Board attributes and herding in corporate investment: evidence from Chinese-listed firms

Abstract

We examine whether board attributes, including board age, gender diversity, board independence, CEO duality, and board size, explain investment herding by using a panel of 1155 Chinese-listed non-financial firms during 1999–2004. Investment herding is measured by the absolute value of the difference between the investment ratio of firm i in year t and the average investment ratio of other firms in the same industry excluding firm i in year (t−1). We find that corporate boards that have more young directors, more female directors, more independent directors, a CEO who is not the chairman of the board, and a larger board are more likely to make investment decisions closer to their peers in the same industry. We also provide evidence that investment herding is positively related to firm performance, suggesting that investment herding does not necessarily hurt shareholders in the Chinese context. We identify that herding in making investment decisions is a possible channel through which some board attributes, such as board age diversity, gender diversity, and board independence, contribute to firm performance.

Keywords:

• board attributes
• herding
• fixed investment
• China

JEL Classification:

• D21
• G31
• G34
• P34

Acknowledgements

We thank seminar participants at SOAS, University of London; Birmingham Business School; Symposium organised by Chinese Capital Market Innovative Research Team at UIBE, Beijing; and participants at the 2nd EJF special issue conference on China's capital market at Durham Business School, particularly our discussant Alessandra Guariglia for constructive comments. Hong Bo acknowledges financial support from the British Academy small research grants. Yanmei Sun acknowledges financial support from NSFC Project no. 71202026. Anonymous referees are acknowledged for comments and suggestions on earlier versions of the paper.

Notes

1. The information on political connections of individual directors is not available in our data set, which restricts us from conducting direct tests of the impact of political connections on investment herding. However, in the empirical analyses of this paper, we control for the state involvement in the firm by adding state ownership in the estimation.

2. We thank Joseph P.H. Fan for bringing this theory to our attention.

3. The information on multiple directorships held by individual directors is not reported by the Chinese official data sources. Such information requires a manual collection and match of multiple directorships for each individual director. Therefore, in the current paper we cannot directly test the impact of multiple directorships on herding. However, we acknowledge that multiple directorships can explain why independent directors are more likely to follow the herd.

4. In this paper, extreme investment decisions refer to the decisions that are deviating from the herd and hence results in large volatility in the firm's performance. Adams, Almeida, and Ferreira (2005) document a positive association between CEO power and the variability in firm performance, arguing that a powerful CEO often makes decisions with extreme consequences resulting in large volatility in the firm's performance.

5. More specifically, we split the firms in the manufacturing industry into 10 subgroups according to its second tier in the official classifications by the CSRC. Under the manufacturing industry, these 10 subcategories are: the food and beverage, textiles and apparel, timber and furnishings, paper and printing, petrochemicals, electronics, metals and non-metals, machinery, pharmaceuticals, and other manufacturing industries.

6. We thank two referees for pointing out the relevance of financing variables in the investment herding equation.

7. We also tried to use a dummy variable of state control in the same estimation, which produces an insignificant result for state dummy in the investment herding equation.