6. Major Findings and Recommendations
Can OECA identify simplified quantitative database analyses capable of capturing the specific and general deterrence effects of environmental monitoring and enforcement in a scientifically rigorous yet efficient manner?
Major Finding 1: Simplified, cost-effective quantitative database methods exist to measure the specific and general deterrence of environmental monitoring and enforcement.
Sections 3 and 4 present such simplified metrics and statistical methods. They are grounded in peer-reviewed research and technically rigorous, yet can be cost- effectively implemented in-house by OECA personnel with modest database and statistical training.
Major Finding 2: When benchmarked against data analyzed in the pre-existing literature, the simplified metrics and methods typically produce deterrence effects approximately equal to those in the literature. Specific deterrence results for our models applied to air compliance status in a 1980-1989 steel industry dataset are typically quite similar to the results found in the peer-reviewed studies Gray and Deily (1996) and Deily and Gray (2007). General deterrence results for our models applied to water discharges in a 1990- 2004 pulp and paper industry dataset are nearly identical to the results found in the peer-reviewed study Shimshack and Ward (2008). General deterrence results for our models applied to water non-compliance status in a 1990-2004 pulp and paper industry dataset are very similar to the collective results in the peer- reviewed studies Shimshack and Ward (2005) and Shimshack and Ward (2008).
How can OECA use the methods and results to measure and/or manage elements of compliance assurance and enforcement programs?
Recommendation 1: In the short-run, OECA and its contractors should apply the simplified models developed in Sections 3 and 4 and benchmarked in Section 5 to approximately 4 additional sector / pollution media combinations (as outlined in Task 4 of the Statement of Work). On average, the easily implemented models seem to reveal similar deterrence effects as their more sophisticated and costly counterparts in the academic literature. Further, applying these models to new sectors, contaminants, and time periods could importantly contribute to the state of knowledge on deterrence. While the model benchmarking for conventional air and water pollutants in the steel and pulp and paper sectors is illustrative, it is quite possible that other sectors and contaminants exhibit different patterns of deterrence. Such deterrence effect heterogeneity would not be surprising, since various sectors, contaminants,