论文部分内容阅读
Abstract：The thesis is to investigate the relationship between state ownership and disclosure quality and whether corporate governance plays its role in moderation of the relationship. Focusing on Hong Kong Exchanges and Shanghai Stock Exchange， the research uses two different data sources to test the same hypothesis and draw different conclusions. The sample is selected from the existing companies by referring to complete and precise information from the year of 2014 to 2016. The total numbers of sample are 68 and 64 companies listed in Hong Kong Exchanges and Shanghai Stock Exchange， respectively. In Hong Kong Stock Exchange， the findings revealed a significantly positive relationship between state ownership and disclosure quality， for hypotheses 3， corporate governance can weaken the relationship between state ownership and disclosure quality. By comparison， in Shanghai Stock Exchange， there is a negative significant relationship between them and corporate governance can strength the relationship between state ownership and disclosure quality. For hypotheses 2， the result showed that the listed companies in Hong Kong have a better disclosure quality.
Keywords： Disclosure quality；Corporate governance；Stock Exchange；Index
1.Introduction
Theinitial research problem was identified between the Hong Kong and Shanghai markets. In 1997， the People's Republic of China officially took over the sovereignty of Hong Kong and governed the region. The mainland continues to adopt socialism while capitalism remained in Hong Kong after 1997. With the tight historical connection with the UK， Hong Kong adopted the corporate governance system that was very similar to the British corporate governance.
Thedisclosure mechanism of the information in mainland China became worse compared to Hong Kong. In the mainland， the annual and financial reports could only be registered and reviewed. In Hong Kong， all the accounting and business details in the listed companies have to go through the supervision of Hong Kong’s Stock Exchange.
Based on theearlier literature， the state-owned enterprises and their owners generally face five corporate governance problems （Crnkovi， Po， 2014）. They are low levels of disclosure quality， poor relationship among the shareholders， fuzzy ownership objectives and poor ownership function.
There are 3 research questions：
·Is there any relationship between the state ownership and disclosure quality？ ·Is information disclosure of Hong Kong Stock Exchange better than that of Shanghai Stock Exchange？
·Does corporate governance moderate the relationship between state ownership and disclosure quality？
2.Literature Review
2.1 Corporate Governance in Chinese SOEs
Unlike in the private sector， whose main goal is to optimize profits， the Chinese state-owned enterprises have a social responsibility as an additional objective. Although they expect to make a profit， they also promote the country to maintain the national economic stability and sustainable development （Wang， 2012）. The key difference of the corporate governance between the state-owned enterprises and the private sector is their proximity to the state power. The SOEs are in a unique position of the same entity， ultimately responsible for the legal and regulatory environments， and directly influenced by the politicians rather than the professional businessmen （Terence， 2014）.
2.2State-owned Enterprises Listed on the Hong Kong Stock Exchange
Some Chinese scholars have pointed out the existing problems in the corporate governance of the state-owned enterprises listed on the Hong Kong Stock Exchange （Cao， 2003）.
·Equity structure is unreasonable. When the state-owned capital operation is in a dominant position and once there is a lack of checks and balances of the other ownership capital， many problems would exist.
·The problem of internal control. When the controlling shareholder is the state， the internal control is often accompanied with the government. The key people and the control focus their power on the major issues.
·The organizational structure of the company. There are many problems such as the independence of the board of directors and the limited role of the board of supervisors in China’s listed companies.
2.3State-owned Enterprises Listed on the Shanghai Stock Exchange
The Chinese mainland stock market has some notable features （Carpenter， Lu & Whitelaw， 2015）. First， it is dominated by the retail investors， accounting for more than 80% of the transactions， which mean that China’s asset management industry is still underdeveloped， resulting in several market efficiency issues. Second， it is a pure order-driven rather than a quote-driven market， while the United States and several other countries have a mixed stock market system. Third， it is a concentrated market， while the U.S. market is fragmented， which may have a vital impact on the market information. 3.Theoretical Framework
This study has employed the agency theory framework. Jensen and Meckling （1976） used the agency theory as a tool to explain the inherent difficulties in the governance structure of the state-owned enterprises in Italy. The agency theory is particularly useful as it contains the factors that may affect the risk disclosure of the companies. It also can be illustrated by agency problem type I and type II. Type I is the agency problem agency problem that arises between the principal as the owner of companies and agents as the manager who is the executor the company’s operations. While the issue of agency Type II is the agency problem that occurs between controlling shareholders and citizens.
