Association for Public Policy Analysis & Management

Panel: Cost-Benefit Analysis in Theory and Practice

(Tools of Analysis: Methods, Data, Informatics and Research Design)

Cost-benefit analysis is now widely used by the federal government, especially in rulemaking, and increasingly by state governments to assess social policies. It has yielded many useful results in fields as diverse as education, criminal justice, environmental, and health policy. It has helped public officials to allocate scarce resources more efficiently and to evaluate the merits of competing public policies. Yet scholars and policy analysts still wrestle with some fundamental questions as they apply cost-benefit techniques to specific policy interventions and proposals. One vexing question involves benefit transfers in which information from a completed study informs the prediction of benefits in a different context or policy domain. Behavioral economists have raised some questions about the propriety of benefit transfers when individuals make decisions that do not appear consistent with maximizing their own utility. Confident use of cost-benefit analysis in the face of these behavioral findings requires guidelines for determining when benefits can be plausibly transferred. Another big question is when to conduct benefit-cost analysis or cost-effectiveness analysis and how to integrate that analysis with independent impact evaluations. Some contexts are more appropriate than others. Considerations may include: outcome specification; treatment contrast; implementation fidelity; the role of mediators; power of the test; and meta-analysis. In short, some research settings lend themselves readily to benefit-cost analysis or cost effectiveness analysis, while others do not. The actual application of cost-benefit analysis techniques can be challenging. Often, it involves splicing original data and secondary data sets, using proxy variables to bridge gaps between policy interventions and outcomes of ultimate interest, judgments about whether to use willingness to pay estimates as the basis for anticipated future benefits, and difficult decisions on whether to include estimates that are statistically insignificant. In this panel, we propose to illuminate theoretical and practical issues that make cost-benefit analysis both attractive and challenging. The ultimate goal of the panel is to improve our understanding of the need for cost-benefit analysis and ways to improve it.