*This is the tenth post in a series on Michael Livermore and Richard Revesz’s new book, Reviving Rationality: Saving Cost-Benefit Analysis for the Sake of the Environment and Our Health. For other posts in the series, click here.
With Reviving Rationality (“RR”), Michael Livermore and Richard Revesz (“L&R”) have produced another important, timely, and provocative work on the use of cost-benefit analysis (“CBA”) in government decisionmaking. L&R recount the successes of the Obama Administration, which was able to deploy its progressive agenda with support from CBA, and present a cautionary tale with respect to the Trump Administration, which unsuccessfully tried to manipulate CBA to support its regulatory agenda. Importantly, the book underscores the value of CBA as a way of providing public accountability for the Trump Administration’s bad policies.
In particular, L&R critically analyze several moves by the Trump Administration to weaken the constraints of CBA, including limiting the scientific evidence that can be considered by agencies in these analyses, truncating the consideration of beneficial impacts of regulations, and using discredited models to generate lower values of the benefits of regulations. The overall effect is devastating to the Trump Administration. At every turn, its moves were met with widespread and unequivocal condemnation by the scientific community, courts, and the public. For those who shudder to imagine a world where the Trump Administration was able to implement more of its wrong-headed policies, the takeaway from the Trump Administration should be more analysis and not less.
And yet, many progressives still call for the Biden Administration to abandon CBA. The Center for Progressive Reform has released several reports calling for President Biden to revamp the Office of Information and Regulatory Affairs (“OIRA”), the office responsible for reviewing agency CBAs, and end reliance on CBA more generally. A concurrent symposium hosted by the Law and Political Economy Project already has several contributors arguing for rejecting CBA or otherwise manipulating the analysis to support progressive ends. How could the lessons from the Trump Administration’s failures have so far fallen on deaf ears?
Thankfully, for now, there are promising signals for the continuation of CBA requirements: in one of his first directives as president, President Biden instructed the Office of Management and Budget (“OMB”) to begin a process for modernizing regulatory review, and key to this modernization is taking into account the distributional consequences of regulations “including as part of any quantitative or qualitative analysis of the costs and benefits of regulations.” That said, as of today, President Biden has yet to nominate someone to be director of OMB or the administrator of OIRA.
L&R and I overwhelmingly agree about the value of CBA in promoting net beneficial and equitable government policy. While I don’t necessarily agree with every argument they make in RR, any quibbles we have are vastly overshadowed by our agreements. In this blog post, I focus on this common ground. In particular, I highlight some additional, implied takeaways of L&R’s book that follow from the fact that the Trump Administration’s “out of bounds” actions were consistently rebuffed by the scientific community, courts, and public opinion.
CBA is not easy to manipulate. CBA requires policymakers to make decisions about data, models, assumptions, and methodology. It is true that many of these decisions cannot be made solely on the basis of science and therefore will reflect a combination of the available evidence and underlying values and policy judgments. For any CBA, then, there is a range of reasonable decisions that experts and policymakers can make and support. In RR, L&R argue that while the Obama Administration tended to make decisions within this reasonable range, the Trump Administration made decisions outside of this range, “kicking at the guardrails that had constrained agency decision making for decades” (p. 33). The scientific community, courts, and the public noticed.
I have previously argued that CBA would constrain some regulatory objectives if an agency purports to rely on CBA to support its actions. In particular, although courts tend to be deferential to supporting agency analyses, if the administrative record already contains a relatively complete prior CBA, it would be easier for courts to evaluate the reasonableness of further CBA-supported regulatory or deregulatory action. For this reason, I predicted that the Trump Administration was going to have a difficult time rolling back regulations issued by the Obama Administration exactly because the Obama Administration used CBA to support many of its policies. And there’s evidence that this has been true. In a recent article, Bethany Davis Noll shows just how much the Trump Administration has lost in court compared to other presidential administrations—and she finds that faulty analysis was one of the commonly asserted grounds for setting aside the administration’s actions. The constraining influence of CBA was felt within the administration, too. For example, in November 2019 the Trump Administration changed its policy on freezing fuel economy standards. According to reports, career and political staffers at the Environmental Protection Agency and the Department of Transportation could not produce a CBA that would justify the freeze and withstand scrutiny.
Relatedly, CBA does not make political decisions and value judgments less transparent by hiding them behind technocratic assumptions. Instead, the experience of the Trump Administration demonstrated how helpful CBA is in highlighting exactly these kinds of decisions. Because the analysis could not be manipulated so easily, the Trump Administration was forced to openly acknowledge political decisions and subjective value judgments—and it was panned by the scientific community, courts, and the public when it did not.
There is room for political decisions and value judgements within the robust practice of CBA when the science isn’t there yet or is indeterminate. CBA is often viewed as a roadblock by progressives because many important categories of benefits are still difficult to quantify and monetize. For one, quantification and monetization of regulatory benefits has greatly improved over time, diminishing this concern. In addition, quantification and monetization has often revealed that the benefits are even more valuable and worthwhile than those who consider them priceless initially perceived. Examples include the use of the Value of Statistical Life, the Reagan Administration’s decision to pursue a stricter standard for phasing out lead in gasoline, and the real value of additional reductions in particulate matter emissions below the health-based, cost-blind National Ambient Air Quality Standard.
But, more to the point, the lack of quantification and monetization has not stood in the way of issuing regulations. In fact, when examining OIRA’s reports to Congress going back to 2010, it emerges that, on average, 46 percent of significant (non-transfer) regulations reviewed by OIRA do not monetize any benefits. (For regulations issued by the Trump Administration, this percentage was even higher—64 percent of regulations.) Given statutory discretion, agencies can pursue their preferred regulations despite largely incomplete CBA as long as the political decisions and subjective value judgments that ultimately drive their decisions are made clear. The only effect of gaps in quantification and monetization is that a new administration that does not share the prior administration’s intuitions about the value of such benefits (or costs) could more easily get rid of the regulations.
