Is Voluntary Pollution Abatement in the Absence of a Carrot or Stick Effective? Evidence from Facility Participation in the EPA's 33/50 Program

We examine whether voluntary pollution abatement programs in which there is no program-specific participation incentive are effective in reducing emissions below what they would have been otherwise. We use data on facility participation in the 33/50 Program and emissions reported to the US EPA's toxic releases inventory (TRI) between 1991 and 1995 for a sample of facilities whose parent firms committed to the program. By focusing on participation by individual facilities we avoid the influence of firm level incentives under the program. The mandatory disclosure of emissions data to the TRI avoids the potential bias evident in voluntarily disclosed data. We find that while facilities with larger total emissions were more likely to participate, there is no evidence of greater participation by facilities that account for a higher share of a parent firm's 33/50 emissions. Although emissions of the 33/50 chemicals fell over the years, we find that participation in the program was not associated with the decline in the 33/50 releases generated by these facilities and the reductions seemed to have occurred for reasons unrelated to the program.