[Paper Review] Cross-shareholding networks and stock price synchronicity: Evidence from China

Fenghua Wen, Yujie Yuan+1 more|arXiv (Cornell University)|Mar 5, 2019

Financial Markets and Investment Strategies86 references42 citations

TL;DR

This paper examines how cross-shareholding networks in China affect stock price synchronicity, a measure of price informativeness. Using network centrality to gauge cross-shareholding intensity, it finds that higher centrality reduces noise trading and enhances price informativeness, especially during market downturns and among large firms.

ABSTRACT

This paper investigates the effect of cross-shareholding on stock price synchronicity, as a measure of price informativeness, of the listed firms in the Chinese stock market. We gauge firms' levels of cross-shareholdings in terms of centrality in the cross-shareholding network. It is confirmed that it is through a noise-reducing process that cross-shareholding promotes price synchronicity and reduces price delay. More importantly, this effect on price informativeness is pronounced for large firms and in the periods of market downturns. Overall, our analyses provide insights into the relation between the ownership structure and price informativeness.

Motivation & Objective

• To investigate how cross-shareholding networks influence stock price synchronicity in the Chinese stock market.
• To assess whether cross-shareholding improves price informativeness through noise reduction rather than firm-specific information production.
• To examine the role of market conditions, particularly market downturns, in amplifying the effects of cross-shareholding on price efficiency.
• To explore whether the impact of cross-shareholding varies by firm size, especially given differences in information environment quality.
• To test whether investor inattention and attention dynamics mediate the relationship between network centrality and price synchronicity.

Proposed method

• Constructed a cross-shareholding network using listed firm ownership data, treating firms as nodes and mutual shareholdings as edges.
• Calculated network centrality (e.g., degree, betweenness) as a proxy for a firm’s position and level of cross-shareholding.
• Used the R² from a market-model regression (with market and industry returns) as a measure of stock price synchronicity.
• Applied multivariate regression models to isolate the effect of centrality on synchronicity, controlling for firm size, ownership concentration, and other controls.
• Conducted sub-sample analyses by market regime (upturn vs. downturn) and firm size to test conditional effects.
• Used alternative measures of synchronicity and robustness checks to validate results, including firm-year fixed effects and alternative network metrics.

Experimental results

Research questions

• RQ1 Does a firm’s centrality in the cross-shareholding network predict higher stock price synchronicity in China’s market?
• RQ2 Is the effect of cross-shareholding on synchronicity driven by noise reduction or by enhanced firm-specific information incorporation?
• RQ3 How does the impact of cross-shareholding on price informativeness vary across market conditions, particularly during downturns?
• RQ4 Are the effects of cross-shareholding stronger for large firms compared to small firms, and why?
• RQ5 To what extent does investor attention modulate the relationship between cross-shareholding and price discovery efficiency?

Key findings

• Firms with higher centrality in the cross-shareholding network exhibit significantly higher stock price synchronicity, indicating improved price informativeness.
• The positive effect of centrality on synchronicity is driven by a noise-reducing mechanism, not by increased firm-specific information production.
• Firms at more central positions in the network experience lower stock price delay, suggesting faster incorporation of information into prices.
• The effect of cross-shareholding on synchronicity is significantly stronger during market downturns, when investor attention to market-wide information increases.
• Large firms benefit more from cross-shareholding in terms of price informativeness, likely due to more stable cash flows and higher information quality demands.
• The results remain robust after controlling for short-selling constraints and alternative measures of synchronicity, supporting the validity of the noise-reduction channel.