For the past 40 years, the process by which the U.S. government issues new agency rules and regulations has been largely premised on demonstrating that the benefits of any regulation “justify” the costs. This is known as cost-benefit analysis (CBA).
While CBA may seem like an esoteric process far removed from people’s everyday lives, it has enormous effects on Americans’ health, safety, and financial security. Rules that produce significant net quantifiable benefits are far more likely to be issued, while those that result in significant net quantifiable costs are more likely to be stymied in the executive branch or struck down in court. CBA, therefore, influences the extent to which federal regulators require power plants to limit the amount of toxic pollutants they put into the air; financial advisers to treat their customers fairly; and businesses to provide adequate safety standards for workers and consumers.
Many progressives have long held concerns that CBA has made it harder to move forward with needed regulatory protections. The process has a significant status quo bias; it often fails to account for environmental impacts and distributional effects; and it makes it far easier for sophisticated parties with significant resources to generate data showing large costs than it is for agencies to accurately capture more diffuse and, at times, unquantifiable benefits to society as a whole. Yet CBA, while imperfect, has proved a constant across administrations. Both sides have played by the same rules, with progressives knowing that conservatives will need to demonstrate that a regulation’s benefits justify its costs, and vice versa.
However, since 2017, conservatives have entirely upended this equilibrium. Finding that progressives have been able to advance important regulatory protections consistent with CBA requirements and that those same requirements would stymie aspects of their deregulatory agenda, conservatives have responded by changing the process. For example, they have manipulated CBA to try to increase the amount of toxic mercury in the air, to attempt to allow employers to steal their employees’ tips, and to try to eliminate protections for American wetlands.
This issue brief outlines how conservatives have rewritten the regulatory process in ways that have left traditional cost-benefit analyses behind. Rather than regulating by the same rules as progressives, conservatives have ignored the rules when they prove inconvenient. Worse, they have sought to make procedural changes that would drastically limit or prevent a progressive presidential administration from advancing commonsense regulatory protections at all.
These breaks from traditional CBA reveal that conservatives see it less as a tool for sound policy and more as a means of preventing progressive regulation when conservatives are out of power. Reckoning with this fact means more than simply looking to undo the current administration’s regulatory abuses. It requires rethinking the centrality of CBA in the federal regulatory review process: While it is an important tool for agencies to utilize in developing sound regulations, it is not the only tool. This issue brief concludes by encouraging progressives to articulate a new vision for the federal regulatory process, rather than simply reverting to the status quo.
The rise of cost-benefit analysis
Widespread use of cost-benefit analysis has largely been a creation of the executive branch, not the legislative one. In 1946, Congress enacted the Administrative Procedure Act (APA), a law that put in place a formal process for federal agencies to follow when enacting regulations.1 Broadly, the APA requires agencies wishing to finalize rules to articulate their policy rationales and allow the public an opportunity to comment.2 The law provide agencies wide latitude to determine their rationales, prohibiting an agency from issuing a regulation on policy grounds only when the decision is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law” or “contrary to constitutional right.”3
It was President Ronald Reagan who, in one of his first acts in the White House, signed Executive Order 12291 and instituted what has since become the standard process by which executive agencies promulgate regulations.4 Two changes are particularly significant:
- Cost-benefit analysis requirement: The order required all agencies to issue a regulation only if “the potential benefits to society from the regulation outweigh the potential costs to society,” with the objective of “maximiz[ing] the net benefits to society.” To demonstrate that the benefits of a proposed rule were to exceed its costs, each agency was to develop a regulatory impact analysis (RIA) for each “major” rule proposed, analyzing a regulation’s predicted quantified and unquantifiable costs and benefits, as well as an analysis of the costs and benefits of potential alternative courses of action. The order also required agencies to “set regulatory priorities with the aim of maximizing the aggregate net benefits to society.”
- Centralized regulatory review: Once an agency had drafted a proposed rule and its RIA, the order required a rule to be submitted to the Office of Information and Regulatory Affairs (OIRA) for review prior to the regulation’s enactment. The staff of OIRA, a small office within the White House’s Office of Management and Budget, would review every proposed rule to ensure its costs did not exceed its benefits. Only when OIRA’s review of a rule was finished would an agency be permitted to finalize and publish it, giving OIRA and the White House unprecedented authority over agencies’ regulatory activities.
