The informational value of Toxics Release Inventory performance

Purpose – The study aims to evaluate the informational value to investors of the Toxics Release Inventory (TRI) as an external outcome measure of corporate environmental performance. Emphasis is placed on the market response differences between three highly polluting industries. Design/methodology/approach – The study uses pooled cross‐sectional, time‐series data and an event study methodology to examine the effects of TRI emissions on abnormal market returns. Findings – There is empirical evidence that market reactions to TRI emissions information vary by industry. Investors reward decreases in emissions in the electric utility industry, but do not penalize increases. In the chemical industry, increases in emissions are penalized, but decreases are not rewarded. Models do not capture any reaction to emissions changes in the pulp and paper industry. These results may be explained by the significant difference between industries in the US percentage of total firm sales. Research limitations/implications – This research analyzes only data from US firms in three industries and evaluates a single measure of environmental performance, TRI. The value of TRI information is measured for one stakeholder group. Future research should attempt to address these limitations. Practical implications – The results suggest that research on the effects of environmental performance on market‐based measures should estimate models by industry, whenever possible. From a public policy perspective, the results suggest that regulators may want to consider alternative methods of reducing chemical emissions beyond TRI disclosure in the chemicals and pulp and paper industries. Originality/value – The study distinguishes between the value of TRI as a “message service” and the content of the “message”. TRI may provide information of value to investors, but performance changes may not be sufficient to merit a price response. The study also specifically addresses industry differences and clearly shows how average coefficients can be misleading relating to this one environmental performance indicator. The use of pooled industry coefficients may lead to inefficient resource allocation decisions within industries and ineffective policies at the regulatory level.