Interest in promoting Pollution Prevention (P2) has been increasing since 1991, following the passage of the Pollution Prevention Act (PPA) of 1990. As part of the PPA, facilities that are subject to the Toxics Releases Inventory (TRI) are required to disclose the number of incremental P2 activities for each listed chemicals from 1991 onward. Though the disclosure is required by the PPA, the adoption of P2 remains a voluntary initiative by firms. To promote P2 ethic among firms, the U.S.EPA has established several voluntary programs and P2 information clearinghouse. P2 technologies were more likely to be facility- and operation- specific and involved considerable information costs and uncertainty. Firms might learn about P2 technologies from their peers through information spillovers. Furthermore, the adoption of P2 technologies might have been motivated in part by regulatory pressures and in part by voluntary commitments. To investigate the role of information spillover, regulatory pressure and voluntary commitment in motivating the adoption of P2 technologies, we focus on the first voluntary program initiated by the U.S. EPA, the 33/50 program. The objective of the program was to reduce the releases of 17 chemicals by 33% by 1992 and by 50% by 1995. It also sought to promote P2 as the preferred method to achieve the reduction. We conduct the empirical analysis on 6974 facilities that were eligible for the 33/50 program from 1991 to 1995. We estimate the number of P2 technologies adopted for 33/50 chemicals and other TRI chemicals at the facility level with respect to program participation, compliance costs to regulations, prior P2 experience by the neighbors on the respective chemicals, and the program participation ratio in the neighborhood. We address the endogeneity of program participation with instrumental variables, and control for location and industry fixed effects. We find that facilities were more likely to learn about adoption of P2 technologies from their industry peers. The direct impact of program participation was only evident for 33/50 chemicals, and the presence of program participants did not significantly motivate P2 adoption in the neighborhood.