§ 7805(b) Time Limits Do Not Apply to Agency Best Interpretations of the Statute (8/11/22)

I recently posted a blog on Guedes v. ATF, 45 F.4th 306 (D.C. Cir. 8/9/22), DCCir here,  and GS here. See Important DC Circuit Opinion That Chevron Deference is Irrelevant if Agency Interpretation is Best Interpretation (Federal Tax Procedure Blog 8/9/22; 8/10/22), here. The essence of this new Guedes case is that an interpretation that is the best interpretation of the statute applies without any deference to the agency interpretation. The best interpretation controls, not because it is the agency interpretation, but because it is the best interpretation. So, for example, even if that best interpretation is in an interpretive regulation, the best interpretation controls. And the best interpretation controls even if the regulation is procedurally invalid. I have made that point (not much discussed in the mainstream claims) in several blog posts. I list only a few:  11th Cir. Invalidates Proportionate Sharing Regulations As Procedurally Arbitrary and Capricious for Failing to Address a Significant Comment (12/30/21; 12/31/21), here; Regulations Interpreting Pre-1996 Code Provisions; Fixing Hewitt (1/6/22; 1/7/22), here; and Sixth Circuit Creates Circuit Conflict with Eleventh Circuit on Conservation Easement Regulations (3/15/22), here.

I focus here on the consequences of this key point—the best interpretation of the statute controls independently of the validity or characterization of the regulation in which the best interpretation may appear. Readers may recall that § 7805(b) limits retroactivity for § 7805(a) regulations interpreting Code sections enacted after the effective date of the 1996 amendments to § 7805(b). For regulations interpreting Code sections enacted before the 1996 effective date, the interpretations may be fully retroactive to the date of enactment of the Code sections being interpreted. So, imagine a hypothetical Code section enacted after 1996, say enacted in 1997, so that § 7805(b) applies to limit the retroactivity of an interpretive regulation. Suppose that, in 2022, the IRS adopts a regulation interpreting the Code section. The regulation qua regulation cannot apply retroactively to 1997. But, the best interpretation can and should apply retroactively to 1997. (I need to clarify what retroactive means in this context; the interpretation has to be within the range of reasonable interpretations since the date of enactment so, in that sense, the interpretation is not retroactive; but the retroactive language is often used to describe the concept of the interpretation later recognized (in the example, in 2022) as the best interpretation applying from the date of enactment.)

The point is that an interpretation in a Treasury regulation (Temporary or Final) otherwise subject to the § 7805(b) time limits will avoid those time limits if the interpretation is the best interpretation of the statute.   Let me repeat that so that it sinks in:  An interpretation that is the best interpretation can and should apply retroactively without regard to § 7805(b). Indeed the best interpretation is effective even if announced in subregulatory guidance (such as Revenue Ruling or Notice). What that necessarily means is that § 7805(b)’s time limits practically apply only to an agency regulation interpretation that is not the best interpretation and thus needs a valid regulation (in this case within the § 7805(b) time limits) for Chevron deference.

This raises some questions:

1. How often are Treasury regulations interpretations the best interpretations? I think most Treasury (and other agency) regulations interpretations are the best interpretations. I have not attempted to do an empirical analysis of that belief, but my anecdotal career-time work with Treasury regulations leads me to that conclusion. In considering “best” interpretation, the court would have to apply Skidmore “respect” in determining the best interpretation.  Caveat: Skidmore is not deference to a less than best agency interpretation. See Really, Skidmore "Deference?" (Federal Tax Procedure Blog 5/31/20; 2/14/21), here.

2. How often does a court apply Chevron deference to a less than best agency interpretation? Supreme Court application of Chevron deference has been spare in recent years, with the Supreme Court clearly avoiding reflexive deference.  . But the Courts of Appeals noise about Chevron a good bit (although less so since they seem to be avoiding reflexive Chevron deference based on signals from Supreme Court Justices). But Court of Appeals noisings (at least some masking what they are doing) about Chevron often are (i) in the context of determining that the agency interpretation is the best interpretation, for which as noted above, Chevron does not control, or (ii) where it is difficult to tell that they were actually deferring to a not best agency interpretation. Is Chevron on Life Support; Does It Matter? (Federal Tax Procedure Blog 4/2/22; 4/3/23), here.

3. If a court is in equipoise as to the best interpretation, then it may apply an agency interpretation if within the zone of equipoise. But even then, the court is not deferring to a less best interpretation. I have written on equipoise in this context. Chevron and Equipoise In Statutory Interpretation (Federal Tax Procedure Blog 5/26/22), here; The Impact of Chevron Deference is Exaggerated (Federal Tax Procedure Blog 8/19/21; 8/21/21), here; and Even More on Skidmore (Including Equipoise as to Interpretation) (Federal Tax Procedure Blog 7/7/19), here.

4. This discussion should show the falsity of the claims that Treasury regulations doing no more than interpreting the statute are legislative in character.  If by incorporating an interpretation in a regulation magically transforms the interpretation into a legislative rule for APA purposes, then the interpretation could not apply independent of the regulation and, as a legislative rule, could not be retroactive.  Legislative characterization of such interpretations for APA purposes is patent nonsense for a lot of other reasons as well.