These two components of the federal regulatory process have become mainstays and have been affirmed in one way or another by every president since. Although President Bill Clinton officially repealed Executive Order 12291 in 1993, he formalized the OIRA review process and cost-benefit analysis requirements in his own Executive Order 12866; this new order did note that the benefits of a regulation must “justify” its costs, rather than “outweigh” them, a recognition that quantified CBA does not capture the full range of relevant costs and benefits.
President Clinton’s executive order put forth a philosophy that, until recently, served as the basis for agency regulations:
In deciding whether and how to regulate, agencies should assess all costs and benefits of available regulatory alternatives, including the alternative of not regulating. Costs and benefits shall be understood to include both quantifiable measures (to the fullest extent that these can be usefully estimated) and qualitative measures of costs and benefits that are difficult to quantify, but nevertheless essential to consider. Further, in choosing among alternative regulatory approaches, agencies should select those approaches that maximize net benefits (including potential economic, environmental, public health and safety, and other advantages; distributive impacts; and equity), unless a statute requires another regulatory approach.5
The two presidents who followed Clinton each issued their own executive orders on analyzing the effects of regulations. These orders amended the procedures and focus of Executive Order 12866 to suit the individual perspectives of these presidents, but they “maintained and even reinforced the basic principles” of cost-benefit analysis.6
In summary, since at least 1981, whenever an agency has proposed a new federal regulation, it has needed to quantify the costs and benefits to the greatest extent possible, including those that are indirect or ancillary, known as co-costs and co-benefits; analyze the unquantifiable costs and benefits; and put forward a recommended agency action for OIRA to review and approve.
The demise of regulatory cost-benefit analysis at conservative hands
Conservatives have upended this status quo after finding that CBA requirements still allow progressives to regulate (albeit less effectively than they might otherwise) and inhibit their own efforts to unwind regulations. As shown by the actions of the current administration, conservatives do not treat CBA as a tool for maximizing the public good through the regulatory process but instead as an excuse for slowing regulation under progressive administrations—an excuse that is quickly discarded when it conflicts with their deregulatory goals.
Imposing an arbitrary ‘regulatory budget’ that ignores regulatory benefits
President Donald Trump and his administration have made clear from the start that their focus was not on ensuring the greatest societal benefits but instead on ensuring the lowest costs to well-connected special interests, regardless of the impact on broader society.
The directive in Reagan’s executive order mandating that, “Agencies shall set regulatory priorities with the aim of maximizing the aggregate net benefits to society” has been the cornerstone of regulation for the past four decades.7 Yet this principle—that the benefits of an action should exceed its costs—has been replaced with a focus only on the costs of actions, regardless of any benefits that could be provided. Cost-only analysis has replaced cost-benefit analysis.
Ten days after his inauguration, President Trump signed Executive Order 13771, “Reducing Regulation and Controlling Regulatory Costs,” which imposed two forms of a “regulatory budget.”8 First, the order imposed a “one-in, two-out” requirement on all new regulations so that whenever an agency proposes a new regulation, “it shall identify at least two existing regulations to be repealed.” Second, the order requires agencies to “cap” the total costs imposed by all their new regulations each year, limiting “the total incremental cost of all new regulations, including repealed regulations.”
Importantly, the order says nothing about the benefits of a regulatory action. Agencies are not directed to maximize net benefits but instead to minimize total costs at the expense of any benefits regulations may bring. In the words of two advocates, “If you followed the logic of the order in your own life … you would not invest in a car, a house, or a college education, even though the benefits of such products are unquestionable.”9 This executive order kicks any semblance of rational rulemaking to the curb. Under the requirements of the order,10 any new rule that is appropriate, necessary, and imposes only minor costs to provide significant public benefits could still only be issued if the agency repeals two additional rules, even if those repeals reduce benefits.
Further, the caps imposed by this executive order can be negative—and in many cases are. The order allows OIRA to set “a total amount of incremental costs that will be allowed for each agency in issuing new regulations and repealing regulations” for each fiscal year. For fiscal year 2020, for example, OIRA has required the Environmental Protection Agency (EPA) and the Department of Transportation (DOT) each to reduce the total costs to the public of their cumulative rulemakings by $40 billion.11 Under the rules of the order and OIRA’s cap, even if DOT wishes to issue a set of rules that imposes no cumulative costs on society but provides billions of dollars’ worth of benefits, it could not do so. Such a requirement completely overturns the premise that the goal of regulations is to maximize public benefit, placing the objective squarely on reducing costs to regulated entities, many of which are politically well-connected.