4.Research Method
4.1 Sample of study
This research compares the different situations in Hong Kong and Shanghai Stock Exchange. The sample is selected based on the existence of the companies and complete information from 2014 to 2016. There are 91 state-owned enterprises listed in Hong Kong Stock Exchange while there are 100 state-owned enterprises in Shanghai （Table 4.1）.
4.3 Research Design
The analysis methodthat will be used in this study is the cross-sectional data analysis. The disclosure index is measured by the formula created by Haniffa and Cooke （2002）， while the relationship between the state ownership and disclosure quality can be measured by the equation， which was created by Omar Juhmani （2013）， and the relationship between corporate governance and disclosure quality is identified using the equation created by Sheila， Ridzwana and Salman （2013）.
4.4 Operationalization of Variables
The dependent variable is the disclosure quality. This is measured by annual report and calculating the disclosure index. The SOEs are defined as those companies that the government owns 20% or above of the shares. The independent variable is the state ownership， where the research selects the state-owned enterprises that are listed on Shanghai and Hong Kong Stock Exchange. he moderating variable is the corporate governance.
The five variables that are under the corporate governance can be measured in five ways. （Sheila， Mohd Ridzwana & Salman，2013）
Institutional investors’ activism： The research calculates the percentage of the capital shares held by institutional investors
Duality leadership structure： This can be measured by checking the type of leadership structure. Board size： This is identified by calculating the total number of the board of directors.
Independent directors： This can be taken by accounting the number of independent directors/total number of board of directors.
Audit committee： The existence of an audit committee is a dummy variable with a value of 1 if the audit committee has financial expert and 0 if otherwise.
This corporate governance index （CGI） by aggregate scoring these 5 mechanisms. The research calculates median values of for institutional investors’ activism， board size and independent directors， if the value is above the median， the sub-index is set to 1 and 0 otherwise. After all the points are added up， the higher the score， the better the corporate governance.
The control variables are：
1.Company size： Natural log of total asset
2.Leverage： The ratio of liabilities and assets.
3.Profitability： Total profit divided by sales net income
4.Industry： The different classifications of industries
4.5 Technique of Analysis
The collected data were coded and analysed using a statistical software programme， stata. First， the formula can measure disclosure quality （Md. Tanvir Hasan & Md. Zakir Hosain， 2015）：
=
where：
This index indicates the level of disclosure for a firm j， where N is the maximum number of relevant subcategories a firm may disclose and Xδ is equal to 1 if disclosed or 0 if not.
Second，this model can examine the relationship between the state ownership and the extent of information disclosure， and the relationship between corporate governance and disclosure quality， when the H1 was measured， this research just uses a part of formula， it means not includes CGI （corporate governance index）：
DSCLSRE= [β0 + β1 STAOW + β2SIZE + β3 LEV + β4 PROF +β5 IND+ （CGI） +μ]
where：
DSCLSRE =extent of voluntary disclosure scores
STAOW =state ownership
SIZE =firm size
LEV =leverage
PROF =profitability
IND =industry
CGI = corporate governance index
μ= error term
5.Findings and Analysis
5.1 Descriptive statistics
Descriptive statistics for all variables before transformation is presented in Table 5.1 and 5.2. In Table 5.1， the disclosure quality （DSCLSRE）， as dependent variable， is 0.45. The mean for the independent variable （STAOW） is 0.62 and moderating variable （CGI） is 2.58. And in Shanghai Stock Exchange， the mean for DSCLSRE is 0.72， 0.23 for STAOW and 2.80 for CGI. （see Table 5.2） As shown in Table 5.1 and 5.2， the values of skewness range from -0.233 to 1.864 in Hong Kong Stock Exchange， while from -0.876 to 1.764 in Shanghai Stock Exchange. The values for kurtosis range from -1.12 to 1.794 in Hong Kong Stock Exchange and from -0.523 to 1.285 in Shanghai Stock Exchange. The values for asymmetry and kurtosis between -2 and +2 are considered acceptable in order to prove normal univariate distribution （George & Mallery， 2010）. The variables including DSCLSRE， SIZE， IND and PROF in Hong Kong Stock Exchange and DSCLSRE， LEV and IND in Shanghai Stock Exchange report the highest values for skewness and kurtosis that deviate from 0 in two stock exchanges. Therefore， they are considered to be non-normal distributions and they need the transformation. Other values of variables are close to 0， indicating a normal distribution.
