INTRODUCTION
THE NECCESITY OF THE RESEARCH
Banking is one of the crucial sectors in the economy Banking system acts as a fuel injection system, directing savings into channels for investments to boost economic competence (Aspal & Nazneen, 2014) It performs all tasks related to the efficient and secure channeling of funds, acting as a link between savers and borrowers (Aspal & Nazneen, 2014) According to Ebong (2005), for any modern economy to experience rapid growth and economic development, an effective financial system is regarded as a necessary and sufficient condition In the daily operation of the bank, to ensure stability in the operation process, banks or financial institutions need to balance capital and risks in assets However, bad management of a bank could bring negative consequences spreading to other areas of the economy, causing financial crisis or a worsening of current fluctuations (Asarkaya & ệzcan, 2007) Therefore, it is important that banks manage their risks properly to ensure the safety of the bank's operations as well as the health of the whole economy In order to strengthen banks’ risk management, BCBS has published the principles and standards to give guidance on the proper capital levels for banks through the Capital Accords called Basel I, Basel II and Basel III The majority of Basel versions' content is devoted to calculating the minimum capital adequacy ratio, which is regarded as an indicator of the amount of safe capital that a bank must make sure to maintain in order to bear certain losses arising beyond expectations in the bank's business (Aspal & Nazneen, 2014) As the result, the Basel Capital Adequacy Ratio Standards application contributes to the stability and effectiveness of the financial system by lowering the probability of bank insolvency (Abba et al, 2018).
Vietnam is currently not a member of BCBS and is therefore not obligated to obey Basel regulations However, a few Vietnamese commercial banks have implemented and applying Basel Accord, or in the process of applying this international banking regulation agreement Remaining the capital adequacy ratio
(CAR) to equal at least 8% is the first pillar of the Basel II Accord along with the
2 remaining two pillars in order to help banks in Vietnam improve their risk management capacity SBV has issued Circular 41/2016/TT-NHNN, which regulates the capital adequacy rate for banks and foreign bank branches and has been in implication since
2020 Although Circular 41 is only "covered" in part by the Basel II Accord, ensuring these fairly stringent requirements necessitates significant efforts on the part of banks. According to data from the State Bank of Vietnam (2021), the average capital adequacy rate of banks in Vietnam is 11.37%, separately with the group of joint stock commercial banks, this rate is at the level 11.38% However, this ratio in Vietnam remains very low when compared to the ASEAN + 5 countries, which range from 16% to 24% (World Bank, 2022) As a result, the increase in capital is weighing on Vietnam's joint stock commercial banks, particularly small-scale banks Banks must increase capital in order to meet the Basel II capital adequacy ratio and progress toward Basel III.
The capital adequacy ratio is maintained at an excessively high level is not very good for banking activities When the capital adequacy ratio is too high, the bank must reserve more capital or invest in assets with lower risks, resulting in inefficient capital use and a decline in profits In contrast, when a bank has a low level of capital adequacy ratio, its ability to deal with the crisis and economic shocks is reduced Consequently, many authors, both foreign and domestic, have been interested in analyzing factors affecting capital adequacy ratio Foreign studies such as those by Dreca (2013), Bateni et al (2014), Aktas et al (2015), and Shingjergji & Hyseni (2016) may be mentioned. The majority of these studies, however, were carried out in foreign commercial banks, so the findings do not accurately represent the situation in Vietnamese commercial banks In Vietnam, there are several studies with the related topic were published by domestic authors, such as Vo Hong Duc et al (2014), Nguyen Kim Chi (2018) and Do Hoai Linh et al (2019) For these domestic paper research, the data used in these studies is relatively outdated compared to the present such as Vo Hong Duc et al (2014) took data from 2007 to 2012, Nguyen Kim Chi (2018) gathered secondary data from 2007 to 2016, Do Hoai Linh et al (2019) used data from 2008 to
2013 The findings of these studies, however, are ambiguous and contradictory with
3 regard to the variables influencing capital adequacy ratios as well as the direction in which the variables have an impact.
Therefore, in order to make proposals that are more in line with currentVietnamese practice, the author implements the research topic " Determinants of the capital adequacy ratio of joint stock commercial banks in Vietnam" The result of this thesis makes some recommendations for bank administrators to issue appropriate policies to improve capital adequacy ratio, with the goal of achieving safety and stability in the operation of the Vietnamese banking system.
OBJECTIVES AND RESEARCH QUESTIONS
The general objective of this research is to examine the factors that affect the capital adequacy ratio of joint stock commercial banks in Vietnam in the period from
In order to achieve the general objective, this research has to implement 3 specific objectives, including:
Firstly, identifying the factors influencing the capital adequacy ratios of Vietnam's joint stock commercial banks.
Secondly, assessing the factors affect Vietnamese joint stock commercial banks’ capital adequacy ratio.
Thirdly, proposing some recommendations on how to improve the capital adequacy ratio of joint stock commercial banks in Vietnam.
Based on the general objective stated above, the thesis suggests the research questions as follow:
Firstly, what are factors that affecting the capital adequacy ratio of Vietnam's joint stock commercial banks in the period 2012 - 2021?
Secondly, how do these factors affect the capital adequacy ratio of Vietnam joint stock commercial banks in the period 2012 - 2021?
Thirdly, which recommendations should be proposed to improve the capital
4 adequacy ratio of joint stock commercial banks in Vietnam?
RESEARCH SUBJECTS AND SCOPE
This research is focus on identifying factors of the capital adequacy in Vietnamese joint stock commercial banks.
Regarding to the research space, the study is anticipated to use secondary data from 31 joint stock commercial banks in Vietnam in the period 2012-2021 However, the author has chosen 24 joint stock commercial banks in Vietnam because the research data of some banks during the aforementioned period was constrained by incomplete information from financial statements and annual reports, as well as banks that were under special control by SBV (Dong A joint stock commercial bank) The list of banks used in the thesis will be shown specifically in Appendix 1.
