Dividend Announcement And Institutional Holdings In China Finance Essay
Dividend has been considered a signal of information which influences stock price positively. However, since Fama and French (2001) found a dividend disappearing phenomenon, many other studies also show a decline ratio in dividend payout and yield. Evidence shows that firms are reluctant to use dividend as signal because the content of information delivered by dividend is decreasing. In the study of the declining information content of dividend announcements and the effects of institutional holdings, Yakov and kefei (2006) confirmed dividend disappearing phenomenon again and propose an explanation by studying the institutional holdings effects based on the data on American stocks market (NYSE/AMEX stocks). In their study, they present that one of explanation for the decline in information content of dividend announcement is the institutional holdings effects. As the increase of institutional holdings, information that should be delivered by the dividend announcement is already incorporated in the stock price, the dividend signal becomes less informative to the market. Since dividend is a costly signal, firms refrain from use it. As before studies about dividend disappearing are all about American market and Europe market, this paper will study the same phenomenon focused on Chinese market, and examine whether the explanations proposed by before studies can be applied to the stock market in eastern country.
China as a large developing country, the economic characters are different from that of the developed countries such American and UK. For example, those stock return predictor variables in U.S have less predictability to forecast Chinese stock return because of the mispricing is more rampant in China than in the U.S, and the predictability of Chinese stock return is weaker due to a lower price efficiency. (Xuanjuan, 2008). Dar-Hsin (2009) et al. study the signal hypothesis in Chinese stock market, and found the result only partly support the signaling hypothesis and the cash dividend are welcome in both situations of dividend increase and decrease. Chinese stock market is still evolving and whole financial market has not become mature enough so that whether some theories proposed by developed countries can also hold in Chinese market still need to be examined.
The purpose of this study is to study the stock price reaction to dividend announcement in Chinese stock market to see whether the disappearing dividend phenomenon also exist in developing country and whether the explanation of institutional holding is cross-sectional consistent in developing country. In this paper, it will examine the three hypotheses proposed by Yakov to see whether it can explain the disappearing dividend from the perspective of institutional holdings in Chinese market. And if it can, then whether this can be a predicted element for Chinese future market trend
This paper will focus on Chinese stock market, using the data from Thompson One Banker Database. The sample is chose from the Shenzhen stock exchange stands for the mainland of China . The study will start with study the Chinese stock price reaction to dividend announcement from 1999 (after the Asian financial crisis) to 2008 (one year before the futures are introduced to Chinese stock market), using regression models to test whether there is a relationship between the dividend changes and firms cumulated abnormal return. I will examine the dividend response coefficient by using model in Yakov study. We expected that there exist a relationship between Chinese stock reaction and dividend changes, and the coefficient ratio is significant. Then, the study will test the three hypothesis proposed by Yakov (2006) about the institutional holding effects on price reaction to dividend announcement to see whether the explanation is also applicable to Chinese stock market. The three hypothesizes are: the greater the institutional ownership, the smaller the reaction of stock prices to dividend news; institutional investors increase their holdings before an dividend increase announcement which lead to a smaller reaction of stock price to dividend news; with the institutional holdings increase, firms are less likely to raise dividend because of the necessary of delivering information by dividend reduce.
Signal has been a major reason for firms to pay dividend, however, it has been observed that there is a decline in the propensity in firms to pay dividend due to the information content of dividend announcement decrease. The phenomenon of “disappearing dividend” has been proved by Fama and French. Fama and French(2001) studied the MYSE, AMEX and NASDAQ non-financial non-utility firms, and concluded that there is a falling propensity to pay dividends after 1978. During the period 1978-1999, firms paying cash dividends fall from 66.5% to 20.8% partially due to the changing characteristics of publicly traded firms. Evidence from Fama and French shows the dividend payout declines after both a positive earnings announcement and a negative earnings announcement, While Dar-Hsin (2009) et al. study the signal hypothesis in Chinese stock market, and found the result only partly support the signaling hypothesis and the cash dividend are welcome in both situations of dividend increase and decrease.
