Policymakers and administrators periodically revise or jettison rules, enforcement priorities, and agency structures for a variety of reasons, from resource constraints to changes in administration. This is particularly the case when presidential administrations change, as evidenced, for example, by the transition from President Carter to President Reagan. 1 See, e.g., Abner J. Mikva, Deregulating Through the Back Door: The Hard Way to Fight a Revolution, 57 U. Chi. L. Rev. 521, 521 (1990) (noting that President Reagan’s program to eliminate regulations initiated soon after he took office and that “[b]efore six months of the Reagan presidency had passed, 180 regulations had been withdrawn, modified, or delayed”). As the current U.S. presidency undergoes one of the most monumental political changes in recent history, proposals to dramatically cut and revoke public laws—not just revise them—have taken center stage. 2 See, e.g., Current Unified Agenda of Regulatory and Deregulatory Actions, Office of Info. & Regulatory Affairs, [http://] (last visited Aug. 10, 2017) (describing the Trump Administration’s approach to reduce regulation).
Health care, immigration, and environmental regulations are among areas slated to receive the most attention, and executive action in these areas has already commenced. 3 See, e.g., Exec. Order No. 13,783, 82 Fed. Reg. 16,093, 16,094–96 (Mar. 31, 2017) (rescinding four climate-related Obama Administration executive orders, withdrawing technical support documents on measuring the social cost of carbon, directing the Secretary of the Interior to withdraw a coal-leasing moratorium on federal lands, and requiring review of various environmental regulations to determine whether to “revise or withdraw” them); Exec. Order No. 13,769, 82 Fed. Reg. 8977, 8978 (Feb. 1, 2017) (suspending entry of immigrants and nonimmigrants from certain countries into the United States for ninety days); Exec. Order No. 13,765, 82 Fed. Reg. 8351, 8351 (Jan. 24, 2017) (directing all agencies with responsibility under the Affordable Care Act to “exercise all authority and discretion available to them to waive, defer, grant exemptions from, or delay the implementation of any provision or requirement of the Act” that imposes fiscal burdens on various entities). President Trump has also indicated a desire to severely slash funding for agencies like the Environmental Protection Agency (EPA). 4 Office of Mgmt. & Budget, America First: A Budget Blueprint to Make America Great Again 41–42,
fy2018/2018_blueprint.pdf [] (showing substantial proposed cuts to the EPA’s budget)(last visited Aug. 10, 2017).

Even without this dramatic shift, there are often fewer partisan driv­ers of regulatory change, with agencies already making difficult choices about prioritizing certain regulations and their enforcement over others. 5 See, e.g., EPA, Unfinished Business: A Comparative Assessment of Environmental Problems Overview Report, at xiii (1987),
2001635G.PDF?Dockey=2001635G.PDF [] (ranking thirty-one environmental problems addressed by the EPA with respect to four different risks—cancer, non-cancer, ecological, and welfare—but not immediately changing regulatory priorities based on this assessment); Nat’l Research Council, Risk Assessment in the Federal Government: Managing the Process 89–93 (1983),
366/chapter/1 [] (describing risk assessment and risk management conducted by agencies such as the Occupational Safety and Health Administration and the Food and Drug Administration).
But there is a threat that if regulatory reform rapidly progresses, typical regulatory capture problems will prevail. 6 The term “capture” in this Piece refers to excessive influence by small, relatively wealthy, organized stakeholders at the expense of a diffusive public that would collectively have more to gain from certain regulations but lack the resources to overcome the transaction costs of organizing and having an adequate voice in the process. For a similar definition, see, e.g., Michael A. Livermore & Richard L. Revesz, Regulatory Review, Capture, and Agency Inaction, 101 Geo. L.J. 1337, 1342–43 (2013) (defining capture as occurring “when organized groups successfully act to vindicate their interests through government policy at the expense of the public interest” and defining “policies that run counter to the public interest” as being “difficult to defend to an informed and neutral observer” on a number of grounds). For an analysis of both “insidious” and more “innocent” versions of capture, see David B. Spence, Federalism, Regulatory Lags, and the Political Economy of Energy Production, 161 U. Pa. L. Rev. 431, 466–67 (2013). Substantial regulatory changes could proceed at the expense of dispersed interests that collectively have much to lose from certain changes—such as important public health and environmental protections—but individually lack the motivation or re­sources to overcome organizational transaction costs. 7 Many members of the Trump Administration are former lobbyists from powerful, wealthy corporations. For example, Donald Trump’s energy advisory team consisted largely of lobbyists for some of the largest fossil fuel companies, such as Halliburton and Devon Energy. See Steven Mufson, Trump’s Energy and Environment Team Leans Heavily on Industry Lobbyists, Wash. Post (Sept. 29, 2016),
2016/09/29/6eb7a2a6-84ec-11e6-ac72-a29979381495_story.html []. For more discussion of the capture threat, and for brief counterarguments, see infra Part II.

Numerous scholarly accounts have explored regulatory transitions, 8 See, e.g., Jack M. Beermann, Presidential Power in Transitions, 83 B.U. L. Rev. 947, 948–53 (2003) (describing the regulatory actions taken by outgoing Presidents and reactions from incoming administrations). dynamism, 9 For a recent account of dynamism and a summary of the dynamism literature, see generally David L. Markell & Robert L. Glicksman, Dynamic Governance in Theory and Application, Part I, 58 Ariz. L. Rev. 563 (2016). and adaptation, 10 See, e.g., Lynn E. Blais & Wendy E. Wagner, Emerging Science, Adaptive Regulation, and the Problem of Rulemaking Ruts, 86 Tex. L. Rev. 1701, 1701–04 (2008) (exploring the obstacles to agencies’ incorporation of new information on science and technology into rulemaking, and proposing changes to both administrative procedures and judicial review of these procedures). and the process that accompanies regulatory change. 11 See, e.g., Merrick B. Garland, Deregulation and Judicial Review, 98 Harv. L. Rev. 505, 511–12 (1985) (describing the judicial review that follows regulatory change during transitions); J.B. Ruhl & James Salzman, Regulatory Exit, 68 Vand. L. Rev. 1295, 1321 (2015) (discussing deregulation as one form of exit). But somewhat unpredictable and rapid shifts in regulation, enforcement priorities, and agency funding might require a fresh look at old tools for evaluating and prioritizing regulations and other agency programs. 12 For accounts of the EPA’s and other federal agencies’ use of various risk-assessment and risk-ranking tools starting in the 1980s, see Sandra A. Hoffmann, Getting to Risk-Based Food Safety Regulatory Management, in Toward Safer Food: Perspectives on Risk and Priority Setting 3, 8–9 (Sandra A. Hoffmann & Michael R. Taylor eds., 2005). See generally Worst Things First?: The Debate over Risk-Based National Environmental Priorities (Adam M. Finkel & Dominic Golding eds., 1994) (providing different authors’ views shared at a conference that addressed various approaches to comparative risk assessment and describing historical national approaches to setting regulatory priorities). For a summary and analysis of proposals for the use of risk assessment in agency regulatory decisionmaking, see David A. Wirth & Ellen K. Silbergeld, Risky Reform, 95 Colum. L. Rev. 1857, 1858–60 (1995) (book review). Agencies have long experimented with various approaches to comparatively assessing risks and associated regulations, 13 See, e.g., EPA, supra note 5; Adam M. Finkel, Should We—and Can We—Reduce the Worst Risks First? in Worst Things First?: The Debate over Risk-Based National Environmental Priorities, supra note 12, at 3, 9 (describing the Office of Management and Budget’s policy in 1991 that encouraged agencies to rank regulations based largely on the cost per life saved); Hoffmann, supra note 12, at 8–9 (describing the EPA’s use of risk assessment and “risk ranking,” which included relatively extensive public participation in addition to scientific analysis in order to establish agency priorities in the 1980s and 1990s). and the need for renewed analysis, potential modification, and application of these approaches seems particularly strong during periods of potentially ex­pansive regulatory reform. 14 As an example of promised regulatory cuts, President Trump has issued an Executive Order directing agencies to cut two regulations for every one they implement. Exec. Order No. 13,771, 82 Fed. Reg. 9339, 9339 (Feb. 3, 2017). Although Administrative Procedure Act challenges to agency actions that carry out this order are likely, and likely to be successful in some cases, the aim of the current Administration to cut regulations is quite strong.

