The Federal Reserve And Market Confidence
Federal Reserve Bank of New York
Princeton University – Bendheim Center for Finance
Matthew C. Plosser
Federal Reserve Banks – Federal Reserve Bank of New York
2016-04-01
FRB of NY Staff Report No. 773
Abstract:
We discover a novel monetary policy shock that has a widespread impact on aggregate financial conditions. Our shock can be summarized by the response of long-horizon yields to Federal Open Market Committee (FOMC) announcements; not only is it orthogonal to changes in the near-term path of policy rates, but it also explains more than half of the abnormal variation in the yield curve on announcement days. We find that our long-rate shock is positively related to changes in real interest rates and market volatility, and negatively related to market returns and mortgage demand, consistent with policy announcements affecting market confidence. Our results demonstrate that Federal Reserve pronouncements influence markets independent of changes in the stance of conventional monetary policy.
The Federal Reserve And Market Confidence – Introduction
“Fed policy never works solely or even mostly through short-term interest rates. Instead, it operates through a constellation of financial conditions: stock prices, corporate bond yields, commodity prices, exchange rates and most critically … the appetite for risk.” (Greg Ip, Wall Street Journal, January 25, 2016)
Modern monetary policy relies heavily on communication as a supplementary policy tool. While central banks typically have direct control over a single interest rate, the mandate to support a well-functioning economy might necessitate the ability to influence asset prices and interest rates at all maturities. The recent experience of major central banks has revealed that monetary policy authorities have unconventional tools to impact asset prices in a low rate environment. In this paper, we evaluate whether monetary policy announcements play this role beyond the recent period of unconventional policy by studying changes in asset prices around conventional monetary policy announcements.
The key empirical challenge in identifying the transmission of monetary policy shocks to asset prices is that interest rate changes are not exogenous to the state of the economy, as monetary policy is set in response to changes in economic conditions. The extant literature has pursued three approaches to overcoming this endogeneity problem: structural vector autoregressions (VARs) (e.g. Christiano et al., 1999), using changes in interest rates orthogonal to the information contained in internal Fed forecasts (Romer and Romer, 2004), and identification using high frequency changes to interest rates around FOMC announcements (e.g. Kuttner, 2001; Nakamura and Steinsson, 2013).
This paper uses a novel methodology that explicitly allows for multiple monetary policy shocks. We construct the monetary policy shocks as principal components of the variation in yields attributable to policy announcements. This is an extension of the Rigobon (2003) heteroskedasticity approach, where instead of focusing on a single interest rate, we consider changes to the variance-covariance matrix of the changes to the entire U. S. Treasury nominal yield curve around scheduled Federal Open Market Committee (FOMC) meetings. The identification assumption is that the difference between the covariance matrix in the benchmark periods and the covariance matrix on announcement dates is the covariance matrix of monetary policy shocks. This approach allows us to jointly estimate multiple policy shocks by inferring information from changes to the whole yield curve. Instead, previous studies have focused on a single policy factor related to changes at shorter maturities (see e.g. Gurkaynak et al., 2005a; Nakamura and Steinsson, 2013), and thus confound the short rate factor with an “expected path” factor.
Federal Reserve and Market Confidence
We identify two distinct monetary policy factors. The primary factor, which explains 59 percent of the variation in yields attributable to announcements, is related to level of longer term yields. We find that this factor correlates strongly with measures of uncertainty, suggesting that monetary policy announcements affect longer term risk premia, rather than long-run expectations of the policy rate. The second factor is the more traditional short rate factor, and explains a further 34 percent of the variation in yields attributable to announcements. Importantly, these two factors are distinct in the data, suggesting that movements in the long-rate related to uncertainty are separate from the classic focus on near short rates determined by the stance of traditional monetary policy.
The monetary policy shocks we identify have large and distinct impacts on the returns to other risky assets in the economy. To estimate the correlation of the policy factors with other assets, we compute the differential covariance of the factors with asset returns on announcement days versus non-announcement days. In the inflation-protected Treasury market (TIPS), we find that the first factor affects the entire TIPS yield curve, with a higher correlation between the factor and changes to TIPS yields than that between the factor and nominal yields. This suggests that changes to inflation expectations alone are insufficient to explain the impact of the long-term monetary factor and that the covariance is better explained by changes in the risk premium or in expectations of the future path of real rates. Fed funds shocks, on the other hand, have little impact on real interest rates but do lower inflation expectations. This is consistent with a negative lead-lag relationship between consumption growth and realized inflation (e.g. Stock and Watson, 1999).
We argue that the long-term monetary factor impacts aggregate uncertainty and not necessarily expectations of the future path of interest rates. We find that the long-term factor is significantly positively correlated with changes to various measures of aggregate uncertainty, including the one month option-implied volatility for S&P 500 (VIX), S&P 100 (VXO) and U. S. interest rate swaps (SMOVE). Thus, the long-term monetary policy shock impacts aggregate volatility and should command a risk premium.
Turning next to equity markets, we find that the long rate shock is negatively related to the aggregate market return on announcement days, but positively related on other days. This is consistent with FOMC communications impacting risk premia in the economy by coordinating beliefs about economic uncertainty, as outlined in section 2, as well as decreased firm-level investment after shocks to macroeconomic uncertainty (see e.g. Lucas and Prescott, 1971; Bloom et al., 2007; Bloom, 2009). The Fed funds shock, on the other hand, does not covary significantly with aggregate market returns neither on announcement nor non-announcement days. Instead, the Fed funds shock is negatively related to the “value” portfolio factor (HML) on non-announcement days and positively related on announcement days, whereas the long-rate factor is negatively related in both periods. These results emphasize that the two monetary policy shocks are distinct in their economic impact, as stocks with disparate risk characteristics react differently to the two shocks.
The two policy shocks also have distinct effects on real activity. We find that the long-rate shock is positively correlated with the trade-weighted U. S. exchange rate, commodity risk premia, credit risk premia and physical currency amounts. An unexpected increase in the Fed funds rate, on the other hand, tightens financial conditions for both households and financial corporations and decreasing money supply.
More generally, innovations in the long-rate factor appear to be orthogonal to innovations in the short-rate factor, suggesting that the mechanism by which the FOMC impacts the Fed funds rate is different from the one through which the FOMC impacts the long rate. The strong relation between the long rate factor and aggregate risk premia on announcement days reflects the ability of monetary policy to influence real discount rates in the long term and, as such, should be discussed explicitly as a potential policy tool.
Federal Reserve and Market Confidence
Market Confidence See full PDF below.