Environmental Self-Auditing : Setting the Proper Incentives for Discovering and Correcting Environmental Harm

Many firms have instituted a policy of conducting their own "environmental audits" to test compliance with a complex array of environmental regulations. Yet, commentators suggest that self-auditing is still not as common as it should be because firms fear that the information they gather will be used against them. This paper analyzes the two-tiered incentive problem raised by self-auditing-viz., incentives to both test for and effect compliance. We find that conventional tort remedies fail to produce an efficient amount of self-auditing. To fix the problem we propose three separate solutions, each with differing informational requirements and efficiency benefits, and each distinct in its own way from current EPA policy. First, we propose that punitive fines be reduced for firms that conduct their own investigation, whether or not the firm has "fixed" the harm that its investigation uncovers. Importantly, we argue that the nature of the self-auditing incentive problem makes conditioning on investigation informationally feasible, since it is the potential observability of investigative effort that produces the disincentive to investigation in the first place. Our second solution conditions on firm disclosure. While this solution allows for additional savings in government enforcement costs, it raises serious informational issues regarding the verifiability of disclosure. Lastly, we consider a solution that we call "inverse negligence," wherein firms are fined additionally for harms that they would have fixed, had they learned about them through investigation. This solution requires neither verifiable disclosure, nor observable investigation effort, but does require additional information about the firm's private cost of fixing harms