Tóm tắt Luận án The determinants of financial statement quality of companies listed on stock market evidence in Vietnam

of financial statement quality of listed companies, through the financial statement measurement approach method of earning quality, especially, the earning quality measurement is based on two basics, which are accounting basic and market basic, along with the up-to-date secondary data (data presented in financial statements) during the period of 2012 to. Therefore, the results of analyzing the financial statement quality of listed companies in this thesis reflect the reality based on the data which listed companies performed on their financial statements, so this thesis is unlike the previous researches in Vietnam whose results were based on the evaluation opinion of users about the financial statement quality. + With the recommended model, many factors influencing the financial statement quality of listed companies are added to the model to test, and the results of this thesis 5 show that 17 of 23 testable factors influence the financial statement quality of companies listed on stock market of Vietnam. + Based on the results of evaluating the reality of financial statement quality from 2012 to 2014, which are detailed for each year and even for each sectors, and the results of the determinants of financial statement quality of companies listed on stock market in Vietnam. These results are becoming an essential science fundamental for the interested users, such as managers, auditors, investors, government agency, State Securities Commission of Vietnam, etc. Therefore, many solutions are going to be addressed for improving the financial statement quality of listed companies in Vietnam, as well as a reference for financial statement users in order to make many related decisions effectively. 6. Thesis structure Research introduction: Chapter I: Overview of research about the measurement of financial statement quality and determinants of financial statement quality of companies listed on the stock market. Chapter II: Theoretical foundations and research model Chapter III: Research methodology. Chapter IV: Research results and discussion Chapter V: Conclusion and proposed recommendations 6 CHAPTER 1 OVERVIEW OF RESEARCH ABOUT THE MEASUREMENT OF FINANCIAL STATEMENT QUALITY AND THE FACTORS EFFECT ON STATEMENT QUALITY OF COMPANIES LISTED ON THE STOCK MARKET. In the main content of this chapter, the author will introduce an overview of researches around the world and Vietnam ab financial statement about the quality measurement method, and the determinants of financial statement quality of listed companies. Therefore, the author reveals the gap of research and recommends the research approach for this thesis. 1.1. The approaches to the financial statement quality measurement method based on the prior researches In any researches relating to the financial statement quality, although the financial statement quality plays the role as the factor influencing the others or the factor influenced by the others, the measurement of financial statement quality has been really important. However, the financial statement quality is difficult to observe directly. For this reason, the prior studies have used many different measurement methods which is able to divide into two main methods: Measurement by the characteristics of financial statement quality and measurement by earning quality. 1.1.1. Financial statement quality measurement by the quality characteristics The method of financial statement quality measurement by the quality characteristics has been used for evaluating many aspects and sizes of financial and non – financial information in financial statements, in order to indentify the necessity of this financial information. This method of financial statement quality measurement means that the financial statement quality is evaluated based on the scales that are built based on the quality characteristics of FASB. These include two fundamental characteristics that are relevance and reliability, two secondary characteristics that are consistency and comparability. On the other hand, the characteristics, which are based on the opinions of the combined project of FASB and IASB, include two fundamental characteristics, which are relevance and faithful representation, and four enhancing characteristics including comparability, verifiability, timeliness and understandability. 7 Over the past few years, there are many researches on the financial statement quality. These researches have analyzed many problems, such as evaluating the quality characteristics, the determinants of the quality characteristics or the quality characteristics based on the opinions of many general users or auditors, investors, etc. To be more detail, it can be easily seen on the research of Jonas and Blanchet (2000), Beest et al (2009), Beest and Braam (2011), Jara et al (2011), Bauwhede (2001), Saheli & Nassirzadeh (2012), Terzungwe (2013), Tasios (2012), Obaidat (2007). The purpose of this above method is not the earning quality measurement and evaluating many specific factors of financial statements, but the main purpose is evaluating the necessity level of information through the quality characteristics. The benefit gained from this method is a research concentration on quality characteristics, whereas the others concentrated on many aspects of the quality management or stock prices. However, the disadvantage of this method is that identifying scales of quality characteristics is very difficult. Moreover, because of this evaluation based on the scales, the other drawback is that the result will not have high reliability since it depends on many subjective factors of the assessors and it has a high identifiable emotion component, for example, this assessor gives high score for this indication point in the financial statement, whereas, the other may not. 1.1.2. Financial statement quality measurement by the earning quality The main advantage of the method of financial statement quality measurement by the earning