Khóa luận tốt nghiệp: Factors Affecting Non-Performing Loans of Commercial Banks in Viet Nam

List of AbbreviationsIMF International Monetary Fund BCBS Basel Committee on Banking Supervision ECB European Central Bank SBV State bank of VietNam VAMC Vietnam Asset Management Company

VIETNAM NATIONAL UNIVERSITY UNIVERSITY OF ECONOMICS AND BUSINESS

FACULTY OF FINANCE AND BANKING

GRADUATION THESIS

Factors Affecting Non-Performing Loans of Commercial Banks in Viet Nam

Ha Noi, 2023

of this thesis.

I would also like to extend my heartfelt thanks to the Department of Finance and Banking, aswell as University of Economics and Business, for creating a conducive learning environmentthat has allowed us to develop and enhance our capabilities Despite my self-prepared

knowledge and limited deductive skills, I acknowledge that imperfections and errors are

inevitable Therefore, I sincerely hope to receive valuable contributions from esteemedprofessors to further improve this graduation thesis

I] express my heartfelt appreciation

I solemnly declare that this is my research work, with the support of my advisor, Th.S CuNguyen Ha Trang The research content and findings presented in this thesis are honest andhave not been published in any previous research work The data in the tables, which servefor analysis, comments, and evaluation, have been collected by the author from varioussources, as indicated in the reference section.

If any form of academic misconduct is discovered, I take full responsibility befor the counciland the outcome of my thesis

Student

Dang Tran Gia Linh

List of Abbreviations

List of tables

List of figures

CHAPTER 1: INTRODUCTION esscssssssssessssesssesseseceseseseaneeeseneeeseeeseseaeauacaneneneneeeseneseaeauaneseneneneeeeeaeas 10

1.1 The urgency Of the tOpic «- «se SH HH 10

1.2 Research Obj€CẨÏV€S «ch HnhHATRHRTRRREESEEESESESESESESESRREEESESESESESESSr 12

1.3 Research Questions «.«- xi0 12

1.4 Scope Of the Study csscssscssssssssssssssssssesesseesseeacseesseesseseeneaeacseseeseseeaeacaeaeeeseseseseeeeaeassesnsesesneanas 12

1.5 Data and reserach methodỌOBV cscsesesesesesksrsisrrtrrrrrrrrrararsrssrssarntsrsrsrsrsrke 12

1.6 Contribution Of StUCY ccccscsssssssesesssesseessesseessseneeseceessesseeseesesacaseseeesseeesesasaeasesneseneeeaees 13

1.7 Structure Of the Study ccscssssssssssssssssesseesesseessssesecaeaesseeseeesesesassrseeeeesesesecaseeeseeseesesenanas 13

CHAPTER 2: LITERATURE REVIEW -« sen 14

2.1 Literature F€VÏW c eSsnHHgHnnnghHhRREEEERERRERRESEEEEEEERRRRSESEEEEEESEIEESESESESESESE 14

2.2 Theoretical framework ssscssssseseseseeeseeeseseeeeeeeeseeseeeeseseeereceeeeseeeeseeeerononaneneeeneneneneaees 17

2.2.1 Definition of non-performing ÏOaTIS -. sscscseeessesesesesesesrsrsrsrsirrrrrarsrsee 17

2.2.2 Factors affecting non-performing ÏOa'S -«e<c<eseseseseseseeeersrrsre 182.2.3 Consequences of non-performing ÏO4'S -.-«-<c<c<eseseseseseseeseeersrsrsre 21

CHAPTER 3: RESEARCH METHODOLOGGY ccccssssessssessseseeesesrreeececenensneseorerernenenecesenenseeeranes 23

3.1 Research dat ecscscscscscsssesseeeceessesssseeeseeseeseesesereeceseeeeseeeeeseesorereeeeeesanesaeoeseeenesenenereereneenees 23

3.2 Research model and hypothesis «-«cscseseksksksrsrsrsrsrkrrrrsrsrrarrrersrsrsrsrsrar 24

3.2.1 Research Variables cssssssessseeeeeeressesseseseneorerorereeeseneseneneoeeerreneceseseneneneronanerens 24

3.2.2 Research Hypothesis ccsssssessesessesecesecssssssseesreaerevecesecsssseseoeeasavevecesseeneneneoraranenens 273.2.3 Research MOE] , -55s+ series 29