This Piece lumps the process of reviewing and prioritizing regu­lations into a broad term called “regulatory triage” and argues that this is an important time for deploying this type of system, although with great caution. Changes to rules and the agencies that write and enforce them must progress in a manner that takes into account the values of a diverse set of stakeholders, including the individuals who could be the most negatively impacted by the loss of regulatory protections. 15 See, e.g., Mark Seidenfeld, Empowering Stakeholders: Limits on Collaboration as the Basis for Flexible Regulation, 41 Wm. & Mary L. Rev. 411, 411–12 (2000) (noting that proponents of “reinvention” of the regulatory process argue that “groups of individuals who share more diffuse interests in the regulatory endeavor than do regulated entities—public interest groups representing purported beneficiaries of the statutes authorizing regulation—must be given equal power in the regulatory mechanism”); id. at 414–26 (describing concerns that failing to adequately incorporate public interest groups in the regulatory process can lead to capture, make it less likely that the public will accept the resulting regulation, and create regulatory uncertainty through lawsuits and other challenges). But see id. at 413 (concluding that efforts to empower stakeholders have failed to create a “truly collaborative regulatory process”). Although the sausage-making process of politics infuses many values into public law, 16 Indeed, there is a well-known debate as to whether the process and its results primarily involve deal making and negotiation on the one hand or reasoned, thoughtful, publicly interested deliberation on the other hand. statutes and regulations—at least as written—are in large part designed to produce public goods, fix commons-based problems such as unfet­tered air pollution, and address externalities. 17 See, e.g., 33 U.S.C. § 1251(a)(1) (2012) (setting the ambitious national goal in the Clean Water Act “that the discharge of pollutants into the navigable waters be eliminated by 1985”); 42 U.S.C. § 7401(b)(1) (2012) (identifying one purpose of the Clean Air Act as “protect[ing] and enhanc[ing] the quality of the Nation’s air resources so as to promote the public health and welfare and the productive capacity of its population”); cf. John S. Moot, Economic Theories of Regulation and Electricity Restructuring, 25 Energy L.J. 273, 278 (2004) (“A public interest theory [of regulation] starts from the uncontroversial normative proposition that regulation should occur when necessary to address ‘market failures’ such as natural monopoly and externality (social costs).”). Politicians and scholars debate the costs and benefits of these rules and how much law, as opposed to reliance on markets, there should be. 18 See, e.g., Ronald K.L. Collins & David M. Skover, The Future of Liberal Legal Scholarship, 87 Mich. L. Rev. 189, 195 (1988) (characterizing the liberal–conservative divide by positing that “liberals believe that unfettered economic markets are limited in their ability to serve the public welfare” in contrast with the “conservative creed of governmental noninterference in private economic and social choices,” among other differences). But many would agree that correcting market failures and providing basic public protections are at least two purposes of existing public laws. Indeed, after the Flint, Michigan, crisis, in which an emergency manager changed a bankrupt city’s water supply in violation of federal drinking water laws and poisoned thousands of children with lead, 19 See Flint Water Advisory Task Force, Final Report 21 (2016), http:// [] (describing several tests showing an increase in blood lead levels in children living in Flint since April 2014).
politicians from both sides of the aisle expressed dismay at the failure of national, state, and local officials to preserve a basic human right for the city’s residents. 20 See Hearing: Examining Federal Administration of the Safe Drinking Water Act in Flint, Michigan, House Comm. on Oversight & Gov’t Reform (Feb. 3, 2016), http:// [] (detailing the full Committee’s criticism of agencies within several levels of government for their role in the Flint water crisis); see also Hannah J. Wiseman, Delegation and Dysfunction, 35 Yale J. on Reg. (forthcoming 2017) (manuscript at 7) (on file with the Columbia Law Review) (discussing the Flint crisis as an example of an environmental issue that seems to garner bipartisan attention).
And President Trump, although voicing conflicting views on the matter, has indicated a basic desire to protect clean air and water. 21 President Donald Trump, Remarks at Signing of Executive Order to Create Energy Independence (Mar. 28, 2017),
03/28/remarks-president-trump-signing-executive-order-create-energy [
L8ZB-KHB8] (“And we’re going to have safety, we’re going to have clean water, we’re going to have clear air.”).

This Piece proposes that as the next wave of political reform sweeps the country, agencies should build from previous lessons regarding comparative regulatory assessment and form stakeholder groups tasked with prioritizing the preservation of rules, agency offices, and enforce­ment efforts that address the greatest societal problems. Agencies should take into account these recommendations as they change rules, offices and staffing, and enforcement priorities.

