Benefit-Cost Analysis (BCA) is now widely known and used, but it is also widely misunderstood – by many of its advocates as well as its detractors. Over the next few weeks I want to examine some of the strengths and weaknesses of BCA as a normative science; and, yes, that phrase is an oxymoron, which is a source of much of the controversy. BCA is an imperfect answer, but often perhaps the best available answer, to the question of how a society should go about making collective but not unanimous choices. Nowhere is its use more contested than in its application to decisions by regulatory agencies.
Although Congress typically does not include BCA as an explicit requirement in statutes that delegate regulatory authority, it has nonetheless become the standard tool by which presidents seek to guide the administrative state. Building on narrower precedents by Presidents Nixon, Ford, and Carter, Ronald Reagan issued the first of a series of executive orders that requires executive branch regulators to conduct BCA. The most recent was signed by Bill Clinton, and probably has a pretty good chance to survive regardless of the outcome of next week’s election.
Courts, too, have found reason, when reviewing administrative decisions, to look for some form of balancing of benefits and costs. To borrow phrases from Schechter Poultry, Panama Refining, and earlier cases, BCA supplies a convenient “intelligible principle” where Congress has failed to supply a different one, and it handily erects “two banks” to prevent administrative discretion from overflowing too far in one direction or another. More recent rulings have found that the failure to consider costs as well as benefits might render a decision arbitrary and capricious under the APA.
In his farewell blog, Michael Greve laments that “The administrative state is having a grand old time because we have disaggregated a once-unified constitutional theory into a jumble of silly little doctrines (‘intelligible principle,’ ‘arbitrary and capricious’) that do not and cannot do any serious work.” I will not pretend that any sort of economic analysis is adequate to repair the damage suffered by our constitutional theory and practice. But, given the existence of an overgrown administrative state and the undeniable need to manage it, I submit that BCA, properly understood, can help to accomplish some serious work.
Twenty years ago Cass Sunstein wrote a paper (and later a book) titled The Cost-Benefit State, tracing and applauding these developments. (I won’t bore you with the arguments over whether CBA or BCA is correct; suffice it to say that lawyers tend to use the former and economists the latter. I’m a chemist, and thus a free agent. The goal of BCA is to maximize the net benefit, defined as benefit – cost. I simply choose to read the minus sign as a hyphen.) In an October symposium, Benefit-Cost Analysis and the Courts, Sunstein announced that the Cost-Benefit State is now upon us. The panelists largely agreed with that assessment, citing a trio of SCOTUS cases that illustrate the trend.
In Whitman v. American Trucking Associations (2001), the court stated that EPA could not consider costs as a factor in setting ambient air quality standards without a clear “textual commitment” in the statute. In a concurring opinion, however, Justice Breyer argued that “other things being equal, we should read silences or ambiguities in the language of regulatory statutes as permitting, not forbidding, this type of rational regulation.”
Then in Entergy Corp. v. Riverkeeper (2009), the court used Chevron deference to rule that “it was well within the bounds of reasonable interpretation for the EPA to conclude that cost-benefit analysis is not categorically forbidden” in setting a standard under the Clean Water Act.
Finally, in Michigan v. EPA (2015), the court read the vague phrase “appropriate and necessary” in a section of the Clean Air Act as a statutory mandate requiring EPA to weigh costs against benefits. Justice Scalia wrote for the majority that “One would not say that it is even rational, never mind ‘appropriate,’ to impose billions of dollars in economic costs in return for a few dollars in health or environmental benefits. . . No regulation is ‘appropriate’ if it does significantly more harm than good.”
Nor is this doctrinal shift likely to pass into history with the passing of Justice Scalia. Despite the 5-4 decision, Michigan actually counted nine votes for the principle that costs cannot be ignored. Writing for the four dissenters, Justice Kagan argued that, unless Congress specifies otherwise, “an agency must take costs into account in some manner before imposing significant regulatory burdens.”
The excellent Penn RegBlog recently hosted a debate over the evolution of BCA as a default principle for evaluating regulatory decisions. In a recent Federalist Society Teleforum, How Should “Administrative Law” be Taught Today, some panelists argued that BCA has become so important to administrative law that it should become part of the core curriculum in our law schools. If so, we should get it right.
In explaining how BCA works, many observers will cite Ben Franklin’s Prudential Algebra, which involves making a list of pros and cons, and weighing them against each other, before making a consequential decision. But that analogy misleadingly suggests an autonomous decision maker. In applying BCA to regulation, I think it is important to stress that BCA is intended not simply to inform a solitary decision maker, but to help solve a serious principal-agent problem. The 2016 Nobel Memorial Prize in Economics was awarded to two economists for their contributions to solving a similar problem in contract theory. How, for example, might a board of directors compensate a corporate CEO to align her incentives with the interests of the shareholders? Who should bear which risks, and to what degree? Such compensation schemes are of little use, however, when we think about the incentives faced by regulators. How can we ensure that public servants use their considerable powers in the service of the public interest? What exactly do we mean by the public interest, anyway?
In subsequent posts I want to explore these questions, and how they shed light on the assumptions (like rationality) that underlie the theory of BCA, and on the institutional arrangements (Who does the analysis? Who reviews it?) that surround its use in administrative law.