The impact of capital structure on financial performance of joint stock commercial banks in vietnam

MINISTRY OF EDUCATION AND TRAINING STATE BANK OF VIETNAM BANKING UNIVERSITY HO CHI MINH CITY HO QUYNH BAO TRAN THE IMPACT OF CAPITAL STRUCTURE ON FINANCIAL EFFICIENCY OF JOINT STOCK

MINISTRY OF EDUCATION AND TRAINING STATE BANK OF VIETNAM

BANKING UNIVERSITY HO CHI MINH CITY

HO QUYNH BAO TRAN

THE IMPACT OF CAPITAL STRUCTURE ON FINANCIAL EFFICIENCY OF

JOINT STOCK COMMERCIAL BANKS IN VIETNAM

GRADUATION DISSERTATION MAJOR: FINANCE - BANKING

CODE: 7340201

HO CHI MINH CITY, 2018

MINISTRY OF EDUCATION AND TRAINING STATE BANK OF VIETNAM

BANKING UNIVERSITY HO CHI MINH CITY

HO QUYNH BAO TRAN

THE IMPACT OF CAPITAL STRUCTURE ON FINANCIAL EFFICIENCY OF

JOINT STOCK COMMERCIAL BANKS IN VIETNAM

GRADUATION DISSERTATION MAJOR: FINANCE - BANKING

CODE: 7340201

SUPERVISOR

MA TRAN KIM LONG

HO CHI MINH CITY, 2018

ABSTRACT

Author Tran, Ho Quynh Bao

Title The impact of capital structure on financial efficiency of

Joint Stock Commercial Banks in Viet Nam

Language English

Instructor Long, Tran Kim

The research presents the experiment on the impact of capital structure on financial efficiency of Joint Stock Commercial Banks in Vietnam The main objective of this study is to determine the overall effect of capital structure on the performance of these banks, establishing the existential relationship between the choice of capital structure of firms and the return on equity owned.Quantitative analysis of this thesis uses sample consisting of 28 Joint Stock Commercial Banks from 2010 to 2017 Research model makes use of panel data and regression made in 4 methods: Pooled Ordinary Least Square (OLS), Fixed Effects Model (FEM), Random Effects Model (REM) and Generalized Least Squares (GLS) Empirical results indicates there is negative relationship between deposits ratio and financial efficiency – represented by ROE, ROA, the reason behind this is that the stagnation of capital mobilization, speaking in another way, because of the asymmetrical development of capital mobilization and credit activities, the more mobilizing capital the banks get, the more inefficiency they perform Moreover, the study also finds out that the ratio of outstanding loans to total assets and financial performance have positive relationship, which reflects the current situation of Joint Stock Commercial banks that it is essential

to speed the credit activity up, utilizing maximum the capital mobilization Therefore, a number of recommendations are introduced to promote the financial performance of these banks

DECLARATION OF AUTHENTICITY

I hereby confirm that I am the sole author of the written these here enclosed and

I have compiled it in my own words

With my signature, I confirm that I have documented all methods, data and processes truthfully I have mentioned all people who were significant facilitators of the work

I declare that all statements and information contained herein are true, correct and accurate to the best of my knowledge

Ho Chi Minh City, December 24, 2018

Ho Quynh Bao Tran

ACKNOWLEGEMENTS

First and foremost, I would like to express my sincere gratitude to my advisor

Mr Tran Kim Long for his continuous support of my study, for his patience, motivation, enthusiasm and immense knowledge He has given me all the freedom to pursue my research, while silently and non-obtrusively ensuring that I stay on course and do not deviate from the core of my research His guidance helped me all the time

of writing this thesis I could not have imagined having a better advisor and mentor for

my study

Next, my acknowledgement would be incomplete with thanking the biggest source of my strength, my family I thank them for putting up with me difficult moments and guiding me to follow my dream of taking this degree This would not have been possible without their unwavering and unselfish love and support given to

me all the times

Finally, best regards to my lecturers, my friends and BUH for their sharing and support throughout my Bachelor program