Any regulatory triage approach must proceed with great caution, however. The last thing on the minds of most agency officials facing further cuts to an already understaffed and underfinanced office is a lengthy process that will only consume more time and money, taking away valuable resources from the most pressing problems. Triage threat­ens to drain these limited resources and to further ossify agencies’ responsibilities for addressing risk. 22 See Thomas O. McGarity, Some Thoughts on “Deossifying” the Rulemaking Process, 41 Duke L.J. 1385, 1401 (1992) (noting how judicial remand of agency action can “send the project spinning off in odd directions or, worse, can consign it to oblivion” due to “limited staff resources”). Thus, as discussed in the triage procedures portion of this Piece, any triage approaches adopted should be relatively streamlined; they should focus on ensuring that a diverse group of stakeholders has a seat at the table but is not unmanageable in terms of size, they should rely as much as possible on existing resources that have analyzed agency programs and the risks that these programs address, and they should be formed with relatively short timelines and without requirements for lengthy reports or other overly time-consuming products or processes. It is not the aim of this Piece to propose a burden­some process that will only further bog down agencies that already have too few resources to work with.

Part I of this Piece proposes how agencies should consider approach­ing the prioritization of regulatory reforms (“regulatory triage”), building from previous prioritization efforts of agencies like the EPA and the literature on comparative risk assessment and similar modes of roughly ranking agency functions. Part II then describes the formal and informal methods that agencies can deploy to form advisory groups and solicit triage recommendations from them, and it proposes factors that agencies should consider in forming groups’ membership. Finally, Part III explores existing laws that already require or support this type of prioritization and that prevent an all-out slashing of regulations without certain proce­dural steps.

I. Models for Regulatory Triage

In light of potential modifications to regulations in numerous policy areas, this Part first describes and defines regulatory triage and then explores models that could inform the substantive questions that agencies might consider addressing within triage. This Piece aims to call attention to the importance of regulatory triage generally and certain aspects of the triage process that might be the most critical rather than to suggest one particular triage approach; it accordingly explores a variety of potential models that could be instructive. Additionally, it bears emphasis at the outset that the Piece does not propose that triage regularly occur; the aim is to suggest a mechanism for agencies facing mandates for cuts. To the extent that procedures like triage can be avoided, this is beneficial; it will save agencies valuable time and resources and will allow them to focus these resources on protecting the public. To the extent that agencies must make sacrifices, however, triage can potentially lessen the damage that may result.

A. Defining Triage

The Piece defines triage as a process in which an agency—if it faced orders to cut staff or regulations or reform enforcement priorities, among other changes—would attempt to devise a plan for prioritizing the regulations, staff, and enforcement priorities to retain. Triage is both substantive and procedural. At the substantive level, most generally speaking, the triage proposed in this Piece involves evaluating regu­lations, enforcement, and other agency activities and comparing their importance. The triage described here is not a formal comparative risk assessment, which is a “systematic attempt at comparing risks in an effort to set environmental priorities” 23 Ralph M. Perhac Jr., Comparative Risk Assessment: Where Does the Public Fit In?, 23 Sci., Tech., & Hum. Values 221, 222 (1998); see also Richard A. Minard Jr., CRA and the States: History, Politics, and Results, in Comparing Environmental Risks: Tools for Setting Government Priorities 23, 24 (J. Clarence Davies ed., 1996) (“Any comparative risk project attempts to answer two fundamental questions by implementing a [comparative risk assessment]: What are the most serious environmental problems here? How can we most effectively address them?”). and which, in its “hard” form, involves quantifying risks with an assumption that they are fungible and can be directly compared. 24 Jonathan Lash, Integrating Science, Values, and Democracy Through Comparative Risk Assessment, in Worst Things First?: The Debate Over Risk-Based National Environmental Priorities, supra note 12, at 69, 75. The triage described here asks that agencies instead consider numerous factors when attempting to determine which pro­grams should be trimmed the least if trimming is mandated. The risk addressed by agency programs is one of these factors, but others include, at minimum, the agencies’ statutory mandates and questions such as which programs most effectively address risks as well as those that address risks to populations that experience disproportionate harm. 25 See Robert D. Bullard, Unequal Environmental Protection: Incorporating Environmental Justice in Decision Making, in Worst Things First?: The Debate Over Risk-Based National Environmental Priorities, supra note 12, at 237, 238–41 (describing disproportionate environmental impacts in minority communities).

Proposals that build from comparative risk assessment provide exam­ples of the many substantive issues—including identification of the most important risks and the most effective solutions to those risks—that can be part of the triage process. In the food safety context, Peter Nelson and Alan Krupnick suggest specific action items involved in this type of prior­itization, which include the following:

Identify current budgetary priorities in need of change and new problems to be considered, using a [comparative risk assessment] approach with public participation.

Choose the solution-based tools, for example, cost-benefit analysis . . . .

Choose the policy options (solutions) to consider for each pro­gram element.

Evaluate each of the program elements using the tools chosen.

Develop final priorities based on technical analysis and other qualitative factors. 26 Peter Nelson & Alan J. Krupnick, Best Things First: Rethinking Priority Setting for Food Safety Policy in Toward Safer Food: Perspectives on Risk and Priority Setting, supra note 12, at 180, 185–86.

This Piece does not propose that the triage process should involve a formal comparative risk assessment or even this broader, yet somewhat specific, approach outlined by Krupnick and Nelson. Within a harried environment of political change, and one in which officials are facing directives to cut, not add, it is too much to ask of agencies to compre­hensively address each and every portion of their programs under a particular metric, such as cost-benefit analysis. Further, formal compar­ative risk assessment has serious limitations, including concerns that in quantifying risks it fails to incorporate other important regulatory values. 27 Lash, supra note 24, at 69–70, 74–76. But certain basic substantive factors seem important if agencies are pressured to cut regulations, budgets, and enforcement priorities—and even entire offices within agencies, such as data-gathering offices—in a relatively rapid fashion, as discussed in the following section.

B. The Substance of Triage

As defined in section I.A, regulatory triage would involve, at a minimum, an agency reviewing its statutory mandates and attempting to identify the most important risks it is tasked with minimizing and mitigating, effective programs to address these risks, and programs that affect certain populations more than others—particularly programs that benefit communities that have long experienced some of the largest risks addressed by the agency.

1. Statutes. — Agencies face legal requirements within the statutes that they implement, and, absent congressional amendment, agencies may not ignore these requirements regardless of the executive directives that they face. 28 See, e.g., In re Aiken County, 725 F.3d 255, 257, 259–65 (D.C. Cir. 2013) (concluding “[o]ur more modest task is to ensure, in justiciable cases, that agencies comply with the law as it has been set by Congress” and rejecting various attempts by the independent agency at issue to ignore Congress’s requirements on the basis of directives from the President). In the environmental context, these statutes often direct agencies to control specific types of harm to a certain degree. For exam­ple, with respect to air quality, the EPA must regulate emissions that in the agency’s judgment “cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare.” 29 42 U.S.C. § 7408 (2012). Agen­cies may not simply jettison regulations and programs that would end the regulation of those pollutants already deemed to endanger public health, although there is talk of the Trump Administration attempting to withdraw certain endangerment findings—specifically, the carbon finding 30 See Dino Grandoni, The Energy 202: What Would Be the Point of Pruitt’s “Red Team-Blue Team” Climate Exercise?, Wash. Post (July 3, 2017),
?utm_term=.9fcc8229cdb6 [] (reporting that EPA Administrator Scott Pruitt has “expressed interest in the idea of formally challenging the scientific consensus that human activity is warming the planet”).
—in order to avoid this regulatory mandate. Any triage process must take into account statutory mandates so long as those mandates remain in place.

