The VSL Is Not Too High.

• Author: Viscusi, W. Kip
• Position: FOR THE RECORD - Value of a Statistical Life - Book review

In his Spring 2018 review of my book Pricing Lives: Guideposts for a Safer Society (Princeton University Press, 2018), Sam Batkins makes four principal points. First, my estimates of the value of a statistical life and those used by government agencies have increased over time and are too high. Second, the result is that "regulators are employing ever-higher figures for the Value of a Statistical Life (VSL) to justify more stringent regulations." Third, trial lawyers are using the VSL to push for greater damages. Fourth, these developments may impose a threat to further economic progress.

In my book I advocate a S10 million figure for the VSL. This number does not in fact represent any kind of quantum leap in the VSL. By way of history, I introduced the VSL into federal regulatory policymaking in 1982 when I was asked by the Reagan administration to resolve a dispute between the Occupational Safety and Health Administration and the Office of Management and Budget over the proposed hazard communication regulation. The $3 million VSL number that I employed then was based on my estimates of the extra pay that workers receive for incurring job-related fatality risks. My estimate was an order of magnitude greater than OSHA's "cost of death" mortality benefit measure, which was based on lost earnings and medical costs.

When converted to 2017 dollars, my 1982 $3 million VSL figure becomes $7.8 million. If we also take into account the increase in real per-capita income and apply an income elasticity for the VSL of 0.5 to 0.6, then extrapolating my 1982 number to current economic conditions produces an estimated current VSL in the $11-$12 million range. My $10 million VSL figure consequently is a bit below the level that would be obtained by taking my 1982 estimate forward.

What then are we to make of the substantial increases in the VSL that have been observed at regulatory agencies during this century? Price adjustments and increased societal income levels undoubtedly play a role but are not the whole story. An additional contributing factor is that before agencies used the VSL to value mortality risks, they relied on the present value of lost earnings as the measure of the mortality risk reduction benefit. Because many agencies anchored on their previous earnings loss estimates, they were slow to fully embrace the estimates of the VSL in the economics literature. As a result of this anchoring effect, the VSL estimates used in Department of...