After Fama and French’s result, the observed decrease in dividends has been confirmed in the Gustavo et al (2002) study. They found that the difference between actual and expected dividend payments tend to become more negative as the firm spends more money on share repurchases. The fact that share repurchase programs have become the preferred method of payout for many firms is consistent with Fama and French’s finding that there is a trend for firms to have a lower propensity to pay dividend. In addition, the market reaction to dividend decreases is not significantly different from zero. Firms that cut their dividends and do not repurchase experience a significantly negative price drop to the announced dividend cut. When investors perceive that dividends are being replaced by repurchases, they view the reduction in dividends as less negative.
Even though there is much evidence showing the dividend is experiencing a declining trend in firms from different periods, the dividend disappearing phenomenon does not exist in all area. Harry (2004) found that dividend paid by industrial firms actually increased over 1987-2000, he explain that “the increase in real dividends paid by firms at the top of the dividend distribution swamps the dividend reduction associated with the loss of many small payers at the bottom”
Although dividend reductions are considered as the last resort for managers, the earnings rebound significantly following a dividend reduction. And unlike the positive reaction to the announcement of a firm restructuring, investor react negatively to dividend reductions due to the long-term implications of the retrenchment activities (Gerald et al. 2010).
The reasons for dividend decline in firms can be summarized for three: the increase in residual claims; the cost of dividend; the declining content of information with dividend announcement. An early explanation for decrease in dividend for firms is the increase of residual claims. Although dividend helps to reduce the agency costs of management and keep the firms’ monitoring at a lower cost in the capital market. Another method to control agency cost is to hold substantial residual claims in the firm. Thus, as the claims increase, dividends would be less valuable to investors and would decrease (Frank, 1984). Aharon and Daniel (1987) use a number of models to explain the stock price reaction to announcements by assuming that private information that held by management is release to the market with these announcements. They result provide evidence of an observed abnormal returns due to the market participants that revise their expectation after announcements. And the analysts update expectation of firm’s future earnings. Because dividend announcement is costly signal to convey information, when dividend becomes less informative, firms are reluctant to use this expensive method to deliver information. Declining in information content of dividend announcement due to the institutional holdings effect explains the dividend disappearing phenomenon (Yakov, 2006).
Eli (2000) found a negative correlation between institutional holdings and the observed post-announcement abnormal returns. And in his study, when institutional holdings is taken as the explanatory variables to the post- announcement abnormal returns, other variables such as transaction costs and firm size loss their incremental explanatory power. Yakov (2006), confirms this result when testing the dividend response coefficient over 1962-2000, he put firm size as a variable that affect the cumulated abnormal return together with the information content of the magnitude of the change in dividend yield, and then found the coefficient of size is insignificant, while the coefficient of dividend response is negative and significant. The result shows that the dividend response coefficient declines over time.
To interpreter the institutional holdings effects, Yakov (2006) proposed three hypothesizes and test them. The first hypothesis is that the greater the institutional ownership, the smaller the reaction of stock prices to dividend news. Secondly Yakov test the impact of increase in institutional holdings on stock price reaction. Finally, he examine the probability of raising dividends for firms with different amount of institutional holdings, assuming that firms with larger institutional holdings are less likely to use dividend as a costly signal to convey information.
Beside the institutional holdings effect, another important factor that affects the stock price reaction to dividend is firm’s size. In the study of size-related anomalies and stock return seasonality by Donald (1983), evidence show that the relation between abnormal returns and size is always negative especially in January. He argued that about half of the average magnitude of the size-effect from 1963-1979 is due to January abnormal returns. Furthermore, Kenneth (2009) suggests that large firms are expected to be engaged in higher volume of activities and their activities are more diverse than small firms, therefore, large firms also have a higher demand for information than small firms. He expects a direct association between the independence of boards and firm size as a means to mitigate the agency problem associated with firm size.