2. Risks. — Beyond statutory mandates, the extensive comparative risk assessment literature and agencies’ past efforts to prioritize regu­latory programs include attempts to determine the most pressing risks addressed by agencies and to somehow rank these risks, 31 See, e.g., EPA, supra note 5, at xiv (noting that the assessment focused only on risk despite numerous other potential means of prioritizing agency action); Hoffmann, supra note 12, at 8 (noting by the end of the 1980s “risk pervaded the discussion of administration of environmental policy”). even if only in a rough way. This makes sense. As a 1987 comparative risk assessment by the EPA notes, “[t]he unit cost of moving ever closer to the point of zero discharge, zero contamination, and zero risk increases exponentially,” and thus the agency must locate the most important risks remaining while also staying true to its statutory mandates. 32 EPA, supra note 5, at i. The report further observes that “[i]n a world of finite resources, it may be wise to give priority attention to those pollutants and problems that pose the greatest risks to our society.” 33 Id. at ii. There is a persuasive case to be made for prior­itizing risks addressed by an agency facing mandates for cuts—while also accounting for statutory mandates—in any sort of regulatory triage process that occurs in response to mandatory cuts.

The challenge, however, is that there are potentially endless means of defining and selecting particular risks to focus on, and the definition of risk will centrally shape the triage process. 34 See, e.g., J. Clarence Davies, Ranking Risks: Some Key Choices, in Comparing Environmental Risks: Tools for Setting Government Priorities, supra note 23, at 9, 15. “Risk” can be defined based on the hard, quantitative scientific evidence of the likelihood that a particular amount of pollution will lead to particular effects of varying degrees, but it also can be shaped by public perceptions about pollutants and their effects, among many other factors. 35 Nat’l Research Council, supra note 5, at 131; Lash, supra note 24, at 77–78. And then there is the question of the type of risk to focus on. As the EPA noted in a large comparative risk assessment completed in 1987, past studies by the Agency tended to focus only on health risks—and specifically cancer-based risks. 36 EPA, supra note 5, at 5. But in 1987 the EPA looked to its statutory mandates in order to determine the risks it should focus on in prioritizing certain risks over others, noting that the “EPA is also legislatively responsible for protecting natural ecosystems and the general public welfare.” 37 Id.

At a minimum, agencies conducting triage should recognize that risk involves not only a scientific estimate of the likelihood of harm occurring to a particular degree but also a variety of judgments with respect to which harms matter more, and how much. 38 See, e.g., Lash, supra note 24, at 70–71, 77–79 (noting that bad regulatory decisions and other types of decisions can result from a lack of scientific facts or a misunderstanding of the facts but also emphasizing the importance of integrating both science and values in comparative risk assessment approaches and making a similar proposal). And agencies can look to a variety of sources to attempt to define the risks that they will address in triage.

Just as agencies should look independently to statutes as part of the triage process to ensure that they do not diverge from any core man­dates, agencies should also look to their enabling statutes—the logical first place for locating and defining the risks that should be considered within the triage process. 39 See supra notes 28–30 and accompanying text. In addition to specific risks identified within these statutes, such as the endangerment language discussed in section I.A, agencies’ enabling acts often begin with a “purpose” section that identifies the primary goals of the statute, including the primary risks to be addressed. 40 See supra notes 28–30 and accompanying text. Agencies’ existing regulations, too, identify the risks that the agency has focused on in the past in order to comply with statutes, as agencies are required to include a preamble and detailed “statement of basis and purpose” when they issue regulations. 41 Kevin M. Stack, Interpreting Regulations, 111 Mich. L. Rev. 355, 360–61 (2012).

Additionally, the substance of many statutes—beyond the statement of basis and purpose—also already contains certain risk-prioritization approaches that could help to guide agencies in selecting triage criteria. For example, under the Safe Drinking Water Act, the agency divides its regulation of the quality of water providers into three categories: commu­nity water systems that serve large populations; “transient” water systems such as those at hotels, roadside stops, and restaurants; and nontransient systems that serve smaller communities. 42 See 40 C.F.R § 141.2 (2016) (defining the categories of water systems); 40 C.F.R. § 141.80 (applying lead and copper maximum contaminant levels only to community water systems and nontransient, noncommunity water systems). The agencies’ rules have varying degrees of stringency for these systems because, for example, community water systems that serve the largest populations could pose widespread risks in the event of contaminated water.

To the extent that agencies undertaking regulatory triage wish to consider risks not explicitly addressed by their enabling statutes, there are other models that provide food for thought. For example, the EPA already has a National Enforcement Initiative, for which the Agency convenes stakeholder advisory groups and then solicits public comments on various enforcement initiatives and whether they should continue; one initiative focuses on the “largest, highest impact sources of pollution,” defining “highest impact,” albeit somewhat vaguely, as pollutants that “significantly impact human health.” 43 Public Comment on EPA’s National Enforcement Initiatives for Fiscal Years 2017–2019, 80 Fed. Reg. 55,352, 55,353 (Sept. 15, 2015). The advisory groups and public comments guide the Agency’s choice of initiatives to focus on. Some states have similar criteria for enforcement priorities. For example, Pennsylvania directs its oil and gas regulatory staff to focus on violations that could “endanger[]health, safety, or the environment.” 44 Pa. Dep’t of Envtl. Prot., Standards and Guidelines for Identifying, Tracking, and Resolving Oil and Gas Violations 8 (2015),
Get/Document-105828/820-4000-001.pdf [].
Next in line for prioritization under this state framework are violations that “cause the need to restore or replace an adversely affected water supply,” followed by “violations that result in the discharge of pollutants to surface or ground waters,” among other priorities. 45 Id. Many states have also completed comparative risk assessments in an effort to identify priorities among potential regulations. 46 See generally Minard, supra note 23. Washington State’s assessment focused on “threats to the environment,” which received a priority rank of one to five. 47 Id. at 36.

