Does Treasury Comply with Administrative Law, Including the APA? I Say Yes. Others Say No. (7/17/21)

In United States v. Kahn, 5 F.4th 167 (2d Cir. 7/13/21), here, the Court held that the FBAR willful penalty in 31 USC § 5321(a)(5), as amended in 2004 to increase the maximum amount of the penalty to 50% in the account(s), is not limited by the FINCen’s failure to update the underlying regulations (adopted in 1987) which, consistent with pre-2004 law, capped the willful penalty at $100,000.  This holding is currently the strong consensus.  I doubt there will be further aberrations on that issue. 

I generally discuss FBAR willful civil penalty issues principally on my Federal Tax Crimes Blog and did so in this case.  See Second Circuit Continues the Strong Consensus Rejecting the Argument that FINCen Regulations Under Pre-2004 Law Limit the Maximum Willful Penalty Prescribed under the 2004 Statutory Amendment (Federal Tax Crimes Blog 7/14/21), here.

I said in the Tax Crimes Blog posting that I might address Judge Menashi’s dissenting opinion which, in my view, is ideologically tinged repeating mantras in legal jargon that serve as proxies against the evils of the administrative state that play so well to the right.  Chevron is a chief bogeyman that Judge Menashi and other judges of similar bent use for the purpose.  Judge Menashi’s Wikipedia entry is here.

Judge Menashi’s analysis, in summary, is that an administrative agency, here FINCen with administration authority delegated to the IRS, should abide by its own regulations regardless of intervening changes in the statute.  Judge Menashi cites a rule called the Accardi principle (sometimes called a doctrine) after United States ex rel. Accardi v. Shaughnessy, 347 U.S. 260 (1954).  FINCen is part of Treasury, but a different part than the IRS.  That issue then permits Judge Menashi to launch into administrative law.

What set me off particularly about Judge Menashi’s dissent is that, although not relevant to his analysis, he repeats Professor Hickman’s claims about the IRS not following administrative law, including the APA.   Judge Menashi thus asserts (p. 3) “The Treasury Department has sometimes evaded standard administrative law principles when enforcing the tax laws” citing in footnote 3 Kristin E. Hickman, Administering the Tax System We Have, 63 Duke L.J. 1717, 1718 (2014) (describing “tax departures from general administrative-law norms”).  Readers of this Federal Tax Procedure blog will recognize that type of claim by Professor Hickman.  The Court of Appeals in CIC Servs., LLC v. IRS, 925 F.3d 247. 258 (6th Cir. 2019), reh. en banc den. 936 F.3d 501 (6th Cir. 2019), rev’d and remanded 583 U.S. ___, 141 S.Ct. 1582 (2021) quoted Professor Hickman and a colleague as claiming that Treasury and the IRS “do not have a great history of complying with APA procedures, having claimed for several decades that their rules and regulations are exempt from those requirements.”  The quote is from Kristin E. Hickman & Gerald Kerska, Restoring the Lost Anti-Injunction Act, 103 Va. L. Rev. 1683, 1686 (2017)).  Fortunately, in the Supreme Court in CIC Servs. the parties submitting briefs (including Professor Hickman as amicus curiae) did not repeat that claim, and the Supreme Court did not make the claim.  Perhaps they steered away from the claim prominently made by the Court of Appeals because the claim is irrelevant to the issue presented (just as it was in the Court of Appeals) and, in my view, the claim is not true.  (I synthesize my conclusions on this at the end of this blog.)  And the claim is not relevant to the issue resolved in Kahn, but despite the claim's irrelevance, Judge Menashi makes the claim.

Most immediately addressing what is relevant to the issue in Kahn, while it is good law that an agency follow its own regulations, I don’t think that was an iron-clad rule applying even when common sense says it should not.  Accardi did not say that agency regulations must apply when there has been an intervening change in the statute.  When, for example, as here, there is an intervening change in the law and Treasury did not give any indication that the content of the regulations under the old statute was meant to exercise discretion allowed the agency under the amended statute.  Common sense tells you—well, at least me (and apparently the majority on the panel)--that Congress’ very intentional amendment in the statute with the maximum 50% penalty was meant to apply after the statute was amended.

Deviating into territory more ideological than relevant, Judge Menashi makes some claims without subtlety.

Judge Menashi starts correctly (p. 2) by quoting Chrysler Corp. v. Brown, 441 U.S. 281, 295 (1979) that “[P]roperly promulgated, substantive agency regulations have the force and effect of law.”  What he doesn’t say is that a “substantive” regulation in the APA meaning is the equivalent of a “legislative regulation,” which is the term more commonly used now.  I think most administrative law scholars would agree that, as enacted in 1946, the APA used substantive rule (regulation) to mean the same thing that scholars now call legislative regulation.  I use the term legislative regulation here to mean the same as substantive regulations. 

