In response to: A Modest Proposal for Reforming the Administrative State
In his provocative Liberty Forum essay, Ilan Wurman proposes a novel solution to the explosive growth of the administrative state. Constitutional conservatives, rather than pursuing their dream remedy (that is, the Supreme Court overturning or severely limiting its prior holdings on the non-delegation doctrine), should accept that agencies will inevitably exercise a blend of executive, legislative, and judicial powers. Once conservatives accept the delegation of legislative power as a political reality, they can seek more realistic measures to curb agencies’ intrusions on the separation of powers. Specifically, they can embrace measures that vest ultimate control of an executive agency’s actions in the branch of government whose power is allegedly being infringed upon under existing frameworks.
Before further considering Wurman’s piece, I should confess that I’m not quite as critical of the current administrative state as he or others might be. I certainly identify with the formalist camp on separation-of-powers questions and agree that, with a few textual exceptions, the Constitution divides the powers of government into three branches, with constitutional violations occurring whenever one branch exercises another’s power, regardless of any functionalist considerations. I consequently celebrate the decision in Immigration and Naturalization Service v. Chadha (1983) striking down the legislative veto, and lament the administrative resolution of cases involving private rights. However, on perhaps the biggest question, or at least the question most relevant to the non-delegation doctrine, I depart from my fellow formalists and am largely comfortable with agency regulations issued under generally worded statutes.
With that in mind, I’m skeptical of one of the first proposals that Wurman explores, under which Congress could essentially turn administrative regulations into proposed rules. Under Wurman’s analysis, we should first accept that agencies really exercise the legislative power when they issue regulations. But rather than raise constitutional concerns and attack the non-delegation doctrine, conservatives should advocate a more realistic measure. For example, under Wurman’s view, conservatives could push for Congress to impose a seven-month waiting period on all agency regulations, or at least major ones. During that period, some or all of the actors involved in passing typical legislation, individually or in unison, could veto the proposed regulations. That is, a single house of Congress could strike the regulations (since the assent of a particular house is needed to exercise the legislative power), or the President himself could retain a veto power (given his constitutionally prescribed role in the legislative process).
Because I consider the promulgation of regulations an executive act, not a legislative one, I would reject this proposal as an improper legislative encroachment on the executive power, and thus one that would actually exacerbate constitutional problems. Additionally, I am not comfortable with this apparent dismantling of Chadha. However, even if I put my constitutional objections aside and focus on practical consequences, I’m skeptical that congressional vetoes of agency regulations, in the manner Wurman prescribes, would further the public interest or serve conservative goals.
Most significantly, the formalized involvement of Congress in the rulemaking process would jeopardize the notice-and-comment procedures in the Administrative Procedure Act of 1946 (APA). Those procedures, I hope fellow conservatives would agree, allow a measure of protection against agency overreach. Perhaps courts have not held agencies as tightly to the requirements of the APA as they should, but the possibility of commenting on proposed actions and challenging an agency in court places some limits on executive actions.
Formalized congressional involvement in agency rulemaking, however, would turn the APA’s notice-and-comment procedures into something of a farce. When an agency proposes regulations and members of the public object to those regulations (or wish to propose alternatives), the public comments, or at least the serious ones, are made with careful regard to the agency’s statutorily described authority. That is, if the authorizing statute, for example, yields three plausible interpretations and the agency proposes to adopt one, an adversely affected party or industry will usually advocate that the agency adopt one of the other interpretations. Or if the implementation of the statute is left to the agency’s discretion, the industry or party may point to the authorizing statute and encourage the agency to abandon the rulemaking project altogether. In either case, the notice-and-comment process is defined by the scope of the statutory language because an agency, or at least a well-behaving one, can be counted on to consider itself bound by that language.
Formalized congressional involvement in rulemaking would upset this framework and undermine the protections afforded to the public by the APA. Unlike an agency, Congress does not create rules by considering competing plausible interpretations of a positively enacted text. Instead, legislators draft rules with regard to a wide range of factors unconstrained by any text. (Put aside the relatively rare circumstance where Congress might push the limits of its constitutional authority.) That is, legislators choose a particular statutory rule not because it reflects a favored interpretation of some higher legal authority, but because it reflects various preferences, political or otherwise.
So what would a house of Congress do, under Wurman’s proposal, when it received proposed regulations from an agency? Would it put on the hat of an executive branch official, and examine whether the proposed language reflected the best interpretation of the authorizing statute? Or would it wear its own hat—that of the legislator—and consider whether the regulations are desirable as a matter of politics or policy? I strongly suspect the latter. In fact, that seems to be the whole point of the proposal: to make agency rules enforceable only through procedures that more closely resemble the typical legislative process, not the notice-and-comment process.
Additionally, although Wurman posits that, under his proposal, legislators would give administrative regulations either a thumbs up (through inaction) or a thumbs down (via a one-house veto), I see no reason why this must be so. Faced with, for example, a 100-page regulation proposal, some parts of which some legislators find objectionable, the Congress as a whole, with the assent of the President, might enact a statute that adopted a revised version of those regulations. Should it do so, Congress would not be bound by the terms of the original authorizing statute—the revised regulations (now a statute) would simply displace or impliedly repeal the earlier enactment. This seems exquisitely unfair to the members of the public who participated in the notice-and-comment process, who advocated for particular results under the constraints of statutory language.
