The Los Angeles basin has the worst air quality in the United States. For two decades, expensive "command and control" technology-based standards were used to regulate air pollution in the basin. Beginning in 1994, the South Coast Air Quality Management District (SCAQMD), in cooperation with the U.S. Environmental Protection Agency (EPA), began a new regulatory program, the Regional Clean Air Incentives Market (RECLAIM), in an attempt to reduce the costs of achieving regional air quality objectives. The salient features of the RECLAIM program are the distribution of tradable air pollution credits and the establishment of markets in which the credits may be bought and sold.

Industry and policy makers in the region are concerned about the ultimate cost of pollution abatement under RECLAIM and the eventual price of RECLAIM pollution credits. Both the cost of abatement and the price of credits will depend upon the pollution abatement technologies available to firms. Initial estimates of the costs of pollution abatement under RECLAIM made by SCAQMD were based on an assumption of fixed-proportions pollution abatement technology. In this paper we present new estimates of air pollution abatement "production functions" for the region as a whole and for eight highly-polluting disaggregated industrial sectors located in the basin.

Our modeling approach differs from previous analyses in that we allow substitution possibilities among air pollution abatement inputs. We estimate a translog stochastic frontier pollution abatement production function for each sector using data on pollution emissions collected by the

California Air Resources Board and data on abatement costs developed by SCAQMD and the California Energy Commission. With substitution, the potential for pollution abatement extends beyond the conservative limits implied by the fixed-proportion assumptions of existing SCAQMD analyses.

Furthermore, with substitution possibilities the estimated marginal cost of pollution abatement may be significantly lower at higher levels of abatement. The estimated pollution abatement production functions may be used to investigate factor demands for pollution abatement inputs, substitution between abatement inputs, and factor expansion paths in abatement activity levels.

Our results indicate that substitution possibilities exist in air pollution abatement for industrial sectors in the Los Angeles Basin. As a result, we expect lower air pollution control costs than originally anticipated at the outset of the RECLAIM program. Hence, we also expect RECLAIM credit prices to rise less quickly, ceteris paribus, than originally anticipated. We compare realized RECLAIM credit prices with our estimates and with SCAQMD projections based on fixed-proportion assumptions.