In his article, "Framing Environmental Policy Instrument Choice,"(1) Professor Richards proposes a taxonomy of environmental policy instruments as a guide in drawing comparisons between the instruments. His logical apparatus provides a useful framework for organizing environmental policies and offers a measure of insight about the relationships among them. Professor Richards is less successful, however, in evaluating the policy instruments, largely because of his attempt to apply transaction cost economics to environmental policy.
In this Comment, I discuss four points in turn. First, Professor Richards frames his discussion in terms of a stylized cost-minimization problem. Although a useful first step, his presentation of the problem as a static one limits his evaluation of the policy instruments with respect to the dynamic effects on technological change. Framing the problem in explicitly dynamic terms suggests several ways in which the cost components and even the constraints of the problem change over time.
Second, Professor Richards draws a distinction between policy instruments that require payment from polluters to the government and those that do not. He argues that whether or not private firms pay for their emissions determines who bears the burden of environmental damages, and he returns to this distinction in evaluating the instruments on the grounds of "equity and environmental justice." This discussion, however, confuses the incentive properties of emissions payments with equity considerations. Such payments do not compensate the victims of pollution; indeed, they are not necessarily even equal to the damages from pollution.
Third, Professor Richards overlooks an important class of policy instruments: ambient taxes and emissions permits. Indeed, the distinction Professor Richards makes between controlling inputs and controlling outputs can be extended to the choice of the "locus of regulation"--that is, at what stage to regulate, along the chain from inputs to emissions to eventual environmental damages.
Finally, Professor Richards devotes considerable attention to what he calls the "governance costs" associated with implementing environmental policy instruments. Here he draws heavily on transaction cost economics, particularly the work of Oliver Williamson. Arguing that the conditions of uncertainty and asset specificity apply equally well to the environmental policy arena as to contractual relationships between private parties, Professor Richards concludes that, in some cases, market-based policy instruments may invite government opportunism, raising their true social costs and leading to a preference for command-and-control regulation.
However, it is by no means clear that the environmental policy arena is analogous to the contractual setting envisioned by the transaction cost theorists. The examples provided by Professor Richards are better illustrations of a cost-revelation problem than the "holdup" problem that is the focus of the transaction cost literature. I suggest a hypothetical hold-up problem in environmental policy and argue that its relevance rests on special assumptions about the regulator's treatment of firms, which do not generally apply to market-based instruments. Moreover, even to the extent that "governance costs" arise in environmental regulation, hierarchical control is unlikely to represent a solution.
II. THE COST-MINIMIZATION PROBLEM
Economic theory is much more comfortable describing efficiency than equity: the former criterion mobilizes the powerful apparatus of cost-benefit analysis, while the latter involves the messy complications of interpersonal comparisons. In evaluating environmental policy instruments, therefore, economists have tended to give greatest weight to efficiency concerns--or, in the absence of information on benefits, to cost-effectiveness concerns. Professor Richards follows suit. …