TABLEOF CONTENTS

ABSTRACT i

DECLARATION OF AUTHENTICITY ii

ACKNOWLEGEMENTS iii

LIST OF ABBREVIATIONS viii

LIST OF TABLES ix

CHAPTER 1: INTRODUCTION 1

1.1 Motivation 1

1.2 Research goal 3

1.3 Research questions 3

1.4 Research subject and scopes 3

1.5 Methodology 3

1.6 Research contribution 4

1.7 Research outline 4

CHAPTER 2: LITERATURE REVIEW 6

2.1 Definitions 6

2.1.1 Capital structure 6

2.1.2 Financial performance 7

2.1.3 The performance of Joint Stock Commercial Banks 8

2.2 Theoretical background 10

2.2.1 Capital Structure Theories of Modigliani and Miller (M&M) 10

2.2.2 Trade-off theory of Capital structure 11

2.2.3 Pecking Order Theory 11

2.2.4 Agency Theory 12

2.3 Previous studies on the relationship between capital structure and financial

efficiency 13

Conclusion of chapter 2 16

CHAPTER 3: RESEARCH DESIGN 17

3.1 Research methods 17

3.2 Data collecting and processing method 17

3.3 Measurement of research variables 17

3.3.1 Dependent variables 17

3.3.2 Independent variable 18

3.3.3 Other controlling variables 18

3.4 Research model 19

3.5 Hypothesis 19

3.6 Research procedure 21

3.6.1 Descriptive analysis 21

3.6.2 Variance Inflation Factor test 21

3.6.3 Stage of regression models 22

Conclusion of Chapter 3 24

CHAPTER 4: EMPIRICAL RESULTS 25

4.1 Descriptive statistics 25

4.2 Correlation Matrix and multicollinearity VIF 26

4.3 Regression results 28

4.3.1 Testing conformability of Pooled OLS, FEM, REM models 28

4.3.2 Choosing the best fitted model 28

4.3.3 Other relevant tests 29

4.4 Regression result 31

4.5 Discussion about research results 32

Conclusion of Chapter 4 36

CHAPTER 5: CONCLUSION AND RECOMMENDATIONS 37

5.1 Conclusion 37

5.2 Recommendations 38

5.2.1 Solutions to improve business performance from the perspective of capital structure 39

5.2.1.1 Steady increase in payables 39

5.2.1.2 Enhancing the efficiency of credit operation 41

5.2.2 Support solutions group 44

5.2.2.1 Financial solutions 44

5.2.2.2 Personnel solutions group 47

5.2.2.3 Technological solutions group 48

5.2.2.4 Management solutions group 50

5.2.2.5 Constructing and complete the legal enviroment 51

5.2.2.6 Assist banks to carry out M & A activities 51

5.2.2.7 Supporting for the process of restructure and modernizing the banking system 52

5.2.2.8 Promulgate regulations guiding the accounting according tointernational accounting standards 52

5.2.2.9 For the State bank 53

5.3 Limitations of the thesis 54

5.4 New approaches in the future 54

REFERENCES 56

APPENDIX 1 60 APPENDIX 2 68

LIST OF ABBREVIATIONS

JSCBs Joint Stock Commercial Banks

SMEs Small and medium enterprises ADI Association of Deposit Insurers

M & A Mergers and Acquisitions

LIST OF TABLES

Table 4.4 Summary of regression results of Pooled OLS, FEM,

REM models

28

Table 4.7 Regression result of the impact of capital structure on

financial efficiency using GLS method

31

CHAPTER 1: INTRODUCTION 1.1 Motivation

For any business, one of the most important financial indicators is the profitability This financial indicator is influenced by many other factors, including capital structure This is understandable because of a fluctuation of loan rates that changes the average capital cost of commercial banks and leads to variable rates of return However, the effect direction of this factor on the profitability of the Joint Stock Commercial Banks is a matter of study

Capital structure, which simply means the capital formation method, usually a blend of loan and equity capital, through which a firm is financed, has been the issue of interest to many researchers in which they aimed to observe the connection of capital structure with performance of firms The decision of how a firm will be financed is imperative to both the managers of the firms and fund suppliers This is because, first

of all, if financing is done through employment of wrong mix of debt and equity, it may acutely influence the performance and even endurance of the firm Secondly, in order to maximize the firm value, managers should undertake capital structure decision, which is a complex task, as because use of leverage varies from one firm to another Therefore, what managers usually do is they try to achieve the best combination of debt and equity in their capital structure Recognizing the importance, there has been many studies which tried to inspect the affiliation of capital structure with performance of firms different researchers Deesomak (2007) found that impact of capital structure on bank performance to be negative In connection with this issue, Modigliani and Miller (1958) primarily stated that, under perfectly competitive capital market conditions, the firm value is free from the influence of capital structure decisions Instead, they argued that the firm value is determined solely by its basic earning power Later, however, they proposed, by taking the effect of tax advantage on debt, that the firm value can be increased by incorporating more debt into the capital

structure and thus the optimal capital structure of a firm should be made up of a

hundred percent of debt (Modigliani and Miller 1963)

Following the above mentioned theories, a chunk of studies observed the impacts of capital structure on firms’ performance even though their findings are mixed These mixed evidences create an avenue for the researcher to explore and establish the influence of capital structure decision on firms’ performance Besides, in the context of developing country like Vietnam, there exist few studies linking between the same Even, none of those studies focused on the banking sector performance For example, Hiep (2017)studied a study on manufacturing companies over the period of 2007-2013 by GMM model In a similar study, which excludes performance of bank sector, by using data of 2010 - 2015, Phuong (2017)conducted whether manufacturing firms are affected by capital structure Long (2016) studied the association of firm value with capital structure choice Authors used data over a period of 2008-2014 for non-financial firms and thus ignored banking sector performance From above discussions it is evident that to date, there is no study focused on linking the issue of capital structure and performance of Joint Stock Commercial Banks (JSCBs) in Vietnam In this present research an endeavour has been taken to explore the link of capital structure decisions to banks’ performance as banking sector is considered as the most sound and dominant sector in Vietnam

From the survey of a number of issues and typical research in the banking sector, it raises a question: Does capital structure have a great impact on the financial performance of JSCB in Vietnam? Starting from the above, the author said that the

study "THE IMPACT OF CAPITAL STRUCTURE TO FINANCIAL EFFICIENCY OF JOINT STOCK COMMERCIAL BANKS IN VIETNAM" is

really necessary Based on the results of this research, some will help the bank's managers to have a holistic view, thereby developing appropriate, long-term strategies

to reduce bad debts, losses and improve operational efficiency, while increasing the competitiveness of the bank in the macro-economic context of the integration period

- Proposing some solutions focusing on capital structure to improve the financial efficiency of JSCB in Vietnam

1.3 Research questions

To fulfill the above reseach’s goal, this thesis’ significant mission is to answer these following questions:

 Does capital structure affect the financial performance?

 If yes, how capital structure affects the financial efficiency of JSCB in Vietnam?

