Stock price anomalies and corporate dividend policies in China

Abstract

The dissertation consists of two stand-alone but interrelated research projects. Specifically, the two studies try to improve our understanding of the anomalous behavior of stock prices and the corporate policy decisions in China. The first study investigates what drives the price disparity between Chinese "twin" shares (A shares traded largely by domestic investors; B- and H- shares traded mainly by foreign investors). Extending the variance decomposition framework of Vuolteenaho (2002), we decompose the unexpected price disparity into two terms: the difference in expected return news and the difference in cash flow news. The results show that the difference in expected return news overwhelmingly dominates difference in cash flow news in driving the variation of the price disparity. This suggests that to a large extent, market or macro news, rather than firms' specific news, moves the price disparity of the twin shares. The reason is that investors in the two segmented markets react to cash flow news similarly, but react to expected return news quite differently. Moreover, consistent with the literature, the returns of A-shares show a much higher variance in expected return news than cash flow news, whereas cash flow news component is the more important driving force of the returns of B-shares. In other words, the foreign shares of Chinese firms behave more like the shares in the US (cash flow news dominates), while domestic shares are more alike those in other emerging markets. The second study examines whether the evidence in prior studies regarding contracting explanations for dividend policies in U.S., Australia and other western countries, is also applicable to a country with weak legal institutional arrangements (e.g corporate governance and different legal and regulatory environment). Using pooled cross-sectional observations of pure A-companies listed on Shanghai and Shenzhen stock exchanges from 1996 and 2006, I examine the relationship between state ownership, growth opportunities and corporate dividend policies decisions. Low growth firms are found to be negatively associated with dividend policies. The results are even stronger for firms with lower state ownership concentration. Firms with high state ownership tend to pay lower dividends suggesting that these firms have more severe agency problems. Overall, the study provides a new perspective of dividend policy in China. Although the governance and the ownership structure in the corporate sector of China differs largely from those in developed countries, the results suggest that to some extent that contracting costs explanations for dividend policies can apply in China.