Taking the positive value of skewness into account， the logarithm transformation was employed on DSCLSRE， SIZE， PROF and IND in Hong Kong Stock Exchange， while DSCLSRE， LEV and IND should be transformed in Shanghai Stock Exchange. Table 5.3 and 5.4 provide the descriptive statistics after the transformation of data. As shown in Table 5.3 and 5.4， all skewness and kurtosis values have normal range. Because there are no extreme valuates， and all of the values are between -2 and 2 （George & Mallery， 2010）.
5.2Hypotheses Testing
5.2.1 First hypotheses： State ownership and disclosure
The first hypothesis （H1） is there is a negative relationship between state ownership and disclosure quality.
The OLS multiple regression analysis shows that the adjusted R of the model is 0.3429 in Table 5.5， indicating that independent variable can explain the dependent variable in thirty-four percent. The analysis found that， in Table 5.5， LOGPROF （0.098） has a significant negative effect on LOGDSCLSRE. Specifically， LOGSIZE （0.007） and LOGIND （0.009） also have negative effects on LOGDSCLSRE， while STAOW （0.005） has a positive effect on it （see Table 5.5）.
The results have revealed that there is a significantly positive relationship between state ownership and disclosure quality （Coef = 0.6840342， P=0.005）. Therefore， in the Hong Kong Stock Exchange， H1 should be rejected.
In Shanghai Stock Exchange， R of the model is 0.3565 in Table 5.6. The Prob （F-statistic） value of the model is 0.0035 and the probability of the constant is 0， showing that the model’s multivariate regression analysis is effective through the constant test.The analysis found that， SIZE， LOGLEV， PROF and LOGIND have no significant influence on DSCLSRE in Table 5.6. Specifically， STAOW （0.065） and LOGDSCLSRE have negative coefficients. In contrast with Hong Kong Stock Exchange， there is a negative significant relationship between state ownership and disclosure quality （Coef = -0.0022313， P = 0.065） in Shanghai Stock Exchange as can be seen in Table 5. 6. Therefore， the result supports H1， and it has been accepted in Shanghai Stock Exchange. 5.2.2 Second hypotheses： Disclosure in different Stock Exchanges
The second hypothesis is information disclosure of Hong Kong Stock Exchange is better than that of Shanghai Stock Exchange.
Based on Table 5.7， the descriptive analysis found that the mean of Shanghai Stock Exchange DSCLSRE is 0.7163635， while the mean of Hong Kong Stock Exchange is 2.06692. In other words， in terms of DSCLSRE， the latter is basically better than the former.
It can be seen that the disclosure quality of Shanghai Stock Exchange is worse than that of Hong Kong Stock Exchange in terms of maximum value （0.8575624<7.473844）， mean （0.7163635<2.06692） or minimum value （0.6497829<2.454989） in Table 5.7. In conclusion， Shanghai Stock Exchange has a lower amount， proving that H2 can be accepted.
5.2.3Third hypotheses： State ownership， Disclosure quality and Corporate governance
Hypothesis 3 is that corporate governance can either strengthen or weaken the relationship between state ownership and disclosure quality.
The analysis found that， LEV and CGI have no significant influence on other factors of DSCLSRE， In Shanghai Stock Exchange， in Hong Kong Stock Exchange， LOGSIZE （0.000）， LOGPROF （0.095）， LOGIND （0.006） and STAOWCGI （0.001） significantly and negatively influence LOGDSCLSRE.
The results demonstrate that the moderating effect of corporate governance （Coef = -0.2289692， p = 0.001） worked in the relationship between state ownership and disclosure quality. Obviously， the corporate governance could weaken the relationship， because the direction of the value is negative. Therefore， H3 has been accepted in Hong Kong Stock Exchange.
After adding corporate governance as the moderating variable into the relationshipbetween state ownership and disclosure quality， the result is also significant （Coef = -0.1088838， P = 0.004）. Compared with H1， the relationship between state ownership and disclosure quality is stronger. Therefore， the result of H3 in Shanghai Stock Exchange is that corporate governance can strengthen the relationship between state ownership and disclosure quality， which has been accepted.
