Chapter 2. The Rational for Risk Regulation
Congress has almost always rejected a cost-benefit test as a regulatory trigger or regulatory standard in risk reduction legislation. Those who are critical of this pattern depend on economic theory to justify their preference for greater reliance on a cost-benefit test in triggering regulation or in setting the level of regulation. Alternatively, they support the use of cost-benefit analysis to analyze regulation before it is promulgated, even though regulators determine the level of regulation based on some other form of balancing or other standard. The equation of costs and benefits can be defended according to standard economic principles. By comparison, the rationale for risk regulation, as constituted by Congress, is not as obvious.
Pragmatic principles furnish this missing rationale. The current structure is pragmatic because, in comparison to a cost-benefit approach, it better reconciles important social values implicated by risk reduction. Pragmatism requires the accommodation of important, widely held social values in risk regulation. The current structure is also more pragmatic than a cost-benefit standard because it better reconciles the bounded rationality of regulators with the goals of risk regulation. Pragmatism also requires that the design of risk regulation take bounded rationality into account.
We begin by comparing the economic and pragmatic justifications for risk regulation. Pragmatic analysis reveals that a purely economic approach to risk regulation threatens two widely held social goals, which pragmatism attempts to accommodate. Risk legislation respects the integrity of human life and the environment by refusing to view such protection solely in terms of the estimated dollar value of the protection. Current laws also reject the idea that the amount of protection to which a person is entitled should be a function of the person’s wealth. At the same time, current laws also recognize that cost should be an important consideration in regulatory policy. Accommodation of these multiple goals is no easy matter, but the fact that current legislation attempts to do so is what makes it pragmatic.
The efficient operation of markets requires that those whose actions injure humans or the environment take such costs into account when they act. For example, if a chemical manufacturer does not pay for the environmental damage to nearby land that results from a manufacturing process, the chemical will sell at a price that does not fully reflect its production costs. The chemical will be priced too low, because one of the costs of production, environmental damage, is a cost that is external to the producer. Economic analysts describe such injuries as “externalities” – because the costs are not part of the “internal” costs that the manufacturer must take into account in making its production decisions – or as “spillover costs” – because the costs spill over to the persons who own the land damaged by the manufacturing process. If the manufacturer paid for the environmental damage, the price of the chemical would be higher and, unless demand was completely inelastic, less of it would be sold. The additional production of the chemical resulting from a failure to internalize environmental costs is inefficient because it consumes resources, such as raw materials and labor, that would be put to other uses better matched to consumer preferences if the chemical were properly priced.
Until Ronald Coase published his famous article “The Problem of Social Cost,” economic analysts assumed that government intervention would be necessary to require producers, such as the chemical manufacturer, to pay for the damage caused by their activities. They favored taxing firms based on the monetary value of the damage done to the environment. Professor Coase, however, posited that, even absent the imposition of a tax, a free market exchange could result in the same reduction in pollution as the tax plan. Coase pointed out that the neighbors of the factory had an incentive to pay it to reduce the environmental damage to their land. The amount that the neighbors are willing to pay depends on the extent to which the pollution damages their land, which (in theory) is the same amount as the tax the government would assess if taxes were used to address this problem. The manufacturer is willing to reduce the level of pollution up to the point where the costs of doing so are greater than the payments it will receive from the neighbors.
Professor Coase acknowledged that this result required that no significant economic impediments prevent the neighbors from negotiating with the chemical manufacturer. When such “transaction costs” are present, they can prevent the negotiations from taking place or alter their outcome. This would occur, for example, if the manufacturing process damaged land owned by thousands of people because it would be prohibitively expensive, if not practically impossible, for all of the individuals experiencing harm to agree on the appropriate approach to negotiating with the manufacturer. When transaction costs have this impact, government intervention may be necessary to obtain an “efficient” outcome.
When the government intervenes, its goal is to replicate the result that would have occurred if the bargaining between the company and its neighbors had occurred in the absence of transaction costs. The government can do this by using a cost-benefit standard to establish the level of regulation. The amount of money that the neighbors are willing to pay to the manufacturer to reduce the environmental damage depends on the value of undamaged land to them. This is the “benefit” in a cost-benefit standard. The manufacturer’s willingness to reduce the environmental damage is a function of the cost to it of reducing the damage. This is the “cost” in a cost-benefit standard. The manufacturer will reduce the environmental damage up to the point where the cost of any further reduction exceeds the amount of money that the neighbors are willing to pay. The government can replicate this result by requiring a reduction in environmental damage up to the point where the costs and benefits of such an action are equated.
