C(c)(3). When calculating the credit under I.R.C. § 41(c)(5), P excluded qualified clinical testing expenses from both its 2014 qualified research expenses and its average qualified research expenses for the three preceding tax years (2011 through 2013).

R audited P’s return and ultimately issued a Notice of Deficiency determining that P overstated its research credit for 2014 by improperly excluding from its computations the expenses P treated as qualified clinical testing expenses for 2011 through 2013.

Served 05/17/23 2

P timely petitioned our Court for redetermination. The case is before us for decision under Rule 122. R maintains that I.R.C. § 45C(c)(2) requires the result reflected in the Notice of Deficiency. P contends that, because of changes in I.R.C. § 41 since its original enactment, I.R.C. § 45C(c)(2) is a dead letter and has no application here.

Held: The text and structure of I.R.C. §§ 41 and 45C(c)(2) as they existed for 2014 require the result reflected in the Notice of Deficiency.

Thomas H. Dupree, Jr., Lucas C. Townsend, Saul Mezei, and John F. Craig III, for petitioner.

Brandon S. Cline, Anna L. Boning, and Naseem Jehan Khan, for respondent.

OPINION

TORO, Judge: In this deficiency case involving the tax year 2014, we consider a question of first impression: Must expenses that are used to determine the orphan drug credit under section 45C1 also be taken into account in determining certain elements of the research credit under section 41, with the result that a taxpayer claiming both credits receives a reduced research credit? The Commissioner of Internal Revenue maintains that section 45C(c)(2) requires this result. United Therapeutics Corporation (United Therapeutics) contends that section 45C(c)(2) is a dead letter (often referred to as deadwood) and has no application here.

Resolution of the case turns on a question of statutory interpretation. Sections 41 and 45C provide credits (originally enacted as temporary credits) that Congress extended and amended many times

1 Unless otherwise indicated, all statutory references are to the Internal

Revenue Code, Title 26 U.S.C. (I.R.C. or Code), in effect at all relevant times, all regulation references are to the Code of Federal Regulations, Title 26 (Treas. Reg.), in effect at all relevant times, and all Rule references are to the Tax Court Rules of Practice and Procedure. We round all monetary amounts to the nearest dollar. 3

over a number of years. The specific question before us is whether we should give effect to section 45C(c)(2) based on the ordinary meaning of its terms or whether we should ignore the provision altogether as a no- longer-effective rule that Congress neglected multiple times to remove from the Code. In interpreting clear statutory text, we normally do not assume that Congress made a mistake in drafting, and we certainly do not assume that it made the same mistake repeatedly. We see no reason to depart from that practice here. We therefore apply section 45C(c)(2) in accordance with its ordinary meaning and, as explained in more detail below, find in favor of the Commissioner.

Background

The parties submitted this case fully stipulated under Rule 122. The facts below are based on the pleadings and the parties’ Stipulation of Facts (including the Exhibits attached thereto). The parties’ Stipulation of Facts with accompanying Exhibits is incorporated herein by this reference.

United Therapeutics, a biotechnology company, is a Delaware public benefit corporation. When it timely filed the Petition in this case, United Therapeutics maintained principal places of business in Silver Spring, Maryland, and Durham, North Carolina.

United Therapeutics focuses primarily on the development and commercialization of unique products to address the unmet medical needs of patients with chronic and life-threatening conditions. During the 2014 tax year and the preceding three tax years (2011 through 2013), the company conducted research and development on potential treatments for pulmonary arterial hypertension (which ultimately leads to heart failure and death) and neuroblastoma (a rare form of brain cancer that predominantly affects children and infants), among other diseases.

For each of the tax years 2011 through 2014, United Therapeutics computed and claimed both the research credit under section 41 and the orphan drug credit under section 45C. Some of the company’s expenses during those years qualified both as qualified clinical testing expenses under section 45C and as qualified research expenses under section 41. With respect to those expenses, United Therapeutics elected to claim the orphan drug credit under section 45C.

In claiming its research credit for the 2014 tax year, United Therapeutics elected to use the alternative simplified credit calculation 4

under section 41(c)(5) and the reduced credit under section 280C(c)(3). 2 When calculating the credit under section 41(c)(5), the company excluded the expenses it had treated as qualified clinical testing expenses for purposes of section 45C from both its 2014 qualified research expenses and its average qualified research expenses for the three preceding tax years (2011 through 2013). In total for 2014, United Therapeutics claimed that it incurred $42,062,405 of qualified research expenses within the meaning of section 41. And it claimed that its average qualified research expenses for the three preceding tax years (2011 through 2013) were $22,605,492. Accordingly, it claimed an adjusted research credit of $2,799,129 for 2014. 3

The Commissioner audited United Therapeutics and ultimately issued a Notice of Deficiency. The Commissioner determined that United Therapeutics overstated its research credit by improperly excluding from its computations expenses it treated as qualified clinical testing expenses for tax years 2011 through 2013.

The parties have stipulated that if (as United Therapeutics contends) the company properly excluded its qualified clinical testing expenses from the calculation of its average qualified research expenses for the three years immediately preceding its tax year 2014 under section 41(c)(5), then its average qualified research expenses for those years (2011 through 2013) would be $22,605,492. Using that amount, United Therapeutics’ research credit under section 41 for tax year 2014 would be $2,799,129.

The parties have also stipulated that if (as the Commissioner contends) United Therapeutics must include its qualified clinical testing expenses for 2011 through 2013 in the calculation of its average qualified research expenses for those years, then its average qualified research expenses would be $49,257,244. Using that amount, United

2 Section 280C(c), which is not at issue, generally provides that a taxpayer’s deductions (or the amounts it would otherwise charge to its capital account) for qualified research expenses must be reduced according to the amount of the taxpayer’s research credit. I.R.C. § 280C(c)(1) and (2). Alternatively, a taxpayer may avoid these requirements by electing to reduce the amount of its research credit pursuant to section 280C(c)(3). Section 280C(b) provides similar rules with respect to qualified clinical testing expenses. 3 The amounts listed in the text differ from the amounts United Therapeutics

reported on its 2014 return because of adjustments agreed on by the parties. 5

Therapeutics’ research credit under section 41 for tax year 2014 would be $1,586,474.

Discussion

Section 38 permits taxpayers to claim a variety of business credits against federal income tax. Among those credits are the section 41 research credit and the section 45C orphan drug credit. United Therapeutics claimed both credits for the 2014 tax year, raising the question of how the two credits relate to each other. We begin with a brief discussion of the history of the two credits and how they interact.

I. The Research Credit

Congress introduced the “credit for increasing research activities” as part of the Economic Recovery Tax Act of 1981 (ERTA), Pub. L. No. 97-34, § 221(a), 95 Stat. 172, 241. “The credit was intended to ‘stimulate a higher rate of capital formation and to increase productivity’” by incentivizing taxpayers to undertake new research. See Hewlett- Packard Co. & Consol. Subs. v. Commissioner, 139 T.C. 255, 258–59 (2012) (first quoting S. Rep. No. 97-144, at 76–77 (1981), as reprinted in 1981-2 C.B. 412, 438–39; and then quoting H.R. Rep. No. 97-201, at 111 (1981), as reprinted in 1981-2 C.B. 352, 358), aff’d, 875 F.3d 494 (9th Cir. 2017). In general, the credit was equal to a percentage of the amount by which a taxpayer’s “qualified research expenses” for the credit year exceeded its average qualified research expenses for the three preceding tax years. ERTA § 221(a). Consistent with its name, therefore, the credit rewarded taxpayers who increased their research expenditures year over year. The credit was temporary and initially applied only to amounts paid or incurred after June 30, 1981, and before January 1, 1986. Id. § 221(d), 95 Stat. at 247.