6. Conclusion
This study seeks to examine the impact of the corporate governance mechanisms on the enterprises’ disclosure quality by taking evidence from the listed state-owned enterprises on the stock exchange of Hong Kong and Shanghai for three years from 2014 to 2016.There are 2 contributions. There is no literature investigated state ownership， corporate governance and disclosure quality in Mainland China and Hong Kong， based on the two different economic systems. Another one is about policy， under the “One-Belt One Road” policy， the research may be useful for investors to decide which market is more worth investing in. The limitations also exist in the research. The research only uses the secondary data and collects the information from the annual report of every company， providing a limitation to the research. References：
[1] Crnkovi， Poega & Su i. 2014. State Ownership and Corporate Governance of Enterprises in Croatia. American Journal of Physiology Gastrointestinal & Liver Physiology （10）： 623-632..
[2] Cao. 2003. Analysis on the Largest Shareholders' Impact on Corporate Governance and Performance. Journal of Economic Research 2.
[3] Carpenter， J. N.， Lu & Whitelaw，R.F. 2015. The Real Value of China's Stock Market. Social Science Electronic Publishing. Working paper.
[4] D George & P Mallery. 2010. SPSS for Windows Step by Step： A Simple Study Guide and Reference， 17.0 Update. Allyn & Bacon， Inc. 2009， （100） ：357.
[5] Jensen， M. C. & Meckling， W. H. 1976.Theory of the Film：Managerial Behavior， Agency costs and Ownership Structure. Journal of Economics，3（04）： 305-360
[6] Md. Tanvir Hasan1 & Md. Zakir Hosain. 2015. Corporate Mandatory and Voluntary Disclosure Practices in Bangladesh： Evidence from listed companies of Dhaka Stock Exchange. Research Journal of Finance and Accounting，6（12）： 2222-2847.
[7] R.M. Haniffa & T. E. Cooke .2002. Culture， Corporate Governance and Disclosure in Malaysian Corporations. Journal of Accounting，Finance and Business Studies，38（03）： 317-349.
[8] Sheila Nu Nu Htay， Ridzwana Mohd Said & Syed Ahmed Salman .2013 .Impact of Corporate Governance on Disclosure Duality： Empirical Evidence from Listed Banks in Malaysia. International Journal of Economics and Management，7 （02）： 242-279.
[9] Terence，C. 2014 .Corporate Governance in Africa's State-owned Enterprises： Perspectives on an Evolving System. Journal of SAIIA Policy Briefing working paper.
Keywords： Disclosure quality；Corporate governance；Stock Exchange；Index
1.Introduction
Theinitial research problem was identified between the Hong Kong and Shanghai markets. In 1997， the People's Republic of China officially took over the sovereignty of Hong Kong and governed the region. The mainland continues to adopt socialism while capitalism remained in Hong Kong after 1997. With the tight historical connection with the UK， Hong Kong adopted the corporate governance system that was very similar to the British corporate governance.
Thedisclosure mechanism of the information in mainland China became worse compared to Hong Kong. In the mainland， the annual and financial reports could only be registered and reviewed. In Hong Kong， all the accounting and business details in the listed companies have to go through the supervision of Hong Kong’s Stock Exchange.
Based on theearlier literature， the state-owned enterprises and their owners generally face five corporate governance problems （Crnkovi， Po， 2014）. They are low levels of disclosure quality， poor relationship among the shareholders， fuzzy ownership objectives and poor ownership function.
There are 3 research questions：
·Is there any relationship between the state ownership and disclosure quality？ ·Is information disclosure of Hong Kong Stock Exchange better than that of Shanghai Stock Exchange？
·Does corporate governance moderate the relationship between state ownership and disclosure quality？
2.Literature Review
2.1 Corporate Governance in Chinese SOEs
Unlike in the private sector， whose main goal is to optimize profits， the Chinese state-owned enterprises have a social responsibility as an additional objective. Although they expect to make a profit， they also promote the country to maintain the national economic stability and sustainable development （Wang， 2012）. The key difference of the corporate governance between the state-owned enterprises and the private sector is their proximity to the state power. The SOEs are in a unique position of the same entity， ultimately responsible for the legal and regulatory environments， and directly influenced by the politicians rather than the professional businessmen （Terence， 2014）.
