Association for Public Policy Analysis & Management

Philosophers have studied the public enforcement of law, and its implications for regulatory enforcement and compliance, since at least Bentham (1789). Although the related literature is now large, nearly all studies explore the use and impacts of traditional inspections and sanctions. This paper explores a policy instrument receiving little attention in the existing literature: mandatory disclosure within an established regulatory enforcement system.

This paper explores if compulsory labeling of regulated pollution point sources influences facilities’ environmental performance. Our specific empirical context is an Ohio program that required all water pollution permits issued after June 2011 to be accompanied by an outfall signage requirement. We believe this is the first paper to rigorously study mandatory outfall signage. Our evidence matters because regulators are responding to constrained budgets and technological innovations by turning to “next generation compliance” tools, often with targeted transparency and policies leveraging information technologies (Giles 2013).

It is an open question whether outfall signage would be expected to affect pollution and compliance. Scholars and policymakers know relatively little about transparency-based compliance tools for two reasons. First, these programs are typically new and data are scarce. We study a comprehensive dataset of Clean Water Act (CWA) permitting events and Ohio signage requirements from EPA and Ohio EPA. We match these data with extensive facility-by-month water pollution violation, discharges, monitoring, enforcement, facility characteristic, and community characteristic data for all major facilities in Ohio and five control states.

Second, credible statistical identification of disclosure program effects can be extremely challenging. The key concern is program adoption and timing may be endogenous. We exploit a natural experiment to attempt to isolate the causal effects of the signage program. Our first generalized difference-in-differences (DID) strategy is a standard intent-to-treat approach using the implementation date of June 2011 to define treatment periods. We compare before/after changes for treatment facilities in Ohio vs. before/after changes for similar control facilities in non-program states. Our second generalized DID strategy uses permit status changes after 6/2011 to define treatments. Since the signage requirement is only officially binding after permits are renewed, and since permit renewals are typically exogenously determined by fixed 5-year cycles that vary across facilities, this approach should minimize common sources of bias. Our third approach uses a difference-in-difference-in-differences (“triple diff”) approach to construct counterfactuals using the effects of permit status changes in other states as well as the effects of pre-program permit status changes in Ohio. We explore the effects of a permit status change after 6/2011 in Ohio, after netting out the effects of permit status changes after 6/2011 in control states –and– the effects of permit status changes in Ohio itself before the program.

We find that the outfall signage program had significant and robust effects on facilities’ conventional water pollution discharges and compliance. Relative to a counterfactual, treated facility violations fell about 1/3 of baseline levels. Average pollution discharges about 8 percent of baseline levels. These results appear to be almost exclusively driven by facilities located in high education areas.