On Balance: Review of “Pricing Lives: Guideposts for a Safer Society” by W. Kip Viscusi

May 25, 2018

Review by James K. Hammitt

Pricing Lives: Guideposts for a Safer Society  (Princeton University Press, 2018) is a tour de force. It provides an entertaining, accessible account of the modern approach to valuing mortality risk. In non-technical prose, it covers the major aspects of the approach and how it should be applied to social decisions.  It shows how US regulatory agencies’ adoption of the ‘value per statistical life’ (VSL) to replace the ‘cost of death’ (i.e., human capital) approach has led to more-protective regulation and argues the VSL approach should be extended beyond regulation, to private-sector decisions about product design and to government sanctions for regulatory violations. It also finds that the values used outside the US are typically far too small; revising these upward would lead to more-protective and more-appropriate regulation in other countries.

 

Over the course of several chapters, the author, W. Kip Viscusi (Vanderbilt University), examines the rationale and controversy surrounding the variation in VSL with individuals’ characteristics (especially age and income), with the cause of death (e.g., traumatic injury, lingering disease), and the source of the hazard (e.g., occupational, consumer-product defect or lack of safety features, terrorism) and the implications of these differences for public policy. He discusses issues as diverse as the difference between ‘identified’ and ‘statistical’ lives and detecting and correcting for effects of publication bias when seeking to develop a value from the literature. For evaluating risk equity, he proposes using a VSL-like measure (equality of the marginal costs of risk reduction across people).

Viscusi is by far the leading contributor to the estimation of VSL and its adoption by US regulatory agencies. As described in this book, early in his career he was asked to resolve a dispute between the Occupational Safety and Health Administration (OSHA) and the Office of Management and Budget (OMB) about the net benefits of requiring firms to label dangerous chemicals used in the workplace. The hazard-communication rule was one of the costliest proposed in the early years of the Reagan Administration. Executing its new authority to review regulatory impact analyses under Executive Order 12291, OMB had found fault with the OSHA analysis and concluded that the rule would not pass the cost-benefit test. Viscusi accepted most of the OMB criticisms, but argued that the mortality-risk reductions should be valued using the $3 million VSL he had estimated rather than the much smaller value from the cost-of-death approach used by OSHA and OMB, and hence the net benefits were positive. Viscusi’s analysis was persuasive and the regulation was adopted.

Since the early 1980s, US regulatory agencies, led by the Department of Transportation and Environmental Protection Agency (EPA), have adopted the VSL approach to valuing mortality risk and have converged (despite emphasizing somewhat different strands of the literature) on a current value of roughly $10 million. In contrast, although many other countries use this approach in regulatory evaluation, Viscusi argues that the values applied are generally much too small, e.g., US$2.1 – 2.4 million in Great Britain. He attributes this in part to governments anchoring on older, cost-of-death estimates and on the limited quality of occupational-fatality-risk data outside the US.

Viscusi emphasizes that the VSL is “not a universal constant” and discusses how it varies with characteristics of the population (income, age) and of the hazard. He notes that it is politically difficult to use different VSLs for regulations within a country, but that in some cases, such as Federal Aviation Administration regulations that affect airline safety, a higher VSL is appropriate as airline passengers have higher-than-average income and pay the incremental costs of regulation through higher ticket prices. He does not, however, argue for using a lower VSL for other products, as consumers may expect a “normal” safety level.

While adoption of the VSL approach has led to more-protective and economically rational regulation, it could contribute to much greater social benefits if used appropriately in private-sector decisions about product design, in the courts, and in government enforcement. Viscusi describes some of the prominent motor-vehicle product-liability suits, ranging from the Ford Pinto fuel-tank-fires to the GM ignition-switch cases. In the 1960s and 1970s, automakers conducted benefit-cost analyses of at least some design options and safety features. Viscusi argues these analyses were flawed, as they valued fatality risks using the cost of death (equal to the cost of liability to the manufacturer) rather than the VSL (though empirical estimates of VSL were not available to the early cases). As a result, manufacturers produced more-dangerous products than their buyers would have preferred, had the buyers known the incremental risks and costs of the alternatives.

But a larger problem is that the manufacturers learned, from bitter experience, that explicitly evaluating risks and costs of alternatives could make them appear callous to consumers’ safety and vulnerable to product-liability suits; failure to choose a safer alternative, when identified, led juries to assess punitive damages of $100 million and more. As a result, automakers (and presumably other firms) have evidently stopped evaluating product-safety features, or at least are careful to avoid leaving any evidence. Viscusi describes how by the 2000s GM’s culture discouraged explicit discussion of product safety. Moreover, GM urged its staff to avoid using words with “emotional connotations,” not only obvious terms like “deathtrap” and “rolling coffin,” but even seemingly neutral words required for discussing risk-cost tradeoffs, such as “safety,” “failure,” and “problem.”

A key problem for manufacturers seeking to make appropriate risk-cost tradeoffs is that jurors (and judges) attempting to determine liability and quantify damages to an identified individual have difficulty weighing the visible, realized harm to the claimant against both the comparatively trivial (per product) costs of alternative designs that might have prevented the injury and the (largely invisible) individuals who would have been harmed by an alternative design. Viscusi suggests some changes to rules of evidence to encourage firms to use benefit-cost analysis in product-safety decisions.

Another way that VSL could improve risk management is by justifying large increases in the fines government can levy for regulatory violations, which are limited by statute. Viscusi notes that the maximum penalty OSHA can issue for violations associated with worker fatality is about $12,500 for serious violations and $125,000 for willful or repeated violations. Other agencies, including the Food and Drug Administration, National Highway Traffic Safety Administration, and EPA also are constrained by statute to penalties that pale in comparison with the economically appropriate incentive.

Pricing Lives: Guideposts for a Safer Society shows how much progress US regulatory agencies have achieved in developing economically rational and more-protective policies by accepting the VSL as the relevant concept for measuring the benefits of reduced mortality risk. It also shows how much further this guidepost can lead toward a safer society if its use is extended to civil liability and enforcement.

 

James K. Hammitt is professor of economics and decision sciences at the Harvard T.H. Chan School of Public Health, director of the Harvard Center for Risk Analysis, and associate of the Toulouse School of Economics. His research and teaching concern the development and application of benefit-cost, decision, and risk analysis to health and environmental policy.