Counterpoint: Tax Exceptionalism from Notice-and-Comment Rulemaking Procedures Is Bad Policy and Bad Law

It has been said that democracy is the worst form of government except for all the others. Much the same can be said for notice-and-comment rulemaking and judicial review thereof as an approach to administrative governance. Critics argue that the procedures and deliberative process imposed upon agency rulemaking by the Administrative Procedure Act (APA), Office of Information and Regulatory Affairs (OIRA) review, and the courts are excessive. All that procedure and process takes time, consumes scarce agency resources, and delays agency regulations that we anticipate will accomplish good things. Further, judicial review to enforce agency compliance gives judges the opportunity to impose their own policy preferences. Yet, as political scientist Rachel Potter has observed, “the United States’ notice-and-comment rulemaking process has repeatedly been held up as a model for bureaucratic decision making across the world,” including by the OECD and the World Bank, precisely because its many requirements facilitate oversight and constrain the otherwise broad discretion enjoyed by agency officials.

Echoing administrative law’s critics, David Weisbach maintains that notice-and-comment rulemaking procedures generally add little value in the tax context. He points to the many Treasury regulations that receive few or no public comments. He observes that Treasury and the IRS have other, better ways of getting the information they need. He doubts that judicial scrutiny of regulatory preambles results in better regulations. Accordingly, he argues that courts should back off and interpret administrative law requirements to allow Treasury and the IRS to be more relaxed than other agencies about how they follow those procedures.

Notwithstanding his framing, in the end, David’s argument is not about tax exceptionalism at all. He has made no real effort to actually distinguish tax administration from other areas of administrative governance, and his complaints are not unique to tax. Normatively, much of my reply can be summarized in two brief statements. First, we impose procedural and process burdens on agencies not for the easy cases but for the hard ones. Second, sometimes (perhaps often) Treasury and the IRS (like most agencies) just don’t know what they don’t know. Turning to doctrinal arguments raised in David’s recent article on the same topic, it is true that administrative law relies more on standards than on bright-line rules, thus affording judges a certain latitude in individual cases. The flexibility those standards offer is not so great as David suggests, however, particularly against a backdrop of longstanding Supreme Court policy favoring administrative law uniformity. Unless and until Congress and the courts move the law in a direction more to David’s liking, we instead should encourage Treasury and the IRS to follow the law that governs them—not half-way or in loosey-goosey fashion but fully, in both letter and spirit, just as they expect taxpayers to follow the tax laws.

No one expects that an agency will get all, most, or even its best information from the public notice and comment process. A map of the regulatory process that OIRA features at regulations.gov includes nine separate steps, with publication of a notice of proposed rulemaking and consideration of public comments received in response designated as steps five and six, respectively. Beyond the bar association reports, industry meetings, and so on that David mentions, the interagency review process facilitated by OIRA before proposed regulations are issued, which Treasury nixed for tax regulations last year, can be a useful source of information before proposed regulations are ever issued. Occasionally, an agency will find itself blindsided by negative comments from a direction that it did not anticipate, prompting the agency to withdraw or dramatically alter proposed regulations. Such cases do happen, even if they are the exception rather than the rule. In terms of information gathering, it is better to think of the notice-and-comment process as serving two functions: as a means of identifying and correcting smaller, more technical errors and omissions, and as final confirmation that the agency has not made a big mistake.

Consider Clint Wallace’s 2017 study of 106 proposed Treasury regulations published in 2013, 2014, and 2015. David makes much of Wallace’s finding that roughly a quarter of proposed regulations received no comments and the median proposed regulation received three comments. But the value of notice-and-comment rulemaking is maximized in the hard rulemakings, not the easy ones. Looking carefully at Wallace’s data, six proposed Treasury regulations in a three-year period were sufficiently controversial that they received more than 500 comments. Two of the six—concerning IRC section 501(c)(4) social welfare organization political activity and substantiation requirements for charitable donations—were withdrawn entirely after Treasury received substantial negative feedback in the notice-and-comment process. Treasury and the IRS held onto another of the six proposed regulations, addressing the treatment of foreign reinsurance companies under the passive foreign investment company regime, for several years before withdrawing and replacing them with a new and overhauled set of proposed regulations after the enactment of the Tax Cuts and Jobs Act. In other words, in half of the most commented upon proposed regulations in the three-year period that Wallace studied, public notice and comment seem to have played a central role in stopping Treasury and the IRS from moving forward with seemingly flawed regulations.

Beyond those few dramatic outliers, Wallace’s data showed another 32 proposed regulations over the same three years that received ten, twenty, fifty, or more comments—numbers that could be significant depending upon the scope of the regulations. For that matter, even one, two, or three comments raising minor, technical issues that Treasury and the IRS failed to identify themselves may prove more relevant to an otherwise uncontroversial regulation than hundreds of comments submitted in response to a controversial proposal.

