At present, there are 16 listed banks in China including which listed on Shenzhen stock exchange and Shanghai stock exchange. Table 1 shows the basic information of the 16 listed banks.

Table 1 show that most of the headquarters of China’s listed banks are located in the economically developed provinces and regions, 94% are concentrated in Beijing, Shanghai, Guangdong, Jiangsu and Zhejiang area, what is more, there are 56% located in Beijing. Besides, there are only 16 listed banks in Shenzhen stock exchange and Shanghai stock exchange, it is quite a small number compared to the thousands of listed companies in China.

1) A small proportion of listed banks do not pay dividends

According to the dividend payment rate data of China’s listed banks from RESSET database, we can calculate the proportion of listed banks which do not pay dividend to shareholders from 2010 to 2013, and the results are shown in Table 2 as follows:

From Table 2, we can learn that the quantity and proportion of listed banks which do not pay dividend is quite small from 2010 to 2013, most of the listed banks would pay dividend to shareholders, the proportion of listed banks which pay dividend even reached to 100% in 2012 and 2013.

2) The dividend payment rate of listed banks is low

Dividend payment rate is different from different countries and different industries, in China, the listed company’s average dividend payment rate is about 30%, and in foreign countries, the average dividend payment rate is around 40%. We calculated the average dividend payment rate of China’s listed banks from 2010 to 2013 based on the dividend payment rate data from RESSET database, and the results are shown in Table 3.

From Table 3, we can see that the average dividend payment rate of China’s listed Banks have a slowly rising trend year by year from 2010 to 2013, but overall the dividend payment rate is low, maintain at the levels of 23% to 27%, lower than the average dividend payment rate of listed companies in China and the average dividend payment rate in foreign countries.

3) State-controlled banks generally have a higher dividend payment rate

Among 16 listed banks in China, there are five state-controlled banks, including Industrial and Commercial Bank of China, Agricultural Bank of China, Bank of China, China Construction Bank and Bank of Communications. We can calculate the average dividend payment rate between 2010 and 2013 for each bank, and the results are shown in Table 4 as follows.

From Table 4, we can see that the top four of average dividend payment rate are all state-controlled banks, their dividend payment rate are all higher than 30%, at a high level of dividend payment rate, while the other listed banks have a relatively low dividend payment rate, Bank of Beijing, Industrial Bank and PingAn Bank are all less than 20%, especially the PingAn Bank, its average dividend payment rate is only 5.08%.

Because the Agricultural Bank of China and the China Everbright Bank was listed in 2010, in order to ensure the integrity of the data, this article selects the data from 2010 to 2013 as samples. The relevant data are gained from the RESSET database data and the annual report of listed banks, and the tool using for data processing and empirical analysis is Eviews 6.0.

There are many factors that can influence a listed bank’s dividend policy, and this article tries to find out the effect of the profitability, the growth ability, the operation capacity, the debt levels and the liquidity of assets by using empirical method. First of all, we need to put forward the following hypothesis:

Hypothesis 1: The stronger the profitability, the higher the dividend payment rate.

Dividend is a way for a company to distribute its surplus, to pay back for its shareholders. Therefore, the profitability of a company is one of the important factors affecting its dividend policy. In general, a company has strong profitability can obtain long-term stability of surplus, and is more likely to have a higher dividend payment rate. On the other hand, a company has weak or unstable profitability is more likely to have a lower dividend payment rate. The indicators reflecting a company’s profitability including return on total assets, return on net assets and earnings per share etc. This article selects earnings per share as an indicator for the profitability of listed banks in China.

Hypothesis 2: The stronger the growth ability, the lower the dividend payment rate.

A company has strong growth ability often need more money for its development, so it tends to remain its surplus within the company instead of distributing to shareholders. Therefore, a company has strong growth ability is more likely to have a lower dividend payment rate. This article selects earnings per share growth rate to represent the growth ability of listed banks.

Hypothesis 3: The stronger the operation capacity, the higher the dividend payment rate.

Operation capacity represents a company’s ability to use money, represents the turnover efficiency of company funds. A company has strong operation capacity can use less money to meet its funding need, therefore it can distribute more surplus to shareholders, that is to say, it often has a higher dividend payment rate. This article selects the total assets turnover ratio as the indicator for listed bank’s operation capacity.

Hypothesis 4: The higher the debt levels, the lower the dividend payment rate.

When a company has high debt levels, it often chooses not to pay dividends or pay fewer dividends, because most of its surplus will be used to make a repayment to creditors. So, the higher debt level a company is, the smaller the probability it will pay dividend to shareholders. This article selects the asset-liability ratio as the indicator for listed bank’s debt levels.

Hypothesis 5: The stronger the liquidity of assets, the higher the dividend payment rate.

There is a positive correlation between a company’s profitability and liquidity of assets, a company has strong liquidity of assets tends to bring high profit, thus the ability to pay dividends is guaranteed, the dividend payment rate is higher. This article selects operating activities cash flow per share as the indicator for listed bank’s liquidity of assets.

Based on the above hypothesis and theoretical analysis, this article will analyze the factors affecting the dividend policy of listed banks in China by using multivariate linear regression model. In this model, the explained variable is dividend payment rate, and the explanatory variables are earnings per share, earnings per share growth rate, total assets turnover ratio, asset-liability ratio and operating activities cash flow per share. The specific information of these variables is shown in Table 5 as follows:

Based on the above definition of variables, we can build the multivariate linear regression model as follows:

In this model, is constant term, is regression coefficients of five Explanatory variables, and is random error.

After building the model, we should test the model. Because the panel data used in this article is “wide and short” type, there is no need to carry out the unit root test, but we still need to carry out the F test to find out whether we should use a mixed effect model or a fixed effect model, the F test results by using Eviews 6.0 are shown in Table 6 as follows:

From Table 6, we know that the P value of F test is less than 0.05, based on the principle of F test, if the P value is less than 0.05, the fixed effect model is appropriate, so the multivariate linear regression model should

¹“+” represents a positive relationship between the variable and the dividend payment rate, and “−”represents a negative relationship between the variable and the dividend payment rate.

be a fixed effect model.

Next we carry out the Hausman test to find out whether we should use a fixed effects model or a random effects model, the Hausman test results by using Eviews 6.0 are shown in Table 7 as follows:

From Table 7, we know that the P value of Hausman test is less than 0.05, based on the principle of Hausman test, if the P value is less than 0.05, the fixed effect model is appropriate, so the multivariate linear regression model should be a fixed effect model.

According to the results of F test and Hausman test above, this article will choose a fixed effects model.

Based on the above model and the test results, we can carry out the multiple linear regression analysis with panel data of 16 listed banks from 2010 to 2013 by using Eviews 6.0, and the results are shown in Table 8 as follows.

As it is shown in Table 8, the R2 of the regression analysis is 0.871056, and the adjusted R2 is 0.811082, so the goodness-of-fit of this model can basically accords with the requirement. Besides, the DW value is 2.226460, which is close to 2, so there is no autocorrelation. In addition according to the P values of coefficients, we know that the coefficient of X2 and X3 is not significant, and they are symbolic contrary to expectations, the another three coefficients of variables are significant under the confidence level of 95% and their symbol are the same as expected.