In recent years, several scientific and political controversies have arisen regarding the market introduction of new products and/or new technologies. Key examples are genetically modified organisms and hydraulic fracturing to extract shale gas. In both cases, a common feature is that the regulatory decision to approve, or to ban, the new product and/or new technology, faces the challenge of balancing large economic profits against uncertain (and possibly huge) environmental and/or health detrimental effects. The stylized fact is that this has lead to large differences in regulations among countries, which may reflect the preferences of various interest groups involved in the process. ¹¹See Vigani and Olper (2013), for genetically modified organisms, and Johnson and Boersma (2012), for hydraulic fracturing to extract shale gas.

This paper aims at representing the government decision-making in such situations, accounting for the intervention of interest groups, in a context of scientific controversy. Specifically, we analyze a Tullock contest (Tullock, 1980), with the two contestants being an industrial and an environmental lobby. The key assumption in our model is that the industrial lobby has private information about the detrimental externality, but can be held liable for damage ex post. We determine the equilibrium of the contest and derive both its positive and normative properties. Importantly, we identify in which circumstances the lobbying activities foster a more or less efficient decision-making by the government, which may justify and help implementing ex ante constitutional rules to frame lobbying activities (Brennan and Buchanan, 1985).

In a sense, Cropper et al. (1992) provide an empirical background to our analysis. They examine the EPA’s decision to cancel or continue the registrations of cancer-causing pesticides between 1975 and 1989. During this period, the final decision followed a two steps procedure, with the EPA first proposing a decision based on a risk-benefit analysis, and the interest groups then contesting it. Cropper et al. (1992) find that the EPA’s decisions indeed balanced risks against benefits, but simultaneously that intervention by special-interest groups was also important in the regulatory process. They conclude that their “findings provide both comfort and concern to those interested in improving the efficiency of environmental regulation” (p. 178).

The main result of this paper is to provide conditions under which it is socially better that the government decides to approve, or to ban, the new product and/or new technology, according to the contest, rather than according to an ex ante cost-benefit analysis, using his prior beliefs. The reason why this can be so, is because the ex post liability for harm induces the industrial lobby to reveal his private information about the environmental and/or health detrimental externalities in the contest. The reason why this is not always so, is because the industrial lobby can be financially insolvent and the contest is a costly scheme. More precisely, we find that the contest outperforms the ex ante cost-benefit analysis, if and only if the industrial lobby can lose in court a sufficient fraction of the profit earned from selling the new product/technology. We check the robustness of our main result with two extensions. On the one hand, we show that the contest should be used more carefully if there exists an incentive to free-ride within the environmental pressure group. On the other hand, we show that the contest can be used more confidently if the environmental lobby also knows the harm prior to the contest.