A Maximum Probability Approach to Short-run Policy

The optimal control literature traditionally analyzes linear-quadratic Gaussian (LQG) formulations of macroeconomic policy design where policy planers seek to minimize the distance between a direct target vector y and a vector of aspiration levels y*, using a suitably dimensioned matrix M as a distance metric. Until recently, neither y* nor M were known, and the arguments within the quadratic loss functions were arbitrarily selected. Since passage of the Humphrey-Hawkins Act, estimates of y* are available, but the metric M, necessary to resolve goal conflicts remains unknown. This paper suggests an approach to ex ante policy evaluation using a Mahalanobis distance metric. The proposed objective is to maximize the probability that the direct target vector y will lie in a region covering y*. By the Gaussian assumption, this problem reduces to minimization of a quadratic form, where the inverse of the direct target vector serves as the effective distance metric. Both the mean and covariance of y are influenced by the design of the direct-target policy. Results of the proposed direct target policy are contrasted with results of an intermediate target policy using stochastic properties of the FRB quarterly model. This comparison suggests that direct targeting is about five times more effective than intermediate targeting