quality is its reliability. Because the measurement by the quality characteristics, which is used by the other research, is based on the opinions of the assessors based on the secondary data (the data in the financial statements and annual reports) or the opinions of the people who answer in the survey (in the researches using the viewpoint survey and the data is the primary data), and this may lead to the result that the evaluation will depend on their opinions, and the research result may not reflect the reality of financial statements of companies. On the other hand, the method of financial statement quality measurement by the earning quality, the data used for this method is based on the avaialable data in the financial statements and annual reports (secondary data) and the evaluation is directly based on these data. Moreover, according to the opinions of the financial statement users, the investors, the debtors and the accounting standard issuers, the method of financial 8 statement quality measurement by the earning quality is really useful. 1.1.2.1. Some models of earning quality measurement based on extraordinary accrual model. There are some models of earning quality measurement based on extraordinary accrual model. The author chooses the model of Jones, which has been adjusted by Dechow et al (1995) and utilizes cross – sectional data based on the research of DeFond & Jiambalvo (1994). The researches of Teoh et al (1998), Myers et al (2003), Xie et al (2003), Hsu & Koh (2005), Aboody et al (2005), Qinghua et al (2007), Jiang et al (2008), Jaggi et al (2009), Alves (2014), Gonzalez et al (2014), Aygun et al (2014), Gonza´lez, J. S., and Meca, E. G. (2014), also used this method to measure the dependent variable, which is earning management in their research. 1.1.2.2. Some models of earning quality measurement based on a value relevance of accounting information model. Among the models based on a value relevance of accounting information to measure earning management, the author has chosen the adjusted EBO model, which is one of two models measuring financial statement quality in this thesis. The researches of Abubakar (2011), Hassan (2012) và Ahmed (2012), also used this method to measure the dependent variable, which is financial statement quality, in their research. There are many prior researches measuring financial statement quality by earning management, including the researches using financial statement quality as an independent variable as well as a role of independent variable, such as the researches of Choi and Pae (2011), Gajevszky (2015), Darabi et al (2012), Hashim (2012) and Alzoubi (2012). Besides, many researches of the determinants of financial statement quality have used earning management to measure this dependent variable, which will be presented in the next content. 1.2. Overview of previous studies about the determinants of financial statement quality by measuring earning quality. 1.2.1. Overview of previous studies in the world There are many research in the world in regard to the determinants of financial 9 statement quality, especially earning quality in financial statements, including internal and external factors. These are Dechow et al (2010), Qinghua et al (2007), Radzi et al (2011), Alves (2014), Hassan (2013), Healy (1985), Gaver et al (1995), Holthausen et al (1995), Abed et al (2012), Waweru và Riro (2013), Ahmed (2013), Aygun et al (2014), Hassan (2012), Habib & Azim (2008), Jamaluddin et al (2009), Klai (2011), Chalaki et al (2012). Besides, there are more researches relating to the determinants of financial statement quality, the author has summarized the research results in the world in the thesis. Consequently, there are many research works on the determinants of financial statement quality in the world in which the financial statement quality is measured by many methods, each researches usually concentrates on one or some factor groups, such as the management of companies, the characteristics of companies, etc. Most of the researches used a regression model and the ordinary least square method (OLS) (Houqe et al, 2010; Radzi et al, 2011; Liu, 2012; Waweru & Riro, 2013; Hassan, 2013; Johl et al, 2013; Aygun et al, 2014), some researches used finite element method (FEM) (Soliman & Ragab, 2013; Rahman et al, 2008), and few researches used the method of generalized least square (GLS) (Alkdai & Hanefah, 2012; Rafiee et al, 2014). 1.2.2. Overview of previous studies in the country In Vietnam, the theme relating to financial statement quality has been also concerned by some researchers, these parlicular researches are listed below: Nguyen Phuc Sinh (2008), Nguyen Dinh Hung (2010), Nguyen Thi Bich Van (2012), Duong Nguyen Hung (2013), Nguyen Thi Hong Nga (2013), Ngo Thu Giang và Dang Anh Tuan (2013), Pham Thi Bich Van (2014), Nguyen Cong Phuong and Nguyen Thi Thanh Phuong (2014), Nguyen Thi Phuong Hong and Mai Thi Hoang Minh (2014), Nguyen Thi Phuong Hong and Duong Thi Khanh Linh (2014), Nguyen Thi Hong Nga (2014), Nguyen Trong Nguyen (2014), Doan Thi Hong Nhung and Vu Thi Kim Lan (2014), Ngo Thi Thu Giang (2014), Pham Ngoc Toan and Hoang Thi Thu Hoai (2015), Le Thi My Hanh (2015), Dang Ngoc Hung (2015), Pham Thi Bich Van (2015), Nguyen Trong Nguyen (2015). Therefore, Vietnamese researchers have been initially concerned about researching on some related theme about financial statements. However, researchers have just analyzed 10 basic contents relating to financial statements. There are few researches on the determinants of financial statement, but they have only analyzed the information disclosure, the determinants analyzed and tested have not been concerned thoroughly, or some researches have also focused on the financial statement transparency. 