• Choose a category
• All
• Classic-Fiction

In the early 1980s, I took part in programs intended to educate members of the media (e.g., television networks, newspapers) about various aspects of environmental and health and safety risks. To prepare participants prior to my talk, I developed a short questionnaire. One question asked: “In considering the clean-up of old hazardous waste sites, rank the following in order of importance: economic cost of the clean-up, potential human life lost or sickness due to the hazard, and potential damage to the natural environment.” All of the approximately 100 respondents ranked economic costs as least important, and every respondent but one had potential human life lost or sickness as most important. In a subsequent discussion, very few of these individuals seemed to think that it would be worth a $2 billion increase in clean-up costs to avoid 20 people being sick for a week and then recovering completely. The point is that it is essentially meaningless to prioritize objectives without understanding how much of each of the objectives ($2 billion and 20 one-week illnesses) they are prioritizing.

One of the earlier questions in the same questionnaire asked: “In evaluating risks to lives of the public, judgments should or should not be made about the relative importance of the lives of different individuals?” The large majority of the respondents replied “should not.” A later question asked: “Do you agree or not that in allocating government funds to reduce risks to citizens, the saving of the life of any individual should be counted equally?” In this case, almost all participants responded that they “agree.” I pointed out that counting the lives of individuals equally was a value judgment and that the responses on these two questions conflicted. This caused some consternation. Then I asked about the relative evaluation of saving children versus the elderly. Most people preferred saving children, which also conflicted with their “agree” response to the question. Many participants became somewhat disturbed, as they recognized that value judgments about the relative importance of saving the lives of different members of the public are essential to health and safety policy decisions.

Another question asked: “If it were known for sure that an improvement in the maintenance of air pollution control equipment on coal-fired power plants would save two expected lives a year, what is the maximum dollar amount that should be expended for this maintenance?” Possible responses were “$100k,” “$1million,” “$10 million,” “$100 million,” and “whatever it costs.” The responses were rather evenly spread across these options. Many people, including many involved in setting policy, do not understand that value tradeoffs between economic costs and expected fatalities are an inherent part of numerous policy decisions, nor do they seem to understand that spending $100 million to save lives in one arena means that this money cannot be spent for other purposes including possibly avoiding many more potential fatalities elsewhere.

UNDERSTANDING THE STRUCTURE OF POLICY DECISIONS

In 1987, I gave a two-hour presentation on probabilities as part of a two-day workshop for earth sciences professionals at the University of California. Preceding my presentation, I asked all eighty-one participants to fill out a questionnaire that asked the following: “Is it true or false that there is a reasonable chance of a moderate to large earthquake in the San Francisco Bay area in the near future?” Then I asked for separate definitions of reasonable chance, moderate to large earthquake, San Francisco Bay Area, and near future. I perused the results and was able to begin my presentation as follows.

You are in complete agreement on the main question: Everyone indicated that the statement was true. However, there were significant discrepancies among the definitions. A reasonable chance ranged from 5 to 92 percent. Does the term moderate to large earthquake include very large earthquakes? About half of you thought so and the other half didn’t. Some defined this as a Richter