In the years following its enactment, Congress extended the credit multiple times and, in at least one instance, allowed it to expire for a year before reinstating it prospectively. 4 When we say that Congress “extended the credit,” we mean that Congress made the benefit applicable to expenses incurred in a period not originally covered by the statute. See In re Grand Jury Subpoenas Duces Tecum, 78 F.3d 1307, 1311–12 (8th Cir. 1996) (holding that the Independent Counsel Reauthorization Act of 1987 was validly reenacted when “Congress

4 The Commissioner provided a helpful table summarizing the relevant

amendments, their enactment dates, and the effective dates covered by the relevant provisions, which we reproduce in the Appendix. See also Suppl. Br. for Resp’t 6. 6

passed [a public law] amend[ing] the sunset provision . . . of the 1987 Act by substituting the year 1994 for the year 1987”). Without these extensions, taxpayers would not have been entitled to any research credit in years following 1986 for incurring the types of expenses the credit is intended to incentivize. Congress finally made the research credit permanent (that is, it removed the provision that limited its application to specific time periods) in 2015. 5

Congress also modified the research credit a number of times after its initial enactment, including by moving the credit to different Code sections, changing the primary method of calculating the credit, and adding new methods for calculating the credit, each on more than one occasion. 6

The version of the research credit in effect for 2014, the tax year before us, was in section 41 (where it remains today). It was extended and amended earlier that year. It describes five methods for calculating the research credit, some that operate as alternatives to each other and some that work in tandem. 7 Each method is different from the others in various respects, but, consistent with the credit’s original design, nearly all the methods include a mechanism to reward taxpayers who

5For a discussion of the budgetary impact of legislation with permanent and temporary effects and the legislative process followed in adopting such legislation, see George K. Yin, Temporary-Effect Legislation, Political Accountability, and Fiscal Restraint, 84 N.Y.U. L. Rev. 174 (2009). See also id. at 199–202 (discussing the initial adoption and subsequent extensions of the research credit). For a broader discussion of temporary legislation, see Jacob E. Gersen, Temporary Legislation, 74 U. Chi. L. Rev. 247 (2007). 6 Significant amendments included, among others, those made by the Deficit Reduction Act of 1984 (DEFRA), Pub. L. No. 98-369, §§ 471, 474(i), 98 Stat. 494, 825– 26, 831–32, the Tax Reform Act of 1986 (TRA 1986), Pub. L. No. 99-514, § 231, 100 Stat. 2085, 2173–80, the Omnibus Budget Reconciliation Act of 1989 (OBRA 1989), Pub. L. No. 101-239, § 7110, 103 Stat. 2106, 2322–26, the Small Business Job Protection Act of 1996 (SBJPA), Pub. L. No. 104-188, § 1204, 110 Stat. 1755, 1773–75, and the Tax Relief and Health Care Act of 2006 (TRHCA), Pub. L. No. 109-432, §§ 104, 123(a), 120 Stat. 2922, 2934–36, 2944. 7 The five methods are (1) the incremental research credit under

section 41(a)(1); (2) the basic research credit under section 41(a)(2); (3) the credit related to energy research under section 41(a)(3); (4) the alternative incremental credit under section 41(c)(4); and (5) the alternative simplified credit under section 41(c)(5). The alternative incremental credit expired for taxable years beginning after December 31, 2008, but remains in the statute. I.R.C. § 41(h)(2). 7

increase their research activity in the current year relative to some earlier baseline defined by the statute.

The alternative simplified method—the method United Therapeutics used in 2014—is a good example. Like the original method for calculating the credit adopted in 1981, the alternative simplified method generally requires a taxpayer to compare its current year qualified research expenses to those it incurred during the three preceding years. See I.R.C. § 41(c)(5). In particular, section 41(c)(5)(A) provides that, subject to an exception not relevant here, a taxpayer’s credit under section 41(a)(1) equals 14% of the amount by which the taxpayer’s current year qualified research expenses exceed 50% of its average qualified research expenses for the three previous years. 8 So, a taxpayer that increases its qualified research expenses in the current year relative to the three-year period (i.e., the baseline) generally gets a larger credit. And the calculation of a taxpayer’s baseline expenses— i.e., the issue before us—can significantly affect the final credit amount.

II. The Orphan Drug Credit

In 1983, approximately two years after first establishing the research credit, Congress enacted the orphan drug credit as part of the Orphan Drug Act, Pub. L. No. 97-414, 96 Stat. 2049 (1983) (codified in relevant part as amended at 21 U.S.C. §§ 360aa–360ee and I.R.C. § 44H). “The Orphan Drug Act incentivizes pharmaceutical companies to develop ‘orphan drugs’—drugs for rare diseases that affect such a small portion of the population that there otherwise would be no financial incentive to research and develop treatments.” Catalyst Pharms., Inc. v. Becerra, 14 F.4th 1299, 1302 (11th Cir. 2021); see also Eagle Pharms., Inc. v. Azar, 952 F.3d 323, 325 (D.C. Cir. 2020). The orphan drug credit was one of the Orphan Drug Act’s financial incentives. The credit could be elected on an annual basis and rewarded taxpayers who, during a taxable year, incurred qualified clinical testing expenses in researching and developing drugs to treat rare diseases. I.R.C. § 44H(a), (b), (d)(5) (1983).

8 Expressed in the form of an equation, the formula for calculating the

alternative simplified credit is as follows: Current year credit = 14% × (X – (50% × ((Y1 + Y2 + Y3) / 3))). In the formula, X represents qualified research expenses for the credit year, and Y1, Y2, and Y3 represent qualified research expenses for the three years preceding the credit year. 8

Like the research credit, the orphan drug credit originally was temporary, with an expiration date of December 31, 1987. I.R.C. § 44H(e) (1983). Congress extended and modified the credit frequently over the years. 9 In at least one instance, Congress allowed the credit to expire before reinstating it prospectively, and it ultimately made the credit permanent in 1997. See infra Appendix; see also supra note 4.

The 2014 version of the credit was in section 45C (where it remains today). It generally permits taxpayers who incur qualified clinical testing expenses 10 and elect to apply section 45C to claim a credit equal to 50% of such expenses for the year, regardless of their expenditures in prior years. I.R.C. § 45C(a), (d)(4). This relatively straightforward computation makes the orphan drug credit a simpler (and more generous) benefit than the research credit, but with a potentially smaller pool of eligible expenses.

III. Interaction Between the Research Credit and the Orphan Drug Credit

As one might expect given the overlapping goals of the research credit and the orphan drug credit, expenses that qualify for one credit may also qualify for the other. Congress recognized this potential for overlap and addressed it in section 45C(c), which provides as follows: 11

Sec. 45C(c). Coordination with credit for increasing research expenditures.— (1) In general.—Except as provided in paragraph (2), any qualified clinical testing expenses for a taxable year to which an election under this section applies shall not be taken into account for purposes of determining the credit allowable under section 41 for such taxable year.

9 Significant amendments have included, among others, those made by the

DEFRA §§ 471, 474(g), 98 Stat. at 826, 831–32, the TRA 1986 §§ 232, 701(c)(2), 1275(c)(4), 1879(b), 100 Stat. at 2180, 2340, 2599, 2905–06, the SBJPA § 1205, 110 Stat. at 1775–76, and the Taxpayer Relief Act of 1997, Pub. L. No. 105-34, § 604, 111 Stat. 788, 863. 10 Qualified clinical testing expenses are defined with reference to qualified research expenses under section 41, subject to certain modifications. I.R.C. § 45C(b)(1). 11 Essentially the same text appeared in the original orphan drug credit at

section 44H(c) (1983). 9

(2) Expenses included in determining base period research expenses.—Any qualified clinical testing expenses for any taxable year which are qualified research expenses (within the meaning of section 41(b)) shall be taken into account in determining base period research expenses for purposes of applying section 41 to subsequent taxable years.

As in effect for 2014, neither section 45C(c)(2) nor any other Code provision defines the phrase “base period research expenses.”

The parties’ dispute turns on the meaning of this coordination rule. A simple example illustrates the stakes.

Assume that each year for four years (2011 through 2014) a taxpayer incurs $50 of expenses that qualify both as qualified research expenses and qualified clinical testing expenses. In each of the same years, the taxpayer also incurs $100 of additional expenses that qualify only as qualified research expenses. The second column of Table 1 below shows the result in the fourth year if, for all four years, the taxpayer claims only the research credit and uses the alternative simplified method to calculate the credit. The third column of Table 1 shows the result if the taxpayer claims only the orphan drug credit. 12

12 For simplicity’s sake, our discussion here does not take into account

section 280C, which operates to further limit a taxpayer’s credits in certain circumstances. Additionally, because the research credit is not elective, we recognize that a taxpayer may never be in position to claim the orphan drug credit alone. We therefore include this calculation for comparison purposes only. 10

Table 1: Research Credit Only or Orphan Drug Credit Only

Research Credit Orphan Drug Credit Only Only

Year 1 Qualified Research Expenses $150 NA 13

Year 2 Qualified Research Expenses 150 NA13

Year 3 Qualified Research Expenses 150 NA13

Years 1-3 Average Qualified 150 NA13 Research Expenses

Year 4 Qualified Research Expenses 150 NA 14

Year 4 Qualified Clinical Testing NA 15 $50 Expenses

Year 4 Research Credit (a) 10.5 16 -0-

Year 4 Orphan Drug Credit (b) -0- 25 17

Year 4 Total Credits (c) = (a) + (b) 10.5 25

In this example, claiming the more generous orphan drug credit results in a larger credit than claiming the research credit despite the smaller pool of eligible expenses.