1.4 Research subject and scopes

 Research subject: This thesis subject is the relationship between capital structure and financial efficiency of JSCB in Vietnam

 Research scopes:

Space: The thesis uses data of 28 JSCB in Vietnam

Time: The thesis uses secondary data collected from the annual financial statements audited and published by JSCB in Vietnam during the period 2010-2017

1.5 Methodology

With the research goal is to study the relationship between capital structure and financial efficiency of JSCB in Vietnam, the thesis applies mainly quantitative method

The thesis combines statistical analysis and multivariate regression analysis to assess the impact of capital structure on the financial performance of JSCB in Vietnam The software used for quantitative analysis is Stata because it provides necessary tools that the author needs

1.6 Research contribution

With research goal “The impact of capital structure on financial efficiency; evidence from 28 Joint Stock Commercial Banks in Vietnam”, the thesis have following contributions:

Academically, the thesis provides a empirical evidence about the relationship

between capital structure and financial efficiency of 28 JSCBs in Vietnam Consequently, it can support related theories and viewpoints about the relationship between capital structure and financial efficiency of 28 JSCBs in Vietnam Moreover, the thesis’ results could be used as an initial basis for further studies in this field

Experumentally, the thesis provides a specific point of view about the

relationship between capital structure and financial efficiency of 28 JSCBs in Vietnam

In consequence, it can give recommendations to managers as well as investors to take appropriate investment decisions

Chapter 2: Literature review

This chapter will provide theoriotical basis for research content and study existed researches related with the topic at the same time Giving out problems related

to captital structure and collecting factors affecting the financial performance in commercial banks as well

Chapter 3: Research design

This chapter will provide information on the research method and design for the thesis Then it describes regression models developed for the study The model investigates the impact of capital structure on financial performance that defines return

on assets (ROA) and return on equity (ROE) as dependent variable and total dept to total assers as independent variables It also considers controlling variables such as firm size, the ratio of outstanding loans to total assets, the ratio of deposits to total assets of the Joint Stock Commercial Banks

Chapter 4: Empirical results

This chapter will present research empirical results after carrying out research model presented in chapter 3 It provides descriptive statistics, correlation matrix for all variables, Variance Inflation Factor test, multicollinearity test, heteroskedasticity test, autocorrelation test, GLS method and overall regression model By going through the results of each variable, the last part of this section discusses and compares results with earlier studies

Chapter 5: Conclusion and recommendation

At the end, this chapter will conclude research issues, give conclusion from the analysis and come up with recommendations for managers, investors and policy makers Besides, the limitations of the research as well as directions for further researches are mentioned

CHAPTER 2: LITERATURE REVIEW 2.1 Definitions

2.1.1 Capital structure

Capital structure is defined as the composition of all the securities the firm issues in order to finance its operations (Brav & Maug, 1998) Capital structure is the way a firm combines equity and debt to gain the maximum value The value of a firm

is therefore defined as the market value of debt plus the market value of equity (Ross, Westerfield, Jaffe & Kakani, 2009) Capital structure of any institution should therefore be well managed to ensure that the firm is remains in operation and it is able

to finance its projects Therefore, the way a bank combines its debt and equity, will define its market value as noted by (Ross et al., 2009)

Capital structure, in terms of scholarship,is understood as a term that describes the source and method of capital formation that an enterprise can use for the purchase

of assets and facilities tangible and supportive for production and business activities1 Moreover, the capital structure is defined as the combination of regular short-term debt, long-term debt, preferred stock, and equity that are often used to finance corporate investment decisions (Tran Ngoc Tho et al., 2007)

In addition, in the financial sector, Brounen et al (2005) asserted that capital structure refers to the way in which a company or business finances its mobilization by combining short-term and long-term capital (including equity, debt and hybrid) Besides, some academic theories in recent years have suggested that capital structures are also referred to as financial leverage or "corporate debt structure" and explain debt swings throughout the company (Çağlayan and Şak, 2010)

As we can see capital structure has been defined by many authors and scholars However, these definitions are explicit and have the same meaning This research work adopts that of Pandey which says “a company’s capital structure refers to its dept level relative to equity on the balance sheet It is a snapshot of the amounts and types of

capital that a firm has access to, and what financing methods it has used to conduct growth initiatives such as research and development or acquiring assets”

From those definition, the concept of capital structure can be defined as in the proportional relation between a firm’s debt capital and equity capital Firms use capital structure usually to fund their business and expand This decision is vital for a firm as it has a direct influence on the risk and return of a firm

2.1.2 Financial performance

Financial performance and financial profitability are frequently used as interchangeable terms, (Burkhardt & Wheeler, 2013) With the increasing number of analyses and research papers referencing financial performances, there is a need to have basic understanding of definition of financial performance and its various measures, (Burkhardt, 2013) Therefore, choosing a particular measure of financial performance depends on how well it meets the intended purpose Financial performance of a bank is defined as its capacity to generate sustainable profitability, (European Central Bank (ECB), 2010)

Financial performance refers to the act of performing financial activity In broader sense, financial performance refers to the degree to which financial objectives being or has been accomplished It is the process of measuring the results of a firm's policies and operations in monetary terms It is used to measure firm's overall financial health over a given period of time and can also be used to compare similar firms across the same industry or to compare industries or sectors in aggregation (Dr Donthi Ravinder, Muskula Anitha, 2013)

Some think that performance is the total market value of a firm or the sum between market value of equity and value of equity options (Cole and Mehran, 1998; Merz and Yashiv, 2007) Either way, firm performance reflects how effectively companies manage their resources

Financial performance, in a 1broad sense, reflects the organization's efficiency and financial management It depends primarily on skill, talent, ingenuity, and motivation Management is responsible for the business operations of a business, making financing decisions, investing and business decisions Management also selects actions, or strategies and plans Financial performance plays an important role in the planning, budgeting, pricing and control of business operations

