by Niklas Schmitz, Winner of the Cambridge Finance Best Student Paper Award 2022, PhD Student, Faculty of Economics, University of Cambridge
During recessions, central banks play a key role to support the economy. They are often concerned with avoiding bad outcomes for macroeconomic growth or mitigating “downside risks”. Stock markets follow monetary policy decisions closely and respond to policymakers’ actions. But what exactly explains the response of stock prices to monetary policy announcements, especially in crisis times?
Empirical evidence on the effect of monetary policy surprises on equity returns shows that the equity premium accounts for a large share of the return response: Policy decisions affect equity returns to a large extent via the equity premium, ie the compensation that investors require to hold the risk associated with stocks in their portfolios (Bernanke and Kuttner, 2005). But why should monetary policy affect the risk premium?
In a recent paper, I argue that monetary policy can affect equity premia during recessions because its interventions reduce downside risks to future macroeconomic growth. The motivation for this channel comes from a long-standing literature of consumption-based asset pricing models. This literature suggests there should be a close link between the (expected) state of the macroeconomy – usually measured via aggregate consumption – and stock returns (Lucas 1978). Since stocks tend to fall in price when the state of the macroeconomy is bad, they offer low returns in times when investors value returns most strongly, which implies the equity premium. Models incorporating the potential for large negative shocks to the economy have emerged as a promising avenue to realistically connect macroeconomic aggregates and stock market returns (Rietz 1988, Barro 2006, Gabaix 2012).
To empirically test this logic of the “downside risk channel of monetary policy”, we require a measure of downside risks to aggregate consumption growth. Inspired by a recent literature in macroeconomics (Adrian et al. 2019), I obtain this measure by forecasting the conditional distribution of future aggregate consumption growth in the United States. This distribution provides the answer to the question: Given the information about economic and financial conditions available today, what are the possible scenarios for future realisations of consumption growth? Uncertain times are reflected by a large spread in the conditional distribution. Distinctly bad times may be represented by a low mean and a long left tail. Based on the estimated distribution, I construct an index of downside risk that reflects the probability of below-average realisations for consumption growth.
This approach has several advantages over existing measures of risk or uncertainty. First, the index is a clean measure of risk perceptions without any risk aversion component. Especially risk measures obtained from financial market data often struggle to disentangle these two components since market prices necessarily reflect both expected risks and the valuation of these risks. To test the logic of the downside risk channel, we want to focus on the first component only. Similarly, since the index measures risks to a macroeconomic variable, it does not pick up any financial stress that does not spill over into the macroeconomy. Other commonly used measures such as the VIX index do not guarantee this. The downside risk index also allows for a distinction between upside and downside risks. An important result of the paper is that the downside risk channel is strong, whereas there is less clear evidence on the importance of “upside risks”.
The estimation results for the downside risk index show that increases in the probability of bad realisations for future consumption growth occur in line with economic intuition: In the build-up of recessions, downside risks generally rise. Downside risks also increase around specific events such as the collapse of Lehman Brothers in September 2008. Following the Great Recession, downside risks remain elevated for several years, during which the Fed continued to implement its Quantitative Easing. The COVID-19 (coronavirus) shock is associated with another rise in downside risk.
Given the downside risk index, I test the existence of the downside risk channel in two steps. First, I study the predictive ability of the downside risk index for equity premia in the US. Second, I test whether Federal Reserve policy affects the downside risk index.
Changes in the downside risk index are a significant predictor of the equity premium. An increase in the downside risk index predicts higher excess market returns over the next three to six months. The predictive ability of the downside risk index is concentrated in crisis times, whereas the association between downside risk and future returns is weak outside of recessions. The index also predicts future returns on industry portfolios. Industries such as healthcare or non-durable goods consumption show a low sensitivity to changes in downside risk, whereas procyclical industries such as manufacturing or finance show a high sensitivity to changes in downside risk.
Monetary policy has a powerful effect on downside risk. This effect is again concentrated in recession, whereas I find no measurable effect of monetary policy changes on downside risk outside of recessions. This suggests that changes in the monetary policy stance are effective at reducing downside risks during recessions, which can support equity prices by reducing the risk premium investors require to hold equities.
The downside risk channel of monetary policy implies central banks have a powerful lever on stock prices that goes beyond any effect on realised or expected macroeconomic growth rates. Downside risks to future growth may play a particularly important role in explaining the effectiveness of monetary policy in crisis times.
Adrian, T., Boyarchenko, N. & Giannone, D. (2019) ‘ Vulernable Growth’, American Economic Review 109(4), 1263-1289.
Barro, R. J. (2006) ‘Rare disasters and asset markets in the twentieth century’, The Quarterly Journal of Economics 131(4), 1593-1636.
Bernanke, B. S. &Kuttner, K. N. (2005) ‘What Explains the Stock Market’s Reaction to Federal Reserve Policy?’, Journal of Finance 60(3), 1221-1257.
Gabaix, X. (2012) ‘Variable rare disasters: An exactly solved framework for ten puzzles in macro-finance’, The Quarterly Journal of Economics 127(2), 645-700.
Lucas, R. E. (1978) ‘Asset Prices in an Exchange Economy’, Econometrica 46(6), 1429-1445.
Rietz, T. A. (1988) ‘The equity risk premium – a solution’, Journal of Monetary Economics 22(1), 117-131.