By Daniel K. Benjamin
People are knowledgeable
about the hazards faced
by individuals in
their age group.
It is widely argued that individuals cannot accurately assess the environmental and workplace risks they face .1 One typical claim is that people poorly assess the hazards of dying: They wildly overestimate the likelihood that they will die of lightning strikes or food poisoning, while grossly underestimating the chances of dying due to stroke or heart attack. This asserted failing has helped establish and expand various environmental, health, and safety regulatory bodies. Mandatory motorcycle helmet rules, automobile air bags, and Superfund site cleanup rules stem directly from the view that people do not have accurate knowledge of the risks they face.
This pervasive claim, the "biased perception hypothesis," is itself flawed, however.2 My colleague William Dougan and I have reexamined the existing body of evidence in this area and found that it misses the issue most relevant to individual decision makers--namely, relevance itself.
Beginning with surveys first conducted by psychologists 25 years ago, researchers have sought to determine if people can estimate population-wide death rates from various causes. On balance, individuals appear to be rather poor at making such estimates. With this finding we have no quarrel. Yet researchers and policy makers have interpreted this finding as implying that people are therefore poor at estimating the chances that they, as individuals, are likely to die of these causes. This interpretation leads to the notion that people must be protected from their misperceptions--presumably by better informed experts in government agencies.
To see the nature of the problem, consider two potential causes of death, accidental falls and deliberate homicides. People of all ages fall down, yet falling is rarely lethal except among the elderly. The chance of dying due to a fall begins to rise sharply at around age 65, and the average age of death due to falls is approximately 74 years. Toward the other end of the spectrum, deaths due to homicide are relatively infrequent except among the young. The overwhelming majority of murder victims are under the age of 35.
In a world in which information--like all other goods--is scarce, one would expect people to economize on its acquisition. In the present context, this means people should specialize in acquiring information about hazards relevant to them, rather than about the hazards faced by some hypothetical or population-wide group. Thus, older people should be acutely aware of the risks of falling down, while younger people should be relatively knowledgeable about the risks of homicide. Moreover, when younger people are asked to estimate the hazards of falling down, for example, we should expect them to understate these hazards for the population, because the risks to them as individuals are so low.
To test these predictions, Dougan and I revisit the data that have served as the foundation for the biased perception hypothesis. We apply the rational expectations approach developed by economists such as Nobel laureate Robert Lucas. This approach is specifically meant to analyze behavior when people must act in the face of incomplete information.
We find that people are, in fact, knowledgeable about the hazards faced by individuals in their age group. In addition, they use these data in useful and systematic ways when predicting fatality rates among the population as a whole. Overall, it appears that people are remarkably unbiased in judging the hazards of activities that are most relevant to them.
This does not mean that people are perfect in judging hazards. The data imply that people do make mistakes--hardly surprising in a world of costly information. The point is that people do not make systematic mistakes, that is, mistakes that might be easily corrected by information readily available to government bureaucrats.
None of this means there is no role for government regulation of environmental, health, or safety risks. But such regulation cannot be justified by claiming that people are systematically erring in their judgments.
Frank Knight, one of the great economists of the 20th century, once observed, "We are so built that what seems reasonable to us is likely to be confirmed by experience or we could not live in the world at all." The evidence Dougan and I have found is consistent with Knight's view. More importantly, the evidence implies that some argument other than biased risk perceptions will have to be invoked by those who would usurp individual responsibility for judging hazards.
1. See W. Kip Viscusi, "The Value of Risks to Life and Health," Journal of Economic Literature, 31 (December 1993), pp. 1912-46 for an exhaustive review of the literature in this area.
2. Daniel K. Benjamin and William R. Dougan, "Individuals' Estimates of the Risks of Death: Part I--A Reassessment of the Previous Evidence," Journal of Risk and Uncertainty, vol. 15, November, 1997, pp. 115-33.
Daniel K. Benjamin is a PERC senior associate and professor of economics at Clemson University. His regular column, "Tangents-Where Research and Policy Meet," investigates policy implications of recent academic research. He can be reached at: email@example.com