Of course, the President could always refuse to sign the new regulations (statutes) into law on the grounds that the public never had a chance to comment, but would he really do so, if the legislation preserved the bulk of his agency’s proposed guidance? I suspect that if the President had a choice between facing a one-house veto of his proposed guidance or signing into law a bill that adopted (say) 80 percent of that guidance, he would embrace the legislative approach. Members of the public who initially commented on the proposed regulations would have been blindsided by the new provisions or by changes in the statutory framework.
Those skeptical of the notice-and-comment process, I suppose, might not be displeased by such results. That is, rules (now statutes) would more likely come about through the legislative process, not the notice-and-comment process. Yet for those of us who consider the notice-and-comment process superior to legislative lobbying, at least in the sphere of matters traditionally delegated to agencies, something will have been lost. That loss should be acknowledged and weighed by those favoring the “Congress-ification” of agency rulemaking.
Admittedly, even under existing frameworks, Congress can nullify agency rules by enacting superseding statutes or by freezing the notice-and-comment process. In the 1970s, for example, the Internal Revenue Service proposed regulations governing the taxation of so-called fringe benefits, touching off a political firestorm. Congress promptly stepped in and placed a statutory moratorium on IRS rulemaking in this area.
In one respect, this moratorium proved successful in asserting the primacy of the legislature—Congress ultimately passed a statute governing fringe benefits. However, the legislative process being what it is, the moratorium created significant administrative problems because Congress waited nearly a decade to exercise its lawmaking power.
If Congress routinely and actively involved itself in the rulemaking area, I suspect that far more delays would occur and public confidence in the notice-and-comment process would slip. Again, this might be the intention of those looking for the Congress-ification of rulemaking. But I’m concerned that much of the angst over excessive agency rulemaking powers stems from a relatively small percentage of hot-button regulatory projects. For the vast majority of projects, I doubt anything good can come from systematic House or Senate involvement in rulemaking.
Wurman posits that under his proposal, most regulations would ultimately take effect without meaningful congressional interference. The failure (success?) of the Congressional Review Act of 1996 (CRA) might help establish that point, because though that act mandates congressional involvement in some rulemaking, it has rarely caused any delay in agency action. However, the CRA demands action through the normal bicameralism and presentment procedures, whereas under Wurman’s proposal, a single house could veto agency regulations. Under the latter model, I suspect far more gridlock would occur.
Again, I recognize that constitutional conservatives might welcome this gridlock, at least for expansive regulations issued under statutes they are not fond of. But there’s also a possibility that a legislative veto could lead to an increase in executive power. Not all regulations reflect a power grab by the government, after all. For example, if a future IRS issued regulations limiting the scope of tax credits and penalties under the Affordable Care Act, a house of Congress sympathetic to so-called “big government” might veto those regulations. If this happened, I suspect that constitutional conservatives would cry foul, even though technically, under Wurman’s view, the primacy of legislative authority would have been respected. In any case the consequences of the Congress-ification of rulemaking would depend very much on the particular composition of the legislature at any given time.
A rethinking of the administrative state and the non-delegation doctrine could thus lead to a world quite different from the one conservatives intend. Most would expect these new congressional restraints on the executive branch to shrink the administrative state, returning us to a world more closely resembling the pre-New Deal era. But the opposite might very well occur.
Is it possible that this would not bother constitutional conservatives either? Maybe they are concerned solely with the stylistic purity of lawmaking, not its content, and would drop all objections to our current system of government if Congress went ahead and formally enacted every single rule in the Federal Register. However, I suspect that conservative unease with agencies relates not only to the style of rulemaking, but also to the frighteningly large number of administrative regulations governing almost every aspect of our lives. Changing a framework such that Congress simply passed all regulations (now statutes) would do nothing to address these quantitative concerns.
As Wurman continues the research paths he outlines in his fine essay, I hope that some of these practical concerns are considered and appropriately weighed. I look forward to his future scholarly contributions.
The ambitious proposal reconsidering the foundations of the modern regulatory state that Ilan Wurman outlines in his thoughtful Liberty Forum essay is not an outlier. There seems to be a growing call—primarily among conservatives and libertarians—to return to first principles and rein in the administrative state. And I’m not just referring to Philip Hamburger’s condemnation…
Most everyone is familiar with the Hans Christian Andersen tale in which only a child is willing to pronounce what everyone knows: The emperor’s clothes are no clothes at all. As the emperor marches through town, he is as naked as the day he was born. His magnificent new outfit is a fiction, but a…
I would very much like to thank professors Chris Walker, Jonathan Adler, and Andy Grewal for their thoughtful and incisive responses to my “modest” (or perhaps not so modest) proposal for reforming the administrative state. Much of their criticisms, I think, will be addressed in the forthcoming, fuller accounts of this idea, but much of…