Applying incentive-based strategies to toxic substance regulation can be complicated. Potential risks to health and the environment can occur at many stages in the life cycle of a toxic substance, and the risks vary among different products and uses of products containing toxic substances. Thus researchers at Resources for the Future recommend that regulatory intervention be focused on specific stages in the life cycle of toxic substances, but warn that intervention must be broad enough to mitigate incentives to adopt production processes and products that could pose greater risks than the processes and products they replace. Despite this and other potential pitfalls, they find that incentive-based strategies such as product labeling and deposit-refund schemes may be desirable for regulating certain stages of the life cycle of some chemicals.
More than 60,000 chemicals enter into the multitude of products and services that contribute to today's lifestyle. Taken together, these chemicals comprise a huge industry. In the United States alone, sales of chemicals earn more than $200 billion a year. The sheer variety, ubiquity, and economic importance of chemicals make efforts to guard against their possible undesirable health or environmental effects a challenge. Traditionally, regulation to create safeguards has taken some form of command and control—most frequently product bans or other mandated restrictions such as product reformulation. But are there incentive-based approaches to environmental regulation that might serve as desirable alternatives to command and control?
A lengthy and expanding economics literature has argued for the general superiority of incentive-based strategies over command-and-control (CAC) regulation, primarily because of the cost savings expected from these strategies. This literature has been further expanded by a recent study by researchers at Resources for the Future (RFF) that considers the advantages and disadvantages of several incentive-based approaches for regulating toxic substances. One approach is to tax toxic emissions or to issue tradable permits for them. Both taxes and tradable permits are aimed at governing chemical production or use. Another approach would employ a system of deposits and refunds aimed at controlling disposal of chemicals, and a deposit-refund scheme in the form of a performance bond on new chemicals aimed at mitigating the toxic effects of chemicals. Yet another approach would use product labeling to provide information on the safe use of chemicals. The RFF study considered these approaches in case studies of four toxic substances that have been the focus of current regulatory concern: chlorinated solvents, which are principally used in dry cleaning, metal degreasing, aerosols, and paint removers; formaldehyde, which is used mainly in pressed woods and plastics; cadmium, which is used in batteries and paints; and brominated flame retardants (BFRs), which are used in plastics and textiles to reduce their flammability.
The RFF study assumed that the goal of regulatory intervention is to cost-effectively capture (internalize) the health and environmental impacts of chemical production and use. Without attempting to draw conclusions about the benefits arising from such intervention, the RFF study presumed that policymakers would regulate the above substances; the researchers' task was to explore alternatives to CAC regulation. The study did not estimate the magnitude of the potential cost savings of incentive-based approaches as compared with CAC regulation (an important topic for future research), but it did suggest some of the key factors that will affect the size of the savings.
Characteristics of toxic substances
The existing literature on incentive-based strategies for controlling pollution generally assumes a fairly homogeneous pollutant associated with one stage of production at an identifiable source—the canonical example might be sulfur dioxide emissions from electric utility plants. Only some aspects of this literature address specific issues, discussed below, that arise in considering regulation of toxic substances.
The RFF study further explored these issues, taking into account three characteristics of toxic substances that complicate a straightforward application of incentive-based strategies for environmental regulation. The first characteristic is the multiple-stage life cycle of chemicals. The potential for risks to health and the environment may occur at many stages in this life cycle—at the minemouth or during production of the feedstock, during production of intermediate products that use chemicals as an input, during use by industry or households, and upon disposal. Thus regulatory intervention to safeguard against risk may be necessary at more than one stage of a chemical's life cycle and may have to take different forms. For example, it might be beneficial to tax intermediate production of a chemical to mitigate air or water pollution and to label the final product in which the chemical is an input in order to inform end users about the potential hazards and proper handling of the product.