2.2State-owned Enterprises Listed on the Hong Kong Stock Exchange
Some Chinese scholars have pointed out the existing problems in the corporate governance of the state-owned enterprises listed on the Hong Kong Stock Exchange （Cao， 2003）.
·Equity structure is unreasonable. When the state-owned capital operation is in a dominant position and once there is a lack of checks and balances of the other ownership capital， many problems would exist.
·The problem of internal control. When the controlling shareholder is the state， the internal control is often accompanied with the government. The key people and the control focus their power on the major issues.
·The organizational structure of the company. There are many problems such as the independence of the board of directors and the limited role of the board of supervisors in China’s listed companies.
2.3State-owned Enterprises Listed on the Shanghai Stock Exchange
The Chinese mainland stock market has some notable features （Carpenter， Lu & Whitelaw， 2015）. First， it is dominated by the retail investors， accounting for more than 80% of the transactions， which mean that China’s asset management industry is still underdeveloped， resulting in several market efficiency issues. Second， it is a pure order-driven rather than a quote-driven market， while the United States and several other countries have a mixed stock market system. Third， it is a concentrated market， while the U.S. market is fragmented， which may have a vital impact on the market information. 3.Theoretical Framework
This study has employed the agency theory framework. Jensen and Meckling （1976） used the agency theory as a tool to explain the inherent difficulties in the governance structure of the state-owned enterprises in Italy. The agency theory is particularly useful as it contains the factors that may affect the risk disclosure of the companies. It also can be illustrated by agency problem type I and type II. Type I is the agency problem agency problem that arises between the principal as the owner of companies and agents as the manager who is the executor the company’s operations. While the issue of agency Type II is the agency problem that occurs between controlling shareholders and citizens.
4.Research Method
4.1 Sample of study
This research compares the different situations in Hong Kong and Shanghai Stock Exchange. The sample is selected based on the existence of the companies and complete information from 2014 to 2016. There are 91 state-owned enterprises listed in Hong Kong Stock Exchange while there are 100 state-owned enterprises in Shanghai （Table 4.1）.
4.3 Research Design
The analysis methodthat will be used in this study is the cross-sectional data analysis. The disclosure index is measured by the formula created by Haniffa and Cooke （2002）， while the relationship between the state ownership and disclosure quality can be measured by the equation， which was created by Omar Juhmani （2013）， and the relationship between corporate governance and disclosure quality is identified using the equation created by Sheila， Ridzwana and Salman （2013）.
4.4 Operationalization of Variables
The dependent variable is the disclosure quality. This is measured by annual report and calculating the disclosure index. The SOEs are defined as those companies that the government owns 20% or above of the shares. The independent variable is the state ownership， where the research selects the state-owned enterprises that are listed on Shanghai and Hong Kong Stock Exchange. he moderating variable is the corporate governance.
The five variables that are under the corporate governance can be measured in five ways. （Sheila， Mohd Ridzwana & Salman，2013）
Institutional investors’ activism： The research calculates the percentage of the capital shares held by institutional investors
Duality leadership structure： This can be measured by checking the type of leadership structure. Board size： This is identified by calculating the total number of the board of directors.
Independent directors： This can be taken by accounting the number of independent directors/total number of board of directors.
Audit committee： The existence of an audit committee is a dummy variable with a value of 1 if the audit committee has financial expert and 0 if otherwise.
This corporate governance index （CGI） by aggregate scoring these 5 mechanisms. The research calculates median values of for institutional investors’ activism， board size and independent directors， if the value is above the median， the sub-index is set to 1 and 0 otherwise. After all the points are added up， the higher the score， the better the corporate governance.
The control variables are：
1.Company size： Natural log of total asset
2.Leverage： The ratio of liabilities and assets.
3.Profitability： Total profit divided by sales net income
4.Industry： The different classifications of industries
4.5 Technique of Analysis
The collected data were coded and analysed using a statistical software programme， stata. First， the formula can measure disclosure quality （Md. Tanvir Hasan & Md. Zakir Hosain， 2015）：
=
where：
This index indicates the level of disclosure for a firm j， where N is the maximum number of relevant subcategories a firm may disclose and Xδ is equal to 1 if disclosed or 0 if not.
