Abstract
Purpose
This study examines the effect of information uncertainty (IU) on stock price crash risk. Although ambiguity aversion theory predicts that investors overweight bad news under uncertainty, thereby producing asymmetric market responses, existing empirical evidence has largely focused on asset returns, analyst behavior and small-scale experimental settings. By mapping ambiguity-aversion predictions directly onto crash risk, this study provides large-sample, market-level validation of these mechanisms using real trading data.
Design/methodology/approach
Using data on China's A-share listed firms from 2009 to 2019, we estimate market-adjusted firm-specific returns with a market index model and measure crash risk using negative conditional skewness (NCSKEW) and down-to-up volatility (DUVOL). IU is captured in two ways: firm listing age as a parsimonious proxy and a composite index constructed from analyst coverage, analyst forecast dispersion, idiosyncratic volatility, share turnover and firm size. Robustness is established through fixed-effects regressions, GMM estimation and propensity score matching.
Findings
Higher information uncertainty significantly increases stock price crash risk, and the results are robust across alternative specifications, estimation methods and IU measures. Mediation analysis rules out two competing channels, namely managerial bad-news hoarding and investor disagreement, supporting the argument that ambiguity itself elevates crash risk through investors' asymmetric processing of information.
Originality/value
This study provides large-sample empirical evidence that ambiguity aversion mechanisms translate directly into stock price tail risk. By demonstrating that IU raises crash risk through asymmetric investor responses rather than through information concealment or opinion divergence, we bridge the theoretical and experimental ambiguity-aversion literature with the empirical crash risk literature.