The second characteristic of toxic substances considered by the RFF study is the marked variation in the distribution of risk of harmful exposure across heterogeneous products and uses. Not all products or uses of a chemical may pose potential risks; nor is the nature of the risk always the same. A product or its use may be harmful to people or to the environment; it may be harmful to society as a whole or merely to the end user; or it may harm one or several environmental media (for example, it may harm air but not water or soil). Thus regulatory intervention may have to be highly product- or use-specific to safeguard against risk without unduly restricting relatively harmless applications of toxic substances.
Because a toxic substance may pose risks to human health and the environment at many stages in its life cycle, regulatory intervention may be necessary at more than one stage and may have to take different forms.
The third characteristic considered by the RFF study is the generally wide scope of substitute products or production processes. Regulating a substance (or one of its products or uses) is likely to induce substitution of another product or production process. Generally speaking, small modifications in the makeup of many chemicals can lead to a substitute product. Although substitution possibilities are an intended consequence of regulatory intervention, the existence of potentially more harmful substitutes for a particular toxic substance or class of substances implies that toxic substance regulation, if too narrow in scope, could unintentionally result in increased levels of environmental and health risk.
Regulation of toxic substances
The above characteristics lead to several observations regarding the regulation of toxic substances. The first is that focusing intervention on specific life-cycle stages is likely to be desirable. A tax targeted at intermediate stages of production or at certain end uses during which risk of exposure is concentrated would better ensure that risks are mitigated than would a tax on all production of a substance. However, a tax targeted at specific production stages or end uses may entail significant administrative and enforcement costs. In contrast, a blunt instrument—such as a tax on all production of a chemical—may be easier, thus less costly, to administer. This is because typically there are fewer producers of chemicals upstream in the production process—at the minemouth or feedstock level—than there are downstream at the level of intermediate production of products containing chemicals, and because there are fewer producers than end users of products. However, such blunt intervention would reduce use of chemicals in those applications for which there are more substitutes rather than in those applications for which risks of human or environmental exposure are greatest.
A second observation regarding regulation of toxic substances is that the most desirable intervention strategies are self-enforcing. The property of self-enforcement is clearly advisable for all types of regulation (whether of the command-and-control variety or incentive-based) and in all circumstances (whether a single source, homogeneous pollutant or a multisource, multimedia pollutant is involved). However, this property is probably of particular importance in the case of toxic substances, given their ubiquity and heterogeneity. Opportunities for eluding the purchase of a permit for or evading a tax on a toxic substance may be easy to exploit. For example, the large numbers of intermediate producers may make it easy to resell a substance ostensibly intended for benign uses—a substance for which no tax would be levied or permit would be purchased—to a producer who uses it in a production process or a final product that poses great risks. Consider, for instance, the possible response to a tax on formaldehyde used in the production of resins that are found in wood furniture and the absence of a tax on formaldehyde used in the production of household cleaning products. This tax, which would reflect the fact that formaldehyde in wood furniture poses more health risks than formaldehyde in household cleaning products, could be evaded if cleaning product manufacturers resold the untaxed formaldehyde to furniture manufacturers as a substitute for the taxed formaldehyde. Opportunities also arise to undermine deposit-refund schemes. For example, some chemicals may be readily and relatively undetectably diluted in order to increase refunds.
For these reasons, the RFF study suggested the use of intervention strategies that might be self-enforcing. These strategies include deposit-refund schemes that are modified to reduce opportunities for diluting chemicals, as well as taxes and permits that allow cost-effective monitoring. Strategies that increase the probability that violators will be monitored in the future might also prove useful.
A third observation regarding the regulation of toxic substances is that such regulation must be broad enough in scope to mitigate incentives to adopt potentially more harmful substitutes. In the state of California, the regulation of select chlorinated solvents without regard to the harmful effects of possible substitutes has led to higher emissions from the substitute solvents and increased risks of exposure to the toxic effects of these emissions. Regulation of brominated flame retardants presents similar possibilities for risk-increasing substitution. The posting of a performance bond or other insurance might be considered to guard against products that pose greater risks than the products they are substitutes for—or to allow compensation to those harmed by these substitutes. Such risk sharing by producers may reduce the large amounts of information that regulators presently must obtain in overseeing new and existing chemicals.