Regarding to the research time, the research scope is in 10-year period from
2012 to 2021 The 10-year study period is the most common in scientific studies This time frame is relatively recent and long enough to account for economic changes.
DATA AND RESEARCH METHODOLOGY
About the research data, the topic uses the secondary data, collected from the reliable sources In particular, secondary data of joint stock commercial banks are gathered from the announced financial statements, annual reports which is from banks’ website and the official website of Vietstock Finance For the macroeconomic factors, secondary data will be collected from World Bank website (WB) After gathering all the necessary data, the thesis will calculate the variables with the support of Excel.
After collecting and processing the data, this thesis uses quantitative methods,estimate panel data regression model with the support of Stata software to determine the impact of factors on capital adequacy ratios of joint stock commercial banks inVietnam There are three data regression model used in this thesis, namely the smallest square method (Pooled OLS), fixed effects model (FEM), random effects model(REM) Then, the thesis conducts the F-test and Hausman test to find the suitable
5 model Additionally, the thesis will take a number of tests to detect defects in the model,including multicollinearity, autocorrelation and heteroskedasticity In case the appropriate model has any problems, the feasible generalized least squared method(FGLS) is employed to overcome those defects, ensure the suitability of the model.
CONTRIBUTION OF THE THESIS
In terms of theory, the study is based on the findings of both foreign and domestic studies on the same topic Despite the fact that there are numerous published results, each one reveals a different conclusion about the impact factors and the level of impact of those factors on capital adequacy ratios This research will re-systematize the theory of capital adequacy, as well as Vietnamese regulations governing capital adequacy ratios and the development of a regression model for analyzing factors influencing this ratio The data in the study will be updated to the end of 2021.
In terms of practice, the study's findings will serve as research materials for authors interested in the same subject The research also provides empirical evidence of the impact of factors on capital adequacy ratios at joint stock commercial banks inVietnam from 2012 to 2021 As a result, bank administrators are advised to implement policies that will increase the bank's capital adequacy rate while adhering to global standards of evaluation.
STRUTURE OF THE THESIS
This thesis, with the topic "Determinants of the capital adequacy ratio of joint stock commercial banks in Vietnam," is divided into five chapters The content of each chapter is presented as follow.
Chapter 1 is an introduction chapter In this chapter, the thesis presents the necessity of the research, research objectives following with the research questions, scope and subject of research, research methodology, contribution and the structure of the thesis.
Chapter 2 of the thesis presents the basic concepts and theories that underlie this topic Specifically, theories about the bank's capital adequacy ratio and the factors that influence it This chapter also systematize research issues through previous studies
6 conducted in Vietnam and other countries.
Chapter 3 describes a research model, propose research hypotheses of bank- related and macro variables, based on the theory and empirical evidence presented in chapter 2 In addition, it also describes research process, data and research methods to determine the factors affecting the capital adequacy ratio of Vietnam Commercial Joint Stock Bank.
Chapter 4 shows the regression results and results of a series of tests to find the suitable model After that, the thesis discusses the study's findings and based on the model, determines which factors actually influence the capital adequacy ratio.
Chapter 5 outlines the main conclusions and policy implications to improve the capital adequacy ratio of commercial banks in Vietnam Furthermore, this chapter will also present the limitations of the study and propose further research directions.
In this chapter, the author has introduced an overview of the research topic,which presents the urgency of the topic, research objectives, research questions,subjects and scope of research, research methods, scientific significance as well as practical meaning of the topic It is the premise for clarifying, more detailed and more specific as well as the extent of the impact of the factors affecting the bank's capital adequacy ratio will be detailed in the following chapters.
THEORETICAL BASIC AND LITERATURE REVIEW
THEORETICAL BASIS OF CAPITAL ADEQUACY RATIO
2.1.1 Concept of capital adequacy ratio
The concept of “capital adequacy” was first mentioned in the middle of the 1970’s due to the instability of the economic situation along with the failures of large bank, in particular Franklin National Bank - one of the major American banks (Koehn
& Santomero, 1980) According to Ebhodaghe (1991), capital adequacy is defined as a situation in which the bank's adjusted capital is sufficient to absorb all losses and cover fixed assets while leaving a comfortable surplus for current operations and future expansion Furthermore, the level of capital must be adjusted when total operational expenses and withdrawal requirements are expected to rise (Onuh, 2002) Yu Min-Teh
(1996) investigated bank capital adequacy as the level at which the deposit-insuring agency would just break even in guaranteeing individual banks' deposits with the premium the bank pays Dowd (1999) discovered in his research that minimum capital standards on financial institutions can be viewed as a means of reinforcing deposit security and the robustness of the banking system Because of the asymmetric information between bank managers and depositors, it is necessary to have government intervention in the financial system to prevent the market failure.
The Capital Adequacy Ratio (CAR), also known as Capital to Risk-weighted Asset ratio is considered as the crucial measurement of the adequacy of capital of a financial institution Berger et al (1995) stated that the capital adequacy ratio represents a correlation between the bank's capital and possible risks The capital adequacy ratio is calculated as a percentage of the bank's own capital divided by the total value of risk assets The Basel Capital Accords established the framework for identifying and calculating capital adequacy ratios These Accords required banks to maintain a base capital of at least 8% of risky assets In addition, the use of minimum capital adequacy ratios promotes financial system stability and efficiency by lowering the likelihood of bank insolvency (Bateni et al, 2014) This ratio plays an important role in determining a depository institution's "safety and soundness", because it helps these institutions create a buffer to face with financial shocks, protect themselves as well as their customers. Consequently, the CAR has been widely used in many countries with the purpose of protecting their economic financial system.
