Environmental Taxation, Information Precision, and Information Sharing

We analyze how environmental taxes should be optimally levied when the regulators and firms face costs uncertainties in a Stackelberg-Cournot game. We allow linear-quadratic payoffs functions coupled with an affine information structure encompassing common and private information with noisy signals. In the first period, the regulator chooses the intensity of emissions taxes in order to reduce externalities. In the second period, facing industry-related and firm-specific shocks, firms compete in the market-place as Cournot rivals and choose outputs. We show that, given costs uncertainties with non-uniform quality of signals across firms, the regulator sets differentiated tax policy. We also examined the social value of information under ex-ante calibrated emissions taxes. We argue that the magnitude of the associated social benefits and costs of more precise private signals hinge largely and fundamentally on the value of the ratio of the slopes of the marginal damage and the marginal consumer surplus. The lack of accurate data clouds the regulatory process by preventing the necessary fine-tuning of the tax rules towards specific environmental circumstances. Finally, we investigate information sharing between polluters and its impacts on welfare. We stress that, when there are threats of severe environmental damages under deep uncertainties, collusion is welfare reducing and may jeopardize the regulatory process. Numerical simulations illustrate the results that the model delivers.