Certain international approaches provide further potential models, although domestic political priorities will of course differ from the values that other nations focus on in the regulatory process. The United Kingdom 48 Dep’t for Bus., Innovation & Skills, Better Regulation Framework Manual: Practical Guidance for UK Government Officials § 1.4 (2015),
regulation-framework-manual.pdf [].
and Canada 49 See Guide to the Federal Regulatory Development Process, Gov’t of Can., http:// [] (last visited Aug. 10, 2017); see also Triage Statement Form, Gov’t of Can.
guides/temp-gabar/tsf-fet-eng.asp [] [hereinafter Canadian Triage Statement Form] (last visited Aug. 10, 2017).
each have a regulatory triage process. Agencies in both countries use this process in the context of regulatory impact assessment—determining which regulations to focus on based on the costs that they will impose on various entities and assessing whether the regulations should be on the “fast track.” 50 Dep’t for Bus., Innovation & Skills, Regulatory Triage Assessment Form (Sept. 27, 2012), (on file with the Columbia Law Review) (last updated Apr. 5, 2016); see also Canadian Triage Statement Form, supra note 49. But in Canada, agencies con­ducting this triage analyze whether the regulation will have benefits to the environment and human health, “public security,” “society and culture,” and “economy, business, and trade.” 51 Canadian Triage Statement Form, supra note 49.

3. Regulatory Programs. — If agencies must trim regulations and programs, then it seems equally important to identify risks as well as regulations and programs—what Nelson and Krupnick refer to as “solutions” 52 Nelson & Krupnick, supra note 26, at 182. —that most effectively address the risks. 53 See Minard, supra note 23, at 24–26 (explaining the components of effective “comparative risk projects”). Here, too, agen­cies facing immediate pressure to cut budgets, regulations, and programs will not have the time to conduct detailed comparisons of their most effective programs or even likely to decide on a particular metric for effectiveness. But several existing sources of information will help agencies conducting regulatory triage obtain rough ideas of effectiveness along several metrics. For example, agencies already are required to conduct regulatory impact analyses of some of their regulations, 54 Nicholas Bagley & Richard L. Revesz, Centralized Oversight of the Regulatory State, 106 Colum. L. Rev. 1260, 1265 (2006) (describing and critiquing the inception of cost-benefit analyses by the Office of Management and Budget under Reagan); Thomas O. McGarity & Rena I. Steinzor, The End Game of Deregulation: Myopic Risk Management and the Next Catastrophe, 23 Duke Envtl. L. & Pol’y F. 93, 118–24 (2012) (describing and critiquing the regulatory impact analysis requirement and how it has evolved through different administrations). which review the benefits those regulations create as well as the likely costs of implementing, and thus there is an existing body of information that can help to inform this effort. There is great danger, though, associated with prioritizing only those regulations deemed cost effective—a metric far too rudimentary to capture public laws that address the greatest risk and yet, by certain measures, are considered “costly.” 55 See supra note 50 (illustrating how high-cost regulatory measures can be assessed differently from low-cost measures); see also Bagley & Revesz, supra note 54, at 1268–70 (noting the dangers of using cost-benefit analysis selectively and focusing primarily on cost); Amy Sinden, Cass Sunstein’s Cost-Benefit Lite: Economics for Liberals, 29 Colum. J. Envtl. L. 191, 218–21 (2004) (exploring the weaknesses of cost-benefit analysis); Amy Sinden, Douglas A. Kysar & David M. Driesen, Cost-Benefit Analysis: New Foundations on Shifting Sand, 3 Reg. & Governance 48, 55–57 (2009) (critiquing cost-benefit analysis in the environmental context). The extensive literature noting the problems and limitations of this type of analysis 56 See sources cited supra note 55. strongly suggests that agencies conducting triage should include a broader metric for effectiveness.

To extend their consideration of program effectiveness well beyond the cost-benefit process, agencies can look to other sources that have already evaluated certain regulatory programs more broadly—including evaluation of whether the programs have achieved their statutory requirements for protecting the public—such as reports prepared by the Government Accountability Office 57 See, e.g., U.S. Gov’t Accountability Office, GAO-17-317, Progress on Many High-Risk Areas, While Many Substantial Efforts Needed on Others 417–29 (2017), [] (noting strengths and deficiencies in the EPA’s programs to address chemicals that may pose substantial risks to public health). and former National Academy of Sciences. 58 See Advising the Nation. Advancing the Discussion. Connecting New Frontiers, Nat’l Acads. of Scis., Eng’g. & Med.,
index.html [] (last visited Aug. 10, 2017) (describing the process through which the National Academies guide policies and the types of reports they prepare).

4. The Distribution of Regulatory Programs. — A final important substantive factor in the triage process is consideration of which popula­tions various portions of regulatory programs protect. In the environ­mental context, a broad environmental justice literature has documented the burdens that certain communities disproportionately suffer 59 Bullard, supra note 25, at 238–41 (introducing the literature and describing the disproportionate impacts). and has noted that the regulatory process already “mirrors the larger social milieu where discrimination is institutionalized.” 60 Id. at 240. There is an equal danger that regulatory cuts could further exacerbate disparities or cut programs designed to reduce those disparities. 61 See, e.g., Michael Fix & George C. Eads, The Prospects for Regulatory Reform: The Legacy of Reagan’s First Term, 2 Yale J. on Reg. 293, 310 (1985) (describing how industry potentially had the most influence in previous processes that assessed dereg­ulation opportunities). Thus, agencies initiating a triage process must be aware of the types of communities that suffer the highest risks addressed by certain programs and the extent to which eliminating or weakening a program will impact these communities.

II. Models for the Triage Process

Beyond the substance of triage, the process through which agencies conduct triage is an important consideration. A quick pace of regulatory revision, and revision that occurs under the threat of substantial cuts to the agency’s programs and resources, presents opportunities for regu­latory failure, including a failure to adequately consider the interests of individuals and communities that regulations benefit. 62 Cf. McGarity & Steinzor, supra note 54, at 139 (“Industry advocates in many high-stakes rulemakings are now willing to spend millions of dollars to achieve their regulatory goals by lobbying agency staff and members of Congress.”); Erik D. Olson, The Quiet Shift of Power: Office of Management & Budget Supervision of Environmental Protection Agency Rulemaking Under Executive Order 12,291, 4 Va. J. Nat. Resources L. 1, 60–62 (1984) (describing the Office of Management and Budget’s capture by industry in the Reagan regulatory reform era). For example, when President Reagan set out to cut numerous regulations, his order for all regulations to be reviewed by the Office of Management and Budget’s (OMB) Office of Internal and Regulatory Affairs (OIRA) on a cost-benefit metric, with the aim of cutting regulations, resulted in a relatively closed process. 63 See Fix & Eads, supra note 61, at 299 (1985). This, in turn, led to criticism regarding “‘OMB officials’ vulnerability to ex parte and unrecorded contacts with regulated entities” and “raised serious doubts about the possible neutrality of the new OMB review process.” 64 Id.; see also Bagley & Revesz, supra note 54, at 1265 (noting that “[s]trong industry group influence may have concentrated OIRA’s attention on reducing regulatory burdens”); Olson, supra note 62, at 60–62 (“Former EPA Chief of Staff Daniel has charged that OMB frequently represents industry arguments to EPA as its own.”). But see Fix & Eads, supra note 61, at 293 (arguing that Reagan’s regulatory activities should not be labeled as a failure despite the deficiencies noted). This Part discusses the concerns associated with the triage process and potential procedural solutions, focusing on the potential for capture primarily on the stakeholders that should be included.

