CFC Invalidates GILTI Gap-Filling Regulation That Avoided Textual Statute Inconsistency (7/7/26)

In Keysight Technologies, Inc. v. United States, ___ Fed.Cl. ___ (7/2/26), the Court held invalid a Treasury Regulation designed to plug a gap in the statutory text. The opinion may be found: CFC here, TN here, GS here [to come].

The Court opens with this sentence projecting the outcome (Slip Op. 1):

When Chevron fell, so too did the presumption that statutory ambiguity favors the agency. Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984), overruled by Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024).

If that were not clear enough as to where it is going, the Court opens the next paragraph (bold face supplied by JAT):

This controversy relates to the Treasury’s self-inflicted fix of a mismatch between foreign subsidiaries with fiscal- or calendar-year tax filing requirements that Congress quietly built into the global intangible low-taxed income ("GILTI") statutory scheme.

And the next paragraph (bold face supplied by JAT):

The Treasury’s antipathy for this inconsistency resulted in the Secretary promulgating Regulation 1.951A-2(c)(5) ("the Regulation").

I won’t parse the quoted text any further (although my bold face may imply something).

The technical issue was whether Treasury could, by interpretive regulation, fix what appears textually to be a timing glitch producing materially different tax result based on differences in tax years between calendar year taxpayers and fiscal year taxpayers. As the Court says (Slip Op. 2): “Enactment of the TCJA created an inconsistency between treatment of certain taxpayers depending on whether they were fiscal-year or calendar-year filers.”

Basically, the taxpayer argued that it, as a fiscal year taxpayer, was entitled to a benefit that a calendar year taxpayer could not achieve. Without wallowing around in the details, it appears to me that an intuitive view of how Congress enacts tax legislation, one can fairly infer that Congress did not intend the result that the regulation sought to forbid and the Court blesses. Congress appears to have had no specifically articulated intent on the precise issue, but Congress generally does not make such distinctions to permit disparities between otherwise similarly situated taxpayers.

The taxpayer has a good literal case based on the statutory text read literally (dare I say rigidly) and without concern as to what Congress might have intended. The Court bought the argument, hook, line, and sinker. But, as tax procedure enthusiasts know, since Helvering v. Gregory, 69 F.2d 809 (2d Cir. 1934), affd. 293 U.S. 465 (1935), meeting the literal terms of the statute does not work if contrary to the purpose behind the statute.

A general concern I had when reading the opinion is that the court overused slogans (aka sound-bites, etc.) which are not conducive to sound analysis. Of course, Loper Bright overused slogans, so the Court was in good company (good might not be the right descriptor).

Obviously, if I were a judge (I am not and never have been, but I often judge judge’s work), I would not have decided the way the Court (a real judge) did. I just make a few comments on the opinion:

1. I say that the regulation was an interpretive regulation. Administrative law enthusiasts will recall that there has been commotion over the years (principally, I think based on misreading Chevron) as to whether Treasury interpretive regulations based on § 7805(a) are interpretive or legislative in the administrative law and APA sense. I have engaged on that issue ad nauseum in the past and don’t propose to re-engage except to state my belief that § 7805(a) authorizes only interpretive (or interpretive-type, such as gap-filling) regulations and does not authorize legislative (legislative-type) regulations. Just to remind readers, legislative regulations are illustrated in the tax law in § 1502 authorizing consolidated return reporting regulations.

2. Interestingly, the Court quotes § 706(2)(A) which permits a court to “set aside” agency action (here interpretation) that is “not in accordance with law.” (Slip Op. 3.) That standard is from the precise review standard held to require deference when a court is in equipoise as to the best interpretation of statutory text in Dobson v. Commissioner, 320 U. S. 489 (1943). Of course, the Supreme Court ignored Dobson in Loper Bright, and so does this Court. But, in fairness to this Court, Loper Bright may have required ex cathedra that the Court follow Loper Bright in ignoring Dobson. (I have an upcoming article to be published in the ABA Tax Lawyer that explores Dobson further.)

3. The Court commotes about the Loper Bright's reversal of Chevron deference but makes no indication whether pre-Chevron deference might have applied. Loper Bright reversed only Chevron deference based on the “fiction” of presumed delegation of interpretive authority through ambiguous statutory text. Pre-Chevron deference was not based on that “fiction.” The Court relegates pre-Chevron deference to a footnote which really does not address the issue of whether pre-Chevron deference survives Loper Bright (Slip Op. 4, n3):

n3 Scholars suggest that courts work older interpretative approaches into their post-Chevron analyses of tax regulations. See Blaine G. Saito, Tax Regulations in A Loper Bright Light, 28 FLA. TAX REV. 530, 534 (2025) ("[T]he general grant of regulatory authority will create a standard that sounds similar to the old National Muffler test and a Skidmore overlay . . . specific grants should provide greater leeway to the Treasury.") (referring to National Muffler Dealers Ass’n v. United States, 440 U.S. 472 (1979); Amandeep S. Grewal, Tax Regulations After Loper Bright, 2024 MICH. ST. L. REV. 1083, 1087-91, 1096-98 (2024) (discussing the standard set by National Muffler and its relationship to tax exceptionalism after Loper Bright).

In this footnote, the Court uses the out-of-date term “interpretative” rather than the currently favored “interpretive;” elsewhere in the opinion the Court uses “interpretive.”