The issue before us is how the research credit is computed when the taxpayer claims both the research credit and the orphan drug credit for each of the relevant years. Table 2 below shows the calculation of

13 Because the computation of the orphan drug credit turns only on qualified

clinical testing expenses incurred in the year the taxpayer elects to claim the credit, qualified research expenses incurred in other years are irrelevant to the computation of the credit. 14 Qualified research expenses that are also qualified clinical testing expenses are reflected in the “Year 4 Qualified Clinical Testing Expenses” line. 15 Because for purposes of this example the taxpayer elects not to claim the

orphan drug credit, qualified clinical testing expenses that are also qualified research expenses are taken into account in the “Year 4 Qualified Research Expenses” line. 16 Applying the formula described in note 8, the credit computation is as follows: 14% × (150 – (50% × 150)) = 10.5. 17 As discussed above, the orphan drug credit for the year is equal to 50% of

qualified clinical testing expenses incurred in the year: 50 × 50% = 25. 11

the research credit (again using the alternative simplified method) and the orphan drug credit during the fourth year in that scenario. The second column reflects United Therapeutics’ interpretation of section 45C(c)(2)—i.e., that qualified clinical testing expenses are not included in calculating qualified research expenses for the three preceding years. The third column reflects the Commissioner’s interpretation of the provision—i.e., that qualified clinical testing expenses are included in calculating qualified research expenses for the three preceding years because of section 45C(c)(2). 12

Table 2: Research Credit and Orphan Drug Credit

United Therapeutics’ Commissioner’s Position Position

Year 1 Qualified Research Expenses $100 18 $150 19

Year 2 Qualified Research Expenses 10018 15019

Year 3 Qualified Research Expenses 10018 15019

Years 1-3 Average Qualified Research 100 150 Expenses

Year 4 Qualified Research Expenses 100 100

Year 4 Qualified Clinical Testing 50 50 Expenses

Year 4 Research Credit (a) 7 20 3.5 21

Year 4 Orphan Drug Credit (b) 22 25 25

Year 4 Total Credits (c) = (a) + (b) 32 28.5

In this example, including the taxpayer’s qualified clinical testing expenses in its historical qualified research expenses (as the Commissioner maintains) reduces the research credit for 2014. But the taxpayer is still much better off claiming both credits than claiming the research credit alone (as shown in Table 1, claiming the research credit alone would result in a benefit of only $10.50, while claiming both credits would result in a benefit of $28.50 even under the Commissioner’s position). In the case before us, the difference between research credit computed under the Commissioner’s interpretation and the research

18 Expenses that are both qualified research expenses and qualified clinical

testing expenses ($50 each year) are ignored in computing the three-year average. 19 Expenses that are both qualified research expenses and qualified clinical testing expenses ($50 each year) are taken into account in computing the three-year average. 20 Applying the formula described in note 8, the credit computation is as

follows: 14% × (100 – (50% × 100)) = 7. 21 Applying the formula described in note 8, the credit computation is as follows: 14% × (100 – (50% × 150)) = 3.5. 22 See supra note 17. 13

credit computed under United Therapeutics’ interpretation for 2014 is $1,212,655.

IV. Application to United Therapeutics

Every year from 2011 through 2014, United Therapeutics, like the taxpayer in our example, incurred expenses that qualified as both qualified clinical testing expenses under section 45C(b) and qualified research expenses under section 41(b). And each year from 2011 to 2014, United Therapeutics elected to claim the orphan drug credit for all these expenses. In 2014, United Therapeutics excluded all qualified clinical testing expenses from its section 41 credit computations (including the calculation of the three-year average for 2011 through 2013). United Therapeutics argues that this approach is required by section 45C(c)(1) and that section 45C(c)(2) is inapplicable.

The Commissioner agrees that section 45C(c) provides the operative rule for coordinating the research credit and the orphan drug credit. He further agrees that section 45C(c)(1) requires qualified clinical testing expenses incurred in 2014 to be excluded when computing qualified research expenses for the credit year (i.e., 2014). But, unlike United Therapeutics, the Commissioner contends that section 45C(c)(2) requires qualified clinical testing expenses that are also qualified research expenses to be included in determining qualified research expenses for the three-year reference period described in section 41(c)(5)(A) (here, 2011 through 2013). For the reasons below, we agree with the Commissioner.

A. The Text and Structure of the Relevant Provisions Decide the Dispute Before Us.

Section 45C(c) provides that qualified clinical testing expenses must be excluded from all section 41 calculations, except that, under section 45C(c)(2), qualified clinical testing expenses that are also qualified research expenses must be included “in determining base period research expenses for purposes of applying section 41 to subsequent taxable years.” The parties agree that the qualified clinical testing expenses at issue are qualified research expenses. So the only question before us is whether “base period research expenses” are relevant to United Therapeutics’ research credit computation for 2014. As we show below, they are.

We begin with first principles. As the Supreme Court has explained, “[i]n statutory interpretation disputes, a court’s proper 14

starting point lies in a careful examination of the ordinary meaning and structure of the law itself.” Food Mktg. Inst. v. Argus Leader Media, 139 S. Ct. 2356, 2364 (2019) (citing Schindler Elevator Corp. v. United States ex rel. Kirk, 563 U.S. 401, 407 (2011)). And, when the statute does not define a term, “we ask what that term’s ‘ordinary, contemporary, common meaning’ was when Congress enacted” the relevant provision. Id. at 2362 (quoting Perrin v. United States, 444 U.S. 37, 42 (1979)). “The people who come before us are entitled, as well, to have independent judges exhaust ‘all the textual and structural clues’ bearing on that meaning.” Niz-Chavez v. Garland, 141 S. Ct. 1474, 1480 (2021) (quoting Wis. Cent. Ltd. v. United States, 138 S. Ct. 2067, 2074 (2018)). “When exhausting those clues enables [the Court] to resolve the interpretive question put to us,” id., “the sole function of the courts—at least where the disposition required by the text is not absurd—is to enforce it according to its terms,” Lamie v. U.S. Tr., 540 U.S. 526, 534 (2004) (quoting Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 6 (2000)).

The term “base period research expenses” is not defined in the 2014 version of section 45C or section 41. Accordingly, we look to the term’s ordinary meaning. And because there is no dispute that the expenses at issue in this case qualify as research expenses for purposes of section 41, we focus on the term “base period.”

The term “base period” has been defined consistently over time. Cf. BP P.L.C. v. Mayor of Balt., 141 S. Ct. 1532, 1537 (2021) (“Whether we look to the time of § 1447(d)’s adoption or amendment, a judicial ‘order’ meant then what it means today . . . .”). In general, it means “a period of time used as a standard of comparison in measuring changes . . . at other periods of time.” Base Period, Webster’s Encyclopedic Unabridged Dictionary of the English Language (1989); see also Base Period, Merriam-Webster, https://www.merriam-webster.com/ dictionary/base%20period (last visited May 9, 2023) (“[A] period of business or economic activity used as a basis or reference point . . . .”). 23 This meaning is consistent with how Congress has used the term “base

23 Combining the individual definitions of “base” and “period” produces the

same meaning. See, e.g., Base, The American Heritage Dictionary (5th ed. 2011) (“15. A line used as a reference for measurement or computations.”); Period, The American Heritage Dictionary (5th ed. 2011) (“1. An interval of time characterized by the occurrence of a certain condition, event, or phenomenon . . . .”); Base, Random House Webster’s College Dictionary (2001) (“7. a starting point or point of departure.”); Period, Random House Webster’s College Dictionary (2001) (“2. a specific division or portion of time.”). 15

period” in other contexts, including in a specific definition provided for limited purposes in section 41(e). 24 Thus, we interpret the term “base period research expenses” to mean research expenses that are incurred during the base period—i.e., the period of time section 41 employs as a standard of comparison (or as a baseline or reference point).

This interpretation is compatible with the structure of sections 45C and 41 and produces a nonabsurd result. It means that, for purposes of computing the research credit under section 41, the taxpayer that makes the election under section 45C must exclude qualified clinical testing expenses incurred in the year for which the election is made when calculating qualified research expenses for that year. See I.R.C. § 45C(c)(1). But the taxpayer must include qualified clinical testing expenses incurred during a reference period (i.e., a base period) prescribed by section 41 in its calculation of qualified research expenses for that reference period so long as those qualified clinical testing expenses also meet the definition of qualified research expenses. See I.R.C. § 45C(c)(2).