3.2.4 Research metho(ỌOBV «series 30

CHAPTER 4: RESULTS AND DISCUSSION căng 31

4.1 Current situation of NPLs in VietNam -. «s<csesekskksrsrsrerirreerrsrrrrrsee 31

4.2 DesCriptive StAatÏSLÍCS con HH HHRRRHRHRRSEEEEEEESEEESESEEESEiE 36

4.3 Correlation I1ATÏX «5-5 sseesesrersrrirEtsExireieiiirrrsriEEsEirrsrsrssisiriiirrsre 37

4.4 REgression T€SUÏS -«c«csesskksrsretsrkrkrEtrsrirsrirrrirrixrrrierrrsrsrrrxrrrrsrsrsrsrsrsre 37

List of Abbreviations

IMF International Monetary Fund

BCBS Basel Committee on Banking Supervision

ECB European Central Bank

SBV State bank of VietNam

VAMC Vietnam Asset Management Company

NPL Non-performing Loan

LLP Loan Loss Provsioning

LGR Loan Growth

LA Loan To Total Asset

ROE Return On Equity

GDP Gross Domestic Product

EPU Economic Policy Uncertainty

OLS Ordinary Least Square

FEM Fix Effect Model

REM Random Effect Model

List of tables

Table 3.1: Names of 19 banks used in the sample Study - ‹-e -«- 23

Table 3.2: List Of VariaÌ©S «e«s« series 26Table 4.4: DeScriptive StatÏSĂÏCS «se hnrnrrkrree 35Table 4.5: Correlation AT ÏX c«c«s«ssssssksksErrrrrererrrrsrrersrsrrrrrrrrararsrsrsrsrsii 36Table 4.6 : Regression reSults «c«cssssxsesesrsrererktrsrsrrrrrrrrrirrsrsrsrsrsrrrrrrree 37Table 4.7: Breush - Pagan Lagrangian t€SK «-«-scseseerrererrrrrrrrrsrsrsrsir 39Table 4.8: Hausman t©SE -. «5< se herg 40

List of figures

Figure 4.1: Non-performing loans ratio in Vietnam from 2007 to 2016 Figure 4.2: Non-performing loans ratio of Vietcombank from 2015 to 2021 Figure 4.3: Non-performing loans ratio of BIDV from 2019 to 2021

AbstractsCredit activity is a crucial operation for banks, significantly contributing to their profitability.However, poor credit quality and ineffective risk management can lead to non-performingloans This research examines data from 19 commercial banks operating in Vietnam duringthe period 2000-2022 to investigate the impact of both bank-specific factors andmacroeconomic factors on the non-performing loan ratio in Vietnamese commercial banks.The methodology employed for estimation is regression analysis using panel data, whichincludes the Pooled OLS, Fixed Effects Model (FEM), and Random Effects Model (REM), alongwith model selection tests to determine the most suitable model for this study.

The research found that out of the seven bank-specific factors and macroeconomic variables

considered, three have a significant influence on non-performing loans in Vietnamese

commercial banks Based on the study's findings, the author proposes several solutionsgrounded in the factors of significance identified in the research to contribute to mitigatingnon-performing loans in Vietnamese banks

CHAPTER 1: INTRODUCTION

1.1 The urgency of the topic

NPLs is a credit risk that not only has a significant impact on the business activities ofcommercial banks, but NPLs also affects the ability of businesses to access capital, therebycausing negative effects on the growth and development of the national economy According

to the International Monetary Fund (IMF) and the United Nations, a debt is considered

non-performing when the interest and/or principal payments are overdue by more than 90 days;

or when interest payments are overdue for more than 90 days but have been capitalized,refinanced, or delayed under an agreement; or when payments due are overdue for less than

90 days Besides, according to the definition provided by the Basel Committee on BankingSupervision (BCBS), there is no specific time frame for determining when a debt is classified

as non-performing Instead, it is considered a non-performing loan when it is overdue, and

the bank determines that the borrower lacks the ability to fully repay the debt without taking

any remedial actions The classification is based on the bank's assessment of the borrower'srepayment capacity rather than a specific time period for overdue payments In Vietnam,non-performing loans (NPLs) are not directly defined but are indirectly determined throughregulations on loan classification, provisioning, and the use of credit risk reserves in thebanking activities of credit institutions According to the classification of Viet Nam’s Bankingsystem, NPLs is divided into 5 main groups, in which customers classified in groups 3 to 5 arethe most risky customers Therefore, non-performing loans in Vietnam currently bearsimilarities to international standards

Generally speaking, the higher the bank's NPLs, the lower its capital profitability,constrain the bank's growth Hence, NPLs is considered as a financial burden for commercialbanks, which can reduce profits and reduce capital of a bank In addition, it might lead to adecrease in capital adequacy ratio and a decline in shareholders’ equity As a consequence, arising non-performing loans might put banks in danger of bankruptcy, which has been done

in research by previous studies (Demirgii¢-Kunt,1989; Barr and Siems, 1994) In addition toits negative impact on banks, NPLs also have significant influence on businesses and the

overall economy Specifically, NPLs can directly affect the overall development of a country

Furthermore, NPLs can lead to cascading effects that undermine development and, in moresevere cases, can result in the collapse of the entire banking system, posing threats to thenational financial security and economy This becomes even more critical as a majority of

10

businesses today are small and medium-sized enterprises (SMEs) whose capital primarilyrelies on loans from financial institutions Indeed, if banks are negatively affected by NPLsand are forced to tighten lending, it can cause a disruption in the flow of operational capital,significantly impacting both individual businesses and the entire economy Overall, if the

NPLs of a bank worsen and are not timely resolved, it can lead to the bank's collapse,triggering a chain reaction within the banking system and the entire economy

Therefore, policymakers need to pay attention and focus more on the issue of NPLs.They understand that NPLs have significant implications for the stability of the financialsystem and economic growth Emipirically, Messai and Jouini (2013) mention that NPLs areamong the leading causes of economic stagnation Therefore, minimizing NPLs is necessary

for improving economic growth By addressing the issue of NPLs, policymakers aim to

safeguard the stability of the financial system, promote economic growth, ensure smoothcredit provision, and support access to finance for productive activities The identification offactors influencing non-performing loans in commercial banks is crucial Exploring thisresearch question can assist policymakers and bank administrators in formulating effectivepolicies and solutions to mitigate risks, reduce non-performing debts, and enhance theoverall efficiency of banking operations

Furthermore, during and after the COVID-19 pandemic, the situation of NPLs inVietnam has become concerning The strong impact of the pandemic has caused economicturmoil, particularly in sectors such as tourism, services, manufacturing, and exports.Businesses have faced financial difficulties due to reduced revenues and the ability to repaydebts, leading to an increase in the number of NPLs within the banking system However, thegovernment and the State Bank of Vietnam have implemented various measures to supportbusinesses and mitigate the effects of the pandemic Policies such as debt restructuring andfinancial assistance have been implemented to help businesses and individuals overcome

financial hardships and reduce the risk of NPLs At the same time, banks have enhanced risk

management and focused on monitoring and addressing NPLs These efforts aim to ensurethe stability of the banking system and minimize the negative impact of NPLs on the businessoperations of financial institutions Although specific data on the post-COVID-19 NPLssituation in Vietnam may change over time, it is certain that the pandemic has had asignificant impact on the NPLs situation in the economy This calls for attention and proactive

11

measures from policymakers and banks to ensure financial stability and restore economicdevelopment in the country.

For these reasons, in order to enhance a correct understanding of the factorsinfluencing the situation of NPLs in commercial banks in Vietnam, especially the importance

of NPLs resolution in the current economy, this study aims to contribute a fresh perspective

on the issue of NPLs While there have been numerous studies on this topic in the past, thisresearch updates the latest data and investigates the factors impacting NPLs to proposesolutions and recommendations for banks and policymakers

Therefore, I have decided to choose the topic "Factors Influencing Non-Performing

Loans in Commercial Banks in Vietnam" as the focus of this research

1.2 Research objectives

The aim of the study is to identify factors that affect NPLs of banks in Vietnam, and to find

solutions and suggestions that can reduce the risks of NPLs

1.3 Research questions

e Whatare the factors affecting NPLs of commercial banks in Vietnam?

e What is the degree of impact of macro and micro-factors on NPL of commercial

banks?

e What are the solutions and recommendations to reduce NPL?

1.4 Scope of the study

The research examines the determinants of NPLs of 19 listed commercial banks in Vietnam

from 2000-2022

1.5 Data and reserach methodology

Regarding bank-level data, the data is collected from financial statements of commercial

banks Besides, the macro-variables is collected from World Development Indicator and

Global Economic Policy Uncertainty Index is collected from www.policyuncertainty.com

(Becker et al.,2016).

After synthesizing the data from the research variables, I run the OLS, FEM and REMregression to examine the factors affecting NPLs of commercial banks

12

1.6 Contribution of study

The study on "Factors Affecting Non-performing Loans of Commercial Banks in Vietnam" canmake a significant contribution to understanding the credit conditions and risk managementcapabilities of commercial banks in Vietnam By analyzing factors related to thecharacteristics of banks such as bank size, credit growth, loan-to-asset ratio, return on equity,credit risk provisions, and macroeconomic variables such as fluctuations in GDP growth ratethis study aims to explain the relationships between these variables and non-performingloans in Vietnamese banks Additionally, the study will incorporate economic policyuncertainty as a variable to further enhance the understanding of these relationships withnon-performing loans Therefore, this research will help banks identify key factorsinfluencing non-performing loans, facilitating the development of solutions andrecommendations to mitigate the risk of non-performing loans of Commercial Banks in

Vietnam

1.7 Structure of the study

The research is divided into five main sections First one will provide an overview of thesubject and objectives of the study The second section will focus on an overview of previousliterature, research gaps and theoretical framework, clarifying issues related to NPLs Thenext sectiondicusses research methods, which will provide a detailed description of howresearch is carried out, including data collection and analysis The fourth section will presentconcrete findings on the factors affecting NPLs of banks Finally, the last section of the studywill also provide proposals for solutions, requests for bank support Finally, the referencesection lists the sources used for reference and conducting research

13

CHAPTER 2: LITERATURE REVIEW

2.1 Literature review

According to Berger and De Young (1997), problem loans are referred to as NPLs, while Alton

and Hazen (2001) define NPLs as loans that are overdue for 90 days or more or have stoppedaccruing interest NPLs are considered problematic or understood as difficult-to-collect debtsaccording to Fofack (2005)