Hiding the true costs of (de)regulation
The rejection of CBA extends beyond ignoring benefits. Perhaps the most significant change conservatives have made to the regulatory process is to shamelessly hide the true costs of deregulation. Any regulation has trade-offs, and policymakers are expected to demonstrate that they fully considered a rule’s impacts by publicly disclosing the costs and benefits. Yet time and again, conservatives have worked to ensure that the costs of their deregulatory policies do not come to light.
The EPA under the Trump administration has likely been the worst offender. In 2017, the EPA, along with the Army Corps of Engineers, began the process of repealing the 2015 rule defining “Waters of the United States” under the Clean Water Act by proposing a new definition backed by a new regulatory impact analysis.12 In creating this new analysis, EPA economists and statisticians reviewed the RIA that the EPA had developed for the 2015 rule and concluded that many of the numbers it had used just two years prior were outdated and created too much “uncertainty” to be usefully considered.13 They wrote: “In the case of the forgone benefits of wetland protection [from repealing the 2015 regulation] the agencies believe the cumulative uncertainty in this context is too large to include quantitative estimates in the main analysis for this proposed rule.” Essentially, in May 2015, the EPA concluded that protecting wetlands would have a quantified benefit of from $306.1 million to $501.2 million,14 but by July 2017, the agency concluded it could not quantify the benefits of protecting wetlands because the data were too old.
In the words of one expert, “They reach a conclusion that wetlands have no value, none that is knowable.”15 In the words of another, then-EPA Administrator “Scott Pruitt is not just cooking the books, he is burning the books.”16
Much as it hid the true costs of deregulating under the Clean Water Act, the Trump administration sought to hide the benefits of an Obama administration regulation under the Clean Air Act that limited toxic mercury emissions from coal- and oil-fired power plants—known as the mercury rule—which can affect neurological development in fetuses and young children. In making a required finding that the rule was “appropriate and necessary,” the Obama administration considered co-benefits—benefits that are ancillary to a rule’s intended purpose—from the mercury rule.17 In this case, that meant considering benefits that were not the direct result of reducing mercury emissions, such as the impact that installing control equipment to reduce mercury emissions would have on other pollutants, notably fine particulate matter (often referred to as soot). From the standpoint of an accurate CBA, these benefits must be considered, as must any co-costs; the benefits to society are not affected by whether they arise from an intended or ancillary effect of the rule. Taking into account both co-costs and co-benefits, the EPA found that its regulation would result in annualized benefits from $37 billion to $90 billion, 99.9 percent of which stems from co-benefits.
However, in seeking to undermine the mercury rule, the EPA redid the analysis in 2020 to exclude co-benefits.18 The results were significant: The quantifiable benefits of the Obama-era rule were now calculated to be roughly $4 million to $6 million per year—orders of magnitude less than under the prior analysis. Given the rule’s estimated costs of $7.4 billion to $9.6 billion, ignoring co-benefits severely weakens the justification for the rule.
The impact of this attack on co-benefits is broader than just the mercury rule. Co-benefits are, rightly, considered in a wide range of rules.19 They are particularly important to Clean Air Act regulations. While scientists know certain pollutants are damaging to human health, determining specific quantifiable costs for specific, pervasive air pollutants can be quite difficult. As one scholar noted, “the EPA has been unable to generate dollar estimates of benefits for the vast majority of the pollutants the agency is charged with regulating because the data simply don’t exist.”20 One notable exception is particulate matter, which is easier to monitor and has negative health impacts that can be observed in short-term studies.21 Thus, reductions in particulate matter have constituted the lion’s share of benefits for many air pollutant regulations; eliminating the consideration of co-benefits would significantly undermine efforts to regulate a wide range of harmful pollutants.
The EPA is not the only agency hiding the costs of deregulation. For example, in December 2017, the Department of Labor (DOL) proposed to repeal a regulation that prevented employers from confiscating tips earned by their employees.22 In that proposal, DOL released an economic analysis where it provided a “qualitative discussion of the benefits and transfers that may result from the proposed rule”; it predicted that “tip pooling may foster service that is customer-focused and promotes a setting where employees get along well, and may increase productivity,” “employers may see a decreased turnover rate amongst [untipped] employees,” and “traditionally tipped employees” would benefit from having a “guaranteed direct cash wage.” The DOL further declared that “the potential benefits and transfers have not been quantified in this” notice.