In terms of narrowness and popularity, financial performance is an indicator reflecting financial performance and is used to forecast income of enterprises The measurement of financial performance can be affected by the target of the enterprise, which is measured by the performance of the business and the development of the stock market Measurement of financial performance by accounting by: return on equity - ROE, earnings per share - EPS; Market price by price to earnings ratio - P / E, market capitalization of equity on the book value of equity - MBRV If the stock market is underdeveloped, measuring market efficiency will not produce good results

If measuring by book value, financial performance shows the relationship between income and other indicators, built from the financial statements of the enterprise, so the results depend on the accuracy and the medium authenticity of financial statements;

So, financial performance can be affected by the target of the enterprise Financial performance measured by ROE is most commonly used

2.1.3 The performance of Joint Stock Commercial Banks

Business performance is a category that shows in-depth development, reflecting the level of utilization of resources at the least cost that results in the highest efficiency Business performance in a bank can be assessed in two ways: qualitative or quantitative On the qualitative side, performance is assessed through the bank's decision-making process in the course of its operations Typically, policies and strategies are considered suitable Is there an effective monitoring and control system?

It is also important to evaluate the professional level of the staff that ensures the

successful implementation of policies, effective implementation of strategies of managers These criteria will help us evaluate whether the bank is operating and developing effectively or not qualitatively

On the quantitative side, there are many models, indicators to evaluate business performance In general, to quantify, determine the performance of business

in all aspects, we often use quantitative indicators Because of the qualitative indicators are subjective of evaluators, the quantitative indicators can probably tell us whether the bank really work effectively or not Quantitative indicators are often used as profit targets, which reflects the business situation generally Researchers often use models such as Internal Rate of Return (IRR), Cash Flow Return on Investment (CFROI), Discounted Cash Flow (DCF); along with other indicators such as Return on Equity (ROE), Return on Assets (ROA), Earning Per Share (EPS), In which, the most widely used indicators are ROE and ROA, as we can see in many previous researches such as Mathewos Woldemariam Birru (2016), Abbadi, S.M & Abu-Rub, N (2012), Salim and Yadav (2012), Tobin’s Q.Manawaduge et al (2011), Hall and Weiss (1967), etc

Return on Equity = Net income / Shareholder’s Equity

ROE is a ratio that provides investors with insight into how efficiently a company (or more specifically, its management team) is handling the money that shareholders have contributed to it In other words, it measures the profitability of a corporation in relation to stockholders’ equity The higher the ROE, the more efficient

a company's management is at generating income and growth from its equity financing

Return on Assets = Net income / Total Assets

ROA is an indicator of how profitable a company is relative to its total assets ROA gives an idea as to how efficient management is at using its assets to generate earnings Calculated by dividing a company's annual earnings by its total assets, ROA

is displayed as a percentage Sometimes this is referred to as Return on Investment (ROI)

2.2 Theoretical background

2.2.1 Capital Structure Theories of Modigliani and Miller (M&M)

In the process of developing theories of capital structure, the initial major contributing research work of Modigliani and Miller (1958) The first point of Modigliani and Miller (often referred to as M & M) in 1958 suggested that enterprise value was independent of capital structure With perfect capital market assumptions such as no transaction costs, a loan-to-loan ratio and the same amount of free debt and taxation are ignored, it means the advantage is that loans are cheaper and less risky for investors; And one disadvantage is that the cost of equity will increase with debt The

M & M argument has proved that the two effects are balanced; That is, debt consolidation gives owners higher returns, but this is what the owner must compensate for the increased risk of a loan to total capital Because this theory is based on limited assumptions and is not very relevant to reality, the application of this theory to the real world is still limited In 1963, M & M launched a follow-up study, reviewing their previous study in 1958 by combining the benefits of corporate income tax as an important part of the decision on capital structure In particular, Modigliani and Miller (1963) state that the cost of equity of a levered firm increases with the debt-to-equity ratio Accordingly, the presence of corporate income tax, the use of debt will increase the value of the business Interest expense is tax deductible; Thus, businesses that are obliged to pay taxes will benefit from the tax shield The higher the debt, the higher the value of the business and the maximum increase when the business is funded by 100% debt

Otherwise, some researchers have criticized M & M (1963) that the value of a business is not grows indefinitely when they use debt, because when it comes to debt, there arises financial distress costs This expense is formed when the profit of the business is not sufficient to cover the debt and interest expenses In addition, leverage businesses also lead to representative costs, which are not just a conflict between shareholders and managers because of the separation of management and ownership of

the business but also the conflict of interest between the lender and the borrower (Jensen and Meckling , 1976)

In summary, M&M theories say that when there are no taxes, a company’s value

is not affected by its capital structure and its cost of equity increases linearly as a function of its debt to equity ratio but when there are taxes, the value of a levered company is always higher than an unlevered company and cost of equity increases as a function of debt to equity ratio and tax rate Despite M&M require certain unrealistic assumptions such as: existence of a totally efficient market with no transaction costs,

no financial distress and agency costs, ability to borrow and lend at the risk-free rate, its about capital structure offer a good starting point in a company’s quest for optimal capital structure

2.2.2 Trade-off theory of Capital structure

Trade-off theory of Capital structure explains why firms are typically funded by

a mix of debt and equity Apart from the existence of benefits from the tax shield from debt, there is also the expense of financial exhaustion With each percentage point increase in debt, as the tax shield benefits increase, the cost of financial exhaustion also increases At the time when borrowing more debt, the current benefit from the tax shield is no higher than the financial distress costs anymore, the debt is no longer beneficial to the business Therefore, when financial managers choose a reasonable loan ratio based on the principle of balance, it is also the optimal capital structure of the business This structural capitalist theory was studied by Kraus and Litzenberger (1973), Myers (1977) and Myers (1984)

