New Foundations of Cost-Benefit Analysis
Cambridge, MA: Harvard University Press, 2006, 256 pp.
Among legal scholars, views about the value of cost-benefit analysis in government regulation range from rampant enthusiasm to outright rejection. Law professors Mathew D. Adler (University of Pennsylvania) and Eric A. Posner (University of Chicago) fall in between these extremes, albeit with considerably more sympathy for the approach than not. In New Foundations of Cost-Benefit Analysis, they portray cost-benefit analysis—defined as a procedure that measures the impact of agency choice on a plurality of aspects of human well-being using a money scale—as an aid to improving the allocation of society’s resources while encouraging transparency and accountability in the decisionmaking process.
At the same time, they are troubled by the fact that the economic methodology is based primarily on the preferences of individuals, as measured by their willingness to pay for gains and their expectation to be compensated for losses. Specifically, Adler and Posner consider the economists’ claim to maximize human well-being (welfare) as flawed, because it is usually infeasible and certainly uncommon to compensate the “losers” from any given policy, and because the marginal gains (or losses) of the rich are counted equally with those of the poor, even though the practical effect on the poor is much greater.
This thin but cogent volume addresses a range of controversies in the field of regulatory analysis, including the importance (or unimportance) of individual rights and nonmonetary values as they relate to the use of cost-benefit analysis in decisionmaking. The authors also examine and largely dismiss claims that cost-benefit analysis inappropriately values goods and services that are “priceless” and that discounting human lives is misplaced in regulatory analysis.
Although fervent in their critique of cost-benefit analysis as currently practiced, Adler and Posner see considerable hope for its future use, if only the overall goal of the analysis could be reconsidered. Specifically, they argue that the preferred goal should be to enhance their own definition of well-being—what they call “weak welfarism”—rather than the one professed by most economists. Citing the extant literature, they argue that human well-being and preference-satisfaction are by no means equivalent. Building on the work of well-known scholars such as Daniel Kahneman, Thomas Scanlon, and Amartya Sen, they believe that preferences must be self-interested in order to affect well-being and that interpersonal welfare comparisons are both feasible and necessary. Economists are generally hesitant to make such judgments, largely because of the difficulty of distinguishing between preferences that are self-interested and those that are not, and because of the inherent difficulty (some might say impossibility) of making interpersonal comparisons of the sort envisioned by Adler and Posner.
The real question, of course, is who defines and ultimately implements the Adler-Posner notion of overall wellbeing, a topic on which the authors make only modest progress. While embracing Khaneman’s optimism about the potential feasibility of measuring pleasure and pain, it is not clear whether the authors advocate a Scanlon-style notion that primary goods are the central basis of well-being or, alternatively, whether they favor an approach focused more on individual freedom and the capability to act on one’s own, as advanced, for example, by Sen. In other words, even if one agrees that strictly economic measures are inadequate, there is still the question of devising an alternative measure of societal wellbeing.
Coincident with their efforts to recast the cost-benefit framework, the authors also launch into a broad-scale defense of this (imperfect) technique, arguing that compared to the safety maximization approach used by the U.S. Food and Drug Administration or other alternative decision procedures, cost-benefit analysis best advances overall wellbeing. They note that although a project or regulation that passes a cost-benefit test does not necessarily improve overall well-being, it is more likely to do so than one that passes only a safety-maximization test. Central to their case is the assertion that cost-benefit analysis enables agencies to take into account most welfare-related effects of regulations without placing an impossible burden on decisionmakers.
The authors explicitly reject the notion that policymaking is irreducibly about values, quintessentially political, and therefore not amenable to cost-benefit calculations. They recognize that when issues are framed in terms of conflicting rights, it becomes virtually impossible to conduct meaningful policy deliberations. For example, even the most ardent moralist would recognize that, as a practical matter, decisions about which pollutant to control and how much control to impose are not readily resolvable on moral grounds.
Adler and Posner also recognize that transparency of the type promoted by cost-benefit analysis is essential for regulatory processes, because agencies may have important bureaucratic or ideological objectives other than economic efficiency. Further, transparency empowers citizens and elected officials to monitor agencies and discipline those that go awry. The authors emphasize the political and institutional context for administrative decision procedures.
Ideas for change
Adler and Posner advance a number of ideas for modifying the cost-benefit framework for better alignment with their weak welfarism approach. For example, they propose that agencies should “launder” preferences that are poorly informed, or what they term “nonideal,” and that they should ignore “disinterested” preferences based on moral views rather than direct self-interest. Under their approach, the preferences of an individual unlikely to ever visit Alaska for a pristine coastline in that state might be given less weight than an Alaskan resident’s desire for employment in the oil industry. Adler and Posner would probably value the direct gain to the resident more highly than the nonresident’s moral concerns. Of course, an environmentalist might view it the other way around.
Adler and Posner also argue that traditional cost-benefit analysis is most likely to diverge from their weak welfarism approach when those who benefit from a project or regulation are much poorer or richer than those who lose. Their solution is to impose a correction to the analysis to reflect the fact that the marginal utility of money declines as income rises. Unfortunately, they offer few specifics about their proposed modifications and fail to account for the extensive empirical literature that has grappled with similar issues.
Two studies are particularly relevant. In the early 1990s, Jon Elster looked at the way in which interpersonal comparisons are made to select students for college admissions, patients for transplants, or workers for layoffs. He found that in some arenas, there is a systemic tension between giving scarce resources to those who need them most and giving them to those who can benefit most from them. This is an important observation because it highlights the complexity of the issues and, some might argue, the futility of trying to develop a single analytical framework to address such diverse concerns.
Some years earlier, Dale Jorgenson and colleagues developed an empirical approach for “exact aggregation” of individual welfare that enables the articulation of efficiency tradeoffs when one moves away from the individual preference framework. Such an approach, although not imposing particular weights or values on the gains accruing to different demographic or income groups, facilitates the consideration of overall well-being as opposed to narrower efficiency concerns. Importantly, it does so in a way that allows the use of a range of considerations and does not constrain the decisionmaker to adopt a fixed, and likely indefensible, set of weights regarding gains and losses to rich versus poor, urban versus rural, etc.
These two studies highlight a real challenge to the Adler and Posner argument. On the one hand, the authors seek to correct what they identify as limitations in the cost-benefit framework, while on the other hand they shy away from addressing some of the tough questions that cannot be ignored if one is to move beyond that framework. A possible response to their critique, one embraced by many mainstream economists, is to acknowledge many of the limitations they highlight while simultaneously advancing pragmatic solutions for protecting the poor from excessive harm, even at the expense of reducing the stated regulatory goals or benefits.
As is widely known, the basic dynamic of political life is often at variance with the efficiency orientation of cost-benefit analysis. Typically, the political process recoils from the laissez-faire nature of traditional welfare maximization. And despite the efforts of some zealots in the mid-1990s to enshrine cost-benefit analysis in the decisionmaking process for federal regulations, most economists shy away from such an approach. Rather, they see cost-benefit analysis as a tool to be used in the process—to be available at the “elbow” of the decisionmaker— not as a rigid rule to replace judgments by competent officials.
Interestingly, Executive Order 12866, issued by President Clinton, replaced the stipulation contained in a prior Reagan presidential order that benefits must outweigh costs with a requirement for “a reasoned determination that the benefits …justify the costs.” Further, agencies are specifically mandated to “include both quantifiable measures . . . and qualitative measures of costs and benefits” and “to 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.” Thus, decisionmakers are instructed to consider a broad range of equity and other factors when applying the results of cost-benefit analyses.
Most economists also recognize that although the technical process of making complex policy decisions is more accessible to those interests that can afford to deploy competent experts, any technique employed in the political process may be distorted to suit parochial ends. As Harman Leonard and Richard Zeckhauser noted some years ago, “cost-benefit analysis can be an advocacy weapon and it can be used by knaves. But it does not create knaves and to some extent it may even police their behavior. The critical question is whether it is more or less subject to manipulation than alternative decision processes.” Their claim is that the ultimate grounding in analytic disciplines affords some protection.
The authors’ endorsement of at least some form of cost-benefit analysis and their rejection of its use as a “super procedure” for decisionmaking puts them squarely in the mainstream of the economics community. Although their focus is primarily theoretical, they also aspire to provide practical advice. Their recognition that cost-benefit analysis tracks only one part of the moral landscape is also consistent with the views of most economists, even if their call for the laundering of preferences and sensitivity to the declining marginal utility of money is not. Overall, the book will please neither of the extremes in the debate. For that reason alone, it is a welcome addition to the literature.