Professor Coase supported this utilitarian approach by his insistence that there is no objective economic basis for blaming anyone for pollution or similar externalities. Although economists before Coase had identified the polluter as the source of the externality, he argued that the problem was “reciprocal” because the neighbors could equally be “blamed” for causing the externality. In this alternative view, the neighbors’ demand for less pollution is an externality that the neighborhood imposes on the factory. Thus, the “problem we face in dealing with actions which have harmful effects,” Coase argued, “is not simply one of restraining those responsible for them.” Rather, “what has to be decided is whether the gain from preventing the harm is greater than the loss which will be suffered elsewhere as a result of stopping the action which produces the harm.”
Because economic theory eschews assessing blame for pollution and other externalities, it is the existence of transaction costs that justifies government action. The government intervenes because the market fails to achieve an “efficient” level of production. The government does not intervene because injury to the environment, or humans, is somehow “wrong.” Instead, injuries to humans and the environment are simply production costs, and the goal is to achieve the “optimal” or most efficient level of human and environmental injury.
Risk regulation balances the interests of risk creators and risk bearers, but not in the manner indicated by economic theory. Congress has generally permitted regulation on the basis of a risk trigger that permits agency action on the basis of risk predictions, rather than requiring conclusive proof that harm will necessarily occur. Thus current risk laws respect the innate value of human life and of the environment by adopting evidentiary standards that make it easier for agencies to engage in protective actions. Moreover, as the last chapter explained, most risk regulation occurs under a constrained or open-ended balancing standard that allows regulation in instances in which a cost-benefit standard may not, or to a degree beyond that produced by a cost-benefit standard. Such statutes thus reflect an affirmative commitment to protecting humans and the environment. This commitment respects the innate value of human life and of the environment because it makes such protection a priority.
Although protection of individuals and the environment is a priority, costs are not ignored. Risk reduction laws require agencies to engage in some form of balancing of the costs of regulation (that is, the interests of risk creators) and the benefits of risk reduction (the interests of risk bearers). Moreover, as the last chapter indicates, even the phaseout of a toxic substance takes costs into account by creating exceptions, or by adopting a long period for the phaseout.
In this manner, current laws usually commit the country to “do the best that it can” to reduce human and environmental injury. Under constrained balancing, for example, regulation is to occur, but only up to the point at which costs are disruptive or extraordinary. Congress normally identifies that point by requiring industry to limit its spillover effects to the level achievable through the use of some aspirational technology, such as “best technology economically achievable” or the technology capable of producing the “lowest achievable emission rate.” Typically, Congress also requires regulators to consider the economic impact of achieving this level of performance along with other factors that may legitimately affect the regulatory entity’s decision. “Congress has, in other words, announced to the world: ‘If we cannot have a perfectly clean workplace and environment, then we shall do the best that we can”. Sometimes this commitment extends to forcing regulated entities to develop more effective technologies than those currently in use as a means of achieving desirable levels of risk reduction.
OSHA’s mandate is a good illustration of the ethical difference between the economic approach and Congress’s commitment to “do the best that we can.” Congress has said that OSHA must promulgate regulations that protect workers “to the extent feasible” from the harmful effects of toxic substances. Under OSHA’s mandate, a standard is technologically infeasible when it requires protective devices that are not generally available under existing technology. A standard is economically infeasible when regulation threatens the long-term profitability and competitiveness of regulated industries. A standard is not economically infeasible simply because it is financially burdensome or adversely affects profit margins.
Critics of risk regulation have criticized OSHA’s mandate because it might be less expensive to compensate employees for occupational injuries or diseases in some cases than to spend money to prevent them. This argument, however, ignores the crucial ethical distinction between preventing fatalities or injuries and compensating the victim or his or her family for a death or disability. Protecting workers under a constrained balancing test may be somewhat more costly to society than after-the-fact compensation, because OSHA may act on incomplete information that overstates the degree of risk. This tilt in favor of protecting workers, however, respects the fundamental value of their lives. It aspires to something more than obtaining the “optimal” level of worker illnesses.
