Essays in monetary policy conduction and its effectiveness: monetary policy rules, probability forecasting, central bank accountability, and the sacrifice ratio

Abstract

Monetary policy has been given either too many positive attributes or, in contrast, only economy-disturbing features. Central banks must take into account a wide variety of factors to achieve a proper characterization of modern economies for the optimal implementation of monetary policy. Such is the case of central bank accountability and monetary policy effectiveness. The objective of this dissertation is to examine these two concerns relevant to the current macroeconomic debate. The analyses are carried out using an innovative set of tools to extract presumably important information from historical data of selected macroeconomic indicators. This dissertation consists of three essays. The first essay explores the causality between the elements of the "celebrated" Taylor rule, using a Structural Vector Autoregression approach on US data. Directed acyclical graph techniques and Bayesian search models are used to identify the contemporaneous causal structure in the construction of impulse-response functions. Further analysis is performed by evaluating the implications of performing standard innovation-accounting procedures, derived from a Structural Vector Autoregression on interest rates, inflation, and unemployment. This is examined whenever a causal structure is imposed vs. when it is observed. We find that the interest rate causes inflation and unemployment. This suggests that the Fed has not followed a Taylor rule in any of the two periods under study. This result differs significantly to the case when the causal structure is imposed. The second essay presents an incentive-compatible approach based on proper scoring rules to evaluate density forecasts in order to reduce the central banks' accountability problem. Our results indicate that the surveyed forecasters have done a "better" job than the Monetary Policy Committee (MPC). The third essay analyzes the causal structure of the factors that are presumed to influence the effectiveness of monetary policy, represented by the sacrifice ratio. Directed acyclical graph methods are used to identify the causal flow between such determinants and the sacrifice ratio. We find evidence that, while wage rigidities and central bank independence are the two major determinants of the sacrifice ratio, the degree of openness has no direct effect on the sacrifice ratio.