The rise of interest in evidence-based policymaking has created incentives for regulatory agencies to demonstrate the overall benefit–cost merits of their policies. An agency can use evidence to choose more cost-beneficial policies, or it can create the appearance of desirable policies by changing the ground rules by which it assesses a policy’s merits.

The Consumer Product Safety Commission (CPSC) recently chose the latter course when monetizing the benefit of mortality risk reductions for children from a proposed safety standard for operating cords on custom window coverings. The cords are currently estimated to be responsible for nine fatal injuries annually. Each of those deaths is a tragedy, but together their loss as measured by typical value of a statistical life (VSL) estimates would not justify the cost of the proposed standard. Instead of accepting that calculus, the CPSC changed its policymaking rules to double—and considers tripling—the VSL to analyze the proposed rule.

Equitable VSL / Mortality costs comprise the most prominent share of life-saving policy benefits, and risks to children are a major focus of CPSC efforts. Doubling the rate at which regulations’ benefits are valued can result in major swings in regulatory policy attractiveness.

Agencies throughout the government use VSL estimates to monetize the mortality risk reductions of policies. The underlying principle guiding benefit assessment for mortality risks and other policies is that it is based on individual willingness to pay for the risk reduction. The principal source of willingness-to-pay values consists of data drawn from actual decisions that people make with respect to mortality risks. Most of the revealed preference estimates are drawn from studies of wage premiums workers receive for mortality risks. There is almost a half century of economics literature documenting the magnitude of the wage premiums workers receive for health risks.

Agencies use this information to apply an average VSL in the range of $11 million to $12 million. These values make no distinctions based on age, income, race, gender, or other personal characteristics. We refer to the practice of valuing risks symmetrically as equitable risk tradeoffs.

There is, of course, potential heterogeneity in the VSL. More-affluent people require higher levels of compensation to incur a given risk. There are also age variations in the estimated VSL amounts across different age groups. Estimates of the VSL for labor market risks display an inverted U‑shaped pattern that peaks in middle age and is lower for a worker aged 20 than it is for a worker aged 60. However, government agencies do not make distinctions related to differences in the VSL by age, but instead treat mortality risks symmetrically, consistent with our equitable risk tradeoff approach.

Estimating VSL for children / How should we think about valuing risks to children? Instead of exploring private willingness to pay to reduce mortality risks, the government might focus on some other measures or consider what mortality risk reductions are worth, such as longevity of the individuals whose lives are being protected. Children have longer expected remaining lives than adults, so a greater quantity of life is at risk. However, government agencies have generally not adopted a length-of-life metric for valuing mortality risks. The CPSC itself acknowledges that the Office of Management and Budget has specifically cautioned against using age-adjustment factors when applying the VSL and that no other agency uses a different VSL number for children. But it also notes that the CPSC is not legally required to follow OMB guidance.

The CPSC goes beyond appealing to a quantity-of-life rationale for doubling the VSL for children. Its justification draws on stated preference surveys, some of which indicate that people might be willing to value children’s lives more highly than the lives of adults. Responses to hypothetical survey questions are often not a useful guide for policy. Stated preference evidence is not as informative as revealed preference data based on actual risk-taking decisions. Stated preference studies can be instructive but are often problematic and are subject to rampant potential biases.

Besides the broader concerns just mentioned, the available evidence with respect to children is quite sparse, particularly compared to the huge literature on VSL more generally. CPSC cites a review of only five stated preference articles that were based on four surveys. Even if the studies are reliable, they constitute a very slim empirical foundation for a major shift in benefit assessment practices.

But the deficiency of the empirical justification offered by the CPSC is even greater. To utilize any benefit value, it is a prerequisite that the analyst demonstrates that it is appropriate to transfer the benefit value from one context to a different situation. The available stated preference studies cited by the CPSC encounter two deficiencies with respect to the benefit transfer issue. First, half of the samples considered focused on populations outside the United States: one in Italy and one in France. Assessments of the VSL vary greatly by country. The age-related differences in the relative value of risks to children may vary as well. Countries have different age distributions, income levels, health care systems, and social norms.

The second benefit transfer deficiency is that none of the types of death considered in these articles are similar to the nature of the deaths addressed by CPSC policies. Cancer, respiratory disease, and foodborne illness deaths are the focus of the surveys, not traumatic injuries regulated by the CPSC, such as children injured by cords from window coverings.

The practical impetus for the CPSC’s effort to use a greater VSL for children is to justify a prospective regulation. Based on a conventional VSL, the benefits for the proposed corded blind regulation fall far short of the costs. Doubling the VSL for children boosts the apparent attractiveness of the regulation, but even that effort to bolster the policy’s benefits does not carry the day. The CPSC then presents a sensitivity analysis indicating that a tripling of the VSL for children would come close to making the benefits greater than the costs. While the CPSC’s proposed guidance for its standard VSL rate when valuing risks to children is to double the average societal VSL, the CPSC may advocate whatever VSL multiple is needed to create the illusion of a desirable policy in order to make undesirable regulations appear to be worthwhile.

At the other end of the age spectrum, the CPSC also considers regulations to protect senior citizens from product injuries. How it will value deaths of seniors may turn out to be even more problematic. In its recent analyses, the CPSC suggested that the potential for using a lower VSL for seniors is an active area of research, which is a topic that we addressed in an earlier Regulation article. (See “What Are 750,000 Senior Deaths Worth?” Winter 2022–2023.) If the CPSC adopts a lower VSL for senior citizens, it will once again use out-of-the-mainstream practices for regulatory analysis. Ever since the outcry that resulted when the Environmental Protection Agency used a “senior discount” to value mortality risks for people over age 65 in its 2003 analysis of the Clear Skies initiative, government agencies have steered clear of devaluing the lives of senior citizens.

Looking ahead / At some point, government agencies may choose to adopt different VSL levels for children or other age groups such as adults over 65. Any future efforts to improve the mortality risk calculations for government regulations affecting children or other demographic groups should be based on solid empirical evidence rather than an attempt to justify regulations that would not otherwise pass muster based on economic efficiency considerations.

Readings

  • Pricing Lives: Guideposts for a Safer Society, by W. Kip Viscusi. Princeton University Press, 2018.
  • “Promoting Equity through Equitable Risk Tradeoffs,” by Thomas J. Kniesner and W. Kip Viscusi. Journal of Benefit–Cost Analysis, forthcoming.
  • “Publication Selection Biases in Stated Preference Estimates of the Value of a Statistical Life,” by Clayton J. Masterman and W. Kip Viscusi. Journal of Benefit–Cost Analysis 11(3): 357–379 (2022).
  • “Valuing Children’s Fatality Risk Reductions,” by Lisa A. Robinson, William J. Raich, James K. Hammitt, and Lucy O’Keeffe. Journal of Benefit–Cost Analysis 10(2): 156–177 (2019).