As a result, there are various classifications of capital adequacy ratios in banks. However, it is possible to generalize that the capital adequacy ratio is the basis indicator for measuring the bank's capital adequacy, reflecting the level of capital that the bank must maintain in accordance with regulations to absorb possible losses during operations, in order to ensure the safety of the bank and depositors.
2.1.2 Standard for measuring capital adequacy ratio
Referring to the capital adequacy ratio, it is impossible not to mention the Basel Committee on Banking Supervision (BCBS), as it is a pioneer in the standardization of the measurement of this index The history of the formation of BCBS is associated with the instability of international currency and banking markets, typically the collapse of the Bretton Woods system and the failure of Bankhaus Herstatt in West Germany To prevent such serious disturbances, in 1974, BCBS was established with the support of G10 countries including Belgium, Canada, France, Germany, Italy, Japan, the Netherlands, Sweden, the United Kingdom, and the United States This committee serves as a forum for member states to collaborate on a regular basis in order to improve financial stability by improving supervisory knowledge and improving the quality of global banking supervision BCBS has issued a number of documents and documents related to banking supervision since 1975 The most prominent of the international standards of banking regulation issued by this committee are three Basel Capital Accord, namely Basel I, Basel II and Basel III As the Accords become more comprehensive, the latter frequently seeks to overcome the shortcomings of previous versions while adapting to changes in financial markets.
Measure capital adequacy ratio according to Basel I
In July 1988, recognizing that the capital ratios of international banks were deteriorating at a time of increasing international risks, BCBS created the Basel I to enhance the stability of the international banking system The main content of this Accord is to set international standards and methods of measuring capital adequacy The Accord was originally intended to apply to banks operating on an international scale but has been welcomed by many countries and widely adopted at banks operating in their respective countries.
In this Accord, the capital adequacy ratio is calculated by the following formula:
Banks are required to maintain this ratio at a minimum of 8% Accordingly, the bank has the best capital when having CAR > 10%, has the appropriate capital level when CAR > 8%, lack of capital when CAR < 8%, lack of capital is pronounced when CAR < 6% and severe capital shortage when CAR < 2%.
Basel I divided the bank's capital into two categories: Tier 1 capital and Tier 2 capital Tier 1 capital is core capital, including permanent shareholders' equity and disclosed reserves (created or increased by appropriations of retained earnings or other surplus, such as share premiums, retained profit, general reserves and legal reserves). Tier 2 capital, which is supplementary capital, including undisclosed reserves, asset revaluation reserves, general loan-loss reserves, hybrid (debt/equity) capital instruments BCBS stipulates that Tier 1 capital accounts for at least 50% of the bank's capital and tier 2 capital is restricted to a maximum of 100% of the value of tier 1 capital Tier 3 capital including short-term subordinated debt, this capital is not taken into account when calculating capital adequacy ratio because of its lowest reliability. Five risk-weighted groups have been built and assigned to asset groups, with rates of 0%, 10%, 20%, 50%, and 100% Basel I has solely focused on credit risk because it is the most significant threat to banks' operations.
Capital adequacy ratio calculation in Basel II
The publication of Basel I along with detailed regulations had great significance for the risk management of commercial banks However, the development of banking activities in the world had made the application of Basel I reveal some major disadvantages Thus, in June 2004, BCBS published Basel II guidelines aiming to refine and reform the version of Basel I The content of this document includes a set of monitoring standards aimed at perfecting risk management techniques and is divided into three pillars related to minimum capital requirements, supervisory review and market discipline According to the first pillar, capital adequacy ratio is set at a minimum of 8% However, risk is calculated based on three major factors that the bank faces: credit risk, operational risk and market risk.
Cavital RWA ( Credit risk )+ RWA (Operational risk) + RWA ( Market risk )
Basel II abolished Basel I's risk approach and replaced it by delineating risk levels on the basis of a more accurate rating of risk levels The Basel II measurement system is more complex and accurately assesses capital safety with standardized and advanced measurement methods than the previous version.
With regard to assess risk-weighted asset when considering credit risk, Basel II provided three compliance approaches: a Standardized approach and two Internal Ratings- Based approach, which are the foundation and advanced The risk weight is tied to ratings provided by external credit assessments agencies under the Standardized approach The Internal Ratings- Based approach employs the bank's own estimates of certain risk factors, and the gap between the foundation and advanced approaches is created based on the risk factors that can be calculated The credit risk regulations of Basel II also include detailed guidance on dealing with securities and minimizing credit risk.
Market risk according to the Basel committee is the risk of a loss in the trading state when prices fluctuate erratically Market risk information will be associated with four types of risks on book transactions: interest rate risk, capital status, exchange rate risk and commodity risk For this type of risk, Basel 2 allows banks to use tier 3 capital to deal with this risk, and tier 3 capital will be limited to 250% tier 1 capital used to hedge market risks There are two different approaches used to measure market risk, namely the standardized approach and the internal model approach.
Compared to Basel I, Basel II brings in operational risk factors to calculate risky assets This risk includes damage caused by inadequate or failed internal processes, by people or systems, or from external events With this type of risk, Basel II offers three different approaches for banks to choose to apply, namely the Basic Indicator Approach, the Standardized Approach and Advanced Measurement Approaches.
Basel III capital adequacy ratio measurement
Responding to the complex situation of the global financial crisis in 2008, BCBS issued the third version of the Basel Accord, called Basel III This framework builds on the previous accords, Basel I and Basel II, and is part of a process to improve regulation in the banking industry While retaining the three-pillar structure previously built in Basel II, Basel III's minimum capital adequacy ratio regulations have been updated In addition, this accord introduces leverage ratios as well as countercyclical ratio to improve the safety of banks' operations, helping banks to cope with economic shocks.