However, it was reported in February 2018 that the DOL had conducted a quantified analysis of the proposed rule that estimated “that potentially billions of dollars in gratuities could be transferred from workers to their employers,”23 meaning that employers would steal their workers’ tips. According to one report, senior DOL officials “ordered staff to revise the data methodology to lessen the expected impact” after seeing the initial analysis. But even with analyses showing reduced transfers from workers to their employers, political leadership at the DOL were “said to have still been uncomfortable with including the data in the eventual proposal” and decided to leave it out entirely. In 2018, Congress amended the law to prohibit the behaviors the DOL was proposing to allow,24 and the rule was never finalized.
Hiding deregulatory costs is akin to ignoring CBA entirely; if important information is left out, then the weighing of costs and benefits will tell policymakers little about the true impact of a rule.
Salting the earth for future regulations
In addition to hiding deregulatory costs, conservatives are seeking to make it harder for future administrations to accurately consider the benefits of rulemaking by imposing new procedural requirements that legally prohibit the consideration of certain types of benefits. Take the EPA’s so-called Censored Science rule, which it proposed in 2018 and is currently seeking to finalize. The proposed rule would restrict the agency’s ability to use many scientific studies in its analyses.
Officially titled the “Strengthening Transparency in Regulatory Science” rule, the proposed rule would generally limit the studies upon which the EPA relies to only those with underlying data and models made “publicly available in a manner sufficient for independent validation.”25 Although requiring independent validation of science may sound like a positive change, the rule would prohibit the EPA from considering many high-quality studies that demonstrate the benefits of environmental regulations using proprietary health data. For example, scientists who study chemicals’ effects on individuals would be required to obtain participants’ express permission to disclose any personal information, so this change would effectively prohibit the EPA from considering such studies already published or underway even if researchers change their participation agreements in the future.26 Further, businesses that provide confidential information for studies are unlikely to allow such information to be publicly disclosed, for past as well as future studies. As one expert put it, “Think of this in the context of coronavirus … Can you imagine data [availability] being the most important thing or do you want scientific research that is robust?”27
The rejection of valid data is one of several conservative policies intended to make it more difficult for future administrations to enact regulations. Several agencies have or are in the process of codifying specific procedures for conducting cost-benefit analyses, including the Financial Stability Oversight Council,28 the Council on Environmental Quality,29 and the Department of Transportation (DOT).30 Although these procedural regulations, colloquially known as “rules on rules,” appear to benignly implement best practices, they create legally binding processes an agency must follow, even in situations where it may not be beneficial or sensible to so do. The result is that if the agency deviates from the rule on rules’ procedures when promulgating a substantive—or nonprocedural—rule, even for good reason, the agency opens itself up to additional litigation risk and provides an opponent of the substantive rule a “hook” by which to sue the agency to overturn the rule. There is no reason for an agency to tie its hands in this manner, unless its goal is to sabotage the ability of future administrations to regulate.
DOT’s rule on rules, for example, requires that it and component agencies, such as the Federal Aviation Administration, use “the best and most relevant evidence and data known to the Department,” yet takes no steps to ensure that agencies have access to such data.31 This is problematic because sometimes the best data are in the hands of the very entity that would be regulated. So this rule would let a company refuse to share data with DOT, then sue DOT for failing to consider the very same data the company refused to provide. DOT’s rule also requires agencies to adopt, in many cases, “the least costly regulatory alternative.” But it says nothing about considering the benefits of a given action. Much like the regulatory budget imposed by Executive Order 13771, agencies would be forced to enact regulations that minimize total costs at the expense of benefits. But unlike the executive order, the rule on rules makes that mandate judicially enforceable.
The EPA has proposed its own rule on rules to govern regulations under the Clean Air Act (CAA). The rule dictates, in extreme detail, how its CBAs must be conducted and permits a reviewing court to review the analyses of EPA scientists.32 The proposed rule states, for example, that when the EPA uses “an epidemiological study … the study population characteristics must be sufficiently similar to those” affected by the proposed CAA rule. Given that no study will exactly match the entire population affected by a regulation, a judge could overturn a CAA regulation simply if she disagrees that a study upon which the EPA relied was sufficiently applicable or if she believed a different study to be more relevant. Furthermore, the rule on rules would require the EPA’s cost-benefit analyses to articulate many specified considerations; again, if a judge decides an analysis for a proposed CAA rule fails to sufficiently explain even one of those considerations, she can overturn the regulation.