CBA does not serve anti-regulatory goals (or pro-regulatory goals, for that matter). It seems strange to have to emphasize this point now, after the Trump Administration’s documented failures to effectively pursue its deregulatory agenda. (And if this comes as a surprise to you, check out the recent report by Cary Coglianese, Natasha Sarin, and Stuart Shapiro that highlights the disconnect between the administration’s rhetoric and the reality of its deregulatory record.) For sure, CBA cannot take all the credit for the administration’s failure. But it certainly did not help the administration. And that’s because the tool is meant to be a neutral aide to decisionmaking, helping highlight moves from the status quo (whether regulatory or deregulatory) that are net socially beneficial based on available evidence. In other words, if there’s no economic or scientific evidence to support a move away from the status quo (in either direction), then CBA will not be helpful. In such cases, pro-regulatory and anti-regulatory administrations could pursue their objectives without CBA’s support, as they often do. Again, in those cases, the only requirement is that any analysis is honest and that political decisions and subjective value judgments are made clear. Of course, the policy will not be as resilient if a future administration holds a different view.
All that a culture of CBA does is encourage both pro-regulatory and anti-regulatory administrations to seek out evidence of the value of their goals—at least if they want any reforms to have buy-in, acceptance, and resiliency. For both kinds of administrations, this means more research into the effects of policy proposals. As L&R argue, the Obama Administration tried to do exactly that, using emerging scientific and economic literature to identify and even monetize new categories of benefits. The Trump Administration could have deployed similar methods to support and achieve its policy goals. For example, the Trump Administration could have devoted time to understanding the impact of persistent job losses for different sectors of the economy, or, more generally, it could have developed a defensible method to systematically conduct and consider distributional analysis. Again, it is not that the goal of reducing regulatory burdens is necessarily unsupportable by CBA or that it’s never a rational government policy; the problem was the Trump Administration’s attempt to use faulty CBA—CBA that deviates from (well-defined and neutral) norms, principles, and guardrails, as L&R document—to support and justify its actions.
In the end, the Obama Administration policies that were most at risk of rollbacks under the Trump Administration were those not supported by relatively complete CBAs. For example, in its proposed rescission of the Hydraulic Fracturing on Federal and Indian Lands Rule, the Trump Administration failed to quantify and monetize any benefits, focusing only on cost savings. Turns out, however, the Obama Administration had also failed to quantify and monetize benefits when it issued the original rule. Its decision to issue the rule notwithstanding the incomplete analysis was based on the administration’s subjective judgment of the net value of the rule’s requirements, a judgment that the administration was undoubtedly entitled to make—and courts would have deferred to that judgment if it were challenged. But, in such cases, the Trump Administration was similarly entitled to make its own subjective judgment—and this was exactly why a district court upheld the administration’s rescission of the rule. It is easier to explain a new administration’s reversal of a policy when the original policy was based on subjective value judgments. That is why L&R advocate for improving retrospective analysis and quantification (pp. 195-205). And that’s why calls to abandon CBA and implement statutory goals without its support are ultimately self-defeating. If a pro-regulatory administration can do it, so can an anti-regulatory administration. The policy oscillation itself is costly. Not to mention the fact that many policies that are particularly important to progressives, such as seriously tackling the threat of climate change, involve sustained commitments over a long time horizon in order to accrue benefits.
Concern for the distributional consequences of regulations does not require abandoning CBA. As several commentators have already pointed out, CBA, as currently conducted by agencies, tends not to consider the distribution of regulatory benefits and costs. This is wrong from a welfare economics perspective—for example, if a benefit accrues to a rich individual, the increase in utility would be lower than if the same benefit were to accrue to a poor individual due to the diminishing marginal utility of income. Ignoring this issue, a CBA might show that a policy that mostly benefits rich individuals and hurts poor ones is net beneficial, but the policy might not actually be worth pursuing from society’s point of view. And the fact that those who benefit from regulations do not compensate those who are burdened by them makes this issue particularly important.
L&R recognize this blind spot in the current practice of CBA, too, and propose several solutions to make CBA more attentive to distributional concerns (pp. 205-209). And, as noted, the Biden Administration has sent early signals that it will take distributional concerns seriously. There is much more I could say about how distributional analysis could be implemented and used in regulatory decisionmaking (and other scholars are weighing in, too). But simply put, for the reasons pursuing net beneficial policy is best done with some CBA, pursuing policy that is attentive to distributional consequences should involve some analysis of benefits and costs that accrue to different groups. Otherwise, in trying to make things better, advocates for equitable regulations could unintentionally make things worse for the groups they seek to help with their regulations.
Finally, CBA, as practiced by agencies, is not perfect. I doubt anyone seriously thinks it is. But it is the best tool we have available to ensure that our government’s actions help society in ways we value. Pro-regulatory and anti-regulatory advocates both push for less analysis so that they could more easily impose their preferences. They attack CBA simultaneously for being easy to manipulate (by the other side), anti-regulatory/pro-regulatory (as relevant), not transparent, and persistently net costly for some groups—but they typically fail to acknowledge that their preferred alternatives all perform worse by these same measures in a polarized and divided society. L&R’s book is important exactly because it reminds us what was at stake during the Trump Administration—and the resilience and effectiveness of CBA as a line of defense against arbitrary and costly government policies. The response should not be to get rid of CBA and its guardrails going forward. The response should be to fortify CBA, improve it, and expand its usefulness into other contexts we care about.
Who benefits from CBA? We all do.
Caroline Cecot is an Assistant Professor of Law at the Antonin Scalia Law School at George Mason University.