The devolved nature of environmental regulation generates rich regulatory variation across regions, industries and time. We exploit this variation, using direct measures of regulation and plant data, to estimate employment effects of sharply increased air quality regulation in Los Angeles. Regulations were accompanied by large reductions in NOx emissions and induced large abatement investments for refineries. Nevertheless, we find no evidence that local air quality regulation substantially reduced employment, even when allowing for induced plant exit and dissuaded plant entry. Regulations affected employment only slightly - partly because regulated plants are in capital and not labor-intensive industries. These findings are robust to the choice of comparison regions. (C) 2001 Elsevier Science B.V. All rights reserved.