Over the last decade, the precautionary principle –“better safe than sorry”– often has been invoked to justify government regulatory action. According to advocates of this principle, we must protect ourselves from potential environmental threats –such as greenhouse gases, nuclear power, or arsenic in drinking water– even if we are not sure exactly what will be gained from doing so. Recent research by law professor Cass Sunstein (2003, 2005) argues that the precautionary principle actually offers no policy guidance and instead can be expected to produce either policy paralysis or perverse policy outcomes.
One widely accepted version of the precautionary principle asserts that “when an activity raises threats of harm to human health or the environment, precautionary measures should be taken even if some cause and effect relationships are not fully established scientifically” (Global Development Research Center 2005). Thus, because nuclear power plants create potential health and safety risks, including some chance of catastrophe, there should be no construction of new plants. Similarly, fossil fuels raise the risk of global warming, so their use should be reduced.
But the conjunction of these two threats highlights a fundamental flaw in the precautionary principle: It urges us to avoid using nuclear energy to guard against nuclear holocaust even as it counsels us to use nuclear energy to guard against climate catastrophe. As Sunstein demonstrates, this sort of internal contradiction in the precautionary principle arises repeatedly when one attempts to apply it to policy making.
Now, it might be argued that the principle really counsels use of solar power or simply less use of energy altogether. But solar power requires massive battery installations that might cause environmental threats as yet unforeseen. Solar energy also uses far more non-environmental labor and capital resources than nuclear or fossil fuel. This leaves us less able to deal with other health and environmental hazards, ranging from cancer and heart disease to arsenic in our drinking water. Together, these unknowns raise the possibility that solar power might have serious adverse consequences. Thus, following the dictates of the precautionary principle, to be safe we must not take that risk.
Even the seemingly innocuous act of using less energy generates its own threats. For example, energy-efficient buildings require more insulation. The manufacture and use of insulation have uncertain, potentially dangerous environmental effects –the most notorious example is asbestos, used as insulation but later discovered to cause asbestosis and mesothelioma. Moreover, some energyefficient buildings have higher levels of unhealthy indoor air pollution than less efficient buildings, leaving us again with the question unanswered by the precautionary principle: Which risk is the right risk?
Arsenic-laced drinking water offers another illustration of the hazards of the precautionary principle. The Bush administration has proposed lowering permissible levels of arsenic in public water systems from 50 parts per billion (ppb) to 10 ppb. This standard, scheduled to take effect in January 2006, may save as few as six lives per year at an annual cost of over $200 million. Worse yet, by raising the cost of public water systems, the regulation will induce some people to rely instead on private wells; but the water from such wells often has far greater arsenic concentrations than the standard that is being replaced, a factor not accounted for by the EPA. Thus, the precautionary attitude that brought us the arsenic regulation is actually likely to kill more people than it spares.
What are we to do? Sunstein argues for a return to what he views as the “gold standard” –cost-benefit analysis. Both the advantages and the disadvantages of actions (or inaction) are weighed against one another, with the nod going to the alternative that appears to yield the greatest net benefit.
Yet many environmentalists raise two precautionary objections to cost-benefit analysis. First, they say, it fails to account for irreversible outcomes, such as the extinction of a species. Second, they argue, it fails to account for the uncertainties of making decisions when there are many unknowns. As Sunstein notes, however, such objections fail.
Consider the act of driving home for the holidays. Although routinely practiced, this behavior is fraught with both uncertainty and potential irreversibility. Some of those who depart on their holiday journey will die in the attempt, victims of fatal automobile accidents. None of us can be sure in advance how many people will die in this manner, much less their identities. And whoever they are, the outcome will be unquestionably irreversible. Nevertheless, the highways will be crowded, as each of us makes our own assessments, balancing the likely costs and benefits, recognizing that some things can never be known in advance.
Sunstein argues that just as we are not paralyzed by uncertainty and irreversibility, policy makers should not be either. Irreversibility simply means that some consequences are permanent; it doesn’t mean that they are infinitely important. Similarly, when uncertainty is present, if we are averse to risk we should build in a margin of safety; it doesn’t mean that margin should be infinitely large.
It is important to account for the duration of costs and benefits, and for all of the effects of our decisions, good and bad. But these principles are easily accommodated by cost-benefit analysis, and neither calls for the paralysis nor inconsistency implied by the precautionary principle. And so my cautionary tale is simple: Contrary to what you might have heard elsewhere, sometimes safe is sorry.
References
Global Development Research Center. 2005. Wingspread Statement on the Precautionary Principle. Online: www.gdrc.org/u-gov/precaution-3.html (cited October 24, 2005).
Sunstein, Cass. 2003. Beyond the Precautionary Principle. University of Pennsylvania Law Review 151(January): 1003-58.
—. 2005. Laws of Fear: Beyond the Precautionary Principle. New York: Cambridge University Press.
