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

Abstract

There is no consensus whether-and if so, how-monetary policy reacts to asset price fluctuations. This paper shows that the U.S. Federal Reserve Bank lowers interest rates in response to an improvement in market fundamentals, thus accommodating 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. However, unlike some recent notions, the paper reveals that the magnitude of asset price bubbles in interest rate policy rules is too small to warrant any economic policy significance. Stock price bubbles and market fundamentals are estimated using a non-linear model of intrinsic bubbles. Stock market volatility is gauged by an Exponential GARCH model which captures the asymmetries inherent in the conditional variance of equity returns. This paper shows that policy can be improved upon by factoring these market indicators into monetary policy decisions. The results are robust to alternative specifications. Innovative robustness checks using non-nested J tests based on the encompassing principle also reveal that the estimated policy rule encompasses competing alternatives as it appropriately describes the Fed Funds rate behavior in the presence of asset price fluctuations.

JEL Classification: E43, E44, E52, G10

Keywords: Monetary Policy, Stock Market Bubbles, E-GARCH, Non-Nested Tests

I. INTRODUCTION

The relationship between monetary policy in the U.S. and asset prices has received increasing attention over the past two decades. However, two recent developments in the 1990s could potentially change the way monetary policy has traditionally been viewed: one is stock market volatility and the other is productivity gains. (1) The bull market that began in 1983 (Balke and Wolhar, 2001) has led economists to ask whether fundamentals have changed or the market is high only because of some "irrational exuberance" as previous Federal Reserve (Fed) Chairman Alan Greenspan noted in his December 1996 speech. (2) A change in market fundamentals in the form of sustainable productivity gains has very different implications from simply a transient misalignment of asset prices (Koenig, 2000). Mistakenly tightening in response to a fundamentalsdriven asset price boom (reflecting productivity growth, for example) could choke off potential growth (Purfield, Oura, Kramer and Jobst, 2006). Moreover, while volatility in part reflects the nature of asset prices (driven primarily by revisions in expectations of future returns), large movements have important implications as they pose a threat to price stability which is the overriding goal of monetary policy. A sharp correction in asset prices, for instance, may leave the real economy unduly fragile and this vulnerability may become an issue for monetary policymakers (Chadha, Sarno and Valente, 2004).

In the light of asset market developments, asset price changes can be important for monetary policy in at least two ways. First, it had been often wondered whether the central bank targets asset prices in addition to its inflation and output stabilization objectives even though it claims it does not. On the other hand, a second concern is whether asset prices only provide an informational role for monetary policy, for example, providing signals about expected inflation (Chadha, Sarno and Valente, 2004) and preemptive policy response. If asset prices do contain vital information about the economy the logical question, therefore, is: Do asset prices influence the design of monetary policy? This paper investigates the implications for monetary policy when equity market imbalances pose an additional concern to U.S. policymakers besides inflation and output variability.

The existing debate about the role of asset prices in monetary policy has revolved more around analyses of policy prescriptions rather than realized monetary policy actions based on the actual interest rate behavior. Should the Fed systematically react to asset price movements can indeed be better answered if we know whether the Fed does actually react to asset price movements. There are episodes in history that suggest that the Fed did respond to asset prices. For instance, the 1987 and 2001 policy easing in response to the stock market crashes and, the series of interest rate hikes that the Fed undertook in the late 1990s as stock prices escalated to unprecedented levels. However, cases like these are most often described as "discretionary" in the rules-versus-discretion debate (Clarida, Gali and Gertler, 1999). This raises the issue of what variables enter monetary policy rules in a dynamic and continuously evolving financial and economic environment. To understand what policy should be, it is helpful to know what policy has been, so that policy mistakes are not repeated as policy lessons are drawn. (policy mistakes can be costly as they could result in macroeconomic instability and welfare loss to society). The way this has been done in the literature is through a monetary policy rule, the so-called Taylor rule (Taylor, 1993), which describes the interest rate behavior in response to inflation and output variables. To the extent that there is some level of secrecy about what the Fed actually does, an appropriate characterization of the observed monetary policy instrument, for example, the interest rate behavior, would help analyze the role of asset prices for monetary policy design.