OVERVIEW OF RELATED STUDIES
The theory of capital adequacy ratios is inherently diverse, with numerous approaches Many researchers from both developing and developed countries around the world have developed many research models in order to identify factors that influence bank capital adequacy ratios In this section, the thesis will recapitulate some of the findings from relevant studies conducted in Taiwan, Iran, Turkey, India, and other countries, which will serve as the foundation for the research model in the following chapter.
Research paper by Büyükşalvarcı & Abdioğlu (2011) had the primary goal was to analyze the factors affecting the capital adequacy ratio of commercial banks inTurkey The paper used secondary data from annual report of 24 different Turkish banks throughout a five-year period between 2006 and 2010 The authors selected 9 specific bank variables, which were Bank size (SIZE), Deposits (DEP), Loans (LOA), Loan loss reserve (LLR), Liquidity (LIQ), Profitability (ROA and ROE), Net interest margin (NIM) and Leverage (LEV) to study With the assistance of Excel and the econometrical software program EViews 5.1, the FEM regression model was established to study the relationship between the independent factors and the capital adequacy ratio At a result, the value of adjusted R 2 was 0.8146 demonstrated that the independent variables explained 81.46% variability of the capital adequacy ratio This paper found out that there were 5 out of 9 explanatory variables, namely, LOA, LLR, ROA, ROE and LEV had statistical relationships with CAR Otherwise, SIZE, DEP, LIQ and NIM did not statistically influence the capital adequacy ratios The study predicted that the explanatory variable LOA would positively affect CAR, but the regression model discovered the opposite results LLR had a positive and significant relationship with the dependent variable CAR This explained that the bank's capital adequacy ratio increases as its financial health deteriorates At 5% significant levels, LEV negatively impact on CAR In this research, ROA and ROE were the variables represented profitability Both of them have a notable impact to capital adequacy ratio, while ROA had a positive impact on capital adequacy ratio, ROE had the adverse impact.
The additional paper took determinants of CAR as the main purpose is study of Dreca (2013) This paper uses the secondary data of 10 selected banks in Bosnia and Herzegovina from 2005 to 2010 Following the completion of the required tests, the OLS model is chosen as the best model for analyzing the relationship between variables The estimated regression’ result demonstrated that CAR was affected by SIZE, DEP, LOA, ROA, ROE and LEV On the other hand, LLR and NIM appeared to have no relationship with capital adequacy ratio Variables SIZE, DEP, LOA and ROA have negative sign of the beta coefficient in the model, which show the negative between these factors and CAR, while variables ROE and LEV are positively related with CAR Whether having a low or high bank CAR was preferable was not determined by the study From the perspective of stability, a higher CAR is preferable, but from the perspective of profitability, a lower CAR is better.
Another research which studied factors influence the capital adequacy of commercial banks was conducted by Moh'd Al-Tamimi & Obeidat (2013) This research paper attempted to identify the most important factors influencing the capital adequacy of Jordanian commercial banks listed on the Amman Stock Exchange Data of this paper was collected from 15 commercial banks operating in Jordan over the period between 2000 and 2008 This study used the Correlation Coefficient combined with Multiple Linear Regression Analysis as statistical analysis method to complete the stated objective Liquidity risks, credit risks, capital risks, interest rate risk, return on equity ratio, return on assets, revenue power ratio were dependent variables in this research The evaluating regression showed that liquidity risks, interest rate risk, return on equity ratio, return on assets were meant to explain the volatility for the CAR of commercial banks in Jordan In particular, interest rate risk and return on equity ratio negatively impact on dependency variables In contrast, the beta value of liquidity risks and return on assets were positive, implying that when these factors increase, CAR will also rise On the other hand, the remaining factors, namely credit risks, capital risks and revenue power ratio had no statistical relationship with CAR.
Aspal & Nazeen (2014) gathered secondary data of 20 Indian private sector banks from 2008 to 2012 A multiple linear regression model is applied to reach the primary goal of identifying factors impact capital adequacy ratio of banks in India The multiple linear regression’s result revealed that Loans (Advances to Assets Ratio),Management Efficiency (Expenditure to Income Ratio), Liquidity (Liquid Asset toTotal Asset Ratio) and Sensitivity (Risk Sensitive Assets - Risk Sensitive Liabilities) have positive statistically significant relationship According to the authors, as loans and advances increase, so will interest income and profitability, giving Indian private sector banks a greater incentive to protect their owners' capital Furthermore, because CAR and liquidity have a positive relationship, an increase in bank liquidity reflects the bank's ability to meet credit demand and cash flow requirements Besides that, the study reveals that Indian private sector banks have higher level of capital than the requirements of Reserve Bank of India The study also discovered that Indian private sector banks have more funds than they need to cover their obligations and have the chance to extend more loans to the general public by defending owner stakes.
Bateni et al (2014) examined relationships between capital adequacy ratio (CAR) and 7 microeconomic factors of selected Iranian banks This study used FEM to identify the relationships between variables The chosen variables were bank’s size (SIZE), loan asset ratio (LAR), risk asset ratio (RAR), deposit asset ratio (DAR), return on asset (ROA), return on equity (ROE), equity ratio (EQR) This research gathered secondary data from the financial statements of six Iranian private banks, whose financial statements were available to access The time span of the study was seven years, from
2006 to 2012 The panel data regression with total 42 observations for each variable was created in order to fulfill the general objective of the paper The final result of this research illustrated that most of variables have the impact on CAR, except for RAR and DAR Regarding to the variables that affect the capital adequacy ratio, while SIZE had a substantial negative relationship with CAR, other factors, namely, LAR, ROA, EQR and ROE positively affected the dependent variable.
