The importance of research topic
Lending is a crucial activity for commercial banks, driving economic growth; however, excessive loan growth can pose significant risks and diminish bank profits (Fahlenbrach, 2018; Foos, 2010) Over-lending has emerged as a pressing concern, as abnormal credit growth signals potential imbalances within the banking system, potentially leading to crises that affect both banks and the broader economy (Elekdag, 2011) While credit growth is essential for banks to enhance market share and profitability, it is vital for them to establish realistic loan growth targets (Chen, 2014) In developing countries, bank credit serves as a key driver of economic progress, with substantial funds from state-managed banks contributing to notable credit growth rates, which are vital national indicators (Vo, 2018) Thus, analyzing the impact of credit growth on bank performance, particularly in terms of profitability and credit risk, is critical for developing strategies that improve business outcomes while ensuring stability in the banking sector and money market.
From 2012 to 2020, Vietnam experienced a significant recovery in credit growth following the 2008 global economic crisis Despite the lessons learned regarding the dangers of excessive subprime credit growth, as noted by Esa Jokivuolle (2017), it remains overly optimistic to believe that these lessons will completely eliminate the risk of future crises This ongoing challenge necessitates continuous research and the development of new models by researchers and policymakers to assess the impacts of credit growth on the economy It is crucial to identify the thresholds at which credit growth becomes problematic to prevent potential economic downturns.
This thesis examines the complex relationship between credit growth and its impact on banking stability, specifically focusing on profitability and credit risk While macroeconomic predictions are challenging due to the numerous influencing factors involved, the author aims to identify the optimal credit growth threshold for commercial banks This insight is intended to guide important policy decisions that will enhance both stability and development within the banking sector.
Relevant research situation and research problem
Research on credit growth and its effects on the banking sector and the broader economy has surged since the 2008 financial crisis, primarily due to the abnormal rise in subprime credit Keeton's (1999) comprehensive study highlights the interplay between credit growth and bank-specific indicators, revealing that increased loan growth can lead banks to lower credit standards, resulting in higher loan losses Conversely, firms may shift financing from capital markets to banks, or productivity gains may enhance return on investment, indicating that rapid loan growth does not always correlate with increased loan losses Further investigations, such as those by Salas and Saurina (2002) in Spain, Hess et al (2008) in Australia, and Foos et al (2010) analyzing a dataset of 16,000 banks over a decade, illustrate the varied impacts of credit growth on banks' profitability and credit risk.
Recent research on credit growth in Vietnam has been diverse and abundant, yielding varying results across different periods Notable studies include Dang (2019), which examines the impact of credit growth on bank performance indicators, and Vo (2018), which investigates the factors influencing credit growth These studies highlight the complexity and multifaceted nature of credit dynamics in the Vietnamese banking sector.
Firstly, there are still few scientific studies studying the impact of credit growth on the performance factors of banks in Vietnam fully including micro and macro explanatory variables
Second, in Vietnam, there are no in-depth studies on effective credit growth thresholds that maximize bank performance indicators, so that accurate policy implications can be made and effective
Research on the credit growth of Vietnam's commercial banking system primarily employs standard static panel data methods, such as fixed and random effects However, these approaches may produce biased results due to unaddressed endogeneity issues Additionally, the studies are often limited to short timeframes, utilize small sample sizes, and fail to comprehensively capture the variables influencing non-performing loans.
To address the research gap and practical needs in the banking sector, this study aims to evaluate the impact of credit growth on bank profitability and credit risk, using reliable data on non-performing loans and advanced estimation methods for accurate results The thesis develops a credit growth model that incorporates both bank-specific and macroeconomic factors while identifying the optimal credit growth threshold Unlike previous studies in Vietnam that primarily relied on conventional static panel data methods, this research employs dynamic panel data estimation and Hansen's threshold model (PTR) to enhance the understanding of credit growth's effects on banking activities.
In 1999, research was conducted to analyze the impact of lagged variables and the endogenous influence within the research model, aiming to identify the optimal threshold for credit growth The findings of this study offer valuable scientific evidence that can assist managers and government agencies in selecting effective strategies to optimize credit growth in Vietnamese commercial banks.
Research Objectives
This thesis investigates the relationship between credit growth and bank profitability, as well as credit risk, providing managers with insights to enhance profits and effectively manage risks To accomplish this overarching objective, the research outlines three specific goals to be addressed.
This study aims to examine the relationship between credit growth and the profitability of Vietnamese commercial banks Specifically, it seeks to answer the critical question of how credit expansion influences the profitability of these banks Additionally, the research will investigate whether an increase in credit growth correlates with enhanced profitability within Vietnam's commercial banking sector.