1.3. The gap of research Through the synthesis of the researches in the world relating to this thesis, it can be claimed that financial statement quality and its determinants are one of the topics specifically concerned by researchers of many countries around the world. Therefore, there have been many researches on this topic. On the other hand, there have been many researches on financial statement quality in one country, but they have many approaches to measure it. Most of these researches also use the quantitative research method to approach based on the prior researches, theoretical foundations and the economic and social characteristics of each country to build and test models. However, these researches are usually analyzed respectively and mostly focus on the factors of company management, each research has only focused on one specific factor group, such as company management, company characteristics and there are researches analyzing few factors, such as one or two factors. With the reality of researches relating to the determinants of financial statement quality in Vietnam, such as the above synthesis summarizing researches in Vietnam, the author has some observations as below: - In some countries around the world, the research on topics relating to financial statement is conducted earlier than Vietnam, however, in recent years, Vietnamese researchers has also concerned more and had many studies relating to this topic. - Researchers have also perceived an importance of research on the determinants of financial statements. However, these researches have just focused on the financial statement transparency or some researches have just analyzed the financial statement disclosure, and the number of these researches is very limited. Therefore, in addition to the achievements, it can be said that a number of Vietnamese researches relating to this topic is very limited, the researches conducted have analyzed only basic contents about the financial statements, there are few researches on the 11 determinants of financial statements but they have just analyzed the financial statement disclosure, the determinants analyzed and tested are not concerned. These researches have not analyzed the financial statement quality profoundly. Besides, the results of evaluating financial statements and analyzing the determinants in most of the researches are based on survey methodology, hence, they have not reflected the reality of financial statements, which are created and designed by companies in Vietnam at present. There are few researches analyzing directly on financial statements and annual reports of companies, but they have just approached the determinants of financial statement disclosure or focused on the financial statement transparency. In the world, although there are many researches on the determinants of financial statement quality and earning quality, most of these focus on analyzing one specific group factors, such as company management, company characteristics, dividend policy, audit quality influencing financial statement quality, etc. There is a limited amount of researches analyzing all these factors in one research, because if researchers analyzed like this, the results of the general research may be more crucial than the specifically research. As a result of the synthesis of the prior researches in the world and in Vietnam, and the analysis of achievements and limitations in these researches, the author has realized that it is necessary to have one research on the determinants of financial statement quality of companies listed on stock market of Vietnam at present. Therein, the financial statement quality is measured based on data presented in financial statements (secondary data) by companies, and measured by many methodologies contemporarily, a number of determinants analyzed has been increased; therefrom, a recommendation will be proposed in order to improve the financial statement quality of listed companies. Moreover, this result is going to be a necessary and important reference for many related users, such as the country authorities, company managers, auditors, investors and many other related users in the case that the reality of financial statement quality of Vietnamese companies has been concerned at present. 12 CHAPTER 2. THEORETICAL FOUNDATIONS AND RESEARCH MODEL The main content of this chapter is analyzing the related theoretical foundations which are the foundation for the author to build the research model and develop the research hypothesis associated with the content of previous chapter. This helps to achieve the main goals of the thesis to assist the readers in approaching the contents of this thesis easier. In this chapter, the author also introduces some definitions and some important terms. 2.1. Theoretical foundations relating to the determinants of financial statement quality. As the contents mention above, there are many researches on the determinants of financial statement quality in the world and the results of these demonstrate that there are many factors influencing the financial statement quality. However, these results are not similar, there is a factor which has influences the quality in this research but has not influenced in the other. Furthermore, there is also a situation that this factor has a possitve effect on financial statement quality in some researches, but in the other researches, this factor has a negative effect. Therefore, in order to develope the research hypotheses from the research model in this study, the author conducts an analysis of some related theoretical foundations, which are the agency theory, resource dependency theory, political cost theory, the signaling theory, exclusive cost theory and the contract theory. 2.2. Research model 2.2.1. Research model: Based on the related prior researches and the quantitative research result, the author has built the research model with many determinants influencing the financial statements, which are the financial statement quality as the dependent variable and 23 independent variables (determinants) divided into five groups of factor. These are (1) group of factors relating to ownership structure including foreign ownership, state ownership, managerial ownership, institutional ownership and ownership concentration, (2) group of factors relating to corporate governance including board size, CEO duality, board independence, number of board meetings, financial expertise level of board and the existence of bonus plan, (3) group of factors relating to capital structure including firm leverage and firm 13 liquidity, (4) group of factors relating to market characteristics including firm size, firm age, list time, list status, auditor type, delay of financial statement and industrial sector, (5) group of factors relating to performance of company including profitability (ROE), firm growth and dividend policy. 