A. Process-Based Concerns

With efforts at regulatory cutting similar to Reagan’s program now emerging, rapid regulatory program changes, in particular, might exacerbate certain forms of failure, including capture and the threat of worsening existing environmental justice challenges. When agencies with already scarce resources must quickly decide which regulations and programs are on the chopping block, this seems to open up the possi­bility that well-organized, wealthy interests will have the opportunity to coalesce around certain positions and potentially have undue influ­ence in the process. 65 See sources cited supra note 62. These interests could potentially advocate for the removal of regulations in a manner that would unevenly benefit them and erode the very purpose of the statute the regulations were designed to implement.

In some scholars’ view, the capture story is not as simple as the public choice account outlined above. Some have persuasively argued that environmental groups, too, have a surprising amount of influence in regulatory processes despite the traditional assumption that dispersed interests, like individuals benefited by environmental regulations, will have difficulty overcoming the transaction costs of organizing. 66 See, e.g., Richard B. Stewart, Environmental Quality as a National Good in a Federal State, 1997 U. Chi. Legal F. 199, 199–208 (discussing a variety of factors that explain why there are strong federal environmental laws in the United States despite the public choice prediction that industry interests would effectively coalesce to oppose these laws and including the consideration of relatively strong nationally organized environ­mental groups as one explanation). Others have noted that simply including various stakeholders in the process, including vulnerable ones, is not enough to prevent agency action that is inadequately protective of the public and unevenly benefits powerful, organized stakeholders. 67 Seidenfeld, supra note 15, at 477. If the relatively disadvantaged stakeholders included in the process lack the resources or expertise to participate in a productive way, then diverse stakeholder representation is little more than window dressing. But in an environment in which agencies are already lacking in resources and cannot train various stakeholders to effectively participate, designing processes that at least include a variety of stakeholders, including vulnerable groups who are protected by regulations, is a start. Other parties concerned about regulatory cuts, such as nonprofit environmental groups like the Environmental Defense Fund (EDF) or Natural Resources Defense Council (NRDC), which already research and publish information about risks, could take the initiative to educate and train these stakeholders and make them as effective as possible. 68 See, e.g., Extensive Research Effort Covers Methane Leaks, Envtl. Def. Fund, [] (last visited Aug. 10, 2017) (describing EDF’s work on methane leakage—an area in which the Trump Administration has proposed rule repeals—and the resulting reports from this research); see also Ralph Cavanagh & Kala Viswanathan, NRDC’s Annual Energy Reports, Nat. Res. Def. Council (Dec. 5, 2016), [] (showing NRDC’s research in the energy area—an area in which the Trump Administration has promised regulatory cuts). For proposed regulatory cuts relating to methane and energy, see Office of Info. & Regulatory Affairs, supra note 2. Additionally, to the extent that there is concern that industry is not the only powerful, wealthy interest that could unevenly affect the process—such as concern that environmental groups could do the same—including representatives from several more-powerful stakeholder groups in the process, including industry, should help to alleviate this concern.

Similar to capture concerns, uneven representation of well-organized, wealthy interests and too little process associated with the cutting of programs and regulations could weaken or eviscerate regulations de­signed to protect certain vulnerable groups, 69 See sources cited supra note 62. as intro­duced in Part I. The following section discusses how agencies should consider forming the groups that would conduct the triage process with these concerns in mind, including the stakeholders that should partic­ipate and how the agencies might go about forming stakeholder groups.

B. Procedural Approaches

Agencies initiating a triage process must consider who will be in­volved, as well as how those involved will provide information to the agency, such as through a formal advisory committee or public meetings convened by the agency.

Due to concerns about adequately representing a variety of interests, agencies should likely not conduct triage through a process that includes only agency officials. Stakeholders should, at minimum, include constit­uents protected by rules, officials from existing government offices, and entities who must comply with the rules. 70 For an example of an agency structure that incorporates this model, see About the National Transportation Safety Board, Nat’l Transp. Safety Bd.,
about/pages/default.aspx [] (last visited Aug. 10, 2017).
Including experts and agency officials is another means for pushing back against capture concerns by infusing an element of relatively neutral fact-finding into the process and ensuring that regulations critical for protecting the public are not cut. And the types of experts to be included should be true experts—not the type of red team–blue team approach recently initiated by EPA Administrator Scott Pruitt, in which individuals who are overwhelmingly outnumbered with respect to the science 71 For the consensus-based evidence of overwhelming scientific data showing climate change caused by human activity, see, e.g., Intergovernmental Panel on Climate Change, Climate Change 2013: The Physical Science Basis 121–30 (2013). are recognized in the process and held up as equal voices. 72 See Brad Plumer & Coral Davenport, E.P.A. to Give Dissenters a Voice on Climate, No Matter the Consensus, N.Y. Times (June 30, 2017),
2017/06/30/climate/scott-pruitt-climate-change-red-team.html (on file with the Columbia Law Review) (“Scott Pruitt, the head of the Environmental Protection Agency, plans to convene a team of researchers to test the scientific premise of human-caused climate change.”).
Rather, the individuals must be experts with respected, peer-reviewed track records when it comes to identifying risks. 73 For the importance of focusing regulation on risk-based data and the challenges of doing this due to industry influence, see, e.g., Thomas O. McGarity & Wendy E. Wagner, Bending Science: How Special Interests Corrupt Public Health Research 16–19 (2008); Wendy E. Wagner, Commons Ignorance: The Failure of Environmental Law to Produce Needed Information on Health and the Environment, 53 Duke L.J. 1619, 1623–25 (2004) (discussing the inadequacy of available data and the need to incorporate more science into agency regulatory processes). Additionally, for vulnerable stakeholders that lack adequate resources or knowledge to defend certain complex rules, 74 See, e.g., Cary Coglianese, Citizen Participation in Rulemaking: Past, Present, and Future, 55 Duke L.J. 943, 966–68 (2006) (describing knowledge-, resource-, and expertise-based barriers to participation). experts and agency officials can fill in these gaps.

