Regulatory costs of mythic proportions

Critiques of risk regulation rely pervasively on estimates of the costs of various federal regulations per life saved. As Professor Heinzerling illustrates in this Article, most of these estimates derive from a single source, a table prepared in the 1980s by an economist at the Office of Management and Budget, John Morrall. That table reports costs per life saved reaching hundreds of millions, even billions, of dollars. These oft-cited estimates are, however vastly higher than the agencies' own estimates of costs and benefits. The divergence in estimates stems from the fact that Morrall adjusted the agencies' figures by discounting future lives saved and, in many cases, greatly decreasing estimates of risk. Without these adjustments, Professor Heinzerling demonstrates, the costs per life saved of the allegedly costliest regulations drop, in virtually every case examined, to less than $5 million. This number compares favorably to currently cited estimates of the monetary value of a human life. Moreover; Morrall's calculations exclude many unquantified benefits of the regulations in question. An assessment of the cost-effectiveness of current risk regulation thus turns on one's opinions regarding discounting, risk assessment, and regulatory purposes. These in turn depend on one's views of the relative worth of lives saved today and lives saved in the future, the appropriate response to scientific uncertainty, and the relevance of unquantified benefits. As Professor Heinzerling argues, these matters involve choices among values about which reasonable people may disagree. Thus, instead of providing an objective basis for setting regulatory priorities and judging the wisdom of regulation, figures on costs per life saved embody the very normative judgments they have been thought to support.