this report’s methods, findings, and recommendations are discussed in detail in the white paper and will not be repeated here.
As the Task 3 statement of work requires, this report identifies statistically rigorous, yet simplified and cost-effective, metrics for the database analysis of the deterrence effects of environmental monitoring and enforcement. Those metrics are then benchmarked against data analyzed in the pre-existing literature to examine if the estimated deterrence effects from simplified models approximately equal those reported in published studies. These two objectives are admittedly narrow, and several concepts important to the broader compliance and deterrence project are not fully explored here. First, this report is not intended, on its own, to establish the deterrence effects of monitoring and enforcement actions. The extensive peer-reviewed literature summarized in the associated State-of-Science White paper more comprehensively explored these questions. Second, many replication considerations represent important areas for future research for task 4, task 5, and beyond. These issues include the necessary sector size, characteristics, and data variability necessary for applying these models on new sectors in statistically valid ways. Additional subjects for future research include the appropriate replication frequency necessary to characterize a given sector’s current deterrence level and the confounding factors (ie. facility size, industrial sub-category, marginal compliance costs) that may enhance specific and general deterrence.
2. Basic Model and Statistical Intuition Theoretical foundations
Analysis of the impact of environmental regulatory activity on environmental performance is framed in terms of deterrence. Pollution sources decide how much effort to invest in pollution abatement by comparing the marginal benefits and marginal costs of polluting. Marginal benefits of polluting or violating reflect increased production possibilities and decreased abatement expenditures. Marginal costs of polluting or violating are the expected damages associated with regulatory activity and possible community and customer backlash. Greater regulatory activity, as measured by recent inspections or enforcement actions, is hypothesized to increase a plant’s expected compliance and decrease a plant’s expected pollution (on average).
Basic empirical model and intuition
The overall empirical strategy for identifying metrics for specific and general deterrence is to link inspections and enforcement actions to subsequent compliance and pollution behavior. The simplified deterrence measurement models developed in subsequent sections examine plant-level data for many plants in a given sector over several years. The dependent variable in our analyses is a 0/1 discrete compliance indicator or a continuous pollution measure for a given plant in a given time period. For example, the 0/1 compliance indicator may signify if a plant is determined to be in significant non-compliance with its conventional water pollution obligations in a given month. An example of the continuous pollution variable is the percent of permitted TSS contaminants discharged by a given plant in a given month. The key explanatory variable