Additionally, involving stakeholders and experts in different portions of a triage-type process can help to alleviate concerns that stakeholders that lack the requisite expertise to inform decisionmaking can still have a voice in a portion of the process. For example, when agencies in Washington State completed a comparative risk assessment, limited stakeholders and experts initially identified the twenty-three highest-priority environmental threats. 75 Minard, supra note 23, at 36. Agencies then solicited comments from the public at large in town hall meetings, and a “group of analysts . . . considered the risk reduction potential and costs of some 300 policy options, many of them proposed by citizens at the public meeting.” 76 Id. at 37.

Numerous agencies have already deployed a variety of procedural approaches to identifying, ranking, and addressing risks, and these approaches provide potential models for forming these stakeholder groups. To name just a few examples, agencies have formed ad hoc independent panels of experts with the help of the National Academy of Sciences, relied on standing committees tasked with ranking risks, and formed statutorily mandated scientific advisory committees. 77 See Nat’l Research Council, supra note 5, at 89–93 (describing National Academy of Sciences’s panels). There are also other potential, although somewhat more formal, models for form­ing triage advisory groups that suggest how agencies could rely on an inclusive group of stakeholders in conducting regulatory triage.

The Negotiated Rulemaking Act of 1990 78 Pub. L. No. 101-648, 104 Stat. 4969 (codified at 5 U.S.C. §§ 561–70 (2012)). and the Federal Advisory Committee Act 79 Pub. L. No. 92-463, 86 Stat. 770 (1972) (codified at 5 U.S.C. app. § 14). formally sanctioned the process of negotiated rulemaking (also called “reg-neg,” for regulatory negotiation), and a variety of agencies used reg-neg in the 1990s. 80 See Cary Coglianese, Assessing Consensus: The Promise and Performance of Negotiated Rulemaking, 46 Duke L.J. 1255, 1255–56 (1997) [hereinafter Coglianese, Assessing Consensus] (noting how “the executive branch . . . visibly supported regulatory negotiation” during the early 1990s). Through reg-neg, an agency, follow­ing federal statutes that enable this form of rulemaking, solicits detailed input from an advisory group of entities the proposed rule is likely to affect before drafting a rule and publishing it for public notice and com­ment. 81 See id. (describing the reg-neg process in greater detail). Although scholars and many agencies have largely dismissed the practice as moot, some agencies still use it to gather stakeholder-based suggestions for the content of regulations. 82 See Hannah J. Wiseman, Negotiated Rulemaking and New Risks: A Rail Safety Case Study, 7 Wake Forest J.L. & Pol’y 207, 226 (2017) [hereinafter Wiseman, Negotiated Rulemaking] (describing two agencies that still use negotiated rulemaking). The EPA also deploys this process to identify certain enforcement priorities. 83 See infra Part III. As part of the reg-neg process, an agency proposing a rule or action appoints a rulemaking committee consisting of up to twenty-five, or possibly more, stakeholders. 84 5 U.S.C. § 565(b). Congress provides that these stakeholders must represent interests “significantly affected by the rule” and that the committee must have “balanced representation.” 85 Id. § 563(a). For example, in the rail-safety context, when recent explosions of oil trains and deaths on passenger trains have attracted public attention, labor unions, nonprofits, and the rail carriers themselves are part of the reg-neg committee. 86 Wiseman, Negotiated Rulemaking, supra note 82, at 252. Through reg-neg, the committee of stakeholders proposes a consensus rule to the agency, the agency submits the rule for public review through the traditional notice-and-comment process, and the agency then adopts or rejects the rule. 87 See Coglianese, Assessing Consensus, supra note 80, at 1257. For a description of the process in the rail-safety context, specifically, see Wiseman, Negotiated Rulemaking, supra note 82, at 246–49.

If there is strong pressure to quickly cut certain regulatory regimes, agencies might not have time to convene formal committees. In this case, agencies can rely on traditional methods of organizing public meetings open to all potential stakeholders—in geographic locations that are convenient for stakeholders—and smaller discussions with affected parties. 88 Agencies have historically done this. For example, before regulating a substance with potentially carcinogenic or other effects, the Consumer Product Safety Commission convened a technical panel, conducted a notice-and-comment rulemaking, and then solicited public comments at meetings. Nat’l Research Council, supra note 5, at 92. For example, in writing the Clean Power Plan for carbon regu­lation, the EPA held eleven “public listening sessions” with thousands of participants around the country and held “one-on-one” meetings with state and tribal officials, industry groups, non-governmental organi­zations, labor unions, and other federal agencies. 89 Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units, 80 Fed. Reg. 64,662, 64,704–07 (Oct. 23, 2015) (codified at 40 C.F.R. pt. 60).

Additionally, agencies undertaking a triage process should consult with existing external government entities that already have done much of the work of prioritizing regulations. These external entities employ experts who evaluate risks and the effectiveness of programs to address these risks, 90 See, e.g., The Investigative Process, Nat’l Transp. Safety Bd.,
investigations/process/Pages/default.aspx [] (last visited Aug. 10, 2017) (describing the experts on the National Transportation Safety Board’s investi­gative team).
and relying in part on recommendations from these types of agencies will additionally help to alleviate capture concerns and to incorporate accurate data into the triage process. These external entities include, for example, the National Transportation Safety Board, tasked solely with identifying the cause of rail, airplane, and other transpor­tation-based disasters and suggesting which regulations are the most important to avoid these disasters. 91 See Nat’l Transp. Safety Bd., supra note 70. Similarly, the National Oceanic and Atmospheric Administration 92 See About Our Agency, Nat’l Oceanic and Atmospheric Admin.,
about-our-agency [] (last visited Aug. 10, 2017).
and U.S. Geological Survey (USGS) 93 About Us, U.S. Geological Survey, [http://] (last visited Aug. 10, 2017).
are governmental agencies that identify certain high-priority environ­mental problems, their causes, and means of addressing them. These and similar agencies could be helpful in supplying certain information about risks. 94 For a collection of USGS hydraulic-fracturing studies, see Energy Resources Program, U.S. Geological Survey,
HydraulicFracturing.aspx [] (last updated Dec. 16, 2016).

Just as there is no single ideal prescription for the substance of regulatory triage, the process, too, will differ depending on the agency, the mandates that it faces, and its resources. But the models described here provide possible starting points.

