The effects of monetary policy, inflation, and interest rates on residential fixed investment : an empirical analysis

This dissertation undertakes an empirical examination of residential fixed investment (RI). The purpose of the research is twofold. The first objective is to document statistically the cyclical nature of RI and its relationship with monetary policy, inflation and interest rates. This is done using a technique that considers potential simultaneous or mutual causality among the variables, vector autoregressions (VAR). The stability of the results for the effect of money on RI is evaluated when nominal interest rates are considered. Also, the stability of the results for the effect of nominal interest rates on RI is evaluated when real rates, anticipated inflation and financial deregulation are considered. The second objective is to consider the implications of these results for the recent theories regarding the limited efficacy of monetary policy and the lack of causality from interest rates to fluctuations in real economic activity (e.g. Sims' theory of post-war business cycles and the Fama-Litterman-Weiss hypothesis). This is a unique way of examining these theories because it involves an economic aggregate (RI) that generally makes cyclical turns prior to those of the overall business cycle. The hypotheses were originally formulated and tested using aggregates or indices that coincide with or lag the general business cycle. A 'leading' series may be important to examine in these cases. as it may directly be affected by monetary policy or interest rates and also act directly as an impulse mechanism for GNP. Sims' conclusions about the behavior of money in the business cycle cannot be challenged with these results. The Fama-Litterman-Weiss hypothesis, however, with respect to RI is rejected. Nominal interest rates are found strongly 'causally' prior to RI. The rate of interest remains 'causally' prior to RI in the era after financial deregulation (albeit with a slightly smaller effect). RI is also shown strongly 'causally' prior to GNP. This research helps reestablish a link between nominal interest rates and real economic activity.