New model predicts stock crashes and jackpots in China’s volatile market
image: This figure shows changes in institutional ownership (IO) for decile portfolios (G01-G10) sorted by values based on the ex ante probability of price crashes (CRASHP) and ex ante probability of price jackpots (JACKP) in the liquid group. At the end of each quarter t, decile portfolios are constructed by grouping high-liquid stocks based on CRASHP or JACKP. G01 represents the portfolio at the bottom 10% of CRASHP (JACKP), and G10 is the portfolio at the top 10% of CRASHP (JACKP). In part (a), we plot the time-series and cross-sectional average of change in IO between the end of quarter t and the end of quarter t-6 for each decile. Parts (b) and (c) display changes in IO around the entry into stocks of the top 10% of values based on CRASHP and JACKP, respectively. For the stocks that enter into C10 (or J10) at the end of quarter t, we present the average number of IO of the stocks in excess of the mean of each measure for all stocks in the same quarter, for six-quarters before and after the entry into C10 (or J10). The sample is from the fourth quarter of 2005 to the third quarter of 2021
Credit: Yi Fang (Jilin University, China) Hui Niu (Nanjing University of Finance and Economics, China)
Background and Motivation
As a key player in global financial markets, the Chinese stock market exhibits unique characteristics driven by high retail participation and regulatory constraints. China Finance Review International (CFRI) brings you an article titled “Crash and Jackpot Probability Anomalies in the Chinese Stock Market”, which investigates the predictability of extreme price movements—both crashes and jackpots—and their implications for asset pricing in China.
Methodology and Scope
The authors develop an enhanced trinomial logit model that incorporates the price-to-sales ratio (PS) and distinguishes between up-market and down-market states. Using data from all A-shares between June 2000 and December 2021, the model predicts the probability of a stock crashing (falling by 50% in six months) or hitting a jackpot (rising by 50% in six months). The study also examines the role of liquidity, institutional ownership, and arbitrage constraints in explaining these anomalies.
Key Findings and Contributions
- The new model significantly improves the distinction between crash probability (CRASHP) and jackpot probability (JACKP), reducing their correlation from 0.26 to –0.04.
- High-liquidity stocks exhibit stronger mispricing effects—contrary to findings in developed markets—due to noise trading and limited arbitrage.
- Institutional investors tend to chase liquid stocks, amplifying price deviations rather than correcting them.
- Portfolios sorted by CRASHP (JACKP) yield significantly negative (positive) risk-adjusted returns, even after controlling for common risk factors.
Why It Matters
This research challenges conventional asset pricing theories by revealing that extreme return anomalies in China are not driven by sentiment or macroeconomic states, but by structural limits to arbitrage and speculative institutional behaviour. The findings are critical for understanding market efficiency in emerging economies and for designing robust investment and regulatory strategies.
Practical Applications
- Investors can use the enhanced logit model to identify overvalued or undervalued stocks based on crash and jackpot probabilities.
- Regulators should focus on improving arbitrage mechanisms and monitoring institutional trading in high-liquidity stocks to mitigate systemic mispricing.
- Academic researchers can build on this model to explore extreme-return predictability in other emerging markets with similar institutional features.
Discover high-quality academic insights in finance from this article published in China Finance Review International. Click the DOI below to read the full-text original! Open access for a limited time!