Policy News and Stock Market Volatility

Abstract

We exploit the text in newspapers and 10-K filings to quantify the drivers of aggregate and firm-level stock market volatility. We first create a newspaper-based Equity Market Volatility (EMV) tracker that moves closely with the VIX and the volatility of returns on the S&P 500. Parsing the underlying text, we then construct forty category-specific EMV trackers. News about commodity markets, interest rates, real estate markets, aggregate activity and inflation figure prominently in EMV articles, with large category-specific variation over time. Policy news is another major source of market volatility: 30 percent of EMV articles discuss tax policy, 30 percent discuss monetary policy, and 25 percent refer to some form of regulation. Trade policy news went from a virtual nonfactor in market volatility to a leading source after U.S.-China trade tensions escalated. Next, we use 10-K filings to quantify firm-level exposures to the same forty risk categories. Combining our newspaper-based measures with our textual analysis of 10-K filings, we obtain monthly firm-level risk exposure measures. Finally, we show that these measures are highly statistically significant in explaining the firm-level structure of realized equity market volatilities at the monthly frequency, even after conditioning on firm effects, time effects and industry-time effects.

Keywords: stock market, equity returns, volatility, uncertainty, government policy

JEL Classification: D80, E22, E66, G18, L50