This is a summary of "Systematic Extreme Downside Risk," by Richard D. F. Harris, Linh H. Nguyen, and Evarist Stoja, published in the Journal of International Financial Markets, Institutions & Money.
The authors propose two new systematic tail risk measures that they recommend investment managers factor into asset pricing: the tendency of an asset to crash simultaneously with market crashes and the impact of hedging mechanisms to protect against downside risk of an asset.
What Is the Investment Issue?
“The turbulence of financial markets over the last few decades has highlighted the importance of tail risk for asset pricing.”
If an asset performs well during market crashes, investors are willing to pay a premium because it will yield high returns when investors’ wealth is low. Alternatively, assets that tend to crash simultaneously with the market are risky, and investors demand higher risk premiums to hold them. Similarly, if investors are able to hedge against extreme downside risk, they will be willing to pay a premium for the hedge asset—that is, a price discount.
The authors thus propose two new systematic tail risk measures for determining the appropriate tail risk premium for investors that have significant power to predict future returns. The first measure is the propensity of an asset to crash simultaneously with the market: extreme downside correlation (EDC). The second measure incorporates the sensitivity of stock returns to hedging mechanisms against extreme downside risk: extreme downside hedge (EDH). The authors carry out the study after controlling for other measures of downside risk, such as downside beta, co-skewness, and co-kurtosis.
How Did the Authors Conduct This Research?
The authors use daily data for the period 1968–2017 for all stocks listed on the NYSE, AMEX, and NASDAQ. They exclude stocks with less than six months of data and stocks with prices lower than $5 or higher than $1,000. The stocks are grouped in quintiles based on each risk measure, and then the equally weighted excess returns of these quintiles are calculated for individual years. The authors use a return quantile as the threshold rather than the exceedance threshold based on the standard deviation of stock returns because returns are not normally distributed.
The authors use expected tail loss (ETL), also known as conditional value at risk (CVaR) or expected shortfall, as a measure of market tail risk. ETL is the expected value of the loss given that the loss exceeds VaR. Extreme downside hedge is estimated by regressing asset returns on a measure of market tail risk.
What Are the Findings and Implications for Investors and Investment Professionals?
EDC positively predicts the next month’s return, and the coefficient is highly significant. But the coefficient of EDH is not significant, because of EDH’s high variation for stocks with very high or very low exposure. Two advantages of EDH are that it can be estimated by using a very short sample and the impact on stock returns of systematic tail risk can be examined at different horizons.
The authors find that the price premium is significantly positive for EDH when the sample size is at least nine months, suggesting that it takes time for investors to be compensated for bearing high systematic tail risk. When tail risk is high, the stock price is lower and even negative in the short term. Over time, the expected return rises to compensate investors for the higher risk.