Prior literature find evidence supporting that NPLs can have adverse impact on banks

(Kargi, 2011; Dimitrios et al 2016; Partovi and Matousek, 2019) In particular, Berger andDeYoung (1997) found that NPLs negatively affect banks’ efficiency and stability since theydeteriorate the quality of assets in a bank Furthermore, Kargi (2011) discovered that asbanks need to allocate financial resources to handle NPLs, their profitability can be negativelyimpacted Decreased interest rates and interest income from non-performing loans can also

have a considerable impact on the bank's revenue and profitability Simiarily, Aiyar et al

(2015) argued that a significant volume of non-performing loans have detrimental effects onbanks, leading to reduced profitability, increased provisioning requirements, and requiresubstantial resources for effective management

There are several factors that lead to bad debts in banks, as demonstrated in previousresearch A body of prior research delves into the macroeconomic factors influencing thedetermination of bad debts (Sutherland and Hoeller, 2012; Liu, 2016; Louzis et al., 2012;Amuakwa-Mensah and Boakye-Adjei in 2015; Zhang et al., 2019) For example, Louzis et al.(2012), by examining the determinants of bad debts using data from bank balance sheets,stated that bad debts in Greece are reflected through macro variables (such as GDP,unemployment rate, interest rate, public debt) and the bank's management quality Similarly,

to illustrate the adverse effects of bad debts on banks and the economy, Amuakwa-Mensahand Boakye-Adjei (2015) concluded that macroeconomic variables like inflation from theprevious year, GDP, GDP growth, and exchange rates significantly affected bad debts in thebanking sector in Ghana Overall, most of earlier studies emphasized the relationshipbetween GDP and bad debts According to Fofack (2005), GDP is considered a significantdeterminant of bad debts in African countries He argues that the relationship betweenmacroeconomic factors and doubtful loans is due to the limited diversity of certaineconomies The argument in these papers is that when real GDP growth is high, leading tohigher income, bad debts decrease significantly, enhancing borrowers’ debt repayment

14

ability Conversely, during an economic downturn, with slow GDP growth, bad debts tend toincrease Indeed, when the economy experiences robust and stable growth, employment andincome rise, creating conditions for customers and businesses to have a better ability torepay debts, thereby reducing the likelihood of bad debts Furthermore, inflation is alsoconsidered as a macroeconomic variable Nirs (2013) research has demonstrated thatincreasing inflation will reduce the real income of borrowers When interest rates for loansare floating, banks can adjust loan interest rates to maintain the real interest rate applied tocustomers, thereby reducing customers’ debt repayment capacity and increasing the baddebt ratio In a study on commercial banks in Vietnam, Vinh (2015) concluded that inflation

has an inverse relationship with bad debts Another macroeconomic variable that affects baddebts is the unemployment rate Specifically, when studying banks in Italy, Bofoni and Ropele

(2011) found that the unemployment rate impacts bad debts As the unemployment rateincreases, bad debts also rise This result is also supported by the research by Louzis et al.(2010) and Messai and Jouini (2013) They also found that an increasing unemployment ratehas a negative impact on household cash flows and increases their debt burden, leading to a

significant increase in bad debts at banks A high unemployment rate can reduce the debt

repayment capacity of individuals and businesses, leading to an increase in bad debts Inrecent studies, economic policy uncertainty is considered a challenging concept thateconomic agents must face when predicting financial and monetary developments (Chi and

Li, 2017) When researching lending decisions and credit risk in Chinese commercial banksfrom 2000 to 2014, they found that frequent and ambiguous changes in economic policiesincreased credit risk The positive correlation between economic policy uncertainty and baddebts is also found in the research by Zhang et al (2019)

Moreover, an extensive of previous research examines the impacts of bank-specificvariables on NPLs (Sutherland and Hoeller, 2012; Garciya-Macro and Robbles-Fernandez,2008; Amit Ghosh, 2015) First, studies conducted in countries such as Spain, Egypt, and theEuropean Union have suggested that loan loss provisioning ratio is positively related to baddebts (e.g., Salas and Saurina, 2002; Dimitrios et al., 2012; Amit Ghosh, 2015) In Messai andJouini's 2013 study, it was also found that banks typically set aside loan loss provisions based

on the assessment of the credit risk level of loans In cases with higher risk, a larger amount

of provisions is set aside The act of provisioning for loan loss directly impacts a bank's