In Albania, the authors Shingjergji & Hyseni (2015) also used the Capital Adequacy Ratio of the banking system as a dependent variable in their study and examined the factors that affect this ratio in the Albanian banking system The relationship between the dependent and explanatory variables was evaluated using an OLS model The data was gathered from Albanian banking system from first trimester of 2007 until third trimester of 2014 Banks in Albania are required to have a capital adequacy ratio of over 12%, much higher than the minimum regulatory of 8% set by the Basel II Therefore, the authors believe that countries such as Albania should be more cautious in determining the capital adequacy ratio In this paper, variables such as return on asset, return on equity, nonperforming loans, equity multiplier, bank size and loan to deposit ratio were put into a linear regression analysis to investigate the impact of them on capital adequacy ratio The regression model revealed that while return on asset and return on equity had no effect on capital adequacy ratio On the other hand, non- performing loans, loan to deposit ratio, and equity multiplier did and in negative direction In contrast to other studies, bank size has a positive impact on capital adequacy ratios, implying that larger banks have higher capital adequacy ratios.
El-Ansary & Hafez (2015) published research paper identify factors that explain the CAR variance in Egyptian commercial banks in order to identify decisions that improve or degrade capital management quality During the period from 2003 to 2013, the authors conducted empirical research on 33 banks in Egypt (including local Egyptian banks, international banks, and locally or regionally Islamic banks) During the interim period from 2003 to 2013, the estimated result showed that while return on assets, liquidity and management quality had positively statistical relationship with CAR, bank size and risk have the inverse impact on CAR Furthermore, the authors conducted two additional models that looked at CAR determinants before and after
2008, the year of the financial crisis The results of the study showed that, in the period before the financial crisis, assets management quality, bank's size, risk and return on equity factors had a negatively statistical impact on CAR ROA variables, in contrast, have a positive statistical effect on the dependent variable After the financial crisis in
2008, variables such as asset management quality, bank size, liquidity, asset quality, and risk became statistically significant factors for CAR While asset management quality, asset quality and liquidity variables have a positive impact on CAR, bank size and risk have the opposite effect.
Aktas et al (2015) established research to investigate the determinants of capital adequacy of banks in the Southeastern European countries from 2007 to 2012 Annual data from total 71 commercial banks in 10 nations, namely, Albania, Bosnia, Bulgaria,Croatia, Greece, Macedonia, Montenegro, Romania, Serbia and Slovenia are used in the study There were 12 independent variables including specific bank factors and environmental factors used to estimate Bank size (SIZE), profitability (ROA), leverage(LEV), liquidity (LQDT), net interest margin (NIM), and bank risk (RISK) were identified as bank-dimensional independent variables, while economic growth(GDP_G), inflation (CPI), real interest rate (RIR), Eurozone stock market volatility index (EURO_VOL), deposit insurance coverage ratio (COV), and governance indicator (GOV) were categorized as macroeconomic variables The author used FGLS regression model and developed two models to investigate the factors that influence banks' CAR in the SEE region Model 1 only used bank-dimensional explanatory variables, whereas Model 2 included macroeconomic variables in addition to bank- related factors All of the explanatory variables in model 1 have statistically significant effects on CAR While SIZE, LEV, and RISK negatively impact on CAR; ROA, LQDT, and NIM have a positive relationship with the dependent variable In model 2, with the addition of macro variables, NIM is no longer statistically significant for CAR. Regarding to the environmental factors; GDP_G and GOV reveal to have negative significant relationship with CAR, whereas the relationship of COV and EURO_VOL and CAR is positively statistical.
Nuviyanti & Achmad (2015) studied the influence of risk-based bank rating components such as good corporate governance, risk profile, and earnings to capital adequacy ratios in 19 Indonesian commercial banks from 2008 to 2013 After doing some test, this study decided to choose EGLS as the appropriate model do assess the data Firstly, the operating expense to operating income ratio and the net interest margin ratio are both indicators of good corporate governance, but their significance to dependent variables are different The estimated model demonstrated that net interest margin ratio positively explains changes in CAR, operating expense to operating income ratio has inverse relationship Secondly, risk profile was represented by non- performing loans and loan-to-deposit ratios Both of them have significant influence with CAR, while the former has positive impact on CAR, the latter revealed to have a negative correlation coefficient with CAR Lastly, the authors used Return on Equity and Return on Asset as proxy for earnings Return on asset impact on CAR in positive direction, whereas return on equity is negative.
In Vietnam, inherit the results of previous studies, some researchers also use capital adequacy ratios as their research goals Some studies can be mentioned as research by Vo Hong Duc et al (2014), Nguyen Kim Chi (2018), Do Hoai Linh et al
Firstly, the study by Vo Hong Duc et al (2014) is one of the studies in Vietnam that examines the factors that influence a bank's capital adequacy ratio The study took research data from 28 commercial banks in Vietnam over a five-year period from 2007 to 2012 Banks chosen to serve as research samples must publish capital adequacy ratios, have charter capital of VND 3,000 billion or more These 28 banks represented approximately 83 percent of charter capital and 70% of the total number of commercial banks at the time of the study About the estimation method, this paper used FGLS regression model to estimate the collected data The study's findings revealed that the liquid assets have the positive impact on the capital adequacy ratio Banks with a high proportion of liquid assets are more likely to reduce defaults, resulting in higher capital adequacy ratios Similar to liquid assets, loan loss reserves also positively affect CAR.
