Toward the usable recognition of individual benefits and costs in regulatory analysis and governance

Abstract

Regulatory agencies in the United States and Europe have well‐deserved reputations for fixating on the total benefits and costs of proposed and final regulatory actions, without doing any more than anecdotally mentioning the subpopulations and individuals who may bear disproportionate costs or reap disproportionate benefits. This is especially true on the “cost” side of the cost–benefit ledger, where analysts exert little effort to even inform decisionmakers and the public that the costs of regulations might be distributed either regressively or progressively. Many scholars and advocates have observed that regulation can increase the efficiency of market outcomes, but caution about its untoward (or suboptimal) effects on equity. Here, we argue that without considering distributional information about costs and benefits, regulatory policies in fact can also cause violence to notions of efficiency, for two reasons: (i) society cannot hope to approach Pareto‐efficient outcomes without identifying those who must lose so that others can gain more; and (ii) because the harm experienced by involuntary risks and by imposed regulatory costs is likely non‐linear in its magnitude (at the individual level), efficiency is, in fact, a strong function of the shape of the distribution of these effects. This article reviews evidence about the distribution of regulatory costs and benefits, describes how agencies fail to incorporate readily available distributional information, and sketches a vision for how they could analyze costs and benefits to promote more efficient regulatory choices and outcomes.

Ethical Dilemmas in Protecting Susceptible Subpopulations From Environmental Health Risks: Liberty, Utility, Fairness, and Accountability for Reasonableness

“…At some point, decision makers will need to decide which subpopulations warrant special protections in specific contexts and how much protection they deserve (Cranor 2008a). These are difficult choices to make, given the tradeoffs among competing values and interests (Viscusi 1992; Cranor and Finkel 2016). Members of susceptible groups who disagree with the outcomes of these deliberations may still regard them as fair and legitimate if they conform to AFR conditions.…”

“…We argue that while these theories 3 provide diverse perspectives about the values and interests relevant to justice, none of the theories offer the specific guidance required for environmental policy decisions involving trade-offs among the competing values and interests they themselves identify as justice relevant. To promote justice in the distribution of environmental health risks, it is therefore necessary to implement a decision-making procedure that balances competing values and interests fairly and legitimately (Cranor and Finkel 2016). We argue that an approach to public decision making known as accountability for reasonableness (AFR) can complement the theories we discuss and has considerable merit for establishing acceptable environmental health risks for the general population and susceptible subpopulations.…”

Demystifying Evidence‐Based Policy Analysis by Revealing Hidden Value‐Laden Constraints

“…Economists widely acknowledge that these post hoc transfers rarely occur in practice. However, they rarely admit that these transfers cannot occur even in theory unless the analysis identifies the winners and losers …”

“…Analysts could also explore the possibility of using a fraction of an individual's wealth as the unit of cost, so as not to ignore the key fact that a cost of, say, $1000 to a very poor person is much more significant than a $1000 cost to a billionaire …”

“…For example, it is very common for cost‐benefit analysts to use “lives saved” as the central measure of regulatory benefits and to estimate “lives saved” by multiplying the average individual risk by the number of people at risk (for example, a risk of one in one million to a population of fifty million people multiplies out to fifty “statistical fatalities”). But this calculation masks two underlying assumptions: (1) that the harm done by a probability that death or disease will occur is strictly proportional to that probability (in other words, a very high probability of harm is never deemed flatly intolerable, and a very low probability can never be set aside as being too insignificant to affect population welfare) and (2) that society should be indifferent between a uniformly distributed probability of harm and a widely varying probability with the same average (for example, between a situation in which one million people all face a risk of 1 out of 1,000,000, versus one in which 100 of them face a risk of 1 out of 100 and the other 999,900 face zero risk).…”