Doctoral Dissertations

Date of Award

8-2000

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Economics

Major Professor

Jean Gauger

Committee Members

Charles Garrison, Hui Change, Ronald Shrieves

Abstract

This study evaluates the relative empirical performance of two weighted monetary aggregation methods and the simple sum method. In particular, the performance of monetary aggregates constructed by currency equivalent (CE) and Divisia (D) indices is compared relative to each other and relative to their simple sum (SS) counterparts. The empirical performance is measured by the ability of these aggregates to explain fluctuations in real output, nominal output, and prices. Further, their ability to predict changes in output and prices is evaluated and compared to the predictions of the standard macroeconomic theory. This is the first study to comprehensively evaluate all the aggregation methods across the conventional four levels of monetary aggregation (M1 through L). Multivariate time series techniques, in particular vector autoregression (VAR) and vector error correction (VEC) models are used. Several VAR and VEC models are constructed and estimated to provide evidence on the empirical differences between CE, D, and SS aggregates. Dynamic simulations of the systems (using impulse response functions, IRFs, and forecast error variance decompositions, FEYDs) suggest that there are important differences between the performance of CE, D, and SS monetary aggregates in empirical applications. At the M1 level of monetary aggregation, results here indicate that the behavior of CE, D, and SS aggregates is similar and consistently weak. At broader levels of monetary aggregation, the empirical differences between CE, D, and SS aggregates are more pronounced, in particular between CE and D aggregates. Evidence from IRFs and FEVDs indicates that currency equivalent aggregates are notably less informative about changes in either real or nominal economic activity, relative to Divisia aggregates. This evidence suggests that CE aggregates are less useful in applied work as a measure of money, and therefore a less useful policy tool than D aggregates. Similar conclusion is drawn when comparing currency equivalent aggregates against simple sum counterparts. Furthermore, the empirical evidence presented in this study shows a close similarity in the behavior of D and SS aggregates in predicting real and nominal economic activity.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS