Research¶
📄 Publications¶
Bubble-crash experience and investment styles of mutual fund managers. Journal of Corporate Finance, 2022, 76, 102262 (with Deming Luo and Yanjian Zhu).
Chinese mutual fund managers who experienced a stock market crash are more value-oriented in their portfolios.
Economic policy uncertainty and mutual fund risk shifting. Pacific-Basin Finance Journal, 2023, 77, 101921 (with Deming Luo and Sainan Jiang).
Mutual fund managers increase risk-shifting activities when uncertainty rises.
Belief dispersion in the Chinese stock market and fund flows. Journal of Banking and Finance, 2024, 166, 107252. (with Yue Fang and Deming Luo).
We construct a text-based degree of disagreement (DOD) about stock market performance among fund managers using deep learning models. In the time series, the DOD negatively predicts market returns. Cross-sectional results show that investors correctly perceive the DOD as an overpricing signal and discount fund performance accordingly. Flow-performance sensitivity is diminished during high dispersion periods.
Visible hands versus invisible hands: Default risk and stock price crashes in China. Pacific-Basin Finance Journal, 2025, forthcoming, 102715 (with Huaigang Long, Cuixia Tao, and Yanjian Zhu).
We revisit the default-crash risk relation in the context of China. We find that firms with higher default risk have lower stock price crash risk both in monthly and yearly frequencies. The negative relation arises from the active involvement of the government before 2014 and creditors after 2014 in corporate governance.
📝 Working Papers¶
In circulation¶
“Extrapolative expectations and asset returns: Evidence from Chinese mutual funds”, 2025.
We study mutual funds' stock market expectation formation and its implications for asset returns using a novel text-based measure of beliefs inferred from Chinese fund reports. Funds extrapolate from recent stock market and fund returns when reporting expectations about future market performance, with more recent returns receiving greater weight. This extrapolative belief is not systematically biased but overextrapolation is: consensus expectations positively predict future market returns, whereas the degree of extrapolation negatively predicts future returns. Moreover, we find a positive time-series relation between expectations and performance for a given fund unconditionally. However, in the cross-section, this relation holds only when funds align their portfolio adjustments with their expectations and when those expectations are accurate.