China International Conference in Finance�Shenzhen, 2025/07

More Stringent, More Efficient?

Short-Selling Activities and Anomaly Mispricing

​

Chuyu Wang and Junye Li

​

Discussant：Zhuo Chen

1

Motivation and Overview

• The paper examines the impact of short-selling activities on mispricing correction for anomaly portfolios in China’s stock market.
• Key findings:
• Short-selling activities show limited effects on mispricing correction for 70 representative anomalies in six categories, with marginal improvements observed only for stocks with intensive short selling activities.
• Informed trading coupled with a high short-selling ratio shows potential in correcting mispricing, while institutional investors seem to aid in detecting mispricing in trading-related anomalies.
• Exploiting two exogeneous regulatory shocks on short selling, one positive and one negative, anomaly portfolios exhibit different return patterns within marginable (available for short selling) stocks.
• Contribution:
• The study provides new insights into the price discovery role of China’s short-selling policy, demonstrating that its potential is limited by excessive regulatory constraints and underdeveloped market participation, both of which impede price efficiency.
• This paper complements the findings of Chu et al. (2020, JF), who show that relaxing short-sale constraints in the U.S. weakens anomaly performance by alleviating mispricing in the short leg.

2

Comment 1: Long-leg or Short-leg?

• Short Selling Constraints: Short-selling constraints of those non-designated stocks should theoretically prevent arbitrageurs from trading overpriced stocks, yet the findings in Panels B and C of Table 3 suggest that the effect is not concentrated on the short leg of anomalies.
• Effect on Short Leg: Panels B and C of Table 5 indicate a possible stronger effect of short-selling constraints on the short leg of anomalies with abnormal excess returns, but more rigorous tests are needed to confirm this.
• Note the definition of “short-selling ratio” as the fraction of shares already borrowed and sold may differ from the intended economic interpretation, i.e., shares available for short selling.
• It would be interesting to explore the patterns of long-leg and short-leg anomalies within designated and non-designated stocks.
• Is mispricing persistent in the long or short leg for the two groups of stocks?
• Are there differences in turnover ratios for anomalies driven by fundamental versus technical factors?
• The Role of Margin Trading: The methodology classifies designated stocks (eligible for both margin trading and short selling) into the treatment group, raising the question of whether the correction of mispricing is influenced by the relaxation of margin trading. Thus investigation long leg and short leg separately may also help.

​

​

3

Comment 2: Alternative Classification of Treatment and Control Stocks

• Propensity score matching (PSM) has limitations such as imbalance, inefficiency, and bias, which should be used with caution.
• Alternative Approaches:
• Exact matching on single but important characteristics such as size, liquidity, or volatility, according to CSRC’s rules on stock marginability.
• Sorting stocks into groups based on characteristics and then forming anomaly portfolios within each group using designated and non-designated stocks.
• The matching characteristics are trading-based rather than accounting-based (e.g., book-to-market ratio or profitability measures), which may not fully capture the differences between marginable and non-marginable stocks.
• Recent Policy Shock: Exploiting the exogenous policy shock of the suspension of Securities Lending Transactions under Refinancing Business in July 2024 could provide more recent evidence and establish a causal relationship between short selling and stock price efficiency.

​

4

Comment 3: Additional comments

• Short-Selling Ratio: The short-selling ratio, defined as the aggregate shares short sold scaled by total A-share outstanding within a month, is a crucial variable. Summary statistics on its cross-sectional and time-series properties would enhance understanding.
• Futures Basis: Short selling becomes more difficult when the futures basis is smaller (more negative), indicating higher hedging cost for investors. Splitting the sample into subsamples based on futures basis could provide additional insights.
• OTC Short Selling: The rise of OTC short selling in China, where shares available for short selling come from restricted shares held by major shareholders, could be an area for further exploration.
• Earnings Management: Reconciling findings with He et al. (2024) on earnings management and price informativeness could shed light on whether short sellers or margin buyers detect firms’ earnings management.
• Analyst Downgrades: The association between short selling activities and analyst downgrades may be less informative in China compared to the U.S., as downgrades are rare, and even upgrades rarely cause significant stock price movements.
• Mutual Funds: The use of mutual funds in the analysis is not necessary, as they are not allowed to short sell, making the underlying driving force of the empirical pattern less clear.

​

​

5