Interpretive regulations are not legislative regulations.  The theoretical distinction between legislative regulations and interpretive regulations at the time of enactment of the APA (and, I submit, now) was:  Legislative regulations are regulations adopted pursuant to a specific statutory grant (like the consolidated return regulations) to an agency to make the law rather than interpret ambiguous statutory text.  Because legislative regulations make the law, legislative regulations have historically been said to have the force and effect of law (or just force of law).  That is because legislative regulations are the law.  The consolidated return regulations are the law.  By contrast, interpretive regulations are regulations adopted with express or general authority to interpret ambiguous statutory text; the statutory text is the law; the interpretation is not the law even when a court may adopt an interpretation—its own or an agency’s--in applying the statutory text.  Interpretive regulations have not historically been said to have the force of law because they are not the law as legislative regulations are.  (There will be comments among the thousands of cases that may differ with these propositions, but I can and do support them in my article with more detail in the revised article I will post later this summer.)

Then, Judge Menashi quotes (p. 2) Perez v. Mortg. Bankers Ass’n, 575 U.S. 92, 96 (2015) as  follows: “Rules issued through the notice-and-comment process are often referred to as ‘legislative rules’ because they have the ‘force and effect of law.’”  Perez does say that, but the statement is not correct as stated if read as claiming that all notice-and-comment regulations are legislative regulations.  Justice Breyer, an administrative law expert, famously said “there are hundreds of thousands, possibly millions of interpretive regulations.”  Under the APA,  legislative rules must be published through notice-and-comment regulations; interpretive rules may be adopted as notice-and-comment regulations. Interpretive rules adopted with notice-and-comment that do no more than interpret ambiguous statutory text are not legislative rules.  And this is true even if a court defers to the interpretation.

Returning to Professor Hickman’s claim adopted by Judge Menashi, Professor Hickman has several related threads, implicit or explicit:  (i) there are no interpretive regulations which have been adopted by notice-and-comment rulemaking; (ii) rules that only interpret the statute (including making interpretive choices within the zone of statutory ambiguity) if adopted as a notice-and-comment regulation are thereby transformed into legislative regulations; (iii) deference (currently the Chevron iteration) gets into the discussion because, as asserted by Justice Scalia, interpretive regulations, if given deference, have the force of law (I discuss this claim below); and (iv) temporary regulations looking forward to notice-and-comment regulations are invalid because when so adopted they are legislative regulations which must be prospective and therefore cannot be effective before adopting with notice-and-comment.

Basically, my claims are (i) regulations that interpret statutory text and adopted with notice-and-comment are interpretive regulations rather than legislative regulations, even when the interpretations are given deference; (ii) interpretive regulations do not have the force of law in the meaning that the term was used in differentiating between legislative and interpretive regulations in the APA (Justice Scalia notwithstanding); (iii) Temporary Regulations interpreting ambiguous statutory text may be applied before notice-and-comment and may indeed be retroactive to the date of the statute (although retroactivity for tax regulations may be limited by § 7805(b)); and (iv) the IRS’s historical practices for interpretive regulations do not violate administrative law or the APA.

Now, let’s look at Justice Scalia’s claim that “Interpretive rules that command deference do have the force of law.” Perez v. Mortg. Bankers Ass'n, 575 U.S. 92,110 (2015) (concurring in the judgment).  Justice Scalia did not claim that giving an interpretation deference transforms an interpretive rule into a legislative rule.  Scalia opens with the statement that he is discussing an interpretive rule; if his “force of law” comment meant that the regulations are legislative, he would not have called them interpretive.  The only claim Justice Scalia makes is that interpretive rules (in this case regulations) given deference have the force of law.  In a general sense, interpretations given deference do have the force of law because the interpretation can control judicial outcomes.  But, as I have noted above, that was not the sense or use of the term "force of law" when and after the APA was enacted.  (In an originalist interpretive universe, which Justice Scalia claimed to occupy (at least when it suited him), the original meaning should prevail.)  Indeed, as the majority noted in Perez (p. 129), “substantive agency regulations have the ‘force and effect of law,’” citing Chrysler Corp. v. Brown, 441 U. S. 281, 295 (1979).  To which I say, of course.  But what the majority did not say was the interpretive regulations have the force of law, or that interpretive regulations are thereby transformed into legislative regulations.

Now back to Professor Hickman’s claim that the Treasury and IRS do not comply with administrative law and the APA.  I believe that the IRS takes the position (as do I) that (i) interpretive rules (whether adopted with notice-and-comment and whether given deference or not) are not legislative rules with the APA requirements for legislative rules (specifically must be notice-and-comment and prospective only);  (ii) interpretive rules may be applied retroactively to the date of the statute (for IRC regulations subject to § 7805(b) limitations on retroactivity), with the result that interpretive rules applying in Temporary Regulations can be applied immediately before final promulgation; and (iv) interpretive rules do not have the force of law in the meaning of the term when it was deployed in relation to the APA and its administrative law context.  If that position is correct, then historic IRS practice in this regard have been compliant with administrative law and the APA.  My understanding is that Professor Hickman and others of similar disposition disagree with those premises.  To which I counter that the premises were the law when and after the APA was adopted and should be the law now because nothing has changed the law.

My SSRN article is: Townsend, John A., The Report of the Death of the Interpretive Regulation Is an Exaggeration (August 23, 2020). Available at SSRN: https://ssrn.com/abstract=3400489 or http://dx.doi.org/10.2139/ssrn.3400489.

Remember that I will have a substantial revision of that article later this summer, perhaps shortly after I published the 2021 editions of my Federal Tax Procedure Book in early August.