2.2.3 Pecking Order Theory

Myers and Majluf (1984) began to study the trade-off theory of capital structure Pecking order theory is viewed as opposed to trade-off theory of capital structure, assuming that there is no apparent optimal capital structure for a company This theory is based on the results of asymmetric information, which leads managers to

better understand the business value, risk, and potential future performance of the business rather than the investor This gives managers the choice between internal and external funding; New issuance of debt securities and equity securities This leads to a grading order, starting with internal capital or retained earnings, mainly reinvested profits; Following by debt issuance and finally new equity issuance (Myers, 1984; Myers and Majluf, 1984)

2.2.4 Agency Theory

The Agency theory is about the relationship between principles (owners) and the agents (managers) Principles engage agents to perform some of their duties on their behalf A typical research of Jensen and Meckling (1976) has explained that this happens because of the separation of ownership and control when the owners of the company have to employ managers to run their business and monitor their performance that they act for the owners’ interests and in asymmetric information environment, agency cost arises due to conflicting interests of the managers and owners The individual utility maximization is a common objective of managers andowners If both parties are concerned about their own utility maximization, managers would not always act for the best interests of the owners, they can retreat from their difficult tasks or act out of self-interest The owners then can implement incentive plans, monitoring cost to align managers’ activities toward their interests or limit their aberrant activities Shareholders (owners) are the one who bear the costs and expenses caused by these conflicting interests To lessen the conflicts, some say managers should be shareholders

of the company so that their interests would be as same as the owners’

Jensen (1986) has published a research in which he stated that dividend payment is used as a tool to neutralize conflicts among managers and shareholders because managers are more likely to retain financial resources instead of paying dividends Managers prefer following growth strategies for their firm as the more developed the company gets, the more power to control those resources they can gain Meanwhile, shareholders are more interested in dividends than retained profits In

consequence, conflicts that arise among them can be solved through dividend payment policy

According to Rozeff (1982), Easterbrook (1984) dividend payment is a way to constrain agency cost by forcing the company to approach capital market to raise external capital, by which gains more supervision La Porta et al (2002) argued that dividend payment can reduce agency cost because (1) managers are forced to seek external capital to replenish funds paid out in dividends, (2) managers are forced to seek external financing for their projects, therefore, they would have to provide more internal information and (3) free cash flow would go up with increased profits and decrease in debts and it would reduce the opportunity for managers to use free cash flow for their perquisites activities

Moreover, many researchers have argued on a point that institutional investors have positive effect on agency problem by cutting down on agency cost through monitoring managers’ activities (Han et al, 1999) and by influencing dividend policy (Eckbo and Verma, 1994)

2.3 Previous studies on the relationship between capital structure and financial efficiency

There are several empirical studies that have observed the association of capital structure decisionwith the performance of firms Some of them have noticed a positive impact, while others have noted that either a negative effect or no effect

 Positive relationship

Saeed, M, Gull, A, Rasheed, M (2013) after empirically carrying out whether the capital structure has impact on the financial performance of banks in Pakistani or not during the period 2007 – 2011 The performance was measured by return on assets (ROA), return on equity (ROE) and earnings per share (EPS) Determinants of capital structure included long term debt to capital ratio, short term debt to capital ratio and total debt to capital ratio The research concludes that there are a positive relationship between determinants of capital structure and performance of banking industry

Saeed et al., (2013) found out that there exists a positive relationship among capital structure and profitability of Pakistani banks As a result of a strong positive dependence of short term debt to capital (STDTC) on all profitability measures (ROA, ROE and EPS) Long term debt to capital (LTDTC) having a negative relationship with return on assets (ROA), return on equity (ROE) and earnings per share (EPS) Total debt to capital and firm size (SIZE) experienced a strong optimistic connection with all dependent variables (ROA, ROE and EPS)

Nikoo (2015) conducted a paper to measure the impact of capital structure on bank performance The analysis is performed using accounting and market measures of performance on a sample of 17Iranian banks over a period of 2009–2014,which published by Tehran Stock Exchange Most researchers used the proxy ROE as accounting performance measure while Tobin’s Q, is used to measure the market performance of the firms The independent variables used in both measures are the bank deposits total assets, total bank loans and other investment and net profit The empirical results indicate that capital structure has a significant positive effect on bank performance

Argumants of Julius B Adesina and Nwidobie (2015) also agreed that there has

an impact of post consolidation capital structure on the financial performance of Nigeria quoted banks By using profit before tax as a dependent variable and two capital structure variables (equity and debt) as independent variables for the time horizon of 8 year with ordinary least square regression, the study determined that there exists to be a significant positive relationship between capital structure and the financial performance of Nigeria quoted banks

Pinto & Quadras (2016) when examining the impact of capital structure on financial performance Indian banks with sample of 21 banks from both public sector and private sector in a period of five years, the experimental research has been found that the results of the hypothesis testing reveals significant impact of debt equity ratio and debt to total assets on the net profit, net interest margin as well as return on capital

employed indicating that capital structure has a significant impact on the financial performance in the banking industry

 Negativerelationship

Vitor, D.A & Badu, J (2012) conducted the the impact of capital structure and performance of listed banks in Ghana and they found a negative relationship between capital structure and financial performance High level gearing among listed banks lead

to over dependency on short term debt due to which there was high lending rate and low level of bond market activities The regression analysis revealed that there is an inverse connection between capital structure and banks performance in terms of return

on equity

Taani, K (2013) examined the impact of capital structure on performance of Jordanian banks To estimate the relationship between capital structure and banking performance multiple regressions is applied on performance indicators and capital structure variable Net profit, return on capital employed and net interest margin was used to measure the bank performance which showed positive relation with total debt, but total debt is insignificant in determining return on equity in the banking industry of Jordon This argument supports the idea that there exists to be a negative and insignificant relationship between capital and firm performance