Second，this model can examine the relationship between the state ownership and the extent of information disclosure， and the relationship between corporate governance and disclosure quality， when the H1 was measured， this research just uses a part of formula， it means not includes CGI （corporate governance index）：
DSCLSRE= [β0 + β1 STAOW + β2SIZE + β3 LEV + β4 PROF +β5 IND+ （CGI） +μ]
where：
DSCLSRE =extent of voluntary disclosure scores
STAOW =state ownership
SIZE =firm size
LEV =leverage
PROF =profitability
IND =industry
CGI = corporate governance index
μ= error term
5.Findings and Analysis
5.1 Descriptive statistics
Descriptive statistics for all variables before transformation is presented in Table 5.1 and 5.2. In Table 5.1， the disclosure quality （DSCLSRE）， as dependent variable， is 0.45. The mean for the independent variable （STAOW） is 0.62 and moderating variable （CGI） is 2.58. And in Shanghai Stock Exchange， the mean for DSCLSRE is 0.72， 0.23 for STAOW and 2.80 for CGI. （see Table 5.2） As shown in Table 5.1 and 5.2， the values of skewness range from -0.233 to 1.864 in Hong Kong Stock Exchange， while from -0.876 to 1.764 in Shanghai Stock Exchange. The values for kurtosis range from -1.12 to 1.794 in Hong Kong Stock Exchange and from -0.523 to 1.285 in Shanghai Stock Exchange. The values for asymmetry and kurtosis between -2 and +2 are considered acceptable in order to prove normal univariate distribution （George & Mallery， 2010）. The variables including DSCLSRE， SIZE， IND and PROF in Hong Kong Stock Exchange and DSCLSRE， LEV and IND in Shanghai Stock Exchange report the highest values for skewness and kurtosis that deviate from 0 in two stock exchanges. Therefore， they are considered to be non-normal distributions and they need the transformation. Other values of variables are close to 0， indicating a normal distribution.
Taking the positive value of skewness into account， the logarithm transformation was employed on DSCLSRE， SIZE， PROF and IND in Hong Kong Stock Exchange， while DSCLSRE， LEV and IND should be transformed in Shanghai Stock Exchange. Table 5.3 and 5.4 provide the descriptive statistics after the transformation of data. As shown in Table 5.3 and 5.4， all skewness and kurtosis values have normal range. Because there are no extreme valuates， and all of the values are between -2 and 2 （George & Mallery， 2010）.
5.2Hypotheses Testing
5.2.1 First hypotheses： State ownership and disclosure
The first hypothesis （H1） is there is a negative relationship between state ownership and disclosure quality.
The OLS multiple regression analysis shows that the adjusted R of the model is 0.3429 in Table 5.5， indicating that independent variable can explain the dependent variable in thirty-four percent. The analysis found that， in Table 5.5， LOGPROF （0.098） has a significant negative effect on LOGDSCLSRE. Specifically， LOGSIZE （0.007） and LOGIND （0.009） also have negative effects on LOGDSCLSRE， while STAOW （0.005） has a positive effect on it （see Table 5.5）.
The results have revealed that there is a significantly positive relationship between state ownership and disclosure quality （Coef = 0.6840342， P=0.005）. Therefore， in the Hong Kong Stock Exchange， H1 should be rejected.
In Shanghai Stock Exchange， R of the model is 0.3565 in Table 5.6. The Prob （F-statistic） value of the model is 0.0035 and the probability of the constant is 0， showing that the model’s multivariate regression analysis is effective through the constant test.The analysis found that， SIZE， LOGLEV， PROF and LOGIND have no significant influence on DSCLSRE in Table 5.6. Specifically， STAOW （0.065） and LOGDSCLSRE have negative coefficients. In contrast with Hong Kong Stock Exchange， there is a negative significant relationship between state ownership and disclosure quality （Coef = -0.0022313， P = 0.065） in Shanghai Stock Exchange as can be seen in Table 5. 6. Therefore， the result supports H1， and it has been accepted in Shanghai Stock Exchange. 5.2.2 Second hypotheses： Disclosure in different Stock Exchanges
The second hypothesis is information disclosure of Hong Kong Stock Exchange is better than that of Shanghai Stock Exchange.