4. After Loper Bright, the Supreme Court has emphasized contemporaneity of the agency interpretation as being an important factor, often citing Skidmore (discussed further below). See e.g.,  Trump v. Barbara, 609 U. S. ___, ___ (2026) (Slip Op. 3 [54 of pdf]) (Kavanaugh concurring, "See Loper Bright Enterprises v. Raimondo, 603 U. S. 369, 394 (2024) (“interpretations issued contemporaneously with the statute at issue, and which have remained consistent over time, may be especially useful in determining the statute’s meaning”)"; and West Virginia v. B. P. J., 609 U. S. ___, ___ (2026) (Slip Op. 24) ("Because HEW’s interpretation of Title IX was “issued contemporaneously with the statute” and has “remained consistent over time,” it is “especially useful in determining the statute’s meaning.” Loper Bright Enterprises v. Raimondo, 603 U. S. 369, 394 (2024); Mullin v. Al Oro Lado, 609 U. S. ___ (2026) (Slip Op. 12) (same). Applying these concepts in Keysight is mixed because the regulations' interpretation was more or less contemporaneously adopted after the TCJA (at least in terms of time trajectories for tax regulations), it is not clear that the regulations have stood "over time." (whatever "over time" means). But query if that defeats application of this Loper Bright work around in Keysight, what would have been the result if the regulation had not been challenged for 10 years, 15 years, etc. Some tax cases do take that long. Why should the "over time" factor be critical for this type  of regulation? 

Added 7/7/26 3:10pm: Immediately after posting this blog, I read a Lawrence Solum posting addressing the type of "10 years, 15 years" hypo I gave above. Those wanting to explore the legal and philosophical uses of such hypos might want to read the post: Lawrence Solum, Legal Theory Musings No. 9: Hypotheticals, Intuition Pumps, and Legal Theory (Legal Theory Stack 7/7/26), here, addressing such bandied-about notions as textualism, purposivism, etc.

5. The principal basis the IRS claimed in promulgating and defending the Regulation was § 7805(a). The scope of permissible interpretation allowed by § 7805(a) consistent with Loper Bright is a big issue. Indeed, in closing, the Court says (Slip Op. 12-13):

The Court agrees that Section 7805(a) grants the Secretary broad authority to prescribe regulations, but that grant alone is insufficient to sustain the regulation at issue. To that end, the Court’s review of the Code has revealed no specific authority — either express or implied — to buttress Section 7805(a).

Again, the Court seems to have been focused on some textual indication of authority to fill the gap other than § 7805(a), which it simply could not find.

6. The Court addresses Skidmore respect (misnamed deference, a misnaming the Court adopts) by saying that the agency interpretation is not persuasive. (Slip Op. 10-12). For what it's worth (not much), I am not persuaded by the Court’s explanation of why it was not persuaded. Others may or may not be persuaded by the Court’s explanation.

7. On statutory interpretation, the Court pays partial allegiance to the textualist article of faith that committee reports discussing the text that was enacted do not count. The Court thus says (Slip Op. 9):

    Finding no express or implied grant of authority in the plain language of the cited statutes, the Court turns next to the legislative history of Section 951A-2(c)(5) for evidence of congressional intent. Defendant cites to a House of Representatives Conference Report on the Tax Reform Act of 1986 to support the premise that Congress was interested in curbing abusive transactions. (Def.’s Cross-Mot. at 28 (citing H.R. Rep. No. 99-841, pt. II, at 578-79 (1986) (Conf. Rep.))). Committee reports do not necessarily reflect the intent of the entire Congress at the time legislation is actually enacted and thus have only limited persuasive value.

The court does not acknowledge (or even seem to recognize) that Conference Reports are the most persuasive form of legislative history. Here is a footnote from the 2026 working draft of my book (may be the same as the 2025 published edition:

n__ Lawrence + Mem'l Hosp. v. Burwell, 812 F.3d 257, 266-267 (2d Cir. 2016) (“highest form of legislative history,” citing Disabled in Action of Metro. New York v. Hammons, 202 F.3d 110, 124 (2d Cir. 2000) and Robert A. Katzmann, Judging Statutes 38, 54 (2014)); Nestle Purina Petcare Co. v. Commissioner, 594 F.3d 968 (8th Cir. 2010) (citing Sierra Club v. Clark, 755 F.2d 608, 615 (8th Cir. 1985) which quoted Demby v. Schweiker, 671 F.2d 507, 510 (D.C. Cir. 1981) (“Because a ‘conference report represents the final statement of terms agreed to by both houses, next to the statute itself it is the most persuasive evidence of congressional intent.’”); and Nw. Forest Res. Council v. Glickman, 82 F.3d 825, 835 (9th Cir. 1996) (“[A] congressional conference report is recognized as the most reliable evidence of congressional intent because it ‘represents the final statement of the terms agreed to by both houses.”). For an instance where the clear conference committee explanation was not accepted to determine the meaning of the statutory text, see Arlington Central School District Board of Education v. Murphy, 548 U.S. 291, 324 (2006), as discussed in Stephen Breyer, Pragmatism or Textualism, 138 Harv. L. Rev. 717, 728 (2024).

And, on legislative history generally, Professor Abbe Gluck of Yale Law School has this outstanding post: Abbe R. Gluck, Justice Jackson reignites the interpretation wars, adding to textualism’s emerging cracks, SCOTUSblog (Jul. 7, 2026, 10:00 AM), https://www.scotusblog.com/2026/07/justice-jackson-reignites-the-interpretation-wars-adding-to-textualisms-emerging-cracks/