Factors that Determine the Efficiency Ranking of Second-Best Instruments for Environmental Regulation

Cost-effective policies allow minimizing the compliance costs associated to reaching a desired environmental quality target. However cost reductions associated to the use of these policies are not always significant. In this paper a conceptual model is developed to analyze explicitly the interaction among the factors that determine the compliance costs under two market based policies (the optimal ambient permit system, APS, and an emission permit system, EPS) and two CAC policies (equal percentage reduction, PER and a uniform concentration standard of emissions, STD). Considering a non-uniformly mixed pollutant the model incorporates explicitly the number of polluting sources; the size, in terms of emissions, of each process; the marginal costs of abatement for each process; the concentration of the emitted pollutant at the source; the transfer coefficient that relates emissions at each location with the impact on environmental quality at the receptor; and the desired environmental quality target. A first question addressed using the model. is how each of these factors affects compliance costs under each policy and as a result how the costs of sub-optimal policies compare with those of the optimal policy. The model shows that each factor affect the relative efficiency of each suboptimal policy quite differently. A second issue addressed is the efficiency ranking of second-best instruments under plausible values of each factor. It is shown that (1) APS is significantly less costly than the suboptimal policies in 45% of the cases; (2)EPS is very efficient in 75% of the cases, particularly when sources are clustered around the receptor; (3) a uniform standard performs well in many common situations; and PER is also efficient in some specific cases; (4) there is a high dispersion of results for cost quotients in some specific cases; and (5) relative compliance costs for PER and STD show extreme variations becoming very inefficient in some cases. Extreme values of the cost quotient for EPS are much lower.