15

profitability and lending process, leading to an increase in the bad debt ratio However, incontrast to this view, Boudriga et al (2009) observed that higher loan loss provisioningseemed to reduce the level of difficult-to-collect debts In addition to the bank-specificvariables influencing bad debts, we can also consider credit growth Analyzing the impact of

credit growth and non-performing loans in the United States using a vector regression model,Keeton in 1999 showed that rapid credit growth was associated with lower credit standards,

resulting in significant lending losses in certain states in the U.S (Bercoff et al., 2002; Salasand Saurina, 2002) demonstrated that credit growth affects bad debts when loans thatexceed the borrower's capacity, provided by the bank, are often considered a determinant ofpoor-quality lending It can be seen that credit growth carries the inherent risk of not

recovering interest and principal, which contributes to an increasingly severe bad debt

situation and significantly affects the bank's operations The factor of bank scale is alsoconsidered in explaining changes in bad debts (Salas and Saurina, 2002) According toDimitrios et al (2012) and Amit Ghosh (2015), bank scale is regarded as an indicator of abank's position in the market, but as the bank scale increases, the potential risk of bad debtsalso increases Conversely, Salas and Saurina (2002) and Hu et al (2004) found an inverserelationship They argue that a larger bank will have more opportunities and experience toimprove risk management and credit processes, thus significantly reducing bad debts.Furthermore, the return on equity (ROE) is another factor contributing to a bank's bad debts.While studying banks in Spain from 1993 to 2000, Garciya-Macro and Robbles-Fernandez(2008) pointed out that a high ROE would lead to higher risks in the future They argued thatprofit maximization policies are often associated with higher risk levels Louzis et al (2012)and Nir (2013) also noted that when a bank's ROE increases, it indicates that the bankoperates very efficiently, and the bad debt ratio tends to be low Conversely, if the ROEdecreases suddenly or remains at a low level, it may indicate difficulties in optimizing asset

efficiency and may increase the risk of difficult-to-collect loans Moreover, the loan-to-asset

ratio also demonstrates a positive relationship with bad debts (Dash and Kabra, 2010) Theresearch by Keeton and Morris (1987) stands as one of the earliest and significant studies onthe losses associated with lending The report concluded that excessive lending in aparticular area is a cause of high bad debts due to the inefficient operation of that sector.Moreover, the authors emphasize that it is the willingness of banks to accept risks that leads

to significant lending losses.In line with this perspective, a high loan-to-asset ratio of a bank

16

can increase the risk of bad debts when a substantial portion of the bank's assets is extended

to customers and businesses as loans Examining the case of Vietnam, Hoa (2021) used dataanalysis from 22 banks operating in Vietnam from 2012 to 2020 and found that the bankscale, the loan loss reserve ratio (LLR), and credit growth rate have a positive impact on baddebts However, the return on equity (ROE) does not have a significant correlation with baddebts

Previous studies have emphasized the relationship between macroeconomic factorsand the specific characteristics of banks in relation to non-performing loans (NPLs) Despiteadvancements in understanding the causes of NPLs, there are still gaps that need to beaddressed Previous research on macroeconomic factors has primarily focused on factors

such as GDP growth, inflation rate, and unemployment rate, while the aspect of economic

policy uncertainty has not been extensively studied Furthermore, studies on economic policyuncertainty have predominantly focused on countries in Europe Therefore, this study aims

to examine the relationship between economic policy uncertainty and NPLs in order toprovide objective assessments and recommendations for reducing NPLs in banks in Vietnam

2.2 Theoretical framework

2.2.1 Definition of non-performing loans

According to the Basel Committee on Banking Supervision (BCBS), NPLs are considered debtsthat are overdue and the borrower does not have the ability to fully repay when the bank hasnot taken any action to attempt recovery According to the International Monetary Fund

(IMF), a debt is classified as non-performing when the interest or principal repayment is

overdue for more than 90 days or interest payments due for more than 90 days have beencapitalized, refinanced, or delayed under a renegotiated agreement Additionally, debts thatare overdue for 90 days but have a solid reason to doubt the borrower's ability to fully repayare also considered non-performing Meanwhile, the concept of non-performing loans by theEuropean Central Bank (ECB) is when interest and principal repayments are overdue formore than 90 days, and the repayment ability is suspected

In Vietnam, according to the State Bank of Vietnam (SBV) with decision number493/2005/QD-NHNN regarding the issuance of regulations on debt classification, non-performing loans are categorized into Group 3 (below standard), Group 4 (doubtful debt),

and Group 5 (likely loss) Specifically, Group 3 debt has overdue periods ranging from 90 to

17

180 days, Group 4 debt has overdue periods from 181 to 360 days, and Group 5 debt hasoverdue periods exceeding 360 days Thus, it can be seen that the current definition of non-performing loans in Vietnam is quite similar to international standards and practices applied

in many countries around the world

Overall, NPLs are typically considered as a result of borrowers experiencing financialdifficulties or being unable to meet their repayment obligations These loans are oftencategorized as non-performing assets on the balance sheets of financial institutions such asbanks The classification of a loan as non-performing varies across jurisdictions and can besubject to specific regulations or guidelines set by regulatory authorities However, the

underlying principle is that non-performing loans represent a higher risk for the lender, asthere is a greater likelihood of default and potential loss of principal and interest payments