Supporting the above statement,Mathewos Woldemariam Birru (2016) found evidence proving the negative relationship between capital structure and financial performance when he examined the correlation among them By using secondary data collected from financial statements, empirical study on financial efficiency of 8 commercial banks in Ethiopia over the past five year period from 2011 to 2015 used fixed effect regression model to estimate the relationship between the capital structure and firm performance measured by ROA and ROE The results show that financial performance, which is measured by both ROA, is dramarically and negatively associated with capital structure proxies such as DER, SIZE and TANG whereas DR have negative impact Therefore, based on the findings gained from the results, the

study suggests recommendations that the commercial banks of Ethiopia should focus

on the proportion of debt used by the bank, the manner of utilizing the resources while expanding the banks and the amount of investment on fixed asset

Conclusion of chapter 2

The purpose of chapter 2 is to cover a summary of theoretical literature on the capital structure Based on M & M Theory, Trade-off Theory, Peckin Order Theory, Agency Theory, the author also builds a robust foundation for this study Moreover, many previous empirical studies are listed to consolidate the framework of the author The results are mixed The chapter ends with the gaps from previous studies and the contribution of this thesis to fulfil those gaps

CHAPTER 3: RESEARCH DESIGN

3.1 Data analysis tool

The data collected for the study is in the form of table data Therefore, when estimating the regression model the data table has many different approaches We can access the model using the pooled model (Pooled), or approach through the Fixed Effect Model (FEM), also known as the regression model LSDV (Least Squares Dummy Variable) and Random Effects Model (REM) In this paper, the author uses the Lagrange Multiplier test to select the appropriate regression model between the two Pooled or REM models; and Hausman testing to decide on the FEM or REM model From the test results, the author select the most suitable regression model

3.2 Data collecting and processing method

The research uses the data collected from the financial statements audited by 28 JSCB in Vietnam in the period from 2010 to 2017 The author excludes data from JSCB without adequacy The data selected from the financial report is the balance sheet of the banks' website as well as the website of vietstock.vn

3.3 Measurement of research variables

3.3.1 Dependent variables

This study uses two indicators: Return on Assets (ROA) and Return on Equity (ROE) as dependent variables to measure the performance of a bank In spite of the study do not have a single measurement that assert the financial performance will be measured by any indicators, the two ROAs and ROEs are often chosen because they are a measure of corporate financial performance, which are widely accepted ROA is calculated by subtracting the after-tax profit from total assets ROE is calculated by dividing profit after tax by equity

Authors such as Grossman & Hart (1986), Gleason & Mathur LK and Mathur I (2000), Phillips & Sipahioglu (2004), Abor (2005), Abor (2007), Ebaid (2009), Saeedi

& Mahmoodi (2011) , San & Heng (2011), Ahmad, Abdullah & Roslan (2012), and,

Mesquita & Lara (2003) Aloy Niresh (2012), John Gartchie Gatsi (2012), Saeed, Gull

& Rasheed (2012), Koranteng (2012), AM Goyal (10/2013) also used ROA and ROE

in their research in order to measure financial efficiency in the business

3.3.2 Independent variable

Based on the Koranteng study (2012), the specific of the balance sheet as well

as the nature of liabilities in the bank is short-term debt; therefore the independent variable of the model in this study is the Total Debt to Total Asset (TD), which represents the capital structure of the bank

3.3.3 Other controlling variables

Other factors are likely to affect the banking performance For that reason, it is necessary to introduce control variables into the model, which is further explained in the model The three control variables included in the model are: The ratio of outstanding loans to total assets (LOANS), the ratio of deposits to total assets (DEPS), size of the bank (SIZE)

Table 3.1: Variable Definition and Measurement

Dependent variables (Financial efficiency)

Return on Assets ROA Profit after tax/ Average total assets

shareholders’ fund

Independent variable (Represent for capital)

Total Debt to Total Assets TD Total liabilities/ Total assets

Control variables

The ratio of outstanding

Loan outstanding/ Total assets

The ratio of deposits to DEPS (Mobilized capital from customers

total assets + Mobilized capital from financial

institution)/ Total assets Size of the bank SIZE Natural logarithm of Total Assets

Source: The author selected variables based on previous studies

3.4 Research model

In order to verify the accuracy of these hypotheses, the models used in this study included:

The impact model of capital structure on the return on assets (ROA)

ROAi,t= β0 + β1TDi,t + β2DEPSi,t + β3LOANSi,t + β4SIZEi,t + ℮i,t

The impact model of capital structure on the return on equity (ROE)

ROEi,t = β0 + β1TDi,t + β2DEPSi,t + β3LOANSi,t + β4SIZEi,t + ℮i,t

Where:

• i: Bank i (i = 1,2…,31)

• t: Time of the year t (t = 1,2…,5)

• ß 0 :constant or the value of ROA/ROE when all values of X are zero

• ß 1 … ß 4 :Slope ofthe independent variables

• ℮ i,t : The error term

• ROA:The return on assets

• ROE:The return on equity

• TD:Total Debt to Total Assets

• DEPS: The ratio of deposits on assets

• LOANS: The ratio of outstanding loans to total assets

• SIZE: The size of the bank

3.5 Hypothesis

Total Debt to Total Assets (TD)

