The main object of the research is to analyze the Toxics Release Inventory (TRI) data in conjunction with other economic, pollution, and regulatory databases to address a set of basic economic issues relating to the generation, management, and environmental consequences of industrial toxic waste. Empirical analyses will be conducted by linking (1) information about State and local environmental regulations; (2) ambient air quality data and emissions estimates contained in the Aerometric Information Retrieval System (AIRS); (3) firm-level financial data; (4) the TRI; (5) the Longitudinal Research Database (LRD) (a Census Bureau data set containing cost and production information for a large panel of manufacturing establishments); and (6) the Survey of Pollution Abatement Costs and Expenditures (PACE) (another Census Bureau data
set). The studies proposed fall into two groups. The first involves analyses aimed at characterizing, both theoretically and empirically, the generation and management of industrial
toxic waste as part of the production process. Among the questions to be addressed are the following: (1) How much variation is there in the pollution intensity of manufacturing establishments? (2) How much of that variation can be accounted for by differences between industries and how much within industries? (3) How does the variation depend on the amounts of capital, materials, and energy used in production, the age of the capital, and the amount of pollution abatement capital in use? (4) To what extend are there economies of scale in pollution reduction: and (5) How strong are the statistical associations among production inputs, industrial toxic air releases, and ambient air quality? The second group of analyses will treat cross-regional variation in pollution regulations as an "experiment" in environmental policy in order to explore establishment-level responses to environmental regulation. Environmental regulations may be expected to increase the costs of disposing of industrial byproducts, which in turn may induce manufacturing establishments to respond in a number of ways that can be studied empirically using the data sets listed above. These responses include (1) relying on less pollution-intensive production processes or inputs; (2) increasing recycling of byproducts; (3) directing capital away from heavily regulated jurisdiction; (4) reducing production of desired outputs, perhaps by changing the scale of manufacturing operations; and (5) shifting of releases to media that are relatively lightly regulated.