Based on Table 5.7， the descriptive analysis found that the mean of Shanghai Stock Exchange DSCLSRE is 0.7163635， while the mean of Hong Kong Stock Exchange is 2.06692. In other words， in terms of DSCLSRE， the latter is basically better than the former.
It can be seen that the disclosure quality of Shanghai Stock Exchange is worse than that of Hong Kong Stock Exchange in terms of maximum value （0.8575624<7.473844）， mean （0.7163635<2.06692） or minimum value （0.6497829<2.454989） in Table 5.7. In conclusion， Shanghai Stock Exchange has a lower amount， proving that H2 can be accepted.
5.2.3Third hypotheses： State ownership， Disclosure quality and Corporate governance
Hypothesis 3 is that corporate governance can either strengthen or weaken the relationship between state ownership and disclosure quality.
The analysis found that， LEV and CGI have no significant influence on other factors of DSCLSRE， In Shanghai Stock Exchange， in Hong Kong Stock Exchange， LOGSIZE （0.000）， LOGPROF （0.095）， LOGIND （0.006） and STAOWCGI （0.001） significantly and negatively influence LOGDSCLSRE.
The results demonstrate that the moderating effect of corporate governance （Coef = -0.2289692， p = 0.001） worked in the relationship between state ownership and disclosure quality. Obviously， the corporate governance could weaken the relationship， because the direction of the value is negative. Therefore， H3 has been accepted in Hong Kong Stock Exchange.
After adding corporate governance as the moderating variable into the relationshipbetween state ownership and disclosure quality， the result is also significant （Coef = -0.1088838， P = 0.004）. Compared with H1， the relationship between state ownership and disclosure quality is stronger. Therefore， the result of H3 in Shanghai Stock Exchange is that corporate governance can strengthen the relationship between state ownership and disclosure quality， which has been accepted.
6. Conclusion
This study seeks to examine the impact of the corporate governance mechanisms on the enterprises’ disclosure quality by taking evidence from the listed state-owned enterprises on the stock exchange of Hong Kong and Shanghai for three years from 2014 to 2016.There are 2 contributions. There is no literature investigated state ownership， corporate governance and disclosure quality in Mainland China and Hong Kong， based on the two different economic systems. Another one is about policy， under the “One-Belt One Road” policy， the research may be useful for investors to decide which market is more worth investing in. The limitations also exist in the research. The research only uses the secondary data and collects the information from the annual report of every company， providing a limitation to the research. References：
[1] Crnkovi， Poega & Su i. 2014. State Ownership and Corporate Governance of Enterprises in Croatia. American Journal of Physiology Gastrointestinal & Liver Physiology （10）： 623-632..
[2] Cao. 2003. Analysis on the Largest Shareholders' Impact on Corporate Governance and Performance. Journal of Economic Research 2.
[3] Carpenter， J. N.， Lu & Whitelaw，R.F. 2015. The Real Value of China's Stock Market. Social Science Electronic Publishing. Working paper.
[4] D George & P Mallery. 2010. SPSS for Windows Step by Step： A Simple Study Guide and Reference， 17.0 Update. Allyn & Bacon， Inc. 2009， （100） ：357.
[5] Jensen， M. C. & Meckling， W. H. 1976.Theory of the Film：Managerial Behavior， Agency costs and Ownership Structure. Journal of Economics，3（04）： 305-360
[6] Md. Tanvir Hasan1 & Md. Zakir Hosain. 2015. Corporate Mandatory and Voluntary Disclosure Practices in Bangladesh： Evidence from listed companies of Dhaka Stock Exchange. Research Journal of Finance and Accounting，6（12）： 2222-2847.
[7] R.M. Haniffa & T. E. Cooke .2002. Culture， Corporate Governance and Disclosure in Malaysian Corporations. Journal of Accounting，Finance and Business Studies，38（03）： 317-349.
[8] Sheila Nu Nu Htay， Ridzwana Mohd Said & Syed Ahmed Salman .2013 .Impact of Corporate Governance on Disclosure Duality： Empirical Evidence from Listed Banks in Malaysia. International Journal of Economics and Management，7 （02）： 242-279.
[9] Terence，C. 2014 .Corporate Governance in Africa's State-owned Enterprises： Perspectives on an Evolving System. Journal of SAIIA Policy Briefing working paper.