The second objective of this study is to examine how credit growth influences the credit risk associated with Vietnamese commercial banks To address this, the research will focus on the question: What is the relationship between credit growth and credit risk in Vietnam's commercial banking sector? Specifically, the analysis aims to determine whether an increase in credit growth correlates with heightened credit risk within the Vietnamese banking system.
The third objective of this research is to identify the optimal credit growth threshold for Vietnamese commercial banks This study aims to answer the critical question: What credit growth level maximizes profitability while minimizing credit risk for these banks?
Research Scope and Subject
This study addresses a gap in previous research by analyzing the effects of credit growth on the operations of Vietnamese commercial banks, specifically focusing on a sample of 20 banks from 2012 to 2020, which includes State-Owned Commercial Banks (SOCBs) and Joint Stock Commercial Banks (JSCBs) This sample represents a significant portion of the 35 existing JSCBs in Vietnam, ensuring its relevance The research aims to explore three main areas: the impact of credit growth on the profitability of Vietnamese commercial banks, its effect on credit risk, and the identification of an optimal credit growth threshold.
Research Methods
To determine the sign and magnitude of the regression coefficients, the quantitative method used in the thesis are estimation methods for the regression model with panel data
This thesis employs a dynamic panel data approach, specifically a two-step system GMM, to investigate the interplay between credit growth, bank profitability, and credit risk It posits that changes in bank-specific and macroeconomic factors can lead to endogenous adjustments in these relationships Given the potential for endogeneity due to the reciprocal nature of credit growth and profitability, traditional static estimators like fixed effects (FEM) and random effects (REM) may yield biased results To address this, the study utilizes the GMM method to manage endogeneity and lagged variables effectively The Sargan-Hansen test confirms that the instrumental variables used meet the model's overidentifying restrictions The analysis and regression coefficient estimations are conducted using Stata 11.0 software.
GMM estimation, an instrumental variable-based technique, offers significant advantages over traditional estimation methods, particularly in scenarios with variable variance Unlike traditional estimates, GMM effectively utilizes moment conditions to deliver precise estimates, even when faced with cross-unit inconsistencies (Hansen).
2000) To test the robustness of the estimate, the study uses the two-step system GMM method developed for the linear dynamic table model (Arellano and Bond 1991; Arellano and Bover, 1995)
To determine the ideal credit growth threshold for banks, the author employs the PTR threshold regression model introduced by Hansen in 1999 This model is favored in financial and macroeconomic analysis due to its simplicity and clear policy implications It assumes a non-linear relationship between credit growth, bank profitability, and credit risk.
New contributions of the study
Compared with previous studies on the same topic that the thesis has referred to, the thesis has new contributions as follows:
This thesis explores the connection between credit growth and the profitability of Vietnamese banks during recent periods of volatility The findings indicate a positive correlation between these two factors, highlighting the motivation behind commercial banks' pursuit of increased credit growth.
To enhance operational efficiency, commercial banks must reduce input costs, which will help them manage loans more effectively and lower non-performing loans A robust credit growth policy is essential, as research indicates a strong positive relationship between credit growth and the non-performing loan ratio This study utilizes the GMM generalized dynamic panel data estimation model to analyze the effects of lagged return on assets (ROA) and non-performing loans Additionally, the findings highlight the significant influence of macroeconomic factors, particularly the real estate market, on banks' profitability and credit risk.
The thesis identifies the optimal credit growth threshold for banks, providing valuable policy insights for both banks and policymakers A key takeaway from this study is that to enhance bank profitability and mitigate credit risk, banks should implement a balanced credit growth strategy that considers various influencing factors.
Research process and structure of the thesis
The research process is carried out as follows: Step one, identify the problem to be researched, the second step is to review the theoretical framework
The thesis comprises five chapters, outlining the process of identifying research gaps through theoretical and empirical results Step three focuses on the research methodology, followed by step four, which involves data collection and quantitative analysis Finally, the results are discussed, leading to key recommendations.
Chapter 1 Introduction This chapter introduced the need as well as the research objectives, scope and object of the research, and the research process
Chapter 2 Theoretical framework and overview of previous studies on bad debt of commercial banks In this chapter, the thesis presents the theoretical framework of credit growth affecting profitability and credit risk from previous documents The main theoretical framework that the study relies on to explain the relationship between specific factors and bad debt is the money supply and demand theory, bank lending channel, monetary policy transmission channel, and hypotheses such as Moral hazard, mismanagement, and the size effect of Kenton's credit supply and demand hypothesis
In 1999, the thesis examines prior empirical studies to uncover quantitative factors that contribute to the development of an empirical model assessing the influence of credit growth on banking operations in Vietnam, along with determining the optimal credit threshold.