FACTORS AFFECTING THE CAPITAL ADEQUACY RATIO
2.3.1 Internal financial criteria of banks
Total bank assets are one of the criteria used to determine the size of a bank The bank size variable (SIZE) is calculated using the logarithm of total asset to achieve statistically significant results According to Büyükşalvarcı & Abdioğlu (2011), the size of banks important due to its relationship to bank ownership characteristics and the banks’ ability to access to equity capital which may reflect the relative importance of managerial risk aversion or bankruptcy cost avoidance Shingjergji & Hyseni (2015) argued that the larger the bank, the more likely it is to ensure a capital adequacy ratio compared to small-scale banks However, Yu (2000) found that larger banks had lower capital adequacy ratios than smaller banks because large banks typically hold a diverse portfolio of deposit claims, making their deposits less risky than those of small banks. According to Alfon et al (2004), the primary objective of small banks in maintaining higher capital levels than larger banks is to fund their long-term business strategy Small banks decide to carry more capital because it is more expensive for them to change their capital in the event of an unexpected capital requirement Many previous practical studies have shown the negative relationship between bank size and capital adequacy ratios such as those of Dreca (2013), Bateni et al (2014), El-Ansary & Hafez (2015), Akta et al (2015) They argued that the larger the bank, the stronger the investment in the business, the riskier assets it will hold, leading to a low capital adequacy ratio. Therefore, this thesis hypothesizes a negative relationship is expected between bank size and CAR.
H 1 : The size of the bank (SIZE) has a negative impact on bank’s capital adequacy ratio
The customer deposit ratio (DEP) is the ratio between the total amount of customer deposit and the total assets Capital mobilization is one of the most basic and important activities of commercial banks, contributing to bringing capital so that the banks can carry out business activities in a normal way According to Weber & Kleff
(2003) bank deposits are a very alluring method of funding the bank, because compared to other sources of funding, such as bonds, loans from business angels and through syndications, deposits require paying relatively lower interest rates Additionally, Dreca
(2014) noted that the bank would maintain a CAR lower optimal capital ratio if the depositor were unable to assess the stability of the banks to which they send money. However, if depositors can be certain that the bank which they are depositing money into is financially stable, they will be more willing to accept lower interest rates, allowing the bank to maintain a sizable source of funds from depositors There have been studies that have found the opposite relationship between the ratio of customer deposits to total assets and the capital adequacy ratio as studied by Shingjergji & Hyseni
(2015), Nuviyanti & Achmad (2015), Vo Hong Duc et al (2014), Nguyen Kim Chi
(2018) They claimed that as the amount of customer deposits increases, the bank will increase its credit operations As a result, the bank will exert greater control over the source of capital in order to ensure liquidity, resulting in a low CAR Therefore, the proposed hypothesis is as follow:
H 2 : Customers deposits (DEP) has a negative impact on bank’s capital adequacy ratio
Another factor affecting the CAR is the loans of commercial bank In this thesis, this factor is expressed through the Loans-to-Asset ratio, which is measured by the total loans to total assets of banks and symbolized as LOAN According to Hassan & Bashir
(2003), this is a very important coefficient because it shows that the relationship between one side is diversification and the other is to establish the first opportunity A high percentage of loans and advances indicates that assets are susceptible to credit risk,especially given the sizeable portion of non-performing assets (Sarker, 2005) Do HoaiLinh et al (2019) stated that the higher this ratio means that a bank may be riskier and more susceptible to defaults Therefore, it will take a larger amount of capital to hedge the risk According to studies by Büyükşalvarcı & Abdioğlu (2011), Dreca
(2013), Shingjergji & Hyseni (2015) indicated that the total loan-to-total asset ratio is negatively correlated with the capital adequacy ratio Therefore, this thesis expects a positive relationship between loans and CAR.
H 3 : Loans (LOAN) has a negative impact on bank’s capital adequacy ratio
The financial leverage coefficient is determined by the ratio between equity to total assets (LEV) In this study, since this factor is determined by equity-to-asset ratio, so a high LEV indicates low leverage and vice versa According to Circular No. 41/2016/TT-NHNN, which provides guidance on how to calculate the minimum capital adequacy ratio, the equity of commercial banks constitutes a significant portion Tier 1 capital According to Workneh (2014), investors will discover that highly leveraged banks (lower equity to total assets) are riskier than other banks, increasing the required rate of return for investors Consequently, the high leveraged banks (Low LEV) may hold less equity capital and find difficulty in raising new equity difficult because of the high cost (Workneh, 2014; Ahmad et al, 2008; Usman et al, 2019) Dreca (2013), Nguyen Kim Chi (2018) found a positive relationship between LEV and CAR As a result, the author expects a positive correlation between financial leverage and capital adequacy ratio.
H 4 : Financial leverage (LEV) has a positive impact on bank’s capital adequacy ratio
Liquidity is considered one of the factors affecting the capital adequacy ratio of commercial banks This factor is calculated by taking the Cash and Cash Equivalents divided by total assets According to Mehranfar (2013), a bank with low liquidity is more likely to experience an unplanned operational mishap that prevents it from meeting its short-term obligations to customers Angbazo (1997) states that as the proportion of funds invested in cash or cash equivalents increases, a bank's liquidity risk declines, leading to lower liquidity premium in the net interest margins Moreover, a higher level of bank liquidity has a favorable effect on the capital ratio by altering the required rate of return on bank shares (Mehranfar, 2013) When the bank ensures the input and output cashflow, it means ensuring liquidity, thereby helping the bank increase profits and capital sources The CAR improves as a result of this This view is also consistent with the studies of Aspal & Nazneen (2014), El-Ansary & Hafez (2015), Akta et al (2015), Nguyen Kim Chi (2018) As a result, the author anticipates a positive correlation between liquidity and capital adequacy.
H 5 : Liquidity (LIQ) has a positive impact on bank’s capital adequacy ratio
The loan loss reserves ratio is also considered as a factor affecting the capital adequacy ratio This ratio is measured by taking loan loss reserves divided by the total loans According to Circular No 11/2021/TT-NHNN, the credit risk provision is considered to be the amount set up and accounted for in operating expenses, to prevent possible losses with the bank's unclaimed debts Blose (2001) believed that loan loss provisioning negatively affects a banking firm, the capital adequacy measures will decline as a result of loan provisioning and the corresponding write-downs When a bank incurs lending losses, it must set aside reserves from its earnings and, if earnings are insufficient to cover the reserves, from its equity, reducing its capital A higher loan loss reserve ratio also indicates that the bank is taking more risks, making it more difficult to raise capital Pham Phat Tien & Nguyen Thi Kieu Ny (2019) examined the capital adequacy ratios of 29 commercial banks in Vietnam between the years of 2013 and 2017 They found that as the loan loss reserves increased, implying that quality of bank loans declined, the risk to the assets of the bank increased As a result, the CAR of banks will decline Similarly, research by El-Ansary & Hafez (2015) and Akta et al
(2015) found that there exists a negative relationship between loan loss reserves ratio and capital adequacy ratio The higher the risk provision ratio the bank has, the lower the capital adequacy ratio Therefore, the study hypothesizes that there is a negative relationship between loan loss reserves ratio and the capital adequacy ratio.
H 6 : Loan loss reserves (LLR) has a negative impact on bank’s capital adequacy ratio
The profitability of commercial bank is another element included in the analysis the relationship with CAR There are a variety of ratios used to assess a bank's profitability In this thesis, profitability expressed through the Return on Asset ratio (ROA), which is measured by taking the net income after tax divided by the total assets. This is an indicator that reflects a bank's ability or efficiency to generate profits from its total assets According to Bateni et al (2014), the bank's return on total assets is negatively correlated with the capital adequacy ratio, because when a bank decides to raise profitability, it must accept portfolio expansion or choose a riskier portfolio. Alajmi & Alqasem (2015) maintained that the higher the profits of local banks the lower the need for more capital to absorb losses Many studies around the world have shown a negative relationship between the profitability and the capital adequacy ratio such as Dreca (2013), Moh'd Al-Tamimi & Obeidat (2013), Nguyen Kim Chi (2018),
Le Hong Thai (2020) They discovered that if a bank wants to earn profits, it needs to increase its risk assets, which can result in lowering capital adequacy ratio Therefore, the study hypothesizes that there is a negative relationship between the return on assets ratio and the capital adequacy ratio.
H 7 : The return on assets ratio has the negative effect on the capital adequacy ratio
Another determinant of CAR is operating cost proxied by the ratio of operating expenses divided by operating income (CIR) When banks are able to offer quality banking services at the most reasonable cost of operation, they are said to be operational(Allen & Rai 1996) More specifically, operational efficiency may be attained by banks when they combine the appropriate inputs and ensure that operating costs are kept to a minimum (Athanasoglou et al 2008) Therefore, the higher CIR is the more inefficiently the operational list was being used, which make the capital adequacy ratio was also declining (Nuviyanti & Achmad, 2015) El-Ansary & Hafez
(2015), when analyzing 33 banks in Egypt between 2003 and 2013, the authors found that before the 2008 world financial crisis that management levels (operating costs) negatively impacted CAR After the 2008 financial crisis, turning operating expenses positively impacted CAR In the study by Aspal & Nazneen (2014), the data was collected from 20 Indian private sector banks between 2008 and 2012 showed a negative relationship between operating costs and capital adequacy ratios Therefore, the author expects a negative effect between operating costs and car capital adequacy ratio.
H 8 : Operating cost has the negative effect on the capital adequacy ratio
RESEARCH METHODOLOGY
RESEARCH PROCESS
The thesis uses quantitative research methods to determine the impact factors as well as the direction the impact on the capital adequacy ratios of Vietnamese joint stock commercial banks, the study uses different multivariate regression methods In order to accomplish the general objectives as well as the specific objectives that have been set out, the thesis conducts a 6-step research process The steps of the process and the content of each step are expressed as follows
Source: Author’s summary Step 1: Identifying research issue
In the first step of the study, the author will clearly define the research objectives, research questions, research subjects, scope of research, and the meaning of research Thus, this step serves as a foundation for developing and conducting research.
Step 2: Generalizing theoretical basic and related empirical research
In this step, the thesis summarizes the theoretical basis about CAR including definition, measuring standards and the meaning of this ratio Additionally, the thesis also illustrates relevant empirical studies in Vietnam and other countries around the world, then discuss the previous studies to identify the research gaps and orient the design of research models for the topic.
Step 3: Setting out research hypothesis and build research model
Based on the theoretical basis and empirical evidence, the topic builds the research hypotheses and measure method for each variable Then, the thesis builds the research model to study factors affecting the capital adequacy ratio of joint stock commercial banks in Vietnam.
The data collected in this thesis includes bank related and macroeconomic data. The thesis collected secondary data from mainstream, reliable information sources. Then, the thesis will then use Excel to calculate the variables in the model.
The author uses STATA 16 software to perform a summary description of the characteristics of the data including the average value, the largest value, the smallest value, the standard deviation of dependent variable and explanatory variables Then, this thesis employs estimation methods such as Pooled Ordinary Least Squares (Pooled OLS), the Fixed Effects Model (FEM), and the Random Effects Model (REM). Following that, the F-test is conducted to choose the appropriate model between OLS and FEM Then, Hausman test is conducted to select the applicable model between FEM and REM To ensure that the model is not violated by hypotheses in model building, inspections of model defects are implemented Specifically, tests of three common defects in quantitative studies: multicollinearity, heteroskedasticity and autocorrelation If the model has heteroskedasticity and autocorrelation, the generalized least squares (GLS) regression model will be used to overcome those defects.
Step 6: Interpreting the findings, stating conclusions and suggesting policy implications
This is the final step of the process Based on the results of regression, the thesis decides to accept or reject the stated research hypotheses and draw conclusions on factors that have the statistical relationship with the capital adequacy ratio of joint stock commercial banks in Vietnam Moreover, the thesis gives some recommendations to answer research questions as well as address the research objectives set out.
METHODOLOGY
In this thesis, the author will also use these 03 methods, namely Pooled Ordinary Least Squares, Fixed Effects Model and Random Effects Model to estimate and analyze the impact of macro and micro factors on the capital adequacy ratios of Vietnamese joint stock commercial banks.