Wallace’s study offered no assessment of the quality of comments received or whether they resulted in actual changes to the proposed regulations. Wallace was more interested in the commenters’ identities than the content of their comments. As to that, although I can’t remember which proposed regulation I was studying, I was struck some years ago by a comment submitted by (as I recall) a random, small-firm CPA from Houston suggesting clarification of one particular sentence in the proposed regulation that, as worded, would have potential unintended consequences given a particular set of facts that one can only presume were associated with one of his clients. Treasury makes changes to proposed regulations in response to comments like that all the time. Maybe David counts that CPA as an “insider” with other methods of communicating his views to Treasury and the IRS. Having been a small firm CPA once myself, I am more skeptical.

A big part of the exceptionalism debate that David ignores is the extent to which Treasury and the IRS are asked to draft regulations outside their core competency of revenue raising and thus must look beyond their usual sources to glean the information they need. In 2010, former Assistant Secretary of the Treasury for Tax Policy Pamela Olson observed that tax administrators are now responsible for “policies aimed at the environment, conservation, green energy, manufacturing, innovation, education, saving, retirement, health care, child care, welfare, corporate governance, export promotion, charitable giving, governance of tax exempt organizations, and economic development, to name a few.” My own study of Treasury regulation projects from 2008 through 2012 found that roughly 30% to 40% of tax regulatory activity concerned social welfare and regulatory matters with only a tangential relationship to revenue raising. Wallace’s article described the bulk of comments received from public interest groups in 2013, 2014, and 2015 as responding to such “nontax” proposed regulations. Tax Cuts and Jobs Act implementation included at least two particularly high-profile projects meeting this description: charitable contributions in exchange for state and local tax credits and IRC section 1400Z opportunity zones. Treasury and IRS efforts to develop and adopt regulations implementing the Inflation Reduction Act’s critical mineral and battery component requirements for the electric vehicles (EV) credit have been especially chaotic: over the past few years, Treasury and the IRS have alternately published, updated, supplemented, withdrawn, and replaced various subregulatory guidance documents and notices of proposed rulemaking as they struggle to navigate not only industry concerns and congressional expectations regarding the EV credit, but also foreign relations with trading partners. Other recent proposed Treasury regulations have concerned topics as disparate as clean hydrogen production, low income housing, and domestic production of semiconductor and semiconductor manufacturing equipment. The clean hydrogen production proposed regulations generated more than 25,000 comments, while most of the rest received at least a few or several dozen. Treasury and the tax bar may have sufficient expertise to evaluate the impact of these regulations on the fisc, but they have limited expertise and perspective on the broader economic, social, public health, and environmental consequences of these proposed policies.

Regardless of the above, providing an agency with information to support its regulation drafting efforts is only one of the many purposes—and arguably not even the most important one—of notice-and-comment rulemaking procedures. Just as significantly, if not more so, notice-and-comment rulemaking is about transparency and accountability—and by the latter, I mean accountability through judicial review for compliance with the law as well as political or public oversight. As with most federal agencies, Congress has given Treasury and the IRS tremendous discretion over how to implement and administer the tax laws. The Internal Revenue Code contains hundreds of specific delegations of rulemaking power as well as the broad, general authority in IRC section 7805(a) to adopt “all needful rules and regulations.” These delegations permit Treasury and the IRS not only to increase or reduce individual tax liabilities but also to narrow or expand benefit eligibility requirements, incentivize or discourage private party behavior, and impose or alleviate tax regulatory burdens. Congress adopted APA procedural requirements not only to provide for public participation in the rulemaking process but also to constrain the exercise of agency discretion through transparency and accountability. In the same statute, Congress set up the courts to be primarily responsible for ensuring that agencies complied.

Merely accepting Treasury and IRS assertions that they obtain the information they need by consulting with bar associations, industry groups, and other insiders and thoroughly considering that input is nearly the opposite of the transparency and accountability the APA envisions. Since the 1970s, the courts have interpreted APA section 553 notice-and-comment procedures, sometimes in conjunction with the APA section 706(2)(A) arbitrary and capricious standard, as requiring agencies to address in their regulatory preambles significant comments received in response to notices of proposed rulemaking. The Supreme Court has recognized this interpretation without question, as has the U.S. Tax Court. Correspondingly, the Tax Court has acknowledged, and no court questions, that Treasury regulations must demonstrate reasoned decision making contemporaneously and in writing to satisfy the APA § 706(2)(A) arbitrary and capricious standard as interpreted by the Supreme Court in Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Insurance Co. and its progeny. According to those cases, the APA requires agencies to identify and justify their discretionary choices in their regulatory preambles. A symbiotic relationship exists between these requirements, in that comments and responses in regulatory preambles together highlight for a reviewing court important aspects of the problems a regulation is attempting to resolve and factors that the agency considered or failed to take into account in making its choices, as well as concerns about whether the agency has exceeded its statutory authority, all of which an agency is supposed to address in its regulatory preambles. By making agencies show their work contemporaneously in this way, courts are able to ensure that agencies take the notice-and-comment process seriously and do not exercise discretionary power arbitrarily.