The same regard for human life is found in open-ended balancing. First, although an agency is required to balance costs and benefits, it is not required to equate them. Thus the agency is not committed to seeking an optimal allocation of injury to humans or the environment. Instead, the agency is free to protect individuals or the environment to an extent that exceeds the level of protection that a cost-benefit standard would provide if the agency determines that, in the particular circumstances, factors other than cost justify the higher protective level. Although this objective is not the same as a commitment to seek the greatest available protection, it still authorizes agencies to protect individuals or the environment in situations in which a cost-benefit standard may block, or at least weaken, such protection.
Second, an agency that operates under open-ended balancing is authorized to regulate even if it lacks sufficient information to quantify all benefits and ensure that they exceed costs. This tilts regulation in favor of protecting humans and the environment for a reason discussed later in this chapter. Because a cost-benefit standard requires more quantification of regulatory benefits than an open-ended balancing test, agencies will have greater difficulty defending their regulatory decisions when they are challenged in court. This difficulty hampers regulation and makes it less protective of humans or the environment. Finally, as discussed next, both constrained and open-ended balancing avoid commodification or the treatment of humans and the environment as commodities – “things” that people are willing to buy and sell.
Prevailing methods of risk regulation respect the innate value of life and the environment in a second important way. They avoid what Margaret Jane Radin has described as “universal commodification.” In universal commodification, Professor Radin explains, “all things desired or valued – from personal attributes to good government – are commodities,” and all social interaction is conceived of as free market exchanges. “In market rhetoric, under the discourse of commodification, one conceives of human attributes (properties of persons) as fungible with owned objects (the property of persons).” Cost-benefit analysis employs universal commodification because it “evaluates all human actions and outcomes in terms of actual or hypothetical gains from trade, measured in money.” In other words, a cost-benefit standard reduces all values to sums of money. Professor Radin identifies why, except for the deepest enthusiast of market rhetoric, this type of commodification “seems intuitively out of place” regarding most social policy:
“One basis for this intuition is that market rhetoric conceives of bodily integrity as a fungible object. A fungible object can pass in and out of the person’s possession without effect on the person so long as its market equivalent is given in exchange; trading commodified objects is just like trading money. To speak of personal attributes as fungible objects – alienable goods – seems intuitively wrong to many people, because they do not conceive of bodily integrity as commodified. ... Bodily integrity is an attribute and not an object. The effect of “detaching” it from the person is non-monetizable. We feel discomfort or even insult, and we fear degradation or even loss of the value involved, when bodily integrity is conceived of as a fungible object.” For Professor Radin, “the way we conceive of things matters to who we are.”
Pragmatism takes this concern seriously. Because it focuses on the actual effects of social policies, pragmatism recognizes that “law, even if it begins as an instrument for the attainment of basic ends . . . can generate its own intrinsic values.” As Thomas Grey reminds us, “Dewey well understood this; that a good system of law was partly constitutive of, not merely instrumental to, a good society.”
Although the legislation does not ignore the cost of reducing risks, it conceives of the process of balancing risks and the costs of reducing them in other than market terms, and it thereby fosters the nonmarket significance of human life or the environment.
The infrequent appearance of risk-based legislation, which proceeds without regard to the cost of compliance, is consistent with a lack of sufficient information about the scope of the risks being targeted or about the effect of various levels of control on reduction of those risks. In a few instances in which Congress has phased out use of a risk-creating substance or activity, the phaseout in effect amounts to a determination that the substance or activity posed unacceptable levels of risk. Although bounded rationality typically prevents drawing confident conclusions about the economic impact of much risk-based regulation, in these instances Congress has taken cost into account in blunt fashion in its determination of the phaseout period. The provisions of the CAA that eliminate over a forty-year period substances that pose threats to the integrity of the stratospheric ozone layer provide an example.
Similarly, analysis based on the new institutional economics suggests that a cost-benefit standard, which requires an agency to have sufficient information to find the optimal level of regulation, is an appropriate choice only under conditions in which an agency is unlikely to be subject to bounded rationality concerning either risk or cost assessments. Under conditions of bounded rationality, another choice is more appropriate. Given the difficulties of ascertaining and quantifying both the costs and benefits of regulation, which we describe in the next two chapters, the decision by Congress to avoid a cost-benefit standard is appropriate.
Thus, institutional economics justifies the popularity of the remaining two options, constrained and open-ended balancing standards. Each reduces the cost of regulatory decision-making by assigning regulators a burden of proof that recognizes and seeks to mitigate the adverse consequences of bounded rationality and opportunistic behavior. The great bulk of risk reduction legislation is based on either a constrained or open-ended balancing model.