According to the findings of Abor (2005), the author found that short-term debt ratios are proportional to the ROE of firms in Ghana In 2007 he consolidated his research by continuing to find that short-term debt ratios were directly proportional to

the ROA of South African companies He argues that this is due to the fact that short term debt has lower expense than long-term debt Similarly with the results of later papers in the banking sector such as Gatsi (2012), Saeed et al (02/2013) and Goyal (10/2013) Moreover, as mentioned, most of the liabilities in the JSCB are short term; therefore, the hypothesis here is:

-H1: There is the positive correlation between short-term debt ratio and return on assets (ROA) and return on equity (ROE)

The ratio of deposits on assets (DEPS)

Deposits are the main fund for the bank to operate as its financial intermediary Bank profits depend on how they seek low-cost financing to provide loans to

customers, resulting in profit from the differences According to the study of

Koranteng (2012), this ratio has the positive correlation with ROA and ROE However, the relationship is the positive or negative, it depends on the cost of capital mobilization, the actual situation as the ratio between mobilization and lending; If too much money can not be loaned, this will also negatively affect the operation of the bank Therefore, this relationship needs to be verified with the data of the JSCB in Vietnam and the hypothesis here is:

H2: There is the positive correlation between ratio of deposits on assets and return on assets (ROA) and return on equity (ROE)

The ratio of outstanding loans to total assets (LOANS)

The main income of financial intermediation as the bank is from interest rates through lending Thus the more lending the bank are, the higher revenue they expect, which means the profits also higher According to Koranteng (2012), the research result also show that the loan outstanding balance and the performance of the bank represented by ROA and ROE have the same impact Therefore, the hypothesis here is:

H3: There is the positive correlation between the ratio of loan outstanding balanceand return on assets (ROA) and return on equity (ROE)

Bank size (SIZE)

Penrose (1959) argued that the larger companies are, the more profit could be made In the banking sector, for this indicator, the research papers have many controversal results In Ghana, Gatsi (2012) in his study found a negative relationship between bank size with ROA and ROE In contrast, the results of Saeed et al (02/2013)

in Pakistan found a similar relationship between bank size with ROA and ROE Goyal's research (10/2013) shows that bank size has a strong correlation with ROA but

is negatively correlated with ROE From these results, it is necessary to re-examine the relationship between size and financial efficiency of banking operations Hence, the hypothesis here is:

H4: There is a relationship between size of the bank and return on assets (ROA) and return on equity (ROE)

3.6 Estimation procedure

In order to the fit of the recommended model and determine the factors influencing capital structure of Joint Stock Commercial banks in Vietnam, the author uses regression analysis on Stata software with Return on Asset (ROA) and Return on Equity (ROE) as dependent variable Meanwhile, independent variables consist of TD (Total dept to total assets) and controlling variables DEPS (the ratio outstanding loans

to total assets), LOANS (The ratio of deposits to total assets), SIZE (Bank's size)

3.6.1 Descriptive analysis

In this paper, at the beginning, descriptive analysis is implemented Results will

be describe including observation, mean, standard deviation, max, min of all variables

in the model

3.6.2 Variance Inflation Factor test

After that, the author uses correlation analysis to show the level of relationships among pairs of variables via correlation matrix If the correlation coefficient is

positive, there is a positive relationship between two variables in a pair; if the coefficient is negative, the relationship is negative

With the purpose of VIF test of the regression model, model Pooled OLS is conducted before carrying out multicollinearity test When the Variance Inflation Factor (VIF) value of one variable is greater than 5, there is strong multicollinearity and this variable needs to be removed (Hair et al, 1995)

3.6.3 Stage of regression models

 Next, Hausman test is conducted to find out whether FEM and REM is suitable

Fixed effects model has a hypothesis that there is correlation between independent variables and error while random effects model does not If the hypothesis that there is no correlation between independent variables is not satisfied, model analyzed with REM model will easily cause endogenous problem and random effects estimation will have inconsistency As such, Hausman test is used by comparing estimations from FEM and REM according to these hypotheses:

Hypothesis H 0 : ßFEM = ßREM

Hypothesis H 1 : ßFEM ≠ ßREM

If p-value < 0.05: reject H0, FEM is more suitable

If p-value > 0.05: accept H0, REM is more suitable

 Redundant Fixed Effects test is then conducted to choose between FEM and Pooled OLS

According to Free (2004), before executing FEM regression, it is necessary to test is there exist to be heterogeneity between cross section units, which means whether FEM model is conformable or not to solve the solution This test is call redundant fixed effects test The hypothesis are as follows:

Hypothesis H0: α1 = α2= … = αn= α; intercept varying between cross section units are alike or in other words, individual specific of each subject does not exist

Hypothesis H 1 : There is at least one of intercepts α1, α2, , αnwill be different depends on any subject

If p-value < 0.05: reject H0 , concluded that there does not exist to be individual specific of each cross section unit FEM is more conformable

If p-value > 0.05: accept H0, Pooled OLS is more suitable

 Heteroskedasticity Breusch Pagan test is applied

In statistics, the Breusch Pagan test, developed in 1979 by Trevor Breusch and Adrian Pagan, is used to test for heteroskedasticity in a linear regression model It is a statistical test that establishes whether the variance of the errors in a regression model

is constant : that is for homoskedasticity

Hypothesis H 0 : There is no heteroskedasticity

Hypothesis H 1 : There is heteroskedasticity

If p-value ≤ 0.05: reject H0, There is heteroskedasticity

If p-value > 0.05: accept H0, There is no heteroskedasticity

 Wooldrigde test is conducted to test for autocorrelation in the residual

Iterated GLS with autocorrelation does not produce the maximum likehood estimates, so we cannot use the likelihood-ratio test procedure, as with heteroskedasticity However, Wooldridge (2002, 282–283) derives a simple test for autocorrelation in panel-data models It tests whether the variance of the errors from a regression is dependent on the values of the independent variables In that case, heteroskedasticity is present

