|
|
|
| Fund Common Ownership Network and Stock Price Comovement |
| ZHOU Yinggang, TANG Chengwei, XU Xingbai
|
| School of Economics/Wang Yanan Institute for Studies in Economics/ Paula and Gregory Chow Institute for Studies in Economics, Xiamen University |
|
|
|
Abstract In recent years, public offering of funds in China has developed rapidly and plays a crucial role in the financial market. As important institutional investors, the behavior of funds can have a significant impact on the correlation between stocks, which is also a key factor in stock pricing and risk. With the continuous expansion of assets under management, the investment behavior of funds may further affect stock prices by influencing the correlation between stocks. A deep understanding and analysis of the relationship between funds and stock price comovement is crucial for the stable development of China's stock market and the safety of the financial system.
Antón and Polk (2014) find that the common ownership of two stocks by US mutual funds significantly increases their future price correlation, enhancing price comovement. This conclusion has also been verified in China. In fact, a network emerges among stocks that are commonly held by funds, which is the fund common ownership network. In this network, where stocks serve as nodes, stocks directly commonly held by funds exhibit comovement. This is akin to the correlation between “neighbors” in the network, representing a first-order network effect, and has been the focus of existing literature. Moreover, different funds often share information on the same large-position stocks, potentially exhibiting a herding effect in asset allocation, and even facing similar fund flow shocks. This leads to correlations in their trading behaviors. Consequently, stocks that are not directly commonly held by funds may also experience indirect price comovement under the influence of the fund common ownership network, representing a higher-order network effect or the influence from multi-layered “neighbors of neighbors”. Existing literature primarily focuses on the correlations between two stocks directly held by funds, neglecting the price comovement within a higher-order network.
Using open-ended active equity funds and their holdings of Shanghai and Shenzhen A-share stocks from 2007 to 2022 as samples, we construct a fund common ownership network with stocks as nodes, and employ a spatial autoregressive model to study the price comovement and risk contagion under this network. We make four main contributions. First, we establish a fund common ownership network based on common shareholding behaviors of funds, enriching the literature related to stock networks. Second, existing literature measures stock price comovement from a correlation perspective, focusing on the similarity between two time series. This approach not only fails to effectively quantify the amplitudes of cross-sectional comovement but also overlooks possible indirect links among a wider range of stocks, which can be solved by a spatial autoregressive model. Third, we analyze how the fund common ownership network influences stock price comovement from three perspectives: fund flows, information sharing, and investor sentiment. Fourth, we show that the fund common ownership network may serve as a contagion channel under extreme market scenarios, through which the suspended stocks can impact unsuspended stocks, thereby enriching the literature on stock market risk contagion.
The research findings indicate that stocks exhibit significant price comovement under the fund common ownership network, supported by a series of robustness tests. We discover that stocks exhibit stronger price comovement effects under networks with higher fund flows and higher information sharing. Furthermore, the higher the investor sentiment, the greater the intensity of stock price comovement within the fund common ownership network. By applying a spatial autoregressive Tobit model to conduct cross-sectional regressions on daily returns during the significant fluctuation in the stock market in 2015, we find that not only do unsuspended stocks exhibit significant price comovement within the fund common ownership network, but large-scale suspended stocks can also impact the prices of unsuspended stocks through this network. Our research points out that the fund common ownership network not only affects stock price comovement but also potentially serves as a channel for risk contagion in extreme events. The conclusions offer insights into the relationship between fund common ownership and stock price comovement, and have implications for preventing risk contagion in the capital market.
|
|
Received: 23 October 2023
Published: 02 December 2025
|
|
|
|
|
| No related articles found! |
|
|
|
|
