Comparing Apples to Oranges: Risk of Cost/Benefit Analysis

This essay is reproduced here as it appeared in the print edition of the original Science for the People magazine. These web-formatted archives are preserved complete with typographical errors and available for reference and educational and activist use. Scanned PDFs of the back issues can be browsed by headline at the website for the 2014 SftP conference held at UMass-Amherst. For more information or to support the project, email

Comparing Apples to Oranges: Risk of Cost/Benefit Analysis

by Barry Commoner

‘Science for the People’ Vol. 12, No. 3, May-June 1980, p. 9-10

Barry Commoner is director of the Center for the Biology of Natural Systems at Washington University in St. Louis. This article is a condensed version of “Should this Sucker Get an Even Break?” which appeared in Hospital Practice, March 1979.

Not long ago there was a curiously symbolic clash between two well-meaning public interest groups, the American Cancer Society (ACS) and the Center for Science in the Public Interest. The subject was lollipops. It seems that certain chapters of the Society were selling lollipops to raise money to fight cancer. Some of the lollipops were red ones, and according to the Center, Red Dye No. 40, which they contained, is a suspected carcinogen. The Center wants the Cancer Society to stop selling the lollipops. The Society has replied: “Until Red Dye No. 40 or any other additive has been declared unsafe and taken off the market, its use in manufactured products is perfectly proper.”
Does the ACS think that smoking cigarettes is “perfectly proper” because they have not been taken off the market? Clearly, the Cancer Society would not dream of raising money by selling cigarettes. Apparently, the Cancer Society believes that smoking cigarettes is a more serious risk than sucking red lollipops. But how does one make such comparisons? 

As the concern about the risks of modern technology to people and the environment has been translated into legislation, a basic idea has emerged – that the best way to evaluate such a risk is to compare it with the associated benefits. This is known as risk/benefit assessment. The most recent legislation based on this concept is the Toxic Substances Control Act (TSCA). This act requires that the Environmental Protection Agency (EPA) administrator establish rules for governing the production and use of chemicals which take into account not only a substance’s toxic hazards but also “the benefits of a substance for a given use or uses and the availability of less hazardous substances for the same uses.” One important section of the TSCA law requires what might be called a meta-cost/benefit assessment. This is a cost/benefit evaluation not only of the substance but also of the EPA decision to regulate it – that is, the social cost of administrative action itself.

Although TSCA does not apply to food additives, it is, nevertheless, the most clear-cut statement of the risk/benefit principle and it is an interesting exercise to apply it to the red lollipop controversy. Let us assume that a small cancer risk is associated with Red Dye No. 40, and that the dye adds no nutritional or taste value, so has no benefit to the consumer. With the dye in it, however, the lollipop might be said to be more “appealing” than blue ones or yellow ones. Translated, this means that they sell better. Hence the red dye is of benefit to the people that manufacture and sell the lollipops, not to the consumers. However, the hazard associated with the dye is directed toward the consumers of the lollipop, not the sellers. Since environmental regulations are designed to protect people who are exposed to toxic substances rather than the people who manufacture and sell them, it would be reasonable to conclude from this assessment that however small the risk of a toxic effect from the dye, it outweighs the benefits, since there are none. But regulatory action against Red Dye No. 40 could be challenged on the basis of economic and social impact — jobs and profits. 

Taking all this into account, how could the Cancer Society, or the rest of us, decide about the relative importance of doing something about cigarettes and red lollipops? How do you compare the importance of controlling one possible carcinogen with another? This is the central question of the growing debate on carcinogens. There is, as yet, no generally accepted logical answer to it; but consider for a moment an illogical answer. 

Suppose we compare the costs of a regulatory action that might be taken against cigarettes or red lollipops — banning them. Banning cigarettes would wipe out a $6 billion industry (in annual sales), whereas banning even all uses of Red Dye No. 40 would eliminate sales of only a few million dollars per year. Clearly, the social costs of banning red lollipops are much smaller than the costs of banning cigarettes. But this fact would, of course, be a palpably illogical basis for action, since cigarettes are more dangerous. The logical fault is obvious: In a risk/benefit assessment what should be compared is risk and benefit associated with the same substance, not risk (or benefit) of Substance A with the risk (or benefit) of Substance B. 

Yet this is just what is now being proposed by the chemical industry. We are told that “we do not live in a risk-free world and a balanced policy on carcinogens must take this into account.” The American Industrial Health Council (an industry lobby) gives us tables comparing cancer risk with other hazards. The tables tell us, for example, that people who frequently fly the airlines incur the risk of fatality with a probability of .0015% per year; that the comparable risk of cancer from an average medical diagnostic X-ray is .001% per year; that the risk of a fatality from playing football is .004% per year; from canoeing, .04%; and from motorcycle racing, 1.8% per year. The report points out that, “Society has chosen not to prohibit any of these activities, or even activities with much higher risks. There are a few activities which pose such a high risk that society has banned them completely (e.g., going over Niagara Falls in a barrel).” 

What is the Health Council trying to tell us? Since a 1.8% risk of death per year (motorcycle racing) is acceptable, but a risk of 100% (Niagara Falls in a barrel) is not, do they think we ought to ban a chemical when its risks lies between these two limits? 

This logic is reminiscent of a much acclaimed report to the Nuclear Regulatory Commission (and just recently repudiated by it) that decided, on the basis of very elaborate computations, that the risk of being killed by a nuclear reactor accident was about the same as the risk of being hit by a meteorite. Such a “risk assessment” approach destroys the logic of the risk/benefit concept just as surely as does the comparison of the regulatory costs of banning cigarettes and Red Dye No. 40. Its proponents would do well to ponder the moral of the Cancer Society’s red lollipops: What counts is the lollipop’s risk in relation to the lollipop’s benefits. When the dye’s contribution to that benefit is zero, the relative risk (however small) becomes, so to speak, infinitely large. This suggests that it may be unnecessary to make elaborate, relative estimates of the risks of many toxic chemicals because, like the color of the lollipop, their benefits are zero.


>> Back to Vol. 12, No. 3 <<