Jaffe and Palmer (1997) present three distinct variants of the so-called Porter Hypothesis. The “weak” version of the hypothesis posits that environmental regulation will stimulate certain kinds of environmental innovations. The “narrow” version of the hypothesis asserts that flexible environmental policy regimes give firms greater incentive to innovate than prescriptive regulations, such as technology-based standards. Finally, the “strong” version posits that properly designed regulation may induce cost-saving innovation that more than compensates for the cost of compliance. In this paper, we test the significance of these different variants of the Porter Hypothesis using data on the four main elements of the hypothesised causality chain (environmental policy, research and development, environmental performance and commercial performance). The analysis is based upon a unique database which includes observations from approximately 4200 facilities in seven OECD countries. In general, we find strong support for the “weak” version, qualified support for the “narrow” version, and qualified support for the “strong” version as well.