If Hypothesis H0: is rejected, which means there exists to be autocorrelation at least a level and in contrary, we accept Hypothesis H0, which means there is no autocorrelation

Hypothesis H 0 : There is no autocorrelation in the residual

Hypothesis H 1 : There is autocorrelation in the residual

If p-value ≤ 0.05: reject H0, There is autocorrelation in the residual

If p-value > 0.05: accept H0, There is no autocorrelation in the residual

CHAPTER 4: EMPIRICAL RESULTS

4.1 Descriptive statistics

Descriptive statistics on variables used in the model is presented in table 4.1 below Statistic data give us a wide-spread view on varibles’ characteristics such as mean value, standard deviation value, minimum value, maximum value From 28 JSCBs, in the period of time from 2010 to 2017, we have 224 observations

Table 4.1 Descriptive statistics

Variable Observation Mean Std Dev Minimum Maximum

Source: Data collected by the author and calculated on Stata software

Through the descriptive statistics table above, it can be seen that the smallest value of the ROA variable is approximately 0 (0.0001) because the banking system in Vietnam has been in the restructuring period, some weak banks perform merger activities – equitization as a result financial indicators can not be shown in a few years

In addition, we also draw the following information:

Firstly, the mean of return on asset (ROA) for the entire sample in the period 2010-2017 is 0.76% on average, with a standard deviation of 0.67% The highest ROA

is 5.51% Secondly, the summary table shows that the return on equity (ROE) has an average value of 8.27% and a standard deviation of 7.71% with the highest ROE is 82.01%

When it comes to independent variables, total dept to total asset ratio of JSCBs from 2010 to 2017 is 89.98%; in which, bank has the highest TD ratio of 1.0894 and the lowest of 0.0947 with standard deviation of 0.0719

As with controlling variables, size has the highest standard deviation of 1.1103, varies from 9.0149 to 13.9997 with the biggest mean of 11.3312 Average deposits to total asset ratio of JSCBs in 8 years is 85.44% with the highest value of 7.6643 and the smallest value of 0.5187 Last but not least, the outstanding loans to total asset ratio has the mean lowest, 10.81%, the value ranging from 0.06% to 79.28% with the standard deviaton is 1.1103

4.2 Correlation Matrix and multicollinearity VIF

 Correlation analysis

Correlation analysis gives us an overview on relationships among variables of the model Explaining correlation between two variables is to study the reaction of one variable when the other variable changes in different values Linear regression can tell whether there is any relationship between two variables or not Correlation coefficient varies from -1 to 1 Results on analyzing correlation coefficient matrix are presented in table 4.2 below

Table 4.2 Correlation coefficient matrix

Source: Data collected by the author and calculated on Stata software

The correlated matrix illustrates that Total Debt to Total Asset (TD) ratio, Total

On Assets (ROA) Meanwhile, it is positive correlation between the ratio of outstanding loans to total assets (LOANS) and ROA In addition, besides the ratio of deposits to total assets (DEPS) has a negative relationship, the correlation between the remaining variables and the return on equity (ROE) are positive

Additionally , correlation coefficients between pairs of variables in the model are not significant (<0.8) and after calculating variance inflation factors (VIF), outcome demonstrates that highest VIF is 1.40 and the VIF of all variables is smaller than five,

so that multicollinearity problem does not have much impact on regression model results

Table 4.3 Multicollinearity Test

Source: Data collected by the author and calculated on Stata software

In summary, the correlation analysis revealed that the selected ratio measures capital structure with both negative and positive correlation with financial efficiency during analysis The correlation matrix allows to determine the correlation between the variables being considered, but does not establish a causal relationship between all independent variables on each dependent variable Therefore, to assess both the impact and direction of the capital on performance, we need to conduct regression analysis

4.3 Regression results

4.3.1 Testing conformability of Pooled OLS, FEM, REM models

Table 4.4 Summary of regression results of Pooled OLS, FEM, REM models

Variables

Pooled OLS (1)

REM (2)

FEM (3)

Source: Data collected by the author and calculated on Stata software

Note: * indicates significance at 1%, ** indicates significance at 5%, *** indicates significance at 10%

4.3.2 Choosing the best fitted model

 Testing conformability of FEM and REM

Hausman test

Hypothesis H 0 : REM is more effective (both FEM and REM are stable)

Hypothesis H 1 : FEM is more stable than REM

Hausman test result (detailed results are shown in appendix 1) shows that

theProb > chi2 equals 0.0291 less than 5%; therefore, we reject hypothesis H0 and FEM (fixed at time and space), as a consequence, is more conformable than REM

 Testing conformability of FEM and Pooled OLS

Hypothesis H 0 : the characteristics do not change (Pooled OLS is more effective) Hypothesis H 1 : the characteristics have changes (FEM is more effective)

Results show that after making regression in FEM method in two situations ROA and ROE: considering the value of PF-test for both ROA and ROE indicators equal 0.000 less than 1% Hence, it is possible to reject hypothesis H0, which means FEM is the conformable method

4.3.3 Other relevant tests

 Heteroskedasticity test

Hypothesis H 0 : There is not detected heteroskedasticity yet

Hypothesis H 1 : The model has heteroskedasticity phenomenon

Table 4.5 Summary of Breusch Pagan Test

Heteroskedasticity Test: Breusch Pagan Included observations: 224

Chi2 (28) = 1436.93

Pvalue>chi2 = 0.0000

Source: Data collected by the author and calculated on Stata software

Breusch Pagan test result indicates that P-value = 0.0000 < 5% so it is concluded that hypothesis H0 is rejected The model does have heteroskedasticity phenomenon