and when risk management institutions are trusted, explaining data better often leads to improved decision‐making. For other risk problems, such as when experts claim the risk is not significant or large, but when people are extremely concerned or outraged, explaining risk information is seldom effective in calming people or reducing outrage. Researchers and risk managers recognized solutions had to be found elsewhere.
Several events influenced developments during Stage 2. One was the passage of multiple laws requiring public participation, consultation, and the right‐to‐know. A second were events that resulted in public outrage. One such event was the nuclear power plant accident at Three Mile Island in Pennsylvania. The accident was the most significant accident in US commercial nuclear power plant history. It raised awareness of the dangers associated with industrial facilities and concerns about caring, competence, honesty, and transparency by industry. Two other significant events in Stage 2 were the ban by the US Environmental Protection Agency (EPA) on the use of the pesticide dichloro‐diphenyl‐trichloroethane (DDT)8 and the hazardous waste crisis at Love Canal, New York.9 President Jimmy Carter’s declaration of a State of Emergency at Love Canal was made following an increase in skin rashes, miscarriages, and birth defects among residents.
Both the DDT and Love Canal events raised public awareness of, and concerns about, the risks associated with agricultural chemicals and the unregulated dumping of hazardous waste. These and related events also heightened awareness of the difficulties and challenges of risk communication and presenting risk data to emotionally charged audiences.
During Stage 2, scientists developed many techniques for explaining and putting risk data in perspective. One of the most popular of these techniques was to present risk comparisons.10 Empirical evidence and theory suggested that risk comparisons could improve public understanding of risks and encourage the adoption of protective behaviors. Researchers hypothesized that risk comparisons could be especially useful in helping people make informed decisions about low‐probability, high‐consequence events, such as major flood or earthquake.11
Researchers also hypothesized that risk comparisons might become counterproductive if the public suspects that they are used to minimize or magnify a problem.12 I provide several examples of risk comparison studies below.
Food risk comparisons. To gain perceptive and improved understanding of the risks posed by food, numerous studies have compared the risks posed by different foods, food products, and food additives.13 One of the earliest and best‐known comparative analyses of the risks of this type were the studies on food risks, diet, and cancer by Professor Bruce Ames and his colleagues at the University of California, Berkeley. These studies compared the cancer risks of foods that contain synthetic chemicals (e.g., food additives and pesticide residues) with the risks of natural foods. An important conclusion of the research was that synthetic chemicals represent only a tiny fraction of the total carcinogens in foods. The researchers pointed out that natural foods are not necessarily benign. Large numbers of potent carcinogens (e.g., aflatoxin in peanuts) and other toxins are present in foods that contain no synthetic chemicals. Natural carcinogens are part of a plant’s natural defense system. Human dietary intake of these natural carcinogens can be as much as 10,000 times greater than the dietary intake of potentially carcinogenic synthetic chemicals in food. However, the many natural anticarcinogens also in food provide partial protection against natural carcinogens in food.
Critics of the work of Ames and his colleagues have argued that his risk estimates are inflated. Critics have also argued against the implicit, and sometimes explicit, argument and risk communication that natural carcinogens in foods deserve greater societal and regulatory attention and concern than synthetic chemicals.14
Energy risk comparisons. One of the earliest and the best‐known studies of energy technologies was a study conducted by Inhaber (1978) for the Atomic Energy Control Board of Canada.15 The study compared the total occupational and public health risks of different energy sources for the complete energy production cycle – from the extraction of raw materials to energy end use. The study examined the risks of eleven methods of generating electricity: coal, oil, nuclear, natural gas, hydroelectricity, wind, methanol, solar space heating, solar thermal, solar photovoltaic, and ocean thermal. Two types of risk data were analyzed: data on public health risks from industrial sources or pollutants and data on occupational risks derived from statistics on injuries, deaths, and illnesses among workers. Alternative sources of energy were compared on the basis of the calculated number of person‐days that would be lost per megawatt year of electricity produced. Total risk for the energy source was calculated by summing the risks for the seven components of the complete energy production cycle: materials acquisition and construction, emissions from materials acquisition and energy production, operation and maintenance, energy backup system, energy storage system, transportation, and waste management.
Inhaber’s report came to the following conclusions:
Most of the risk from coal and oil energy sources is due to toxic air emissions arising from energy production, operation, and maintenance.
Most of the risk from natural gas and ocean thermal energy sources is due to materials acquisition.
Most of the risk from nuclear energy sources is due to materials acquisitions and waste disposal.
Most of the risks from wind, solar thermal, and solar energy sources arise from the large volume of construction materials required for these technologies and the risks associated with energy backup systems and energy storage systems.
The most controversial aspect of Inhaber’s report was the widely communicated conclusion that nuclear power carries only slightly greater risk than natural gas and less risk than all other energy technologies. Inhaber reported, for example, that coal‐based energy has a 50‐fold larger worker death rate than nuclear power. The report also communicated that, contrary to popular opinion, (1) nonconventional energy sources, such as solar power and wind, pose substantial risks; and (2) the risks of nuclear power are significantly lower than those of nonconventional energy sources.
Inhaber’s report can be criticized from several perspectives. For example, the study mixed risks of different types, used risk estimators of dubious validity, made questionable assumptions to cover data gaps, failed to consider future technological developments, made arithmetic errors, and double‐counted labor and backup energy requirements. Perhaps the most important criticism of Inhaber’s study was methodological inconsistencies. For example, while the study considered materials acquisition, component fabrication, and plant construction in the analysis of unconventional energy sources and of hydropower, the study did not follow the same approach for coal, nuclear power, oil, and gas. Furthermore, the labor figures for coal, oil, gas, and nuclear power included only on‐site construction, while those for the renewable energy sources included on‐site construction, materials acquisition, and component manufacture.
Despite these criticisms, Inhaber’s research represented a landmark effort in the literature on risk communication and risk comparisons. It made a significant conceptual contribution by attempting to compare, and communicate, the risks of alternative technologies intended to serve the same purpose. Also important was Inhaber’s observation that risks occur at each stage in processes and product development, from raw material extraction, manufacturing, and use, to disposal. Inhaber’s central argument was that
