*This post is part of a symposium on Modernizing Regulatory Review. For other posts in the series, click here.

A comparison between the current Circular A-4 and the draft revised Circular A-4 publicized for comment by the Office of Management and Budget (OMB) shows that many of the proposed revisions are small textual changes that seem at first blush merely to improve readability or to elaborate, and thus clarify, existing practices. As any good lawyer knows, however, even small textual shifts sometimes make big changes. Such is the case with the discussion of pre- versus post-statutory baselines in Section 4 of the draft revised Circular A-4. 

Meaningful regulatory impact analysis under EO 12866 requires an agency to determine an appropriate baseline for comparison. In both its existing and proposed forms, Circular A-4 calls for a “no-action” baseline, asking what the world will look like if the proposed regulations are not adopted. No-action is not merely synonymous with the status quo, but rather requires an assessment of current and future trends that in turn may be influenced by a variety of different factors. Hence, both versions of Circular A-4 clearly anticipate the need for a certain flexibility in application, with words like “reasonable,” “appropriate,” and “feasible” doing much of the heavy lifting. One presumes as a result that some amount of discussion and negotiation will take place between promulgating agencies and OIRA staff regarding case-by-case details.  So far, so good.

The current Circular A-4 then goes on to offer the following advice:

In some cases, substantial portions of a rule may simply restate statutory requirements that would be self-implementing, even in the absence of the regulatory action. In these cases, you should use a pre-statute baseline. If you are able to separate out those areas where the agency has discretion, you may also use a post-statute baseline to evaluate the discretionary elements of the action. (Emphasis added.)

The draft revised Circular A-4 offers a slightly longer version that initially seems quite similar but is importantly different:

In general, an agency’s first regulatory action implementing a new statutory authority should be assessed in a manner that accounts for the effects of the statute itself—that is, assessed against a pre-statutory baseline. However, in some cases substantial portions of a regulation may simply restate statutory requirements that are self-implementing even in the absence of the regulatory action or over which the agency clearly has little (or no) regulatory discretion. In these cases, you may use a post-statutory baseline in your regulatory analysis, focusing on the discretionary elements of the action and potential alternatives.  Such an analysis should be accompanied by a brief description of and citation to the relevant statute. If you plan to use a post-statutory baseline for a regulation, you should consult with OMB as early as possible in the process of developing your regulatory analysis, including about how to describe—in sufficient detail—the post-statutory baseline that is being used. (Emphasis added.)

In other words, read carefully, where “substantial portions” of a regulation “simply restate statutory requirements” that would be/are “self-implementing,” the revisions to Circular A-4 shift from requiring a pre-statutory baseline with the possibility of additional analysis using a post-statutory baseline to a blanket permission to use a post-statutory baseline alone.  

What does this change accomplish, and why does it matter?  OMB has offered no explanation for the change, so I can only make an educated guess:  OMB has decided to acquiesce to pressure from the Treasury Department, particularly Treasury/IRS, to exclude most of its regulations from OIRA review without the more transparent and politically-controversial act of reversing the 2018 Memorandum of Agreement (MOA) that subjected most Treasury/IRS regulatory actions to the requirements of EO 12866 and OIRA review for the first time.  

As I documented here, prior to the 2018 MOA, pursuant to a pair of memorandum agreements negotiated between OMB and Treasury in 1983 and 1993, virtually all Treasury/IRS regulations were exempt from EO 12866 and OIRA review.  As a result, according to reginfo.gov, Treasury/IRS submitted only 56 regulatory actions to OIRA for review in the almost 40 years between OIRA’s inception in 1981 and the signing of the MOA in 2018. Of those 56 regulatory actions, the vast majority (44, or 78%) occurred prior to 1998. Treasury/IRS submitted no regulations at all to OIRA for review between 1998 and 2010, and only 12 regulations in the decade preceding the MOA. By comparison, in the five years since the 2018 MOA, again according to reginfo.gov, OIRA has submitted 97 regulatory actions to OIRA for review.  

Treasury/IRS does not want so many of its regulations to be subject to the transparency of regulatory impact analysis or the scrutiny of the intra-governmental review process that OIRA facilitates. Hence, to the best of my knowledge and understanding, from the 2018 MOA to the present, OIRA and Treasury/IRS have argued about whether a pre-statutory or post-statutory baseline is appropriate for most tax regulatory actions. The Treasury/IRS position, expressed in the Internal Revenue Manual in the Internal Revenue Manual, is that most tax regulations are interpretative rules “because the underlying statute implemented by the regulation contains the necessary legal authority for the action taken and any effect of the regulation flows directly from that statute.” Also in the Internal Revenue Manual, Treasury/IRS claims that, “[g]enerally, the underlying Internal Revenue Code section imposing the tax or providing for the collection of a tax will provide an adequate legislative basis for the action in the regulations,” such that tax regulations merely “provide a mechanism to implement the Internal Revenue Code provision passed by Congress.” Applying this understanding, if a revised Circular A-4 allows an agency to rely solely on a post-statutory baseline, then many fewer Treasury/IRS regulations would be considered economically significant, and thus subject to OIRA review, than under the pre-statutory baseline requirement of the current Circular A-4. 