3.2.1.1 Pooled ordinary least square (Pooled OLS)
To begin, the thesis will perform the Pooled OLS regression model This is the most commonly used model for analyzing panel data sets The Pooled OLS model has the general formula as follow:
Y it = α+ β 1 X 1it + β 2 X 2it + +β n X nit + ε it Where:
Yit: dependent variable. α: intercept β1, β2, , βn: the structural parameters
X 1it , X 2it , , X nit : the independent variable ε it : the error/disturbance term
According to Pham Thi Tuyet Trinh (2016), this model is assumed in terms of all coefficients are constant over time and by units This means that this model assumes that the cross-sectional units (commercial banks) will be identical in corporate culture,business methods, management methods, operating environment This is a fairly limitation and unrealistic determination that does not properly reflect the relationship between the dependent variable and the independent variable of each object (Pham ThiTuyet Trinh, 2016) As a consequence, this model can give incomplete results and distort the reality of the relationship between independent variables and dependent variables.
According to Pham Thi Tuyet Trinh (2016), the fixed effect model (FEM) examines the correlation between the residual of each unit and the explanatory variables, assuming that each entity has distinct characteristics that can affect the explanatory variables As a result, the model can estimate the net effects of the explanatory variable on the dependent variable by controlling and separating the effect of distinct characteristics from explanatory variables.
The simple FEM model can be written as follow:
Y it = α i + β 1 X 1it + β 2 X 2it + + β n X nit + u nit The coefficient of αi does not change over time However, αi consists of the intercept and the missed variables of each cross unit, known as the subject-specific parameters Therefore, the appearance of αi helps distinguish the characteristics of each bank, this difference is due to the differences in the management and operation policies of the bank, as a result, FEM solves the problem of missing variables (Pham Thi Tuyet Trinh, 2016).
The random effects model (REM) assumes the individual-specific effects are uncorrelated with the independent variables This is the distinction between REM and FEM, with the latter assuming that individual-specific effects are correlated with independent variables As a result, if the unit differences affect the dependent variable, REM will be preferable to FEM (Pham Thi Tuyet Trinh, 2016) In which each entity's balance (which is unrelated to the explanatory variable) is treated as a new explanatory variable.
Y it = α + β 1 X 1it + β 2 X 2it + + β n X nit + w it
With: wit = εi+ unit ε i : error component of cross-section, or individual-specific. unit: idiosyncratic term, other error component of time series and cross-section.
The thesis will run a number of tests in selecting the most appropriate regression model, including the F-test and the Hausman test The thesis will select the most suitable model based on the results of such tests and present the estimated results in Chapter 4.
Firstly, the thesis uses F-test to select the appropriate regression analysis model between FEM and Pooled OLS The F-test has the null hypothesis H0 as follows: All the dummy parameters in the model equal 0 so OLS is more suitable After running the regression model according to the FEM method, if the P-value < 0.05 then with a meaningful level of 5%, the H0 hypothesis is rejected, choose FEM over OLS (Gujarati
Secondly, with the purpose of choosing the more preferred model between FEM and REM, this thesis conducts Hausman test to select In fact, this test is used to consider whether there is a correlation between residual and independent variables The null hypothesis is that there is no correlation between residual and variables, so REM is the preferred model If the P-value is less than 0.05 then with a meaning of 5%, the H 0 hypothesis is rejected, thereby choosing FEM instead of REM, however, if the P-value is greater than 0.05, then with a meaningful level of 5%, we accept H 0, and REM is chosen to estimate.
3.2.3 Check the defects of the model
3.2.3.1 Detecting multicollinearity in the model
Multicollinearity is a statistical concept that describes the correlation of several independent variables in a model When developing a regression model with two or more variables, it is preferable to exclude independent variables with a linear relationship If multicollinearity occurs, the result of the regression will be skewed or misleading, the confidence interval is abnormally expanded To detect the multicollinearity, the author will base on a statistical technique called the variance inflation factor (VIF) VIF of a explanatory variables in excess of 10, that variable is considered to be highly collinear (Gujarati & Porter, 2009) The author will remove variables that has VIF exceed 10, in order to get rid of the problem.
3.2.3.2 Checking heteroscedasticity in the model
Heteroscedasticity take place when the standard deviations of a predicted variable are non-constant when measured over different values of an independent variable or in relation to prior time periods Currently, there are many tests to detect heteroscedasticity in the regression model such as Breusch-Pagan-Godfrey test, Goldfeld-Quandt test, Glejser test, and White test In this thesis, based on the results of the Hausman test, the author uses Modified Wald test (for FEM) or Breusch and Pagan Lagrangian Multiplier test (for REM) These tests have the same pair of null hypotheses
H0: heteroscedasticity does not occur in the model.
H 1 : heteroscedasticity occurs in the model.
If the P-value < 0.05, reject the H 0 hypothesis and accept the H 1 hypothesis.
Autocorrelation is one of the possible defects in panel data When this phenomenon occurs, the R 2 estimates are too high compared to reality, the estimate is skewed, adversely affecting the regression model In this study, the author detects the autocorrelation phenomenon by using Wooldridge test with the hypothesis
H 0 : The model does not have the autocorrelation.
H 1 : The model has the autocorrelation.
If P-Value chi2 = 0.1187 (V_b-V_B is not positive definite)
4 Testing defects in the model
Breusch and Pagan Lagrangian multiplier test for random effects car[bankl,t] = xb + u[bankl] + e[bankl,t]
Va- sd = sqrt(Var) ca r 001502
Test: Var(u) = 0 chibar2(01) = 21.67 Prob > chibar2 = 0.0000
xtserial car size dep loan lev liq llr roa cir gdp exr
Nooldridge test for autocorrelation in panel data HO: no first-order autocorrelation