Judicial assessment of the compliance of regulatory preambles with APA sections 553 and 706(2)(A) is highly dependent upon the specific record and the facts and circumstances of the particular regulatory action at issue. I expect that Treasury and the IRS will win some such challenges and lose others. I suspect that many Treasury regulation preambles are probably good enough to pass muster. Treasury and the IRS have both pursued their own version of notice-and-comment rulemaking all along, addressing at least some comments, even if somewhat sparsely. Many regulations have received few or no comments, making their preambles relatively straightforward. The above-described standards for APA compliance really are not that high a bar. Treasury and the IRS have ramped up their explanatory efforts in recent years in response the post-Mayo Foundation trend in judicial review of tax regulatory actions. Still, even David has acknowledged that some Treasury regulations might have difficulty meeting the requirements of APA sections 553 and 706(2)(A), as demonstrated by recent cases. Most agencies reacted long ago to the jurisprudence interpreting the APA’s procedural requirements by expanding their preambles to respond to more comments, identify areas of discretion, and justify their choices, rather than expose their regulations to increased litigation risk. Until relatively recently, Treasury and the IRS did not do that, preferring instead to speak in generalities. At least through 2012, the Internal Revenue Manual expressly instructed IRS regulation drafters not to discuss alternatives or justify their choices. As Patrick Smith has written, Treasury and IRS policies in this regard suggest the potential for a more widespread and systematic vulnerability of Treasury regulations to APA process and procedure challenges.

David questions whether so much judicial scrutiny improves regulatory quality. The quality of a regulation often is in the eye of the beholder, so objectively demonstrating the contribution of these requirements to regulatory quality is difficult. Government and academic reports and studies have recognized, however, that economic analysis under Executive Order 12,866 and similar statutory provisions requiring consideration of regulatory alternatives and the benefits and costs of each contributes meaningfully to the quality of the decision making process, particularly when agencies engage in such analysis early and often in developing regulations. Many of those same reports and studies acknowledge that economic analysis of that sort overlaps considerably with judicial expectations under APA section 706(2)(A) and State Farm. To the extent that David is concerned about inefficiency or delay, empirical research suggests that the more standard procedural approach may actually speed up the policymaking process for most rules. At a minimum, for regulation projects that raise more complicated and politically controversial issues (like the ongoing effort to implement the Inflation Reduction Act’s critical mineral and battery component requirements for the EV credit, mentioned above), the more ad hoc approach that Treasury and the IRS have followed for gathering information and issuing guidance piecemeal does not seem to avoid procedural slowdowns.

Perhaps the best argument for encouraging Treasury and the IRS to embrace adherence to APA notice-and-comment rulemaking procedures and reasoned decision-making requirements, rather than a more grudging and minimal compliance, is that they are the law. It is axiomatic that federal government agencies are bound to follow the laws that govern them. Among government agencies, Treasury and the IRS ought to be particularly sensitive to the importance of following the APA’s rules care. Treasury and the IRS expect taxpayers to comply with both the letter and the spirit of the tax laws; the IRS is not known for its leniency if a taxpayer means well but fails in that regard. The average taxpayer may not pay too much attention to the niceties of Treasury and IRS noncompliance with the APA. But it is rather hypocritical for Treasury and IRS officials to scorn tax shelter promoters and tax scofflaws for looking for loopholes and twisting the letter of the law to their own ends when Treasury and the IRS too often approach APA compliance similarly. It is truly unfortunate that many taxpayers regard particularly the IRS and its administration of the tax laws with skepticism, suspicion, and even disdain, to the detriment of the tax system. Under such circumstances, one might reasonably expect Treasury and the IRS to hold themselves to a higher standard.

In this regard, I want to address one particular set of doctrinal arguments from David’s longer article on this topic. David insists that the past Treasury and IRS practice of issuing temporary Treasury regulations without pre-promulgation notice and comment, and usually without claiming good cause, complies with the APA. I will stand mostly on past work in disagreeing with David’s arguments, but three strike me as worth a brief comment here: that many temporary Treasury regulations “might” be exempt from notice-and-comment procedures as interpretative rules, that IRC section 7805(e) authorizes this practice, and that the Supreme Court’s decision in Little Sisters of the Poor v. Pennsylvania excuses any perceived defect in the practice as harmless error.