Take section 41(c)(5), which sets out the method for calculating the alternative simplified credit, as an example. As discussed above, that provision requires a taxpayer to compare its qualified research expenses during the current year to the expenses it incurred during “the 3 taxable years preceding the taxable year for which the credit is being determined.” I.R.C. § 41(c)(5)(A). The three-year period described in the provision is a period of time that is being “used as a standard of comparison in measuring changes.” In other words, the three-year period is a “base period” within the ordinary meaning of that phrase. And so, for a taxpayer that made the section 45C election for each of the three years included in the base period, section 45C(c)(2), interpreted according to its ordinary meaning, requires that the taxpayer’s qualified clinical testing expenses (that are also qualified research expenses) be

24 Section 41(e) describes a longstanding method of calculating the research

credit that is not at issue in this case: the basic research credit under section 41(a)(2). The basic research credit generally is calculated by using the amount by which a taxpayer’s payments for basic research during the year exceed its “qualified organization base period amount.” I.R.C. § 41(e)(1)(A). And the calculation of the qualified organization base period amount depends in part on certain categories of expenses incurred during the “base period,” I.R.C. § 41(e)(3)–(5), which is defined (for purposes of subsection (e)) to mean “the 3-taxable-year period ending with the taxable year immediately preceding the 1st taxable year of the taxpayer beginning after December 31, 1983,” I.R.C. § 41(e)(7)(B). Consistent with the definition we describe above, therefore, the three-year “base period” set out in section 41(e)(7)(B) is a period of time that the statute employs as a standard for comparison. 16

included when calculating qualified research expenses during that period. 25

This result follows from the text of the relevant provisions, and there is nothing unreasonable or illogical about it. Working together, the two statutory provisions (section 41(c)(5) and section 45C(c)(2)) require taxpayers who have elected the generous orphan drug credit for prior years to account for that prior-year benefit in calculating their research credit for the current year. One can conceive of many reasons why Congress might have taken such an approach, 26 which is identical to the approach both parties agree it adopted in the original orphan drug credit.

This analysis resolves the issue before us. See Lamie, 540 U.S. at 536 (stating that the Supreme Court will follow the plain meaning of a statute so long as it produces a result that is not absurd). But before concluding, we address certain arguments United Therapeutics raises.

B. United Therapeutics’ Contrary Arguments Are Not Persuasive.

United Therapeutics resists the straightforward reading of sections 41(c)(5) and 45C(c)(2) set out above based on two principal arguments. First, it maintains that the phrase “base period research expenses” should be read as a defined term. And, second, it argues that a consistency rule in section 41(c)(6)(A) trumps the coordination rule in section 45C(c)(2). Despite United Therapeutics’ skillful presentation, neither argument carries the day.

1. “Base Period Research Expenses” Is Not a Defined Term.

We turn first to the claim that the phrase “base period research expenses” should be read as a defined term. As we have already said,

25 Expressed in terms of the formula in note 8, our interpretation of section 45C(c)(2) requires qualified clinical testing expenses incurred in the credit year to be excluded from X. But qualified clinical testing expenses incurred in the three years preceding the credit year that also are qualified research expenses must be included in Y1, Y2, and Y3, and must be taken into account in the three-year average against which X is compared if the taxpayer claimed the orphan drug credit in years 1, 2, and 3. 26 Concerns about the cost of the research credit would be one example. See

infra note 43. 17

sections 41 and 45C as in effect for 2014 do not define that phrase. Nor does any other provision of the Code in effect for 2014. Why then does United Therapeutics contend it is a defined term?

United Therapeutics’ claim rests on a prior version of the research credit provision. Specifically, when Congress first adopted the research credit in 1981, its computation required the calculation of “base period research expenses.” 27 That term was defined in then section 44F(c)(1). 28 When Congress first adopted the orphan drug credit in 1983, it used the same phrase—“base period research expenses”—in section 44H(c)(2). This, United Therapeutics argues, demonstrates that the phrase “base period research expenses” as now used in section 45C(c)(2) must have the defined meaning provided by old section 44F(c)(1). The argument fails for several reasons.

a. Predecessor Statutes May Not Be Used to Manufacture Ambiguity.

To begin, we are not interpreting either the research credit or the orphan drug credit provisions as each existed in 1981 and 1983, respectively. Those provisions would not entitle United Therapeutics to the research credit in 2014 because on their face they applied only to expenses incurred long before 2014 and offered no credits whatever for 2014. See ERTA § 221(d)(1) (“The amendments made by this section shall apply to amounts paid or incurred after June 30, 1981, and before January 1, 1986.”); I.R.C. § 44H(e) (1983) (“Termination.—This section shall not apply to any amount paid or incurred after December 31, 1987.”). Instead, the provisions at issue here are section 41 and section 45C(c)(2) as they read in 2014. And by then Congress had removed from the Code the definition of the term “base period research expenses.” OBRA 1989 § 7110(b), 103 Stat. at 2323–25.

As the Supreme Court has explained, “[t]he starting point in discerning congressional intent is the existing statutory text, . . . and not the predecessor statutes.” Lamie, 540 U.S. at 534 (emphasis added).

27 The original research credit was calculated using the amount by which a taxpayer’s qualified research expenses for the tax year exceeded its “base period research expenses.” I.R.C. § 44F(a) (1981). 28 Section 44F(c)(1) (1981) provided in part as follows: “For purposes of this

section . . . [t]he term ‘base period research expenses’ means the average of the qualified research expenses for each year in the base period.” And the base period was “the 3 taxable years immediately preceding the taxable year for which the determination is being made.” Id. para. (2)(A). 18

We interpret undefined terms in the existing text in accordance with their ordinary meaning at the time Congress adopted them. Niz- Chavez, 141 S. Ct. at 1480. And when that meaning is clear and produces a nonabsurd result, our analysis is finished. See Lamie, 540 U.S. at 534.

Here, the adopting time is either the last time (before 2014) Congress made relevant substantive changes to the orphan drug credit or the time Congress extended the research credit to apply to expenses incurred in 2014. As to the first option, one possible choice is 1996, the year when Congress reinstated the orphan drug credit, moved it, and made it subject to the limitations applicable to general business credits. See infra pp. 24–25. A second possible choice is 1997, when Congress made the credit permanent (that is, applicable to qualified clinical testing expenses incurred in subsequent years, including 2014). As to the second option, the relevant time is 2014, the year when Congress made the research credit applicable to qualified research expenses incurred in 2014. But regardless of which option is chosen, by the relevant time, the definition of the term “base period research expenses” provided in old section 44F(c)(1) (and later in old section 30(c)(1) and section 41(c)(1), see the research credit history discussed in note 33 below) had been missing from the Code for seven years at the very least. 29 Accordingly, United Therapeutics’ argument that the existing text of section 45C(c)(2) somehow still incorporates the old definition is incorrect. 30

29 We point to these alternate timeframes because the coordination rule of section 45C(c)(2) could be viewed either as a limiting condition on the orphan drug credit (i.e., a taxpayer electing to claim the more generous benefits of the orphan drug credit must in effect accept a haircut to its otherwise available research credit) or as an inherent condition of the research credit (i.e., the research credit is calculated a certain way when a taxpayer meets a specific condition, namely, that it elected to claim the orphan drug credit during a year included in the base period). We need not decide here which of these alternatives is the proper one as the outcome in this case is the same under either. 30 United Therapeutics also invokes Treasury Regulation § 1.41-3A in support

of its position. Specifically, it argues both (1) that the regulation confirms “base period research expenses” is a concept applicable only to years before 1990 and (2) that the 2001 redesignation of the regulation reflects agreement by the Department of the Treasury (Treasury) and the Internal Revenue Service (IRS) that the concept no longer applies. But, as United Therapeutics concedes, the regulation says on its face that it does not apply for taxable years after 1989. See Treas. Reg. § 1.41-1(b); T.D. 8930, 2001-1 C.B. 433, 443, 66 Fed. Reg 280, 289 (Jan. 3, 2001). If predecessor statutes do 19

The analysis above faithfully follows the Supreme Court’s analysis in Lamie, 540 U.S. 526, where the Court considered a question arising under the Bankruptcy Code. In Lamie, a bankruptcy attorney had sought compensation under section 330(a)(1) of the Bankruptcy Code, 11 U.S.C., which governs court awards of professional fees. His application was denied, and a challenge followed. The attorney’s argument turned on the text of 11 U.S.C. § 330(a) before and after an amendment made by the Bankruptcy Reform Act of 1994 (1994 Act), 108 Stat. 4106. 31

The Supreme Court described the attorney’s argument as follows:

[The debtor’s attorney] argues that the existing statutory text is ambiguous . . . . He makes the case for

not cast doubt on the meaning of an existing statute’s text, see Lamie, 540 U.S. at 534, then we fail to see how a predecessor regulation could do so. Moreover, for years after 1990, Treasury and the IRS simply have not spoken regarding the meaning of “base period research expenses.” The 2001 redesignation of the regulation was simply an acknowledgment that the research credit had been amended. Silence by Treasury and the IRS is no concession as to the nature of the amended statute. In other words, administrative confirmation that a regulation interpreting a predecessor statute applies to the period the predecessor statute was in effect does not constrain future interpretations of another statute. 31 Before the 1994 Act, 11 U.S.C. § 330(a) read as follows (emphasis added to highlight text later deleted): (a) After notice to any parties in interest and to the United States trustee and a hearing, and subject to sections 326, 328, and 329 of this title, the court may award to a trustee, to an examiner, to a professional person employed under section 327 or 1103 of this title, or to the debtor’s attorney— (1) reasonable compensation for actual, necessary services rendered by such trustee, examiner, professional person, or attorney . . . and by any paraprofessional persons employed by such trustee, professional person, or attorney . . . ; and (2) reimbursement for actual, necessary expenses. Pursuant to the 1994 Act, § 224(b), 108 Stat. at 4130, 11 U.S.C. § 330(a)(1) was amended to read as follows: (a)(1) After notice to the parties in interest and the United States Trustee and a hearing, and subject to sections 326, 328, and 329, the court may award to a trustee, an examiner, a professional person employed under section 327 or 1103— (A) reasonable compensation for actual, necessary services rendered by the trustee, examiner, professional person, or attorney and by any paraprofessional person employed by any such person; and (B) reimbursement for actual, necessary expenses. 20

ambiguity, for the most part, by comparing the present statute with its predecessor. Thus, he says the statute is ambiguous because subsection (A)’s “attorney” is “facially irreconcilable” with the section’s first part since

“[e]ither Congress inadvertently omitted the ‘debtor’s attorney’ from the ‘payees’ list, on which the court of appeals relied, or it inadvertently retained the reference to the attorney in the latter, ‘payees’ list.” Brief for Petitioner 17.

Similarly, with respect to the missing conjunction “or” he says,

“[t]here is no apparent reason, other than a drafting error, that Congress would have rewritten the statute to produce a grammatically incorrect provision.” Ibid.

This is the analysis followed by the Courts of Appeals that hold the statute is ambiguous. . . . One determines ambiguity, under this contention, by relying on the grammatical soundness of the prior statute. That contention is wrong.

Lamie, 540 U.S. at 533–34.

The Court went on to observe:

The starting point in discerning congressional intent is the existing statutory text, see Hughes Aircraft Co. v. Jacobson, 525 U.S. 432, 438 (1999), and not the predecessor statutes. It is well established that “when the statute’s language is plain, the sole function of the courts—at least where the disposition required by the text is not absurd— is to enforce it according to its terms.” Hartford Underwriters Ins. Co. v. Union Planters Bank, N. A., 530 U.S. 1, 6 (2000) (internal quotation marks omitted) (quoting United States v. Ron Pair Enterprises, Inc., 489 U.S. 235, 241 (1989), in turn quoting Caminetti v. United States, 242 U.S. 470, 485 (1917)). So we begin with the present statute. 21

Id. at 534. And turning to that “present statute,” the Court noted: “The statute is awkward, and even ungrammatical; but that does not make it ambiguous on the point at issue.” Id.

The “present” statutory provisions before us (those in effect for 2014) are not in the least bit “awkward” or “ungrammatical.” In these circumstances, there is even less reason than in Lamie to consult predecessor versions of the statute.

In short, United Therapeutics invites us to reject the ordinary (not to mention straightforward and nonabsurd) meaning of an existing statute in favor of a predecessor definition that Congress removed from the Code in 1989. Seeing nothing in the existing statute’s text that authorizes such a rejection, we decline.

b. Even the Predecessor Statutes Do Not Require United Therapeutics’ Preferred Result.

As a further point, we are not persuaded that United Therapeutics’ argument works even on its own terms. That is, even if we were to conduct the relevant statutory analysis as of 1983, the time the orphan drug credit and the coordination rule at issue here were first adopted, we would not be sure that Congress used the phrase “base period research expenses” as a defined term. At that time, section 44H(c)(2) (the predecessor of section 45C(c)(2)) read as follows:

Expenses included in determining base period research expenses.—Any qualified clinical testing expenses for any taxable year which are qualified research expenses (within the meaning of section 44F(b)) shall be taken into account in determining base period research expenses for purposes of applying section 44F to subsequent taxable years.

Note that, when addressing “qualified research expenses,” Congress was careful to indicate that it meant such expenses “within the meaning of section 44F(b).” But when it addressed “base period research expenses,” Congress did not direct the reader to the specific definition in section 44F(c)(1). Courts presume that when Congress includes certain language in one provision but omits it in another, the inclusion and exclusion are intentional. See Loughrin v. United States, 573 U.S. 351, 358 (2014) (“We have often noted that when ‘Congress includes particular language in one section of a statute but omits it in another’— let alone in the very next provision—this Court ‘presume[s]’ that Congress intended a difference in meaning.” (quoting Russello v. United 22

States, 464 U.S. 16, 23 (1983))); see also Henson v. Santander Consumer USA Inc., 137 S. Ct. 1718, 1723 (2017) (same). All the more so when the relevant language is missing in the very same sentence. 32 See Loughrin, 573 U.S. at 358. Thus, textual clues from 1983 support the view that the phrase “base period research expenses” should be given its ordinary meaning, rather than a special, defined, meaning.

c. Other Principles Refute United Therapeutics’ Position. 33

To complicate matters further for United Therapeutics’ position, repeals by implication are disfavored. See Posadas v. Nat’l City Bank of N.Y., 296 U.S. 497, 503 (1936) (“Where there are two acts upon the same subject, effect should be given to both if possible.”); see also id. (discussing the standard for implied repeals); Lockhart v. United States, 546 U.S. 142, 149 (2005) (Scalia, J., concurring) (same). When, in 1989, Congress amended the research credit to delete the definition of “base period research expenses,” it left the same phrase in section 28(c)(2) (the predecessor of section 45C(c)(2)) untouched. United Therapeutics maintains that this congressional action rendered section 28(c)(2) inapplicable. But it is not clear to us why section 28(c)(2) should be interpreted as having been left with no work to do (that is, as having been effectively repealed by the changes in the research credit) since 1989 when, as we discuss above, it is not difficult at all to apply the text

32Note also that the definition provided in section 44F(c)(1) explicitly states that the definition is provided “[f]or purposes of this section.” Given that limiting phrase, one would expect Congress to tell us if it wished to give an undefined term in another section the same meaning. 33 Summarizing a few changes Congress made to the research and orphan drug credits between 1983 and 1989 helps provide context for the discussion that follows. In 1984, Congress reorganized the credits by “group[ing them] together in [a] more logical order.” DEFRA § 471. The orphan drug credit (previously found in section 44H) was moved to new section 28, and the research credit (previously found in section 44F) was moved to new section 30. Id. § 471(c), 98 Stat. at 826. Then, in 1986, the research credit was moved yet again, this time to section 41 (where it remains today). TRA 1986 § 231(d)(2), 100 Stat. at 2178. One reason for the change was to treat the research credit in the same manner as other business credits. Id. § 231(d)(1), 100 Stat. at 2178. With these changes, the definition of the term “base period research expenses” came to be found in section 41(c)(1). Then, in 1989, Congress amended section 41(c)(1) in its entirety, which resulted in the definition of the term “base period research expenses” (previously included in section 41(c)(1)) being removed from the Code altogether. But Congress left the phrase “base period research expenses” in section 28(c)(2) (the predecessor of section 45C(c)(2)) untouched. 23

of that section and its successors to the amended text of section 41. 34 United Therapeutics says that there is a difference between a provision’s having been made inapplicable and implied repeal. On the facts of this case, we are unable to see what that distinction would be. 35

Even if we were to overlook the law’s aversion to implied repeals, United Therapeutics’ position runs afoul of another “‘cardinal principle’ of interpretation.” See Loughrin, 573 U.S. at 358 (quoting Williams v. Taylor, 529 U.S. 362, 404 (2000)). In reading the Code as it applied for 2014, we “must give effect, if possible, to [its] every clause and word.” 36 Id. (quoting Williams, 529 U.S. at 404); Advoc. Health Care Network v. Stapleton, 581 U.S. 468, 478 (2017) (same); see also Sutherland v. Commissioner, 155 T.C. 95, 104 (2020). Our interpretation of section 45C(c)(2) follows this principle. United Therapeutics, on the other hand, reads section 45C(c)(2) as a dead letter. The Code’s text and structure do not support, let alone require, such a reading.

34 That Congress not only left section 28(c)(2) intact in 1989, but also renumbered it later when it moved the orphan drug credit to section 45C, see infra pp. 24–25, further undercuts the view that the provision was impliedly repealed. 35 United Therapeutics also faults the Commissioner for not pointing to “[any] evidence, let alone clear evidence, indicating that Congress intended to amend the limited exception set forth in section 45C(c)(2) to apply to the new and different section 41 research credit when Congress overhauled section 41 in 1989.” Pet’r’s Answering Br. 13. But there was no need for Congress to amend section 28(c)(2) (the predecessor of section 45C(c)(2)) or section 45C(c)(2) itself to apply to changes in the research credit. The existing text, which (as we have explained) did not use a defined term, is sufficiently broad to cover new methods of determining the research credit. This fully explains why Congress both left the provision in the statute in 1989 and did not change it thereafter, including in 2006 when it adopted the alternative simplified method that United Therapeutics used in 2014. See also infra pp. 27–29. 36 As the Supreme Court has maintained for nearly 150 years, we are not at liberty . . . to deny effect to a part of a statute. No rule of statutory construction has been more definitely stated or more often repeated than the cardinal rule that “significance and effect shall, if possible, be accorded to every word. As early as in Bacon’s Abridgment, sect. 2, it was said that ‘a statute ought, upon the whole, to be so construed that, if it can be prevented, no clause, sentence, or word shall be superfluous, void, or insignificant.’” Petition of Pub. Nat’l Bank of N.Y., 278 U.S. 101, 104 (1928) (quoting Washington Market Co. v. Hoffman, 101 U.S. 112, 115 (1879)). 24

A more in-depth look at the history of the relevant provisions further refutes United Therapeutics’ position. 37 We have already discussed changes Congress made to both credits from 1983 to 1989. See supra note 33. Between the 1989 amendments to the research credit that United Therapeutics highlights and the end of 2014, Congress amended the research credit at least 16 times 38 and the orphan drug credit at least 14 times. 39 Many of the amendments to both credits were minor, but others were significant.

As an example, Congress, which had previously renewed the research credit and the orphan drug credit every few years, allowed them both to expire effective July 1, 1995, for the research credit and December 31, 1994, for the orphan drug credit. It revived the credits in 1996, but not retroactively. Thus, there was a period from 1995 to 1996 when neither credit was available. See I.R.C. § 41(h)(1)(A) (1996) (providing that section 41 “shall not apply to any amount paid or incurred . . . after June 30, 1995, and before July 1, 1996”); I.R.C. § 45C(e)(1) (1996) (providing the same for “any amount paid or incurred . . . after December 31, 1994, and before July 1, 1996”). And while Congress made the orphan drug credit permanent in 1997, it continued to extend the research credit every few years, sometimes retroactively, until ultimately making the credit permanent in 2015. See Consolidated Appropriations Act, 2016, Pub. L. No. 114-113, div. Q, § 121(a), 129 Stat. 2242, 3049 (2015).

Moreover, when Congress revived the credits in 1996 following their lapse, it simultaneously made changes to both. For example, Congress modified the orphan drug credit by moving it from section 28 to section 45C, thereby subjecting it to the rules and limitations that apply to general business credits, see I.R.C. § 38, changing the termination and carryback provisions to reflect the credit’s lapse, and making other conforming amendments, see SBJPA § 1205. With respect to the research credit, Congress modified the definition of the term “base

37 The “history [we] have in mind here . . . [is] the record of enacted changes

Congress made to the relevant statutory text over time, the sort of textual evidence everyone agrees can sometimes shed light on meaning.” BNSF Ry. Co. v. Loos, 139 S. Ct. 893, 906 (2019) (Gorsuch, J., dissenting) (citing United States v. Wong Kim Ark, 169 U.S. 649, 653–54 (1898)). For a discussion of the amendments, see Kendall B. Fox et al., Research and 38

Development Expenditures, 556-3rd Tax Mgmt. (BNA) at X.B (Sept. 30, 2019). 39 Nearly all the amendments were made by the same statutes that amended

the research credit. See supra note 38. 25

amount” as it applies to startup companies, provided for the election of the alternative incremental credit, and increased the credit available for certain contract research expenses, among others. See id. § 1204, 110 Stat. at 1773–75. And Congress continued to tinker with the research credit over the years, including, as United Therapeutics points out, adding the alternative simplified credit as an option for calculating the credit in 2006. See TRHCA § 104(c), 120 Stat. at 2935.

All of this goes to show that, between 1989 and 2014, Congress had a number of opportunities to delete or modify the reference to base period research expenses in section 45C(c)(2) if it was in fact deadwood. These opportunities included, among many others, 2014 (the amendments that made the research credit available for 2014 and made a conforming change to section 45C), 2006 (the amendments that added the alternative simplified method to the research credit and updated a provision of the orphan drug credit), and 1996 (the amendments that resurrected both credits and made other changes). See infra Appendix. But with every amendment, Congress left section 45C(c)(2) intact.

Congress’s choice in this regard, a choice that it made over and over in the years leading up to 2014, suggests that it was happy with the text of section 45C(c)(2), including the reference to base period research expenses. New York ex rel. N.Y. State Off. of Child. & Fam. Servs. v. U.S. Dep’t of Health & Hum. Servs.’ Admin. for Child. & Fams., 556 F.3d 90, 99 (2d Cir. 2009) (even “edit[s that] may appear small” are “sufficient” to “demonstrate[] that [one statutory provision] did not escape Congress’s notice at the time it amended [another statutory] provision” and a contrary inference would be “unreasonable”).

This statutory history also explains in part why United Therapeutics’ reliance on Wisconsin Central, 138 S. Ct. 2067, is misplaced. That decision undercuts, rather than supports, United Therapeutics’ position here. In Wisconsin Central, the Supreme Court was called upon to interpret a term contained in a statute that, in relevant part, had been left unchanged since its adoption in 1937. In giving the relevant term the meaning it had in 1937, the Court observed:

Written laws are meant to be understood and lived by. If a fog of uncertainty surrounded them, if their meaning could shift with the latest judicial whim, the point of reducing them to writing would be lost. That is why it’s a “fundamental canon of statutory construction” that words generally should be “interpreted as taking their ordinary, 26

contemporary, common meaning . . . at the time Congress enacted the statute.” Perrin, 444 U.S., at 42. Congress alone has the institutional competence, democratic legitimacy, and (most importantly) constitutional authority to revise statutes in light of new social problems and preferences. Until it exercises that power, the people may rely on the original meaning of the written law.

Wis. Cent. Ltd., 138 S. Ct. at 2074.

What the Supreme Court observed should happen through the legislative process is exactly what Congress has done with respect to the research and orphan drug credits. That constitutionally authorized body has repeatedly “exercise[d] [its] power” “to revise” the terms under which the research and orphan drug credits are made available “in light of new social problems and preferences.” Id. And, as we have already explained, the statute that made the credit available for the year at issue also left in place the coordination provision that United Therapeutics urges us to read as a nullity. So, following United Therapeutics’ lead would require that we ignore “the ordinary, contemporary, common meaning” of the duly enacted statute that gave United Therapeutics the very benefit it seeks. That we will not do. “[A] judge’s job [is] only to apply, not revise or update, the terms of statutes.” Id. (citing Wis. Cent. Ltd. v. United States, 856 F.3d 490, 493 (7th Cir. 2017) (Manion, J., dissenting), rev’d and remanded, 138 S. Ct. 2067); see also New Prime Inc. v. Oliveira, 139 S. Ct. 532, 539 (2019) (interpreting the undefined term “contract of employment” as used in the Federal Arbitration Act based on that term’s meaning at the time of the Act’s adoption in 1925 when (unlike here) the relevant provisions had been left unchanged by Congress).

United Therapeutics’ reliance on Wisconsin Central is misplaced for another, perhaps more fundamental, reason. That case, the cases on which it relied, and the cases that followed it all concerned the proper interpretation of an undefined term. They all answered the question “what should a court do when the statute does not define the meaning of a relevant term at the time of its enactment?” They neither confronted nor answered the question “what should a court do when Congress removes from the statute a definition that might have been viewed as supplying the meaning of what a party claims to be a ‘defined’ term that remains in the statute?” 27

Our case implicates the latter question. In United Therapeutics’ telling, the phrase “base period research expenses” was a defined term when it was first adopted and retains that defined meaning even after Congress eliminated the relevant definition from the statute. United Therapeutics cites no authority for this proposition.

Nor does its position make sense in light of the concerns that animate Wisconsin Central and like cases. As the Supreme Court observed in New Prime Inc., 139 S. Ct. at 539:

[I]f judges could freely invest old statutory terms with new meanings, we would risk amending legislation outside the “single, finely wrought and exhaustively considered, procedure” the Constitution commands. INS v. Chadha, 462 U.S. 919, 951 (1983). We would risk, too, upsetting reliance interests in the settled meaning of a statute. Cf. 2B N. Singer & J. Singer, Sutherland on Statutes and Statutory Construction § 56A:3 (rev. 7th ed. 2012).

The circumstances before us do not involve the Court’s giving an undefined statutory term a meaning different from the ordinary meaning it would have had at the time of its adoption, thus interfering with the “single, finely wrought and exhaustively considered, procedure” for amending a statute. They involve instead a party inviting the Court to treat an undefined term as if it were defined, ignoring a congressional enactment that eliminated the potentially relevant definition from the Code, contrary to the considerations set out in New Prime Inc., and further ignoring repeated amendments to the statute. In short, neither Wisconsin Central nor any other authority United Therapeutics cites supports what United Therapeutics asks us to do.

Also weighing against United Therapeutics’ position is the 111th Congress’s enactment, in 2010, of a new credit that relied on the same language as that used in section 45C(c)(2). Specifically, as part of the Patient Protection and Affordable Care Act, Pub. L. No. 111-148, § 9023(a), 124 Stat. 119, 877 (2010), the 111th Congress enacted a new “qualifying therapeutic discovery project credit” under section 48D. And it included in the new credit, in a paragraph entitled “Denial of a double benefit,” the following coordination rule, with language nearly identical to that in section 45C(c):

(i) In general.—Except as provided in clause (ii), any expenses taken into account under this section for a 28

taxable year shall not be taken into account for purposes of determining the credit allowable under section 41 or 45C for such taxable year. (ii) Expenses included in determining base period research expenses.—Any expenses for any taxable year which are qualified research expenses (within the meaning of section 41(b)) shall be taken into account in determining base period research expenses for purposes of applying section 41 to subsequent taxable years.

I.R.C. § 48D(e)(2)(C) (2010). If section 45C(c)(2) has been a dead letter since 1989 because of its reference to “base period research expenses,” one would not expect Congress to have used the same language for a new credit in 2010. And that same Congress, which was far closer in time to 2014 than the 1989 Congress whose actions United Therapeutics invokes, later amended section 45C itself without modifying the coordination rule in subsection (c)(2). See Tax Relief, Unemployment Insurance Reauthorization, and Job Creation Act of 2010, Pub. L. No. 111-312, § 731(b), 124 Stat. 3296, 3317; cf. New Prime Inc., 139 S. Ct. at 540 (“More confirmation yet comes from a neighboring term in the statutory text.”).

Of course, we do not consider the actions of the 111th Congress as deciding the meaning of statutory provisions adopted by prior or future Congresses. See United States v. Price, 361 U.S. 304, 313 (1960) (“[T]he views of a subsequent Congress form a hazardous basis for inferring the intent of an earlier one.”). We mention them only to highlight that the atextual reading of section 41 and section 45C that United Therapeutics presses here (that the phrase “base period research expenses” refers to a concept that has been inapplicable from 1989 on) does not appear to have been shared by the 111th Congress or the Senate Finance Committee. 40

United Therapeutics insisted at the oral argument we held on January 25, 2023, that the coordination rule of old section 28(c)(2) and its successor section 45C(c)(2) remained in the Code because of an “oversight.” Tr. 28:5. In its view, Congress’s “failure to delete it was not a deliberate choice that [Congress] wanted this section to continue to

40 See S. Rep. No. 111-89, at 363 n.149 (2009) (describing the provision that

later became section 48D and noting that “[a]ny expenses for the taxable year that are qualified research expenses under section 41(b) are taken into account in determining base period research expenses for purposes of computing the research credit under section 41 for subsequent taxable years” (emphasis added)). 29

have life and applicability. It was simply a failure to make a corresponding change to section 45C when it overhauled section 41.” Tr. 28:14–18. This, United Therapeutics says, “[h]appens fairly often.” Tr. 28:12. But we do not interpret statutory enactments by assuming that Congress made mistakes and failed to express in the statutory text what it wished to accomplish. To the contrary, “[w]e ‘must presume that [the] legislature says in a statute what it means and means in a statute what it says there.’” Dodd v. United States, 545 U.S. 353, 357 (2005) (quoting Conn. Nat’l Bank v. Germain, 503 U.S. 249, 253–54 (1992)); see also Russello, 464 U.S. at 23 (“We would not presume to ascribe this difference to a simple mistake in draftsmanship.”); United States ex rel. Totten v. Bombardier Corp., 380 F.3d 488, 496 (D.C. Cir. 2004) (Roberts, J.) (“In the final analysis, we can remain agnostic on the question whether Congress intentionally left the presentment requirement in [the relevant statute] or simply forgot to take it out. The suggestion that Congress may have ‘dropped a stitch,’ [United States ex rel. Yesudian v. Howard Univ., 153 F.3d 731, 738 (D.C. Cir. 1998),] is not enough to permit us to ignore the statutory text.”).

In short, United Therapeutics’ contentions concerning the statutory provisions as they existed before 1989 and the changes made in 1989, while understandable in light of the outcome it wishes to achieve, do not provide valid reasons for ignoring the straightforward and ordinary meaning of the statutory text that applies for 2014.

2. The Consistency Rule of Section 41(c)(6)(A) Does Not Require a Different Outcome.

United Therapeutics next argues that the consistency rule of section 41(c)(6)(A) mandated its approach. Congress added the consistency rule to section 41 in 1989, when it replaced the definition of “base period research expenses” with a new “base amount” concept. Because the consistency rule pertains to the calculation of the base amount, some background regarding that concept is useful in understanding the rule.

Since the 1989 amendments to the research credit and through 2014, section 41 has provided that one component of the credit is an amount equal to 20% of the taxpayer’s qualified research expenses for the year over the “base amount.” I.R.C. § 41(a)(1). The base amount is the product of the taxpayer’s “fixed base percentage” and its average gross receipts for the four years preceding the credit year. I.R.C. § 41(c)(1). In general, the fixed base percentage is “the percentage which 30

the aggregate qualified research expenses of the taxpayer for taxable years beginning after December 31, 1983, and before January 1, 1989, is of the aggregate gross receipts of the taxpayer for [those] years.” I.R.C. § 41(c)(3)(A).

The consistency rule applies to the calculation of a taxpayer’s fixed base percentage. It provides that “the qualified research expenses taken into account in computing such percentage shall be determined on a basis consistent with the determination of qualified research expenses for the credit year.” I.R.C. § 41(c)(6)(A).

The consistency rule (as it appears in the statute) refers only to the fixed base percentage and does not, on its face, apply when calculating the alternative simplified credit. But, after Congress enacted the alternative simplified credit in 2006, Treasury and the IRS promulgated a regulation that extended the consistency rule. The regulation provides as follows:

Treas. Reg. § 1.41-9 Alternative simplified credit. .... (c) Special rules. . . . .... (2) Section 41(c)(6) applicability. [Qualified research expenses] for the three taxable years preceding the credit year must be determined on a basis consistent with the definition of [qualified research expenses] for the credit year, without regard to the law in effect for the three taxable years preceding the credit year. This consistency requirement applies even if the period for filing a claim for credit or refund has expired for any of the three taxable years preceding the credit year.

United Therapeutics argues that this rule requires consistency in calculating the two components of the alternative simplified credit—i.e., that it does not permit qualified clinical testing expenses to be excluded in qualified research expenses for the credit year but included for the three preceding years. Again, we disagree.

First, to the extent United Therapeutics relies on the statutory consistency rule, that provision does the company no good. As United Therapeutics appears to recognize, the consistency rule in section 41(c)(6)(A) applies only when calculating a taxpayer’s fixed base 31

percentage, a concept that had no relevance in calculating the company’s alternative simplified credit.

Second, we disagree with United Therapeutics’ interpretation of the regulation, which says simply that taxpayers must apply the same definition of qualified research expenses to the credit year and the three preceding years even if there has been a change in law. In other words, if the definition of qualified research expenses, which is provided in section 41(b), changes during the relevant years, the regulation requires taxpayers to apply the credit year definition in identifying its qualified research expenses for all four years. Nothing in the regulation purports to override the coordination rule of section 45C(c), which does not address the definition of qualified research expenses other than by referring back to section 41(b). Rather, section 45C(c) provides a special rule for how a certain category of qualified research expenses—those that are also qualified clinical testing expenses—must be treated after they are identified.

