This paper pursues the same line of research, which is to understand whether the intervention of interest groups in the political decision-making process can help to design better regulations, or not. The closest papers are Aidt (1998), Aidt and Hwang (2014), and Graichen et al. (2001). Our principal contribution is to introduce the issue of scientific controversy about the environmental and/or health detrimental externalities, as represented by the private information of the industrial lobby, and to highlight the importance of the ex post liability system.

More specifically, we depart from Aidt (1998) and Aidt and Hwang (2014), by using a Tullock contest, instead of a common agency model of politics. Importantly, this means that the lobbying activity is viewed as costly in our setting, whereas it only involves transfers of numeraire in Aidt (1998) and Aidt and Hwang (2014). As a result, there is less chance that the intervention of the interest groups be optimal in this paper. Finally, we depart from Graichen et al. (2001) by going further into the analysis, to investigate the effect of the contest equilibrium on the social surplus. Surprisingly, Graichen et al. (2001) did not consider this question.

The paper is organized as follows. Section 2 presents the model. Section 3 determines the equilibrium of the contest game. Section 4 and 5 respectively derive the positive and normative properties of the outcome of the contest. Section 6 amends the model, by assuming in turn the possibility of free-riding within the environmental group and the possibility that both lobbies know the damage. Section 7 concludes. Proofs of most results are contained in an Appendix.