Chapter 3 Research model, research method, and research data Starting from the theoretical framework in Chapter 2, inheriting the empirical models of the above studies, this chapter will build empirical models of the thesis, including the model of credit growth affecting the bank profitability and risk and credit growth threshold model Finally, variable measurement and data mining sources are also detailed in this Chapter
Chapter 4 Analysis of research results Based on the experimental model and collected data of 20 Vietnamese commercial banks, the thesis uses the supporting tool Stata 11.0 software to perform the tests and estimate the regression coefficients of the variables in the model Then discuss the experimental results based on the theoretical foundation of research and comparison with previous studies in order to interpret the results logically This result provides evidences to help answer the research questions of the thesis
Chapter 5 Conclusion and solutions This chapter has summarized the main experimental results associated with the research objectives of the thesis From there, the thesis provides some policy implications to control credit growth through the effects of credit growth on profitability and credit risk of banks as well as the minimum credit growth threshold superior These suggestions are expected to provide more references for policymakers when looking for solutions to control credit growth At the same time, this chapter also recognizes some limitations that the thesis has not yet resolved This is also the last chapter of the thesis.
Concept of credit growth, profitability and credit risk
Credit growth is a multifaceted concept that can be interpreted in various ways, influenced by the researcher's perspective According to the International Monetary Fund, International Financial Statistics, World Bank, and OECD GDP, credit growth refers to the domestic credit extended by the financial sector, which encompasses all credit to different sectors on a gross basis while excluding net credit to the central government The financial sector includes monetary authorities and deposit banks, and credit growth is typically expressed as a percentage of GDP (Dutta and Wunsch-Vincent, 2017).
In Vietnam, credit extension, as defined by Article 3 of the Law on Credit Institutions 2010, refers to agreements that allow individuals or organizations to utilize funds with the obligation of repayment The growth of bank credit is measured by the increase in total credits granted annually, serving as a crucial indicator for the State Bank's monetary policy management This growth is influenced by various monetary policy tools, including compulsory reserves and open market operations, which regulate the money supply and directly impact credit growth The State Bank of Vietnam (SBV) enforces strict control over credit growth by setting annual targets and monitoring credit institutions to ensure system safety and manage non-performing loans This approach aims to balance the money supply within the economy, facilitating business borrowing while maintaining macroeconomic stability Consequently, commercial banks must carefully select and manage qualified loans, focusing on borrowers with repayment capacity to optimize interest rate differentials.
2.1.2 Concept of profitability and credit risk
In commercial banking, operating results indicate how effectively resources are utilized to maximize outcomes while minimizing costs The assessment of a bank's performance varies widely within the banking sector, influenced by the specific objectives of the analysis Key profitability metrics include Return on Total Assets (ROA), Return on Equity (ROE), Net Interest Margin (NIM), Non-Interest Margin (NNIM), and Cost to Income Ratio (CIR).
Bank profitability and performance can be assessed using various financial ratios, each serving a distinct purpose The Net Interest Margin (NIM) and Non-Interest Margin (NNIM) ratios differentiate between interest and non-interest income, while the Cost-to-Income Ratio (CIR) evaluates performance through operating cost management Return on Assets (ROA) and Return on Equity (ROE) ratios measure the efficiency of a bank's business operations A low ROA indicates poor asset utilization, potentially due to inflexible investment policies or high operating costs, whereas a high ROA signifies effective management and a sound asset structure ROA also reflects management efficiency, showing how well a bank generates profits from its assets Research has shown that a higher ROA correlates with better resource management and income generation In contrast, the ROE ratio assesses shareholder returns on bank investments, providing insights for investors when comparing stocks within the same industry A higher ROE indicates more efficient use of equity and a balanced approach to equity and loan capital.
In this thesis, the author utilizes Return on Assets (ROA) as a key indicator of bank profitability, reflecting the efficiency of the bank's investments relative to its assets ROA provides insights into the bank's ability to generate income from its assets, excluding cash and fixed assets, and highlights effective asset management amidst economic fluctuations A high ROA signifies strong business performance and a well-structured asset portfolio, showcasing the bank's adaptability to economic changes To enhance ROA, banks must focus on increasing profitable assets, particularly through lending, which serves as the primary source of profit, albeit with associated risks.
Credit risk, as defined by Decision 493/2005/QD-NHNN and Circular 02/2013/TT-NHNN, refers to the potential loss faced by credit institutions when customers fail to meet their financial obligations This risk can arise from the customer's inability to fulfill part or all of their debt commitments Key metrics for assessing bank credit risk include the overdue debt ratio, non-performing loan ratio, and the adequacy of provisions for potential losses In this study, the non-performing loan ratio is selected as the primary indicator of a bank's credit risk.
There is currently no universal definition of non-performing loans (NPLs), as different central banks and international organizations define the term based on their specific criteria and perspectives on debt The World Bank characterizes non-performing loans as subprime loans that are overdue and raise doubts about the debtor's ability to repay This situation often arises when debtors declare bankruptcy or liquidate their assets According to this classification, subprime loans include those that are 90 to 180 days overdue or have been renegotiated, while debts are deemed doubtful if recovery is uncertain, typically when they are overdue between 180 to 360 days.
The Basel Committee on Banking Supervision (BCBS) does not establish a specific maturity period for classifying a loan as "bad," as this can vary across countries, with definitions ranging from 30 to over 180 days past due A loan is deemed non-performing when it is overdue, and the bank determines that the borrower is unlikely to repay the full amount without intervention The BCBS avoids emphasizing a strict overdue timeframe because different countries have varying standards for reporting non-performing loans, with some considering loans non-performing after just 31 or 61 days As a result, while 90 days is a commonly accepted benchmark, it is not a universally applied criterion for classifying non-performing loans.
Non-performing loans (NPLs) are a complex concept with varying definitions across international organizations Generally, NPLs are identified based on two key factors: a payment overdue by more than 90 days and the borrower's questionable ability to repay This definition aligns with the approach taken by the State Bank of Vietnam in classifying non-performing loans.
In this study, a non-performing loan is defined as a loan payment, either of interest or principal, that is overdue by more than 90 days, raising concerns about the borrower's repayment capability This definition aligns with the guidelines set forth in Circular 02/2013/TT-NHNN, issued on January 21.
2013 by the State Bank of Vietnam, ―non-performing loan‖ is debts belonging to groups
According to Articles 10 and 11 of this Decision, Group 3, which pertains to substandard debt, encompasses debts that have been overdue for 90 to 180 days and debts that have a restructured repayment term of less than 90 days based on the new terms established.
Doubtful debts are categorized as those overdue between 181 to 360 days, and debts with restructured repayment terms that are overdue from 90 to 180 days based on the new schedule Group 5, identified as potentially losing capital, encompasses debts overdue for more than 360 days, debts frozen pending government intervention, and debts that have been rescheduled but are overdue for over 180 days This analysis also incorporates the non-performing loan ratio to the total outstanding balance, as reported in accordance with Vietnamese accounting standards for commercial banks, to support the quantitative research findings.
In Vietnam, credit institutions classify debts based on qualitative and quantitative methods as outlined in Decision 493/2005/NHNN and Circular No 02/2013/NHNN However, the majority of these institutions prefer the qualitative approach According to these regulations, debts are categorized into five groups: Group 1 - Qualified debt, Group 2 - Debts needing attention, Group 3 - Subprime debt, Group 4 - Doubtful debt, and Group 5 - Debts likely to lose capital Notably, debts classified in groups 3, 4, and 5 are considered non-performing loans.
Theoretical basis of credit growth
Friedman's modern money demand theory
In 1956 Milton Friedman developed Friedman's monetarist theory in his famous paper
The "Quantity Theory of Money: A Reconfirmation" highlights that the demand for money is influenced by the same factors affecting any asset's demand, as articulated by Friedman He introduces the Money Demand Equation, which posits that money demand is determined by individual resources and the expected returns on alternative assets, alongside currency expectations In this model, a positive relationship is indicated by the sign (+), while a negative relationship is denoted by the sign (-).
Md/P = f (YP, rb - rm, re - rm, πe - rm)
Yp = The variable that Friedman uses to measure wealth is income in the long run rm = Expected return of money
Tb = The bond's expected return re = Expected return of the stock πe = expected inflation rate
The demand for money is positively correlated with wealth, as measured by Friedman's long-run income variable (Yp), which remains constant regardless of economic cycles Friedman categorizes assets into three main types: bonds, stocks, and commodities Individuals tend to prefer holding these assets over cash due to the higher expected returns they offer compared to holding money Consequently, the demand for money decreases as the gap between the expected returns of these assets and the expected return on currency widens.
