Precautionary principle and the cost benefit analysis of innovative projects

Public authorities often invoke the precautionary principle to ban or postpone the development of innovative projects with uncertain but potentially harmful irreversible impacts on the environment or health. As stated in the Rio declaration, the precautionary principle suggests balancing costs and benefits associated with irreversible decisions while taking account of the perspective to acquire better but costly information in the future. Though the real option theory seems to be an appropriate tool to deal with the precautionary principle it has two important limits with this respect. First it focuses on Markovian processes rather than on Bayesian learning. Second, it disregards the role of preferences whereas preferences are at the core of the seemingly linked concept of precautionary saving. The article is an attempt to circumvent these two limits. A canonical model of Bayesian real option expressed in terms of intertemporal utility maximisation is presented and solved. The optimal decision rule is discussed in the light of the precautionary principle. It is then shown how to switch consistently to an equivalent problem expressed in terms of costs and benefits.