In these cases, core principles of CBA—that expert agency analysts should use CBA as a tool to make policy decisions—have been shunted aside to try to limit the ability of regulators to put commonsense safety protections in place and to grant judges the opportunity to second-guess agency decision-makers.
Toward a progressive vision for the federal regulatory process
Conservatives have substantially deformed CBA in their zeal to repeal commonsense regulatory protections. But the status quo that conservatives abandoned was itself significantly flawed.
While progressives have learned to work within the confines of CBA, it is a process that has clear anti-regulatory biases. Large, wealthy corporations have the resources to conduct detailed analyses of every potential cost of a proposed regulation, and frequently overestimate those costs. They can hire so-called experts to produce studies backing their preferred options and high-priced lawyers to attack agency analyses in court. These same corporations also frequently lobby Congress to make it harder for agencies to regulate.
Meanwhile, benefits can be hard for agencies to quantify; even when experts know that an action will be beneficial—such as removing a toxic pollutant from the air—they may not be able to precisely quantify the entirety of its impact. Precise quantification can be particularly difficult for diffuse benefits that accrue to everyone. Too frequently, the inability to quantify diffuse benefits means that marginalized communities pay the steepest price, whether in poor air quality, consumer fraud, or unsafe working conditions.
In addition, CBA does not properly account for distributional effects, meaning the way in which federal rules affect different people in society differently. Rules that significantly shift benefits from the working class to wealthy corporations are not neutral; they can have serious ramifications for equity and fairness. Considering exactly who is affected and how is critical to understanding whether a rule should be issued.
The solution is not to jettison CBA entirely, however. There is certainly value in agencies understanding the relative costs various rules would place on regulated parties and in being able to compare the quantifiable benefits of different actions when determining what to prioritize. But CBA has always been recognized as an imperfect vehicle, one that is necessarily constrained by the limits of the data available to agencies when addressing challenging issues. Rather than relying solely on CBA, policymakers must recognize that CBA is just one of many tools to be utilized when determining whether to enact a particular rule.
While discussing the full extent of a new progressive vision is beyond the scope of this issue brief—and is a topic the authors will address more fully in a brief to follow—envisioning a regulatory process less focused on CBA is not as daunting as it first might seem.
Agencies should still use CBA, and OIRA could provide guidance that agencies should consider in conducting their analyses. For example, OIRA could ensure that agencies pay greater attention to distributional impacts, particularly the effects of their rules on marginalized and vulnerable communities.
Rather than have part of OIRA’s central mission be the probing review of agency CBAs, it should focus greater attention on its coordinative role in ensuring that a broad range of viewpoints are considered in the federal regulatory review process. This would be particularly beneficial for marginalized and vulnerable communities, which are too frequently disadvantaged by CBA and shut out of the rulemaking process. It is critical for OIRA to ensure a fair balance of relevant stakeholders, so those who suffer the brunt of inadequate regulation are heard as loud as the business interests that tend to dominate the rulemaking process.
In addition, OIRA should utilize its unique governmentwide view of regulatory actions to help agencies explore creative ways to address issues using novel approaches.
These changes will help reflect an understanding that the economy requires rules of the road. The focus should be on finding ways to improve how the system works for everyday people, rather than viewing with a measure of suspicion any new federal regulations.
By disavowing cost-benefit analysis in order to make deregulation easier and imposing new procedural requirements—making regulation more difficult—conservatives have fundamentally rewritten how the government writes federal regulations. Playing by these new rules, progressives would never be able to accomplish any of their goals.
Yet simply restoring the previous status quo would mean that progressives continue to play by a set of rules initially put in place to limit their ability to effectively regulate large corporations, while conservatives play by another that lets them eschew analysis in favor of eliminating even the most carefully crafted regulations.
Prior to the current administration, progressives were concerned that the regulatory system was not working and was instead preventing commonsense protections that would have great benefit for the broader public. Rather than reestablish this flawed system, progressives should update the processes by which the government promulgates regulations. By articulating a new vision for the regulatory process, progressives can better ensure that the federal government’s rules reflect, and benefit, the public as a whole and that the process does not provide outsize favor and influence to the wealthy and well-connected.
Todd Phillips is a government lawyer who focuses on regulatory process. Sam Berger is the vice president of Democracy and Government Reform at the Center for American Progress. He was a senior attorney at the Office of Management and Budget during the Obama administration.