Financial institutions closely monitor their non-performing loan portfolios and make efforts

to recover the amounts owed This may involve restructuring the loan terms, negotiatingwith the borrower, or, in some cases, pursuing legal actions to recover the outstandingamounts The management and resolution of non-performing loans are important for theoverall financial health of banks and the stability of the financial system

2.2.2 Factors affecting non-performing loans

There are several micro-level factors influencing non-performing loans (NPLs) stemmingfrom the specific characteristics of banks such as the loan loss provisioning, loan to total assetratio, return on equity, loan growth, and bank size (Hasan and Wall, 2004; Dash and Kabra,2010; Louzis et al., 2012)

First, loan loss provisioning is the amount set aside and accounted for in operationalcosts to anticipate potential losses An increase in risk provisioning allows banks to augmenttheir reserves, minimizing the impact of non-performing loans on the bank and the financial

system, thereby improving risk management Hence, the loans losses reserves reflect the

general attitude of the banking system to control risks According to Messai and Jouini(2013), in evaluating credit risk, banks allocate provisions accordingly, where higher risktranslates to larger provisions Studying the period from 1993 to 2000, Hasan and Wall(2004) found that higher levels of non-performing loans are associated with high levels ofloan loss provisioning The positive correlation between loan loss provisioning and bad debts

18

is indicated in the research by Amit Ghosh (2015) Therefore, based on literature, I developthe first hypothesis that loan loss provision is positively related with NPLs.

Regarding the loan-to-asset ratio, banks with a high loan-to-total asset ratio cancontribute to a significant increase in non-performing loans (Sinkey and Greenwalt, 1991).Typically, the loan-to-total asset ratio measures the bank's dependence on lending compared

to the total assets it owns If credit risk is not managed effectively, an increase in the total asset ratio can lead to a rise in non-performing loans Poor risk management may result

loan-to-in loans beloan-to-ing granted to customers who lack the ability to repay Accordloan-to-ing to Dimitrios et

al (2012), they argue that if credit institutions wish to accelerate the momentum of creditactivities to boost profits, they will accumulate more debt Consequently, while this may

increase profits, it also raises non-performing loans During this period, banks often loosen

lending operations and credit management In line with that perspective, Dash and Kabra(2010) and Nir (2013) also demonstrate a positive correlation between the loan to toatalasset ratio and the bad debt ratio Therefore, I hypothesize that the loan-to-asset ratio haspositive relationship with NPLs in the case of Vietnam

Moreover, a high return on equity (ROE) is often associated with the level of NPLs.According to Louzis et al (2012), when bank’s profitability improves, it helps them minimizebad debts in their loan portfolio Indeed, banks with a high ROE may find it easier toaccumulate reserves for loan loss provisioning to offset potential bad debts Additionally,banks with stable profitabily can effectively control credit risk, leading to a reduction in baddebts As found in the research by Nir (2013) and Hu et al (2004), the return on equity has

an inverse relationship with bad debts Thus, studying the case of Vietnam, I expect that ROEhas an inverse relationship with NPLs

Additionally, rapid and poorly managed credit growth can lead to an increase in performing loans within the financial system (Salas and Saurina, 2002; Dash and Kabra,

non-2010) Rapid credit growth can exert significant pressure on borrowers, causing them to

excessively use financial resources or focusing lending on substandard borrowers This canresult in an inability to ensure repayment, with customers failing to repay on time, leading tonon-performing loans Some studies such as Salas and Saurina (2002) and Dash and Kabra(2010) suggest that high credit growth increases non-performing loans Besides, Amit Ghosh(2015) and Nir (2013) have demonstrated that credit growth and bad debts are positively

19

correlated with each other Hence, in this study, I hypothesize that loan growth has a positivenexus with NPLs.

Previous studies have identified bank size as a significant factor influencing the Performing Loans (NPLs) of commercial banks According to Hu et al (2004), large-scalebanks often possess more experience in dealing with fluctuations, while small-scale banksmay struggle to address issues due to lack of experience and human resources Thus, smallerbanks often exhibit a higher ratio of non-performing loans in their loan portfolios compared

Non-to larger banks Thus, I hypothesize that bank size also has a positive relationship with thenon-performing loans (NPLs) ratio

Apart from to the bank-specific factors, macro-level factors such as GDP growth also

influence non-performing loans (Fofack, 2005; Messai and Jouini, 2013; Zhang et al., 2019).GDP growth plays a crucial role in the level of non-performing loans in the economy Thestrength of a nation's development is often reflected through GDP growth GDP growth istypically accompanied by increased income and the ability of customers to repay debt In arobustly developing economy, customers have a better ability to repay, reducing the risk ofnon-performing loans When the economy grows strongly, the income of individuals andbusinesses increases, improving the debt repayment capability of customers, ultimatelyleading to a reduced risk of non-performing loans Conversely, when the economy enters astate of recession or crisis, all economic activities slow down, causing a decrease in theincome of individuals and businesses Borrowers may struggle to repay their debt, and banksmay find it difficult to recover debts from borrowers, resulting in a significant increase innon-performing loans When demonstrating that GDP growth explains changes in bad debts,the research by Louzis et al (2012); Thao and Dan (2018) have indicated a negativecorrelation between GDP growth and bad debts As a result, GDP growth rate might have anegative relationship with the non-performing loan (NPL) ratio in the case of Vietnam

Furthermore, economic policy uncertainty can be another macro-factor impactingbanks’ NPLs Chi and Li (2017) stated that economic policy uncertainty is described as thedifficulty when economic agents have to anticipate financial developments and policiesaffecting credit risk issues Economic policy uncertainty can arise from various sources,including changes in government policies, shifts in political ideologies, trade disputes,geopolitical tensions, or unexpected events Due to the uncertainty and ambiguity in

20

economic policies or continuous and frequent policy changes, banks may misallocate creditresources, leading to a gradual decrease in borrowing activities Economic policy uncertaintycan increase uncertainty in the business and investment environment This uncertainty cancreate opportunities for lending to unclear or unfeasible projects, potentially leading to an

increase in non-performing loans Additionally, the unclear nature of economic policies can

make banks more cautious in lending, necessitating careful credit risk management to reducethe risk of non-performing loans Using a panel dataset of 18 countries worldwide from 1985

to 2013, Caglayan and Xu (2019) found that economic policy uncertainty increase credit risk,indicating that economic policy uncertainty is positively related to bad debts Therefore, Iexpect that economic policy uncertainty (EPU) has a positive relationship with the non-

performing loan (NPL) ratio of Vietnamese commercial banks

2.2.3 Consequences of non-performing loans

The emergence of NPLs not only affects the operations of banks but also has an impact

on businesses and the development of the economy For commercial banks, they may facecapital erosion or reduced profitability due to NPLs, as they are unable to receive timelyinterest payments and incur additional costs such as provisions, management expenses, andother related costs in handling NPLs In cases where the profits are insufficient, banks have

to offset the losses by using their own capital As a result, the scale of operations of thesecommercial banks is also affected

Furthermore, the repayment capacity of the bank is significantly impacted One of theprimary activities of a bank is capital mobilization and lending The process of capitalrecovery is delayed due to the non-recovery of timely granted credits, while the bank still has

to meet the full and timely repayment of deposits This situation undermines the assurance

of the capital adequacy ratio, leading the bank to the risk of bankruptcy Moreover, a high

NPL ratio or the inability of the bank to repay principal and interest on time will also affect

customer confidence, gradually eroding the trustworthiness of the bank This loss ofcustomer trust further diminishes the financial strength, jeopardizing the stability anddevelopment of the entire banking system

Non-performing loans also have a significant impact on businesses (Faith Macit 2012;Louzis et al., 2012) Typically, businesses have easy access to funding from financialinstitutions, but when these institutions have a high level of NPLs, businesses may face

21

difficulties in accessing loans or extending their current borrowings, increasing thelikelihood of falling into bad debt Furthermore, NPLs can create significant financial pressure

on businesses Businesses with multiple unpaid loans have to pay interest and allocate asubstantial amount of resources to manage and resolve NPLs, which can affect their profitsand cash flow Activities such as investment and business expansion are significantly reducedwhen confronted with NPLs This can have a negative impact on the ability to expandproduction, acquire new assets, and develop the market Additionally, the reduction in profitsand the value of the business due to the influence of NPLs can make it difficult to attract andmaintain shareholders NPLs can also increase the cost of capital and interest rates forbusinesses Banks tend to raise interest rates in response to higher NPL levels, which can

further burden businesses

On the other hand, the banking system and the economy have a close relationship witheach other All activities of financial institutions are directly related to the economicdevelopment of a country They serve as a source of funds for individuals, organizations, andbusinesses According to Reinhart and Rogoff (2011), non-performing loans (NPLs) can be

an indicator of the onset of a banking crisis When bad loans increase and credit crisis occurs,the economy and finance of a country are threatened by the collapse of banks Bad loansreduce the capacity of banks to provide capital to businesses and individuals Difficulties inborrowing money and making new investments lead to a decline in business activities andinvestment, thereby reducing the pace of economic growth The risk exposure of banks alsoincreases when NPLs occur This can lead to pressure for higher interest rates or increasedcost of capital to compensate for hidden risks, thereby increasing the financial burden onbusinesses and households Moreover, bad loans negatively impact the payment capacity andcreditworthiness of banks and other financial institutions, potentially causing instability inthe financial system and affecting the overall financial and economic processes Additionally,bad loans can pose challenges to a country's economic policies Governments may need torespond to the negative effects by promoting incentivizing measures, restructuring, oreffective financial management to stabilize the economic situation Ultimately, the confidence

of investors and foreign investments in the economy will diminish Financial uncertainty andrisks can lead to the withdrawal of foreign investments, weakening the economy andreducing attractiveness to investors

22

CHAPTER 3: RESEARCH METHODOLOGY

3.1 Research data

In this study, bank-level data is mainly collected from the financial reports of 19 listed

Vietnamese commercial banks during the period from 2000 to 2022 I start the sample period

from 2000 due to the limited availability of data before this year, while 2022 is the last yearfrom which data is available to collect Additionally, this timeframe encompasses the financialcrisis period The rationale for opting for 19 listed banks is their provision of ample data

spanning the period from 2000 to 2022 Table 1 presents the list of 19 commercial banks in

the study.