Because end uses of oil-based paints and other products that contain formaldehyde are associated with few third-party health effects, the most appropriate way to regulate them may be product labeling
Regulatory intervention at various stages of the life cycle of a toxic substance must be approached with caution. In its examination of incentive-based strategies for regulating formaldehyde, the RFF study illustrated the potential for such intervention to increase risks of harmful exposure to the substance at another stage of the life cycle. It suggested that product labels and product standards might reduce consumers' potential exposure to formaldehyde vapors from various household products. However, the study considered the possibility that producers of these prod-ucts would reduce such exposure by retaining the products in warehouses and delivering them to retailers only when vapors had dissipated to the point that the products would meet emissions standards for acceptable levels of formaldehyde in households. Thus an unintended effect of intervention at the end-use level might be that warehouse workers would be exposed to levels of vapors higher than those prior to intervention. Similarly, incentives to encourage recycling of chlorinated solvents and cadmium could have unintended consequences by reducing society's exposure to these chemicals during disposal, but increasing emissions of the chemicals during recycling.
Another consideration that arises with regard to regulatory intervention at various stages of a chemical's life cycle is the need to coordinate such intervention with existing regulations that affect the production, distribution, use, and disposal of toxic substances—for example, limits on emissions of toxic substances into the air and water, rules for transport of these substances, and standards for occupational health and safety. Intervention to mitigate the public's risk of harmful exposure to brominated flame retardants at the disposal stage of the BFR life cycle and existing fire safety regulations that induce the use of BFRs illustrate this need. The former would have to be coordinated with the latter in such a way that the health risks arising from the disposal of BFRs are balanced with the benefits of using products in which BFRs are an input. The RFF study noted but did not explicitly consider the combined effects of existing regulation and regulations outlined in its case studies. Instead, it assumed that the effects of each were separable.
The preceding discussion suggests that it is inappropriate to recommend a single regulatory approach for all toxic substances. Although two or more such substances may have similar characteristics, it is generally necessary to conduct a case study of each substance to determine the stage or stages of the life cycle at which intervention might be desirable. This is so even when the risk of harmful exposure occurs at the same stage or stages of the life cycle of two or more substances with similar characteristics.
Regulation aimed at a specific stage in the life cycle of a toxic substance must be coordinated with existing regulation so as to balance health risks with the benefits of using products containing the substance.
As already noted, among the incentive-based approaches to regulating toxic substances analyzed in the RFF study are product labeling and deposit-refund schemes. The former may be desirable for regulating end uses of formaldehyde, while the latter may be desirable for regulating the recycling and disposal of chlorinated solvents.
Product labeling for formaldehyde
In many applications of the substances considered by the RFF study, third-party effects are negligible—that is, risk of harmful exposure is limited to the end user rather than extending to society as a whole. This is generally the case with products containing formaldehyde.
Exposure to formaldehyde arises principally through the emission of vapors from products. Because it is product users who are primarily exposed, few third-party health effects are associated with consumption of the products. The near absence of such effects suggests that the most appropriate form of regulatory intervention may be one that increases the supply of information to consumers about both the possible health effects of formaldehyde and the actions that consumers can take to mitigate these effects. One way to increase this supply is to label products that contain formaldehyde with facts about health effects and mitigating actions. Although the benefits to consumers from access to information about the possible health effects of formaldehyde may differ markedly from one individual to the next, anecdotal evidence suggests that, in general, the gains are likely to be large. If so, labeling may be preferred to bans or other restrictions on products containing formaldehyde.
Consumers of products that contain formaldehyde risk the greatest exposure to formaldehyde; thus product labeling appears to be the most appropriate way to regulate this substance.