Treasury/IRS’s assertion that the effects of most of its regulations come from the statute rather than its own actions is legally untenable for at least two reasons. First, like most contemporary regulatory statutes, the Internal Revenue Code—although often quite detailed—contains hundreds of delegations to Treasury/IRS of discretionary authority to adopt rules and regulations to elaborate and shape statutory requirements and accomplish statutory goals. Some of those delegations are relatively specific, but many of them quite broad, including the authority extended by 26 U.S.C. § 7805(a) to “prescribe all needful rules and regulations for the enforcement of” Title 26. The regulations that Treasury/IRS promulgates pursuant to its statutory rulemaking authority do not merely reiterate or paraphrase statutory text. Rather, as I have explained at greater length elsewhere, Treasury/IRS regulations routinely add copious substantive details that could not be gleaned from the statute using traditional tools of statutory interpretation and reflect discretionary choices by the agency.  In turn, those discretionary choices do not merely add to or subtract from tax collections but also incentivize or disincentive behavioral choices by private parties to whom the regulations apply and, consequently, have broader economic and other consequences.  

Second, as I have explained repeatedly in articles, essays, and amicus briefs (e.g., here and here), Treasury/IRS regulations simply do not qualify as interpretative rules for Administrative Procedure Act (APA) purposes under any contemporary standard.  To summarize briefly, black-letter administrative law holds, and has held for decades now, that legislative rules are legally binding and interpretative rules are not. No one seriously questions that all Treasury regulations are legally binding, whether issued under an express grant of rulemaking power in a particular Internal Revenue Code provision or the general rulemaking grant of 26 U.S.C. § 7805(a). As interpreted by Treasury/IRS, the Internal Revenue Code imposes penalties upon taxpayers and their advisers if they do not comply with Treasury/IRS regulations. The Supreme Court recognized in Mayo Foundation v. United States that Treasury/IRS regulations issued pursuant to § 7805(a) carry the force and effect of law. Lower courts have embraced the Court’s guidance repeatedly since. A different understanding of Treasury/IRS regulatory authority did exist at the time the APA was adopted, but that different understanding was predicated on (1) now-rejected perceptions that the nondelegation doctrine of Article I, §1 of the Constitution did not allow § 7805(a)-type regulations to be legally binding (for more on that, see, e.g., here, esp. pp. 1104-1113), plus (2) very different approaches to the tax legislative process and in how Treasury exercised its general rulemaking authority in the 1940s relative to today. 

No one should think that the shift in draft revised Circular A-4 to a post-statutory default for most tax regulations would be limited to and merely support the traditional revenue-raising function of Treasury/IRS and the Internal Revenue Code. Over the past few decades at least, Congress increasingly has used the Internal Revenue Code, administered by Treasury/IRS, as its preferred vehicle for incentivizing or discouraging certain activities and for pursuing other regulatory and social welfare goals. For example, for fiscal years 2019 through 2023, the Joint Committee on Taxation has identified in the Internal Revenue Code more than 250 separate tax expenditure items across 17 different categories—e.g., Energy; Agriculture; Transportation; Community and Regional Development; and Education, Training, Employment and Social Services—averaging more than $1.6 trillion per year. A sizeable plurality of Treasury/IRS regulation projects (30-40%, according to one five-year study) concern these tax expenditure items and other regulatory and social welfare programs. The resulting regulations bear, at best, only a tangential relationship to the tax system’s traditional revenue-raising function. If the programs and provisions implemented by these regulations were included in statutes other than the Internal Revenue Code and administered by agencies other than Treasury/IRS, no one would question the appropriateness of a pre-statutory baseline and potential OIRA review.  

In summary, OMB should not acquiesce to the dubious claims of Treasury/IRS that the effects of most tax regulatory actions flow directly from the statute, and thus that a post-statutory baseline is appropriate for the tax context. If OMB wants to return to exempting tax regulatory actions from OIRA review and the dictates of EO 12866, OMB should do so transparently and not bury that choice in obscure language regarding pre- and post-statutory baselines in the draft revised Circular A-4. Otherwise, OMB should return to language more closely approximating the expectations of the current Circular A-4, emphatically require agencies to use a pre-statutory baseline for all regulatory actions, and merely allow agencies to offer supplementary analysis using a post-statutory baseline. 

Kristin E. Hickman is the McKnight Presidential Professor in Law, Distinguished McKnight University Professor, and Harlan Albert Rogers Professor of Law at the University of Minnesota. The author of this post also served as Special Adviser to the Administrator of OIRA from 2018 to 2019.