The line between legislative rules and nonlegislative ones is notoriously “enshrouded in considerable smog.” In suggesting that many temporary Treasury regulations might be interpretative rules, David principally focuses on case law. But since the Supreme Court’s 2011 decision in Mayo Foundation, no court has held—indeed, I do not believe a single judge has contended, even in concurrence or dissent—that any Treasury regulation, temporary or final, is an interpretative rule exempt from notice-and-comment rulemaking requirements. The U.S. Tax Court has repeatedly rejected such arguments using reasoning that would apply to virtually any Treasury regulation. The Internal Revenue Manual continues to insist that most Treasury regulations are interpretative rules on grounds that Congress makes all of the major policy choices and leaves the IRS with little discretion except for minor gap filling. Given the hundreds of rulemaking delegations in the Code, the tremendous discretion they confer upon Treasury and the IRS, and the extensive real-world impact of Treasury and IRS regulatory choices, the courts’ repeated rejection of claims that Treasury regulations are interpretative rules is understandable. If Congress truly made the major policy choices and left Treasury and the IRS with only a minor gap-filling role, our arguments over the substantive content of Treasury regulations would be substantially less intense. Notably, neither the IRS nor the Department of Justice seems inclined these days to assert in court that Treasury regulations are interpretative rules, I expect because they know a losing argument when they see it and prefer to avoid even more explicitly negative precedent.

Instead, the litigation over the legislative/interpretative line has shifted to evaluating the proper characterization of at least some IRS subregulatory guidance. Relying on the Sixth Circuit’s analysis in Mann Construction, Inc. v. United States, several courts have found IRS Notices regarding listed transactions to be legislative rules subject to notice-and-comment rulemaking procedures. In Bullock v. Internal Revenue Serv., a federal district court reached the same conclusion regarding a revenue procedure that altered exempt organization donor reporting requirements. I do not mean to argue that every IRS subregulatory pronouncement is or should be subject to notice-and-comment rulemaking requirements, although I have articulated my concerns about litigation exposure as well as my suggestions for fixing the problem elsewhere. But given the now-extensive post-Mayo Foundation body of jurisprudence holding otherwise, continuing to claim that even a meaningful subset of Treasury regulations might qualify as interpretative rules is wishful thinking that effectively (if unintentionally) encourages Treasury and IRS officials who would rather not pay so much attention to notice-and-comment rulemaking procedures to disregard what the courts are telling them. It is time for proponents of tax exceptionalism from general administrative law requirements, doctrines, and norms, as well as Treasury and the IRS, to move past this line of argument.

Courts also have rejected, admittedly less frequently, the argument that IRC section 7805(e) exempts temporary Treasury regulations lacking a valid claim of good cause from pre-promulgation notice-and-comment rulemaking procedures, with no court to my knowledge reaching the opposite conclusion. Again, the government is able to avoid potential negative precedent on this issue by hurriedly finalizing any temporary Treasury regulation so challenged and then move for dismissal on mootness grounds. Read maximally (as David does), the Supreme Court’s conclusion in Little Sisters of the Poor v. Pennsylvania regarding the procedural validity of a Department of Health and Human Services interim-final regulation suggests that Treasury’s past use of temporary regulations (with notice and comment after the fact, and without a valid, contemporaneous good cause claim) might pass judicial muster as harmless error. In fact, my initial reaction to the Little Sisters decision was similar to David’s, albeit with more concern. Yet, the government tried David’s argument in defending the IRC section 245A temporary regulations in the Liberty Global litigation and lost that issue. Lower courts have continued to invalidate and enjoin other agencies’ temporary or interim-final regulations for lack of notice and comment or good cause for forgoing those procedures. As I and other administrative law scholars have written elsewhere, good reasons exist for interpreting the Little Sisters decision more cautiously than David does.

Administrative law is a dynamic field with many open questions and near-constant, high-stakes litigation. Whereas the federal tax laws are very rule-oriented, the APA and administrative law jurisprudence more often use broad and messy standards and principles like “the force of law” or “arbitrary and capricious” that are not easily reducible to clear instructions for either courts or agency regulation drafters. Within established parameters, reasonable minds can and do differ over the meaning, scope, and applicability of statutory and doctrinal administrative law requirements as well as whether those requirements are the best way to achieve congressional and public policy goals. Meanwhile, we all aim to achieve a system of tax administration that is effective, efficient, and fair. I firmly believe that tax administration will be improved by learning from other agencies and embracing general administrative law requirements, doctrines, and norms, particularly as Congress continues to turn to Treasury and the IRS to implement and administer social welfare and regulatory programs with only a tangential relationship to revenue raising. Nevertheless, I welcome ongoing discussion and debate over these issues.