This paper uses the Edmonds-Reilly model to explore an alternative approach for using energy- economic- environmental models when analysing future CO2 emissions. This approach - conducting probabilistic policy experiments - can be used to investigate the effectiveness of various policy options in the context of uncertainty. The analysis builds on work by Nordhaus and Yohe (1983) and Edmonds et al. (1986). A key feature of using a probabilistic approach is that it offers both analysts and policy- makers an opportunity to move away from arguing about which scenario is the 'right', best-guess scenario, and towards a discussion of which strategies are effective across an wide range of possible futures. This paper both develops a methodology for conducting probabilistic policy experiments and presents the results of five preliminary experiments using this approach.

© 1995 Inderscience Enterprises Limited