For the Chinese market, several previous study of Chinese stock reaction have focused on investors’ sentiment. Meijin (2004) et al. indicates that sentiment has significant impact on the return and volatility of stock. It has proved that there exists a positive relationship between sentiment and stock return but negative correlation between sentiment and market volatility in China. (Li Xindan, 2008). The reaction of stock price also is affected by other factors, such as the market situation, releasing proportion and firms’ size. In the study of price reaction in Chinese market, Li SF (2008) et al. found that CAR in falling market was larger than in rising market and CAR of small size stocks went up after release announcement.
Fenghua, Y et al. (2004) applied three factor model to the sample from Chinese equity market during the period 1996-2002, and implied that size explains the cross-sectional difference in returns, but contrary to findings for U.S. market. Because Chinese different economic environment, in the Chinese capital market, the book-to-market ratio is useless in explaining the cross-sectional differences in equity returns, but the size-effect continues to be useful.
Another study by Qi bing (2006) et al. shows that, in Chinese capital market, there is a significant negative relationship between institutional investor holdings and stock volatilities with the control of sizes of the listed stocks. They also studied the decrease of stock volatilities for higher and lower institutional investor holding stocks from the period of 1999 -2001. The results shows that the volatilities of high - institutional stocks decrease significantly from the first period to the second period, whereas the volatilities of lower institutional holding stocks show no significant decrease.
Did Chinese firm reduce to pay dividend as a signal to convey information from 1999 to 2008?
Is there a dividend disappearing phenomenon in Chine from 1999 to 2008?
How does Chinese stock price react to dividend announcement during the period from1999 to 2008?
Are there any relationship between dividend changes and stock price and is that consistent?
Whether the three hypotheses by Yakov apply to Chinese stock market?
Dose institutional holdings influence stock price reaction in China?
What other factors influence the results?
By analysis the stock price reaction to dividend announcement in Chinese stock market, the paper is to examine whether there is an obvious the disappearing dividend phenomenon in China, and to study the trend of stock reaction in Chinese stock market in both past and future.
To explain the disappearing dividend and test the explanation proposed by previous study in western countries, whether there are compliant to Chinese market. Whether an increase in institutional holdings leads to firms’ reduction on dividends payout in China.
To compare the influence of institutional holdings on Chinese stock market and the American market in the previous study.
To find out whether the institutional holdings effect may contribute to a trend prediction to stock price fluctuation in the future.
As a sample for this study, I choose Chinese stocks from Shenzhen stock exchange. The period of stock I am going to study starts from 1999 after the Asian financial crisis, to 2008 just before the futures are introduced to Chinese stock market. The data will not include the financial service sector and public sector.
First step is to examine if dividend changes do have an influence on stock price. Whether there is a linear relationship between the dividend changes and the stock reaction.
According to Fama(1998), the mean and variance of the time series of abnormal portfolio returns can be used to test the average monthly response of the prices of event stocks. I use the average quarterly cumulative abnormal returns to measure the stock price reaction. And the changes of dividend are calculated by the quarterly dividend per share of current quarter minus the ratio of previous quarter. I calculate the CAR for both dividend increase and decrease separately and then use the OLS model to test the linear relationship.
Where Y is the CAR, β is the coefficient, X is the dividend changes, and ε is residuals
The second step is to examine the reaction of stock price to dividend announcement over time, I compare the cumulative abnormal returns on the day of dividend announcement (day 0) and the day after dividend announcement (day 1). Then run a regression between the CAR and the years, to see whether there exists a downturn trend for Chinese stock reaction to dividend announcement. The regression follows the model by Yakov, which study the stock reaction to dividend increase and decrease separately.
CAR=a + b*Y
Where CAR is cumulative abnormal returns and Y is from 1999 to 2008. After the regression, I am going to test it whether the coefficient b is significant, and calculate the R2, to see the explanatory power for Y.
