Are policy variables exogenous? The econometric implications of learning while maximizing (Q1202130)

The subject of this study is concerned with important links between rational expectations theory and the theory of learning. On the background of equilibrium consequences of the rational expectations hypothesis the relationship between learning and econometric exogeneity of policy variables is investigated. A paradigmatic model is presented in which the government maximizes the discounted sum of tax revenues, subject to a constraint imposed by a Laffer curve which involves a parameter initially not precisely known, but about which a process of Bayesian learning can be started. The study focuses on the behavior of a rational agent with less than perfect information. His information set is refined by means of learning using observations generated by the environment. In chapter I, the policy maker's problem is formulated as a stochastic multiperiod optimization problem. The properties of the model are established in chapter II. The conditions for the occurrence of active learning is related to the curvature of the convex value function in beliefs arising in the dynamic optimization approach. Chapter III explores the implication for econometric practice and gives an answer to the question posed in the title. Learning affects the exogeneity of policy variables, active learning is proven to be a distinct cause of time inconsistency of optimal plans. Chapter IV presents the results of a simulation analysis where active and passive learning is contrasted. Several variants of exogeneity tests are applied to data generated under different assumptions on the mode of learning. The research study represents an important contribution to the discussion of rational learning and demonstrates the impact of learning in different forms of exogeneity. The book is best recommended to researchers in the scientific fields of macroeconometric theory and applications.