Third, to the extent United Therapeutics intends to use the statutory consistency rule as a textual clue supporting its reading of section 45C(c)(2), that effort also comes up short. We see no conflict between the statutory consistency rule and section 45C(c)(2). Instead, we read the statutory consistency rule the same way Treasury and the IRS do in their regulations, and the same way the U.S. Court of Appeals for the Fifth Circuit read it in Trinity Industries, Inc. v. United States, 757 F.3d 400 (5th Cir. 2014). That is, the rule simply requires that the definition of qualified research expenses, which Congress has changed over the years, be applied consistently across the credit year and the years in the reference period. See Treas. Reg. § 1.41-3(d)(1) (requiring qualified research expenses “[to] be determined on a basis consistent with the definition of qualified research expenses . . . for the credit year, without regard to the law in effect for the taxable years taken into account in computing the fixed-base percentage or the base amount”); see also Trinity Indus., Inc., 757 F.3d at 411–12 (“In sum, the consistency rule calls for consistent application of the [qualified research expense] definition across the base period years and the claim year . . . .”). 41 This straightforward reading of the statute gives effect to both section 41(c)(5) and section 45C(c)(2), unlike United Therapeutics’ preferred reading. See Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612, 1624 (2018) (“When confronted with two Acts of Congress allegedly touching

41 United Therapeutics cites Trinity Industries in support of its position, but

that case interprets the consistency rule the same way we do here. 32

on the same topic, this Court is not at ‘liberty to pick and choose among congressional enactments’ and must instead strive ‘to give effect to both.’” (citing Morton v. Mancari, 417 U.S. 535, 551 (1974))).

Moreover, even assuming for the sake of argument that the consistency rule of section 41(c)(6) conflicts with the coordination rule of section 45C(c)(2) in certain circumstances, 42 section 45C(c)(2) would prevail under “the specific governs the general” rule of statutory interpretation. See RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 566 U.S. 639, 645 (2012) (“[I]t is a commonplace of statutory construction that the specific governs the general.” (quoting Morales v. Trans World Airlines, Inc., 504 U.S. 374, 384 (1992))); see also Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 183–88 (2012). Section 41(c)(6) is a general rule that applies to all taxpayers computing the incremental research credit, whereas section 45C(c)(2) applies only to taxpayers who elect to claim the orphan drug credit in addition to the research credit. Accordingly, section 45C(c)(2), which may marginally reduce the overall benefit of both credits for taxpayers who claim the orphan drug credit, is the more specific rule in this context and would control in the event of a conflict.

3. Policy Considerations Cannot Change the Clear Directive of the Relevant Provisions.

United Therapeutics also appears to offer a policy argument in support of its position. Its opening brief observes that, in adopting the orphan drug credit,

Congress aimed to encourage the development of desperately needed treatments. See H.R. Rep. No. 101-247, at 1199 (noting that the House Committee on the Budget “modified the method of calculating a taxpayer’s base amount in order to enhance the credit’s incentive effect”). That objective would be frustrated by reducing the section 41 research credit based on a company’s incremental investment in clinical testing of orphan drugs.

42 To reiterate a point we made above, the statutory consistency rule applies

only to taxpayers claiming the incremental credit, which relies on the base amount computation. So, even under United Therapeutics’ interpretation, it would not create a conflict in this case since United Therapeutics claimed the alternative simplified credit, not the incremental credit. See supra note 7. 33

Pet’r’s Opening Br. 12–13.

But, “[a]s [the Supreme] Court has explained, ‘even the most formidable’ policy arguments cannot ‘overcome’ a clear statutory directive.” BP P.L.C., 141 S. Ct. at 1542 (quoting Kloeckner v. Solis, 568 U.S. 41, 56 n.4 (2012)). Moreover,

[t]hat a law might temper its pursuit of one goal [for example, the encouragement of desperately needed treatments] by accommodating others [for example, minimizing the budget impact of an incentive provision like the research credit 43] can come as no surprise. Often legislation becomes possible only because of such compromises. Often lawmakers tread in areas fraught with competing social demands where everyone agrees trade-offs are required.

Id. at 1539. In the final analysis, we agree that, as United Therapeutics notes in its Answering Brief at 17, “‘[t]he judicial function is confined to applying what Congress has enacted after ascertaining what it is that Congress has enacted.’ Local 1976, United Bhd. of Carpenters & Joiners v. NLRB, 357 U.S. 93, 100 (1958). Congress’s policy aims are best served by applying the statute according to its terms . . . .” That is precisely what we do here.

V. Conclusion

For the reasons stated above, the case must be resolved in favor of the Commissioner.

To reflect the foregoing,

Decision will be entered for respondent.

43 Concern over the cost of the research credit is a common theme in the

materials that accompany the legislation governing the research credit. See, e.g., H.R. Rep. No. 101-247, at 1199–1200 (1989), as reprinted in 1989 U.S.C.C.A.N. 1906, 2669– 70 (explaining that changes were made “at the lowest possible revenue cost”); see also Yin, supra note 5, at 199–202. 34

APPENDIX

Research Credit 44

Date of Effective Dates Legislation Enactment Begin End

Economic Recovery Tax Act of 1981, August 13, December July 1, 1981 Pub. L. 97-34 1981 31, 1985

Tax Reform Act of 1986, Pub. L. 99- October 22, January 1, December 514 1986 1986 31, 1988

Technical and Miscellaneous Revenue November 10, January 1, December Act of 1988, Pub. L. 100-647 1988 1989 31, 1989

Omnibus Budget Reconciliation Act of December 19, January 1, December 1989, Pub. L. 101-239 1989 1990 31, 1990

Omnibus Budget Reconciliation Act of November 5, January 1, December 1990, Pub. L. 101-508 1990 1991[ 45] 31, 1991

Tax Extension Act of 1991, Pub. L. December 11, January 1, June 30, 102-227 1991 1992 1992

Omnibus Budget Reconciliation Act of August 10, June 30, July 1, 1992 1993, Pub. L. 103-66 1993 1995

Small Business Job Protection Act of August 20, May 31, July 1, 1996 1996, Pub. L. 104-188 1996 1997

Taxpayer Relief Act of 1997, Pub. L. June 1, June 30, August 5, 1997 105-34 1997 1998

Omnibus Consolidated and Emergency Supplemental October 21, June 30, July 1, 1998 Appropriations Act, [1999,] Pub. L. 1998 1999 105-277 Ticket to Work and Work Incentives December 17, June 30, Improvement Act of 1999, Pub. L. 106- July 1, 1999 1999 2004 170

44 The tables are reproduced from Respondent’s Supplemental Brief pp. 6–7. 45 The effective date of the relevant provisions was January 1, 1990. Omnibus

Budget Reconciliation Act of 1990, Pub. L. No. 101-508, § 11402(c), 104 Stat. 1388, 1388–473. 35

Working Families Tax Relief Act of October 4, December July 1, 2004 2004, Pub. L. 108-311 2004 31, 2005

Tax Relief and Health Care Act of December 20, January 1, December 2006, Pub. L. 109-432 2006 2006 31, 2007

Emergency Economic Stabilization October 3, January 1, December Act of 2008, Pub. L. 110-343 2008 2008 31, 2009

Tax Relief, Unemployment Insurance December 17, January 1, December Reauthorization, and Job Creation Act 2010 2010 31, 2011 of 2010, Pub. L. 111-312

American Taxpayer Relief Act of 2012, January 2, January 1, December Pub. L. 112-240 2013 2012 31, 2013

Tax Increase Prevention Act of 2014, December 19, January 1, December Pub. L. 113-295 2014 2014 31, 2014

Protecting Americans from Tax Hikes December 18, January 1, *Made Act of 2015, Pub. L. 114-113 2015 2015 Permanent 36

Orphan Drug Credit

Date of Effective Dates Legislation Enactment Begin End January 4, January 1, December Orphan Drug Act, Pub. L. 97-414 1983 1983 31, 1987 October 22, December Tax Reform Act of 1986, Pub. L. 99-514 N/A[46] 1986 31, 1990 Omnibus Budget Reconciliation Act of November 5, January 1, December 1990, Pub. L. 101-508 1990 1990[ 47] 31, 1991 Tax Extension Act of 1991, Pub. L. December 11, January 1, June 30, 102-227 1991 1992 1992 Omnibus Budget Reconciliation Act of August 10, July 1, December 1993, Pub. L. 103-66 1993 1992 31, 1994 Small Business Job Protection Act of August 20, July 1, May 31, 1996, Pub. L. 104-188 1996 1996 1997 Taxpayer Relief Act of 1997, Pub. L. August 5, June 1, * Made 105-34 1997 1997 Permanent

46 The effective date of the relevant provisions was January 1, 1983. TRA 1986 § 1879(b)(3), 100 Stat. at 2906. 47 The relevant section does not appear to include an effective date provision.

Omnibus Budget Reconciliation Act of 1990, § 11411, 104 Stat. at 1388–479.