Improving Regulatory Impact Analysis: The Role of Congress and Courts, Part 3, by Reeve Bull and Jerry Ellig

by Guest Blogger — Thursday, Apr. 11, 2019

Part III—The Role of the Courts

Much as agencies prepare higher quality RIAs when Congress issues relatively specific instructions directing them to perform certain types of economic analysis and to take account thereof when issuing proposed rules, the courts also more carefully scrutinize agencies’ rulemaking records in the face of such specific statutory language.

In the 2018 paper, we consider how courts analyze agencies’ economic analyses prepared under the same set of statutory standards under which we assessed agencies’ RIAs (mandate to select a specific regulatory alternative, requirement to consider specific economic costs or benefits, cost/benefit consideration requirement, economic/technological feasibility standard, and statutory silence concerning costs and benefits).

Our findings at the extreme ends of the spectrum are rather predictable.  Courts very carefully parse agencies’ economic analyses when Congress has directed the agencies either to select a specific regulatory alternative (e.g., the “least burdensome” option) or when Congress has set forth very specific economic benefits and costs that the agency must analyze in formulating its rule.  At the other end, courts defer almost completely to the agencies with respect to their economic factfinding when the governing statute makes no mention of economic costs or benefits but the agency analyzes them nevertheless (whether because it has voluntarily chosen to do so or is required to perform such an analysis pursuant to executive order).

The more interesting findings come in the middle ranges.  When Congress has either directed the agencies to “consider” costs or benefits (without setting forth specific economic costs/benefits that must be assessed or directing the agency to select a specific alternative favored by the economic analysis) or to select an economically and technologically “feasible” approach, the level of judicial scrutiny varies widely from case to case.  In some instances, the courts defer almost completely to the agencies, simply ensuring that the record contains some evidence concerning economic costs or benefits without assessing the quality of the evidence or the cogency of the agency’s reasoning.  In other cases, the courts rigorously assess the quality of the agency’s decisionmaking, sometimes going beyond the statutory language to require things like consideration of a sufficient range of regulatory alternatives.

Agencies, for their part, appear to respond to the incentives created by judicial review. Our econometric analysis finds that a prior appeals court evaluation of an agency’s analysis for a similar regulation issued under the same or a predecessor statute is positively correlated with analysis of alternatives, benefits, and costs.  The econometric results are largely consistent and robust using several different definitions of the prior court decision variable.  And these results are consistent with our examination of cases in an earlier  (2017) paper (law review version available at 69 Admin. L. Rev. 725 (2017)). There we found that when courts remanded regulations to agencies for reasons related to the quality of the agency’s economic analysis, the revised regulation was accompanied by an economic analysis that attempted to remedy the original deficiency.

Though our results do not necessarily prove causation, they are consistent with the existence of a positive feedback loop: courts carefully scrutinize agency analyses conducted under the stricter standards, and agencies then prepare higher quality analyses.  They also suggest that Congress should be scrupulously clear in what it intends agencies to accomplish.  Though a wishy-washy standard such as “cost consideration” may seem like an attractive compromise solution, it is likely to yield results that are disappointing to both proponents and opponents of rigorous judicial review, with some courts applying a standard every bit as strict as that seen under the more rigorous analysis requirements and others deferring almost completely to agencies.  Presumably neither side wants inconsistency.

In addition to imposing targeted economic analysis requirements in certain regulatory programs, Congress can provide for cross-cutting judicial review of agency RIAs.  This may or may not be accompanied by a “super-mandate” that requires agencies to perform some type of economic analysis for all rules or all major rules and to justify those rules in benefit-cost terms (an example is the Regulatory Accountability Act, which requires that the benefits “justify” the costs for major rules).