Regarding macroeconomic factors, GDP growth rate data is obtained from the WorldDevelopment Indicators Additionally, data on economic policy uncertainty factors werecollected from the website www.policyuncertainty.com (Becker et al., 2016) for the periodfrom 2000 to 2022

Table 3.1: Names of 19 banks used in the sample study

Number Stock Bank name

code

23

OCB Orient Commercial Joint Stock Bank

¬ =® SHB Saigon - Hanoi Commercial Joint Stock Bank

¬ bh STB Saigon Thuong Tin Commercial Joint Stock Bank

¬ N SSB Southeast Asia Commercial Jont Stock Bank

¬ Ww TPB Tien Phong commercial Joint Stock Bank

EIB Vietnam Export Import Commercial Joint Stock Bank

¬ uo VIB Vietnam international Commercial Joint Stock Bank

¬ ® CTG Vietnam Joint Stock Commercial Bank for Industry and Trade

¬ N MSB Vietnam Maritime Commercial Joint Stock Bank

¬ oe) VPB Vietnam Prosperity Joint Stock Commercial Bank

¬ Ne) TCB Vietnam Technological and Commercial Joint Stock Bank

3.2 Research model and hypothesis

3.2.1 Research variables

I choose the non-performing loan ratio as the dependent variables Alton and Hazen (2001)

refer to non-performing loan as loans that are infeasible, overdue in principle and interest

payments for over 90 days, indicating loans that individuals finds difficulty in repaying ontime or are entirely unable to repay Non-performing loans are caculated as the ratio of non-performing to the total loan portfolio value (Basuki et al., 2021)

As for the independent variables, following the previous research, I have includedspecific bank characteristics and macroeconomic variables in my research (Alam et al., 2019)

First, I include the loan loss provisioning ratio (LLP), which is calculated as the credit riskprovisioning cost divided by total outstanding loans, as one of the factore determining NPLs

According to Messai and Jouini (2013) and Doan (2015), the loan loss provisioning ratio isincluded to capture the relationship between credit risk provisioning and bad debts.Furthermore, credit growth (LGR) is another factor that might influences a customer's debt

24

repayment capacity and can explain changes in bad debts (Salas and Saurina, 2002).Therefore, I have incorporated the credit growth variable into the model, defined as the loangrowth rate of the following year compared to the previous year, as used by (Dash and Kabra,2010; Nir, 2013) Moreover, prior literature stated that ROE might influence the level of NPLs,therefore, I also include this variable into the model Specifically, following previous studies(Louzis et al., 2012), the return on equity is calculated as net income divided by equity.Another distinctive bank characteristic is the loan-to-asset ratio (LA), representing the ratio

of outstanding loans to total assets (Heffernan and Fu, 2008) This ratio reflects a bank'slending efficiency, as a high loan-to-asset ratio indicates that the bank allocates a significantportion of its assets to lending activities Additionally, I control for bank size because it can

influence a bank's lending behavior In this study, the bank size (Size) is measured as the

natural logarithm of total assets, as done by (Vo et al., 2020)

Regarding macroeconomic variables, economic growth (GDP) is a precise indicator ofthe scale of an economy and is probably the single best indicator of economic growth (Basuki

et al., 2021) Following extant literature, I include GDP growth rate, which is the differencebetween the one year's real GDP and the prior year's real GDP divided by the prior year's realGDP, as one of the macro-factor impacting NPLs of banks In addition, as to test whethereconomic policy uncertainty influence banks’ NPLs, I also include economic policyuncertainty index constructed by Baker and et al (2016) Economic policy uncertainty index(EPU index) is collected from www.policyuncertainty.com Specifically, EPU index has gainedwidespread use and is based on articles from major newspapers containing keywordsreferring to economic policy uncertainty However, the EPU index by Baker and et al (2016)

is monthly data Therefore, to construct the yearly data to match with bank andmacroeconomics variables, I following prior literature to use the arithmetic averagingapproach to covert the original monthly index into an annual EPU measure (Karadima and

Louri, 2021; Jun et al., 2023)

25

Table 3.2: List of variables

Non-performing loan divided by total loan

The natural logarithm of total asset

The growth rate of total loan

Loan loss provisions divided by the total loan

Total loan divided by the total asset

Net income divided by the total equity

26

Financial

reports ofcommercialbanks

Financialreports ofcommercialbanks

Financialreports ofcommercialbanks

Financialreports ofcommercialbanks

Financialreports ofcommercialbanks

Financial

reports ofcommercialbanks