Next I will test the dividend response coefficient which measures the information content variation to the changes in dividend yield---DDIVY (Bernheim and Wantz, 1995). The model developed by Yakov (2006) has introduced other variables together with the DDIVY to the CAR, such as firm’s size, long-term dividend yield. The model is as follow
Then, I will test the three hypothesizes proposed by Yakov for the declining stock price reaction to dividend news to see whether it also apply to Chinese stock market. For the first hypothesis, it is important to calculate the annually average institutional holdings over the period. And group the firms with different amount of institutional holdings into three: firms with low institutional holdings, firms with medium institutional holdings and firms with high institutional holdings. To run the regression, I follow the pooled time-series and cross-sectional regression model by Yakov, and use Fama-MacBeth regression for CAR in each year.
Where INST is the percentage of institutional holdings before the dividend increase announcement every quarter,
α2 is what I looking for which is the coefficient of institutional holdings and CAR, if the sign of α2 is positive, it support the hypothesis 1 that the greater the institutional ownership, the smaller reaction of stock prices to dividend news. LTYLD is long term dividend yields.
SIZEN is the size of firms. Kenneth (2009) proved that large firms with higher volume of activities and than smaller firms would have high demand for information than smaller ones. Therefore, firms’ size can explain the cross-sectional difference in returns (Fenghua et al., 2004) and the relation between abnormal returns and size is always negative (Donald, 1983).
AGE is firm’s age, since mature firms are likely to pay higher dividends (Yakov 2006), less surprise are included in a dividend increase. ILLIQ is the illiquidity of the firms. LTVOL is long-term volatility, ROA is return on assets and finally REP is stock repurchases for the year before the dividend increase and announcement. This relationship between REP and CAR was studied by Grinstein and Michaely (2004), but the result was mixed.
To test the effect of institutional holdings on the dividend response coefficient, I use the similar models but the second variable becomes DDIVYj*INSTj .
For the hypothesis 2, I examine whether the Chinese institutional investors with private information will buy stocks before a good new is announced which lead to a reduction in stock price reaction because the information have been incorporated in the stock price. Here I followed the event study type analysis by Yakov (2006). Yakov measure the excess changes in institutional holdings by using the quarterly change in INSTj subtract the average quarterly change in INST for all firms:
E∆INSTj = ∆INSTj- ∆INSTm
The period covers the quarter that announce the dividend and two quarter before and one quarter after the announcement quarter. Through comparing the means of E∆INSTj in different quarter, and their significance, it can figure out whether the hypothesis stands in Chinese market.
The last hypothesis need to be test by examing how does the institutional holdings affect the probability of dividend increase. The model I use is follow the ordered probit model by Christensen and Prabhala (1995), but made some changes according to Chinese economic environment.
Lj= Θ0 +Θ1 INSTj+Θ2 LTYLDj +Θ3 DIFYLDj +Θ4 SIZENj +Θ5 PRCj +Θ6 LTVOLj+ ej
where DIFYLDj is the difference between the short and the long term dividend yield, STYLD is the most recent quarterly dividend by the stock price at the end of the month prior to the dividend announcement, multiplied by four (annualized). PRC is the stock price at the end of the quarter before the dividend announcement. If the Θ1 is negative, it indicates that higher institutional holdings reduce the likelihood of dividend increase and support the third hypothesis.
Submit proposal to supervisor by 3 March 2011
Get feedback on proposal and discuss with supervisor by 12 June 2011
Submit comprehensive reading list with abstract by 26 June 2011
Finish literature review by 24 July 2011
Complete data collection by 24 July 2011
Complete data analysis by 17 August 2011
Complete empirical analysis by 20 August 2011
Complete findings editing by 24 August 2011
Complete result written by 28 August 2011
Complete the final version editing by 30 August 2011
Print the paper by 31 August 2011
Submit dissertation by 1 September 2011
The Thompson One Banker Database is the main resource for the date collection and analysis parts. The Web of Science and other databases provided by University of Leeds on the library website as well as the Google Scholar help to get access to previous literature review.
Need help with your literature review?
Our qualified researchers are here to help. Click on the button below to find out more:
In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: