Stock market volatility and public information flow: A non-linear perspective

Abstract

The relationship between the level of stock market volatility and public information flow is non-linear, resembling a bell-shaped function. Medium levels of information flow generate heightened volatility, whereas weak and strong information flows do not, regardless of whether news are negative or positive. This novel empirical finding is established in a new realized GARCH model with time-varying intercept, measuring changes in the overall volatility level, which is governed by a new measure of daily macroeconomic news flow. We also device a test for model specification. States of medium information flow are characterized by elevated disagreement about the future stance of the economy compared to states of weak or strong information flow, such that our findings are explained by disagreement equilibrium-based models. We confirm our findings on international data.