Does U.S. monetary policy react to asset prices? Implications of stock market bubbles, volatility and productivity

Indian Journal of Economics and Business, Dec, 2007 by Rajeev Sooreea

This paper attempts to investigate this issue. It examines the implications of monetary policy when asset prices pose an additional concern to U.S. policy makers besides inflation and output stabilization. When analyses of stock markets and monetary policy have been undertaken, adequate distinction has not been made between asset price bubbles, market fundamentals and stock market volatility. The current study highlights that these issues are indeed different and could have different policy implications. It decomposes the stock market into three components, namely, stock market bubbles (which captures the irrational exuberance component of asset prices), market fundamentals (which is essentially productivity driven) and stock market uncertainty, proxied by asset returns volatility (which is driven by shocks and revisions in expectations). The response of monetary policy, through the Fed Funds rate instrument, is then analyzed with respect to these three stock market elements.

The contributions of this study are indeed threefold. First, this paper uses a nonlinear model of intrinsic bubbles to isolate market fundamentals from asset price misalignments. The estimated measures of fundamental values of stocks and asset price bubbles are then used in an augmented Taylor rule to assess the Fed's response. As fundamental values are essentially driven by productivity, this paper also uses a more direct measure of productivity growth to assess the robustness of the Fed's reaction function. Productivity has surprisingly not been modeled explicitly in Taylor rules, and this paper shows, that unlike standard variables that enter in Taylor rules, productivity or market fundamentals can have very different implications for monetary policy. Second, this paper estimates a measure of asset returns volatility by using an Exponential Generalized Autoregressive Conditional Heteroskedastic (E-GARCH) model. The attractive feature of this model is that it captures the asymmetries-the leverage effects -that are inherent in the volatility of stock returns. This estimated measure of volatility is then also used as an indicator of stock market uncertainty in the Taylor rule. Finally, for robustness checks, this paper uses formal non-nested J tests (Davidson and MacKinnon, 1993) based on the encompassing principle of Mizon and Richard (1986) in an innovative way to test the estimated reaction function against existing rival models.

The key conclusions of this study are as follows. There is evidence to suggest that the Fed lowers interest rates in response to an improvement in market fundamentals. This means that the Fed tends to accommodate productivity gains which essentially signal low future inflationary pressures. The Fed also responds to higher stock market volatility by reducing interest rates in order to inject liquidity in the market and boost business confidence. This could be observed especially in periods of high stock market volatility such as the 1987 and 2001 stock market crashes. However, unlike some recent notions, the paper shows that asset price bubbles are not economically significant for monetary policy. The magnitude of asset price bubbles in interest rate policy rules is too small to warrant any significance from an economic policy perspective. The paper's results are robust to alternative specifications and the non-nested tests reveal that the estimated reaction function encompasses competing alternatives (outperforms rival models) as it appropriately characterizes the Fed Funds rate behavior in the presence of stock market developments.