Does pollution prevention reduce toxic releases? A dynamic panel data model

We investigate the effectiveness of voluntary pollution prevention activities in reducing toxic releases from facilities that reported to the U.S. Environmental Protection Agency's Toxics Release Inventory from 1991-2001, using generalized method of moments dynamic panel data models that recognize the potential endogeneity of the pollution prevention adoption decision on toxic releases. We find that pollution prevention adoption had a negative impact on toxic releases. The estimated coefficients suggest that the effect of pollution prevention adoption is substantial, but short-lived, dissipating within 4 to 5 years. However, a continual adoption of pollution prevention techniques leads to lower steady-state releases, with estimated reductions between 35% and 50%.