2.2.2. Developing the research hypotheses Based on the research model, the results of related researches, which are in a summary table of the determinants of financial statement quality from the prior researches, and the analysis of the relationship between these factors and financial statement quality from the above theoretical foundations, the author builds 23 hypotheses. CHAPTER III. RESEARCH METHODOLOGY To achieve the goal of one research, the research methodology is an important content where it contributes greatly to the research efficiency of this thesis. Therefore, the author has focused on the research methodology of this thesis, the content of this chapter will present the research model and research methodology chosen by the author. 3.1. Research methodology and research process 3.1.1. Research methodology In this thesis, the author has used the mixed approach in this research. From the research problem, the author has observed the related prior researches on the measurement methodology of financial statement quality, the determinants of financial statement quality. Therefrom, the author has built and tested the model of the determinants of financial statement quality of listed companies in Vietnam. However, besides the factors added to the model based on the prior researches, according to the opinion of the author about characteristics of Vienam, the author intends to add four new factors to the model, which are not researched in the prior researches. In order to have a basic for a decision whether it is necessary to add these factors in the model, the author has used the qualitative research methodology through the data collection method which is the discussions with the professionals (5 professionals) about the problem, which is whether four new factors suggested by the author should be added to the model. Furthermore, the author has 14 designed more open – ended questions for more opinions from the professionals with the content, which is “Besides four above factors, are there any factors influencing the financial statement quality? Why?”. After having the discussion results with the professionals, the author presents the official research model and develops the research hypotheses. Then, the author uses the quantitive research methodology to build the scales for variables, collects data and tests the hypotheses. Unlike the related prior researches using the quantitive research methodology, this research uses the mixed approach, which means that the main methodogy is the quantitive research methodology, along with the qualitative research methodology, because this research has added new factors, therefore, in the author’s opinion, the mixed approach is suitable. 3.1.2. Research processs As described above, the thesis has used the mixed approach, so the research process of this thesis is determined as a mixed process, which combines the qualititve research and quantitive research, with four specific steps in the following: Step 1: Summarizing the prior researches relating to the financial statement quality measurement, the determinants of the financial statement quality and the completed synthesis of the theoretical foundations relating to the determinants of the financial statement quality. According to these bases, determining the gap of research and proposing the research model of the thesis. Step 2: Collecting data for the quantitative research Step 3: Colleting data for the qualitative research. Step 4: Analyzing, discussing the research result, commenting and proposing the recommendations. 3.2. Regression model and variable measurement in the model 3.2.1. Regression model In order to test 23 hypotheses presented in chapter 2, and since the author measures the financial statement quality by two methods with two different models, the author uses two multivariate linear regressions. In model 1, the dependent variable, which is the financial statement quality, is measured by the modified model of Jones. In model 2, the 15 financial statement quality is measured by the modified EBO model. Model 1: FRQ1it = β0 + β1OFOit + β2OSTit + β3OMAit + β4OINit + β5BLOCKit + β6BSit + β7DCit + β8BINit + β9MEETit+ β10EXPERit + β11BONUit + β12LEVit + β13LIQit + β14CSIZEit + β15AGEit + β16TIMEit + β17LSTit + β18ATYPEit + β19TILIit + β20INDit + β21ROEit + β22GROit + β23DIVit + εit (1) Model 2: FRQ2it = β0 + β1OFOit + β2OSTit + β3OMAit + β4OINit + β5BLOCKit + β6BSit + β7DCit + β8BINit + β9MEETit+ β10EXPERit + β11BONUit + β12LEVit + β13LIQit + β14CSIZEit + β15AGEit + β16TIMEit + β17LSTit + β18ATYPEit + β19TILIit + β20INDit + β21ROEit + β22GROit + β23DIVit + εi (2) 3.2.2. Variable measurement in the model 3.2.2.1. Dependent variable measurement Measured by the adjusted model of Jones. AACit/At-1= α1(1/At-1) + α2[(∆REVt - ∆RECt)/At - 1]+ α3(PPEt/At-1) + εit (6) Measured by the adjusted EBO model. Pit = β0 + β1BVit + β2EPSit + β3EPS1it + εit (8) 3.2.2.2. Indepent variable measurement The main regression model of this thesis includes 23 independent variables. The name, the symbol and the measurement of these variables are described in thesis. 3.3. Research design 3.3.1. Research sample size On December 31st 2014, the total amount of listed companies in both two stock exchange of Vietnam is 622, including 303 listed companies on Ho Chi Minh City stock exchange and 359 listed companies on Ha Noi stock exchange. In this thesis, the author has collected two groups of samples. The first one is used to collect data for the measurement of the dependent variable, which is the financial statement quality; with this 16 group, the author applies to choose the non random sample, which means that from listed companies having the annual reports (including financial statements) on two stock exchange, the author chooses the one having sufficient information to collect. Besides, the author eliminates all financial companies, banks and insurance companies, because the current regulations on the preparation and presentation of financial statements of these companies are not homogenous with the other companies in this sample. In table 3.2, the result indentifies that 394 companies in each year 2012, 2013 and 2014 are used and the total amount of sample companies used in 3 years are 1,182 companies. Based on these 394 sample companies, firstly, the author also applies to choose the non random sample and 123 companies for each year 2012, 2013 and 2014 are chosen and the total samples of three years is 369. However, 86 observations of these do not have the sufficient information for 23 independent variables (since some information is voluntarily collected), so the last sample includes 283 observations as the official samples for this thesis. 3.3.2. Data collection To test the hypotheses of this thesis, the author has collected data from many different sources, such as the annual reports (including financial statements). With the collected data for the independent variables, if they are not presented on the annual reports, the author would collect from the website of listed companies, or from the website of Vietstock or the direct information from Vietstock. 3.3.2.1. Information collection relating to the measurement of the financial statement quality. In this thesis, since the author used two measurement methods of financial statement quality (the dependent variable) contemporarily, the data is collected from many different sources. The data referring to the measurement of financial statement quality by the adjusted model of Jones is collected from the balance sheet and the income statement in financial statements. The data referring to the measurement of financial statement quality by the EBO model is collected from the website of Vietstock. Particularly, the information relating to stock prices on March 31st in the year, which is after the year on research, the Vietstock company directly provides information for the author. 17 3.3.2.2. Information collection relating to the determinants The collected data relating to the determinants of financial statement quality (the independent variables) is collected from the financial statements and annual reports of sample companies. The data referring to the related variables of ownership and corporation governance mechanism is collected from the annual reports, the data referring to the capital structure is collected from financial statements. The data referring to the market is collected from both financial statements and annual reports, and lastly, the data referring to the company efficiency, including ROE and development prospects, is collected from financial statements, the other data referring to dividend policy is collected from annual reports of companies. 3.3.3. Process of analyzing data Through STATA 13 software and Excel 2013, the following technical for analyzing will be used to analyze the data in order to aim the research goal, which is analyzing the determinants of financial statement quality. CHAPTER 4. RESEARCH RESULTS AND DISCUSSION Researching on the determinants of financial statement quality has been concerned by many researchers around the world since the information in financial statements is necessary for the economic decisions of many users. Howerver, the number of researches relating on this topic is very limited. In this chapter, the reality of financial statement quality of companies listed on stock market of Vietnam is analyzed, moreover, the determinants of financial statement quality of these listed companies are also analyzed. 4.1. Analysis of the reality of financial statement quality of companies listed on stock market of Vietnam at present. 4.1.1. Analysis of the reality of financial statement quality measured by the earning management. Research results show that the financial statement quality measured by the earning management of listed companies in Vietnam is at an average level compared to the other countries in the world, the mean of the earning management of listed companies in 18 Vietnam is lower than Egypt, Iran and Malaysia of which the corresponding means of the earning management are 0.9875, 0.7410 and 1.7699. However, the average earning management of listed companies in Vietnam is higher than the other countries like Turkey, Jordan and Australia of which the corresponding means of earning management are 0.0077, 0.1330 and 0.0624, and these approximate to some countries of Latin America of which the average earning management is 0.2370. The mean of the earning management of companies listed on stock market of Vietnam during 2012 – 2014 is 0.2032428. During this period, the financial statement quality in 2013 is the lowest that the highest mean of earning management is 0.2887321, in 2012, the average earning management is lower than 2013 with 0.1829587. As a result, the earning management has increased from 2012 to 2013, which means the financial statement quality has decreased from 2012 to 2013. However, in 2014, the financial statement quality has increased significantly compared to 2012 and 2013, the mean of earning management has decreased to 0.1380377. 4.1.2. Analysis of the reality of financial statement quality measured by the value relevance of accounting information Research results show that the R2 value of the EBO model (the adjusted EBO model) has gradually increased over years, which the R2 value in 2012 is only 34% (37%), the one in 2013 is 41% (42%), the one in 2014 is also 40% (41%), and the one during 2012 – 2014 is 38% (39%), however, the R2 value has still been lower than other countries in the world, this result approximates with Britain, Germany and Itatia of which the corresponding R2 value is 35% (40%), 40% và 43% (48%) and highly lower than other countries, such as Nigeria, Kuwai, Norway, Japan, Sri Lanka and the United State of which the corresponding R2 value is 94% (92%), 75%, 66%, 65%, (65%), (56%) và 54%. It can be claimed that the financial statement quality measured by the value relevance of accounting information of listed companies in Vietnam has not been really high. 4.2. Description of the determinants of financial statement 4.3. Results of correlation analysis and regression analysis 4.3.1. Correlation analysis The analysis result indentifies that the factors observed in this research have the low 19 correlation coefficients, so it is unlikely to have the multicollinearity phenomenon. 4.3.2. Regression analysis results of model 1 Before analyzing the regression model 1, the author has conducted OLS estimation of the adjusted model of Jones, which measures the dependent variable (the financial statement quality) in this model 1, the research result demonstrates that the modified model of Jones are appropriate. 4.3.2.1. Appropriate model selection Based on the results of estimation and comparison among three models, which are Pooled OLS, FEM and REM, it is shown that the OLS estimation is the most appropriate estimation for the model 1. 4.3.2.2. Test for appropriate model (1) The independent variables are not correlated with each other (it does not have the multicollinearity phenomenon). The result identifies that it is unlikely to have the multicollinearity phenomenon. (2) Assumption of normally distributed residuals The result indentifies that the residuals of this research model are normally distributed. (3) Assumption of constant variance The result find that this model has the changing variance. (4) Assumption of autocorrelation. The result find that the model has first order autocorrelation Conclusions: Based on the test on four assumptions of the model, two assumptions are not violated, which are the multicollinearity phenomenon and the assumption of normally distributed residuals. However, the remaining two assumptions is violated, which is the assumption of autocorrelation and the assumption of constant variance. To overcome the violation of these assumptions, the author has to conduct the regression by the Feasible generalized least squares estimation (FGLS), this estimation method is as same as the research method conducted by Rafiee et al (2014). 4.3.2.3. Results of the beta and the statistical significance of model 1 The result of study show that, 11 of 23 independent variables entered to the model 20 influence the financial statement quality. Among 11 variables influencing the dependent variable (the financial statement quality), there are five variables having the positive effect on the financial statement quality, which are CEO duality, board independence, financial expertise level of Board, the existence of bonus plan and list time has positive effect on the financial statement quality. There are six variables having the negative effect on the financial statement quality, these variables are institutional ownership, firm liquidity, firm size, industrial sector, firm profitability (ROE) and dividend policy. The remaining variables, including foreign ownership, state ownership, managerial ownership, ownership concentration, board size, number of board meetings, firm leverage, firm age, list status, auditor type, delay of financial statement and firm growth do not influence the financial statement quality in terms of the statistical significance. 4.3.3. Regression analysis results of model 2 Before analyzing the regression model 2, the author has conducted the OLS estimation of the adjusted EBO model, which measures the dependent variable (the financial statement quality) in this model 2, the research result demonstrates that the modified EBO model is appropriate. 4.3.3.1. Appropriate model selection Based on the results of estimation and comparison among three models, which are Pooled OLS, FEM and REM, it is shown that the REM estimation is the most appropriate estimation for the model 2. 4.3.3.2. Test for appropriate model (1) The independent variables are not correlated with each other (it does not have the multicollinearity phenomenon). The result identifies that it is unlikely to have the multicollinearity phenomenon. (2) Assumption of normally distributed residuals The result indentifies that the residuals of this research model are normally distributed. (3) Assumption of constant variance The result find that this model has the changing variance. (4) Assumption of autocorrelation. 21 The result find that the model has first order autocorrelation Conclusions: Based on the test on 4 assumptions of the model, two assumptions are not violated, which are the multicollinearity phenomenon and the assumption of normally distributed residuals. However, the remaining two assumptions is violated, which is the assumption of autocorrelation and the assumption of constant variance. To overcome the violation of these assumptions, the author has to conduct the regression by the Feasible generalized least squares estimation (FGLS), this estimation method is as same as the research method conducted by Rafiee et al (2014). 4.3.3.3. Results of the beta and the statistical significance of model 2 The result of study show that, 10 of 23 independent variables entered to the model influence the financial statement quality. Among 10 variables influencing the dependent variable (the financial statement quality), there are three variables having the positive effect on the financial statement quality, which are firm leverage, auditor type and delay of financial statement has positive effect on the financial statement quality. There are seven variables having the negative effect on the financial statement quality, these variables are foreign ownership, firm size, firm growth, industrial sector, list status, firm profitability (ROE) and dividend policy. The remaining variables, including state ownership, institutional ownership, managerial ownership, ownership concentration, board size, CEO duality, board independence, number of board meetings, financial expertise level of board, the existence of bonus plan, firm liquidity, firm age and list time do not influence the financial statement quality in terms of the statistical significance. 4.4. Synthetic analysis results from two models and discussion The result summary of two models has find that there are 17 variables influencing to the dependent variable, which is the financial statement quality, including foreign ownership, institutional ownership, CEO duality, board independence, financial expertise level of board, The existence of bonus plan, firm leverage, firm liquidity, firm size, list time, list status, auditor type, delay of financial statement, firm growth, industrial sector, firm profitability (ROE) and dividend policy. The remaining variables, including state ownership, managerial ownership, ownership concentration, board size, number of board meetings and firm age have not influenced the financial statement quality of listed companies in Vietnam. 22 CHAPTER 5. CONCLUSION AND RECOMMENDATIONS 5.1. General conclusion In conclusion, the author not only focuses on the results relating to recommendations proposed in order to improving the financial statement quality, but also advises the related objects when using the financial statement information. Based on the analysis results in chapter V, the reality of financial statement quality through two measurements, which are based on the accounting basis and the market basis, of listed companies in Vietnam is not really high. About the determinants of the financial statement quality based on the research results, 14 of 23 tested factors influence the financial statement quality, and the results of the sign effect are multidimensional, and there are some results which are similar to the other researches, but there are also some results different from the other researches in the world, because the economic and social characteristics of Vietnam are different from the other countries. 5.2. Recommendations relating to the financial statement quality of companies listed on stock market of Vietnam 5.2.1. Recommendation proposed for the management agencies - The Ministry of Finance needs to promote the completion of regulations on the preparation and presentation of financial statements, such as the accounting standards, the accounting system, accounting law and other related regulations. - The Ministry of Finance needs to firmly manage the independent audit. + Continuing enhancing the external quality of the management agencies with the independent audit companies, especially the audit companies accepted to audit the financial statements of listed companies. + Continuing to complete the regulations relating to the operational audit, such as the auditing standards, the law on independent audit and the other related regulations. + The Ministry of Finance needs to increase the amount of audit companies accepted to audit the financial statements of listed companies, but have to ensure the sufficient quality in order to satisfy demands on the audit service of listed companies. + Strictly conducting the sanctions for violations of auditors and audit companies. 23 - A collection of criteria of the annual reports needs to focus on the financial information quality, especially the earning quality because this is a highly important criterion, which significantly influences the economic decisions of the objects using information in the financial statements as well as the annual reports. - The results of this research are rather essential for the management agencies when using information in the financial statements of listed companies in Vietnam. - The results are also the essential reference for stock exchanges in Vietnam. The stock exchanges need to focus on the criteria of financial statement quality measurement in their collection of criteria. 5.2.2. Recommendation proposed for audit companies - In the general meeting of shareholders, they may propose the decision of large audit companies, which have a high reputation and independence, in order to conduct an audit for companies. - The audit companies, which are not Big 4, need to enhance the audit quality when conducting the financial statement audit for the customers, especially for the listed companies, in order to protect the benfites of company owners, and enhance their trademark. - The auditor or the representative of audit companies has to request the listed companies whose financial statements are audited to invite to the annual General meeting of shareholders, in order to present the opinions about some problems related to the financial statements in the case that the audit report has significant audit exceptions. - When the independent audit companies in Vietnam conduct the financial statement audit for the special customers, especially for the listed companies, the audit companies have to enhance the independence of auditors, because if the auditors are more independent, the benefits would be inclined towards the public instead of the customers. - The audit companies should not use every ways to sign a contract with customers, including accepting the low audit fee, which may not offset the expense for conducting an audit contract in accordance with the auditing standards, and the companies have to ensure to conduct a high – quality audit. 24 - Have to improve the quality control of the independent audit internally. 5.2.3. Recommendation proposed for listed companies - Enhancing the supervisory and management role in improving the monitoring report of the Board of directors and the supervisory Board to the directors. - Have to invite the audit company and auditors to participate in the general meeting of shareholders in order to enhance the independent audit role. - The listed companies, especially the shareholders (the capital owners) wanting to enhance the reliability of financial statement information and protect their benefits, have to be more cautious about choosing the independent audit companies to audit their financial statements. - The companies have to increase the investment in human resources as well as facilities to quicken the completion time for the financial statements but ensure the financial statement quality. 5.2.4. Recommendation proposed for other users of the financial statement information The objects using information in the financial statements, especially the earning criteria, have to use this research result as a crucial reference, in order to observe the determinants of financial statement quality of listed companies in Vietnam to make the right economic decisions for them, as detailed below: - With the banks and the debtors: For the purpose of decreasing the credit risk for companies whose profits are unrealistic while loss is real in the credit appraisal process, the banks should not only observe the profit presented or significantly depend on the financial information of companies to make decisions, but also focus on other information, such as the feasibility of business plans (because this is the main source of repayment of companies) or the credit reputation of companies to observe the repayment ability as well as the willingness to repay of companies. Besides, the banks, when evaluating the solvency of customers to make a lending decision, should not only depend on the information in financial statements but also focus on many other references in the credit appraisal process. - With the investors and other users of the financial statement information: This 25 research result is fairly useful for the investors and other users of the financial statement information of listed companies in Vietnam. The users may use this research result as the reference to evaluate the reliability of information in financial statements, especially the earning information, so it can help the users to be more careful when depending on these information to make the economic decisions. 5.3. Some limitations and proposed research directions in the future In this research, although the amount of factors entered to the model are more than the other researches in the world and in Vietnam, including five factor groups which are group of factors relating to ownership structure, corporate governance, capital structure, market characteristics and performance of company, there are still many factors influencing the financial statement in Vietnam such as capital markets, State's policy, culture, politics, etc, including many important groups. Furthermore, the research has also measured the financial statement quality by earning quality based on the accounting basic and market basic, with each these basics, the author has chosen one representative measurement model, however, there are many other measurement models for each those basics. Therefore, the future researches may analyze more factors beside the factors tested in this research model and the future research may also use the other models to measure the financial statement quality. 26 CONCLUSION The information in financial statements of companies, especially the listed companies in Vietnam, is concerned by many users to make the economic decisions and use for many other works of the related objects. The research result of this thesis identifies that the financial statement quality measured by the earning management based on two basics, which are the accounting basics and market basics, of listed companies in Vietnam has still been lower than other countries in the world. Besides, one of the important contents about reaching the main goal of this thesis is testing the model of the determinants of financial statement quality of listed companies on stock market in Vietnam. The research result identifies that more than half of 23 tested factors influence the financial statement quality of listed companies and there are some factors having positive effects whereas some factors have negative effects and some factors do not influence the financial statement quality. Therefore, in the author’s opinion, the research result of this thesis and some recommendations proposed based on the research result are going to be an essential science fundamental for the related users, such as the company managers, auditors, investors, the State Securities Commission of Vietnam, the management agencies, etc., in order to suggest some solutions for improving the quality of financial information quality of listed companies in Vietnam, moreover, based on this research result, the objects using the financial statement information also make related decisions efficiently. 27 LIST OF RESEARCH PROJECTS HAVE BEEN PUBLISHED BY AUTHOR RELATED TO THE THESIS 1. Nguyen Thi Phuong Hong & Mai Thi Hoang Minh, 2014. The perceptions of users and preparer of various aspects of financial reporting in Viet Nam. Journal of economics & Development, 6/2014. 2. Nguyen Thi Phuong Hong & Duong Thi Khanh Linh, 2014. The perceptions of the qualitative characteristics of financial reporting of the Vietnam enterprise at the moment. Accounting & Auditing Review, 6/2014. 3. Nguyen Thi Phuong Hong & Nguyen Thi Kim Oanh, 2014. The factors affecting selection of accounting policies of Vietnam Enterprise at the moment. Financial & Monetary Market Review, 6/2014. 4. Nguyen Thi Phuong Hong & Ha Hoang Nhan, 2015. The factors affecting auditor independence: Evidence in Vietnam. Journal of economics & Development, 5/2015. 5. Nguyen Thi Phuong Hong and Nguyen Thi Thuong, 2015. Board Characteristics Affecting Earnings Management: An Empirical Investigation in Vietnam. Internationnal Conference on Accounting, ICOA 2015, 22 May 2015. YOKOHAMA National University, Aston University, Economic University Danang. 6. Nguyen Thi Phuong Hong, 2016. Proposed the recommendations to enhance the quality of financial statements of companies listed on Vietnam stock market. Nationnal Conference on Accounting, Auditing of Vietnam - 20 years Innovation and Integration, April 2016. Vietnamese Association of Accountants and Auditors, Accounting & Auditing Review.