In some instances, industry has already voluntarily undertaken labeling of products containing formaldehyde. The Consumer Product Safety Commission has been encouraging industry to undertake additional voluntary labeling of particular products in which formaldehyde is an input. The U.S. Department of Housing and Urban Development requires labeling of building materials that emit formaldehyde if the materials are used in the construction of homes.
Whether information about the possible health effects of exposure to products that contain formaldehyde is undersupplied and, if so, whether government should encourage or require additional labeling, depends on a host of factors. In any case, when consumers are informed about actions to mitigate harmful exposure to a product through product labeling, they can choose to continue use of the product or find a substitute for it. Thus labeling is one way to preserve consumer choice and at the same time safeguard against health risks.
Deposit-refunds for chlorinated solvents
Hazardous exposure to chlorinated solvents can arise through improper disposal of these solvents, which may lead to groundwater contamination. Existing regulations under the Resource Conservation and Recovery Act are designed to eliminate this exposure by restricting land disposal of spent solvents and sludges that remain after chlorinated solvents are recycled. These regulations require solvent waste to be incinerated at a licensed hazardous waste incinerator or treated and disposed of at a hazardous waste landfill. They also call for a system of manifests for tracking transport of this waste. The regulations have raised the cost of proper solvent disposal, perhaps giving recyclers an incentive to illegally dump solvent waste.
To remove the incentive for illegal disposal and thus reduce the social costs of such disposal, a deposit-refund system might be imposed on all parties that accept spent solvent for recycling and disposal. Such a system would encourage recyclers of chlorinated solvents to use the most socially efficient method of disposal by raising the costs of illegal disposal and rewarding appropriate disposal practices.
A deposit-refund system for recycling and disposal of chlorinated solvents would remove incentives to illegally dispose of these solvents and would encourage a reduction in solvent emissions during recycling.
Under the deposit-refund system, the solvent waste handler would be required to pay a deposit to the government for every pound of spent solvent accepted for recycling and disposal. This deposit would be exchanged for proof of recycling or proof of legal disposal at a licensed hazardous waste facility. The deposit-refund would be set equal to the difference between the social marginal cost of illegal disposal and the private marginal cost of illegal disposal, and the deposit would be refunded on all solvents that are either recycled or properly disposed of at a licensed hazardous waste disposal facility. Since recyclers would not receive a refund on any solvents emitted during recycling, the deposit-refund system would give them an incentive to reduce emissions into the air. Because some emissions will always occur during recycling, the deposit-refund system, which is self-financing, should yield some net revenue to the government. This revenue could be used to cover some of the administrative costs of the deposit-refund program.
Future research
The RFF study suggested several directions for future research on incentive-based approaches to regulating toxic substances. One is analysis of the costs and benefits of these approaches and of alternative command-and-control approaches. This would indicate the size of the savings to society accruing from incentive-based regulatory intervention and identify those likely to benefit or suffer (through loss of employment, for example) from such intervention. Another suggested area of research is exploration of the interactive effects of multiple layers of regulation of a substance. This would include examination of the effects of intervention at more than one stage of a substance's life cycle, as well as the consequences of such intervention for different environmental media (air, water, and soil) and for activities related to the production, use, and disposal of toxic substances—hazardous materials transportation, for example. Still another direction for research would be consideration of how a particular regulatory approach would affect the future technology choices made by firms, particularly with regard to the relative health and environmental effects of substitute production processes and products.
Molly K. Macauley is a fellow in the Energy and Natural Resources Division at RFF. Karen L. Palmer is a fellow in the Quality of the Environment Division at RFF. This article is based on research conducted by Macauley, Palmer, and former RFF fellow Michael D. Bowes. RFF will publish the research under the title Using Economic Incentives to Regulate Toxic Substances in late 1992.
A version of this article appeared in print in the June 1992 issue of Resources magazine.
A version of this article appeared in print in the June 1992 issue of Resources magazine.
