Measuring the time-varying impact of conventional monetary policy on stock markets via an identified multivariate GARCH model

Abstract

This paper proposes a new approach to quantifying the impact of the short-term interest rate on the stock market, which is important to policy-makers. A multivariate GARCH model is considered, in which unexpected changes in Fed funds rates are used for identif ication. This approach combines the merits of events studies (information on exogenous shocks) with those of the time-series model. It permits the estimation of time-varying monetary policy effects on the stock market. Our results show that a cut of 25 basis points in the interest rate would induce a median increase of 1.78 percent in the equity index. In periods of high credit risks, the policy effect is stronger and the variation of the policy effect also increases. This pattern has become even more stronger since 2009.