Cass Sunstein's Cost-Benefit Lite: Economics for Liberals

Abstract

Book Review: Cass R. Sunstein, The Cost-Benefit State: The Future of Regulatory Protection (ABA Section of Administrative Law and Regulatory Practice, 2002) Cass R. Sunstein, Risk and Reason: Safety, Law, and the Environment (Cambridge University Press, 2002). In these two books, Professor Sunstein promotes cost-benefit analysis (CBA) as the presumptively correct decisionmaking tool for virtually all environmental health and safety regulation. His claims are both normative and descriptive. By avoiding the usual welfare economics defense of CBA, his normative claim sets aside the external critiques of CBA (like the incommensurability problem), and offers instead a set of "pragmatic" justifications: 1) CBA will rationalize government decisionmaking and reduce the undue influence of interest groups that exploit "cognitive distortions" that cause ordinary people to inaccurately assess risks; 2) CBA will increase transparency by forcing regulators to evaluate and describe the full array of consequences flowing from proposed regulations. These justifications are ultimately unconvincing because Professor Sunstein fails to take seriously the central concern of the internal critiques of CBA - indeterminacy. Ironically, when he tests his theories in practice with a careful and cogent case study of EPA's CBA of arsenic in drinking water, he discovers indeterminacy of an astonishing magnitude. He concludes that reasonable people might peg the benefits of the arsenic rule as low as $13 million or as high as $3.4 billion. Though he remains staunchly committed to CBA, the indeterminacy of CBA fatally undermines both of his "pragmatic" justifications. On the descriptive side, Professor Sunstein claims that the United States is becoming a "Cost-Benefit State" and that, as part of this trend, federal courts are adopting a set of "cost-benefit default principles" under which they apply a presumption in favor of CBA to their review of agency decisionmaking. A close look at these cases, however, reveals that in many instances results that he attributes to the "default principles" are more easily explained by reference to the specific statutory language at issue or as classic and not particularly remarkable applications of traditional administrative law principles. Moreover, on the two occasions on which the "default principles" have been tested in the U.S. Supreme Court, they have been roundly rejected.