III. Laws that Support Regulatory Triage

In a climate of mandated trimming of agencies and programs, the Administration is likely to be skeptical of the need for triage. But as agencies facing mandates review existing rules and departments and prioritize those to cut and preserve, several laws provide support for the importance of initiating triage or a similar process. In the environmental context, the strongest of these laws is the National Environmental Protection Act (NEPA), which requires that any “major [f]ederal actions significantly affecting the quality of the human environment” be accom­panied by a detailed statement of the “environmental impact of the proposed action” and that alternatives to the action—alternatives with fewer effects—be considered. 95 42 U.S.C. § 4332 (2012). Courts have affirmed that this must be a very detailed statement—often hundreds of pages—and that NEPA applies to numerous agency actions. 96 See, e.g., Richard Lazarus, The National Environmental Policy Act in the U.S. Supreme Court: A Reappraisal and a Peek Behind the Curtains, 100 Geo. L.J. 1507, 1510 (2012) (noting that NEPA has “prompted the preparation of approximately 34,000 draft and final [environmental impact statements] and successfully prevented at least hundreds, and likely thousands, of actions from causing unnecessary damage to the nation’s environment”). Even changes that might seem minor are considered “major federal actions” subject to NEPA. 97 However, if the President acts under inherent constitutional authority that does not derive from Congress, it is sometimes not considered a major federal action that requires NEPA review. See, e.g., Sisseton-Wahpeton Oyate v. U.S. Dep’t of State, 659 F. Supp. 2d 1071, 1078 (D.S.D. 2009). Thus, independent presidential cuts of agency budgets could sometimes avoid NEPA review. In a recent example, under the Bald and Golden Eagle Protection Act, wind energy developers were required to receive a permit that ensures they will not unduly impact eagle populations. 98 See Shearwater v. Ashe, No. 14-CF-02830-LHK, 2015 WL 4747881, at *2 (N.D. Cal. Aug. 11, 2015). In 2012 the Fish and Wildlife Service attempted to extend the length of this permit from five to thirty years without conducting a NEPA review. 99 Id. at *5. The Agency argued that this was merely an “administrative” or “procedural” change in the permitting process, but a U.S. district court rejected this argument, finding inad­equate support for it. 100 Id. at *11, *16–19. Opinions at the circuit court level have rejected similar agency arguments, 101 See, e.g., California ex rel. Lockyer v. U.S. Dep’t of Agric., 575 F.3d 999, 1014 (9th Cir. 2009) (“[T]he USDA’s characterization of the State Petitions Rule as ‘merely procedural in nature and scope’ was unreasonable . . . .”). demonstrating that seemingly minor changes in regulatory programs will still require NEPA review.

When an agency rescinds a regulation that protects or otherwise impacts certain environmental resources and fails to prepare an environmental impact statement demonstrating why those resources no longer require protection, courts are more likely to invalidate the rescission. For instance, when the Fish and Wildlife Service attempted to rescind regulations that created roadless areas in national forests—regulations promulgated under the Clinton Administration 102 Id. at 1006. —the Ninth Circuit invalidated this effort, indicating that the court could not “condone a marked change in roadless area management without environmental analysis.” 103 Id. at 1016; see also David H. Becker, Changing Direction in Administrative Agency Rulemaking: “Reasoned Analysis,” the Roadless Rule Repeal, and the 2006 National Park Service Management Policies, 30 Environs: Envtl. L. & Pol’y J. 65, 92 (2006) (“Although the Forest Service offered some justifications for its regulatory rescission, those justifications could not satisfy the ‘reasoned analysis’ standard for agency changes of direction.”). Although NEPA does not require triage, it provides support for reviewing the impacts of rule and enforcement changes before they occur. Of course, administrations intent upon cutting regulations are also likely to cut NEPA review as much as possible. 104 See, e.g., Charles H. Montange, NEPA in an Era of Economic Deregulation: A Case Study of Environmental Avoidance at the Interstate Commerce Commission, 9 Va. Envtl. L.J. 1, 6 (1989) (noting that, under President Reagan, the staff of the agency that oversees NEPA “was cut from nearly fifty to thirteen” and that “[t]he number of [environmental impact statements] released by federal agencies in the 1980s plummeted in comparison to the comparable figures in the 1970s”). But NEPA is not an optional statute; absent revision, agencies must follow it. Although an administration might aim to weaken NEPA and agencies’ use of it, judicial review will limit the extent to which this can occur.

Within and beyond the environmental context, agencies that substantially revise rules or enforcement priorities might also violate the Administrative Procedure Act by failing to fulfill their nondiscretionary duties or acting in an arbitrary and capricious manner under the statute they are charged with implementing; 105 5 U.S.C. § 706 (2012). careful triage of regulatory priorities will help to ensure that agencies provide an adequate basis for regulatory changes and carry out the mission of the many acts that they implement. Extensive regulatory reform often cannot move forward without revision of the many statutes that have steered agency activity for decades or, at minimum, a somewhat detailed agency explanation for its change of course and how this change comports with the statute. The State Farm decision, which guides courts reviewing agency modification and rescission of rules, found that “an agency changing its course by rescinding a rule is obligated to supply a reasoned analysis for the change beyond that which may be required when an agency does not act in the first instance.” 106 Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 42 (1983). Under State Farm, to avoid acting in an arbitrary and capricious manner under the Administrative Procedure Act, the agency must amass a record supporting its decision, including documentation of the “relevant data” relied upon by the agency. 107 Id. at 43. It also must provide a “satisfactory explanation for its action including ‘a rational connection between the facts found and the choice made.’” 108 Id. (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168 (1962)). Indeed, the majority in State Farm did not follow then-Justice Rehnquist’s reasoning that a change in presidential administration provided an acceptable basis for an agency’s actions. 109 Id. at 59 (Rehnquist, J., concurring in part and dissenting in part); see also Kathryn A. Watts, Proposing a Place for Politics in Arbitrary and Capricious Review, 119 Yale L.J. 2, 6 (2009) (noting that the majority did not focus on this political justification for agency action and instead analyzed whether the agency provided adequate “technocratic justifications”). Courts and scholars have since read State Farm to mean that “influences coming from one political branch or another cannot be allowed to explain administrative decisionmaking” and that regulatory changes must instead be “expert-driven.” 110 Watts, supra note 109, at 6–7.

These and other statutes do not directly require triage, but they show that agencies cannot take substantial program-cutting measures such as eliminating two regulations for every new regulation that they promulgate without providing relatively extensive support and explana­tion for their actions.


In an era of rapid and volatile political change, reaching consensus on the regulatory regimes to retain or jettison will be exceedingly difficult. But in the coming months and years—when regulatory reform could progress at a relatively fast pace—this process must move forward in a manner that protects the core values embodied within public law. Harried reforms, in particular, are at risk of regulatory capture or of cutting regulations that protect communities currently burdened with the worst environmental problems. 111 See supra notes 59–61 and accompanying text. A regulatory triage process will provide an essential convening forum for dispersed interests and could help to avoid capture; it also will help to maintain recognition of the statutory mandates and the importance of assessing the various popu­lations benefited by regulations if agencies are forced to cut certain regu­latory programs.