Even assuming that Congress imposes no such overarching statutory economic analysis requirement, agencies are already required to prepare RIAs pursuant to executive order in many instances, and Congress could explicitly provide that these are reviewable.  In the 2017 paper, we explore what such a standard for judicial review would look like.  We note that the federal appeals courts have reviewed agencies’ economic analyses in a wide array of decisions issued over the last thirty years.  The vast majority of these cases featured some form of statutory economic analysis standard similar to those explored above, but a handful of decisions assessed agencies’ economic factfinding even in the absence of such a statutory requirement

A key argument against judicial review of RIAs is that many judges are generalists who do not have the background necessary to scrutinize an expert agency’s economic analysis.  But our analysis shows that courts have quite capably reviewed agencies’ analysis of the underlying problem, alternatives, benefits, and costs.  There is no clear pro- or anti-regulatory bias in the decisions: the number of cases resulting in an outcome favorable to regulation (i.e., striking down a rule challenged as insufficiently protective or upholding a rule challenged as overburdensome) roughly equal the number resulting in an outcome unfavorable to regulation.

At present, however, courts do not consistently choose to examine an agency’s economic analysis carefully when the analysis is challenged under the “arbitrary and capricious” standard. In the paper, we explore how Congress might amend the Administrative Procedure Act (APA) to clarify that RIAs are reviewable in all instances.  First, we recommend that Congress add a definition of “RIA” to the APA, clarifying that the term encompasses any analysis that assesses the nature and significance of a regulatory problem, analyzes alternatives approaches, and examines the benefits and costs of a proposed rule and its alternatives.  Second, we recommend that Congress amend the APA to explicitly state that an RIA becomes part of the rulemaking record on judicial review and that the court should assess the findings thereof under the “hard look” version of the arbitrary and capricious standard.

In many respects, such a reform would merely codify trends that are already developing in the caselaw.  Courts have assessed agency RIAs even in the absence of a statutory economic analysis requirement (though the relevant executive order provides that its terms are not judicially enforceable), and some scholars have even suggested that some basic benefit-cost analysis requirement is implicit within the APA.  Explicitly providing for judicial review of RIAs would also have the salubrious effect of creating an enhanced role for stakeholders, who would have a greater incentive to submit economic evidence to the agency during the rulemaking process and then to challenge specific rules in court if they believe the agency has reached an irrational result.



Regulatory reform is a worthwhile endeavor, and the U.S. will do well to consider a variety of possible reforms that draw on all three branches of government.  Yet any reforms must be sensitive to the constitutional system of government under which we operate.  For instance, the U.S. will never construct anything nearly so comprehensive as the economic analysis apparatus of the EU Commission, which applies benefit-cost analysis to the equivalent of both U.S. statutes and regulations, for the simple reason that the U.S. Constitution does not mandate that all legislation pass a benefit-cost analysis.  In this light, efforts to recreate international innovations are likely to fall short.

In our recent papers, we explore a variety of reforms that would draw upon the U.S.’s comparative strengths, enhancing the role of stakeholders and the courts to bring greater accountability to agencies.  At the end of the day, all of these reforms depend upon Congress’s reasserting its role as the primary policymaking body, as only Congress can direct agencies to conduct more rigorous economic analyses and the courts to more carefully scrutinize agencies’ rulemaking records.  This may be grounds for pessimism, given the fact that Congress has ceded greater and greater power to the executive branch over the past few decades.

Yet there is reason for hope: perceived executive overreach by both Democratic and Republican Administrations has led to modest but important efforts to reassert Congress’s central legislative role, including Senator Mike Lee’s Article I Project.  Our goal in these papers has been to provide a roadmap by which Congress can reclaim its policymaking role in the regulatory space, providing greater direction to agencies as they assess the economic effects of proposed rules and enlisting the help of the courts and regulatory stakeholders to ensure accountability.


For part 1 of this series, see here. For part 2 of this series, see here.

Reeve Bull is Research Director of the Administrative Conference of the United States. The views expressed in this essay are those of the author and do not necessarily represent the views of the Administrative Conference. Jerry Ellig is a Research Professor at George Washington University’s Regulatory Studies Center.

Cite As: Author Name, Title, 36 Yale J. on Reg.: Notice & Comment (date), URL.

This entry was tagged .

Leave a Reply

Your email address will not be published. Required fields are marked *