The financial accelerator theory, proposed by Bernanke and Gertler (1995), posits that minor fluctuations in financial markets can lead to significant economic shifts, creating a feedback loop that amplifies economic booms and busts, a concept central to Keynesian macroeconomic theory Increased demand can drive substantial investment spending, resulting in rapid economic growth, while decreased demand may lead to significant cuts in investment and a downturn in business activity Business expectations regarding future consumer demand are crucial in influencing these dynamics This theory is essential for understanding prolonged recessions, as the acceleration effect interacts with the investment multiplier effect, further intensifying both economic expansions and contractions.
When a central bank increases the money supply, it often leads to investors channeling funds into real estate rather than production, potentially sparking a real estate boom This surge can inflate property values beyond their true worth, resulting in decreased liquidity Consequently, if real estate prices drop, investors may struggle to repay bank loans, leading to a financial crisis.
The theory of monetary policy transmission channel for bank lending
Monetary policy influences credit through lending and balance sheet channels, as outlined by Bernanke and Blinder (1998) When monetary policy tightens, banks raise interest rates and tighten credit terms, limiting credit supply and discouraging risky investments This shift not only impacts borrowers' access to credit but also affects money supply, interest rates, and inflation Borrowers with poor financial health face higher costs and stricter credit conditions, which can adversely affect their investment and spending capabilities Consequently, tight monetary policy directly influences borrowers' balance sheets by increasing interest payment costs, reducing net cash flows, decreasing the value of collateral, and leading to lower consumer spending and business revenues.
2.2.2 Theory of bank-specific factors
Kenton's theory of credit growth
Keeton (1999) elucidates the intricate relationship between credit growth and non-performing loans, highlighting how these factors can move in both the same and opposite directions This dynamic is illustrated by shifts in the supply curve within the loanable funds market, where the expected rate of return is influenced by the credit standards established by banks A rightward shift in the supply curve indicates that commercial banks are more inclined to lend by lowering their credit standards, thereby allowing borrowers with higher financial risks to access funds, albeit with the expectation of reduced profitability.
When commercial banks loosen lending conditions to increase credit, it often leads to a decline in credit quality over time Specifically, a decrease in credit standards initially boosts credit growth but ultimately diminishes quality Factors influencing this relationship include shifts in the demand curve and labor productivity A rightward shift in the demand curve, driven by changes in an enterprise's capital structure, enhances cash flow and the borrower's repayment ability, thereby maintaining credit quality Additionally, increased labor productivity signals positive borrower prospects Consequently, credit growth can positively correlate with non-performing loans As illustrated, when loan demand arises from the need to optimize capital structure, banks tend to tighten credit standards after recognizing a pool of qualified borrowers, which helps mitigate risks associated with lending to financially unstable entities This process ultimately leads to improved credit quality, as increased credit growth fosters higher standards and reduces non-performing loans.
Increased labor productivity can lead to favorable lending conditions, allowing commercial banks to adjust credit standards flexibly for borrowers This adaptability enables banks to increase credit availability, positively impacting borrowers' financial performance As a result, credit growth may either decrease or increase non-performing loans, illustrating the dynamic relationship between labor productivity and credit standards.
The "size effect hypothesis" posits that larger banks offer greater diversification opportunities, as noted by Salas and Saurina (2002) In contrast, the "too big to fail" hypothesis suggests that large banks tend to assume excessive risks by leveraging more loanable funds, leading to higher rates of non-performing loans due to a lack of market discipline and reliance on government protection (Stern and Feldman, 2004) This excessive leverage often results in lending to lower-quality borrowers Boyd and Gertler (1994) highlight that during the 1980s, the "too big to fail" policy encouraged larger US banks to maintain riskier portfolios Furthermore, the "Loan portfolio diversification hypothesis" indicates that diversification opportunities are linked to credit quality, suggesting an inverse relationship between loan portfolio diversification and the non-performing loan (NPL) ratio, as effective diversification mitigates credit risk.
Review of previous studies
Research on the relationship between credit growth and bank profitability in the world has some remarkable results:
Foos (2010) presents comprehensive evidence on the relationship between abnormal loan growth and the riskiness of individual banks, utilizing Bankscope data from over 16,000 banks across 16 major countries between 1997 and 2007 The study tests three hypotheses regarding the impact of past abnormal loan growth on loan losses, bank profitability, and solvency, while controlling for year- and country-specific effects The findings reveal that past abnormal loan growth significantly increases subsequent loan losses with a lag of two to four years, aligning with aggregate data on the link between loan growth and losses in individual countries Additionally, the research indicates that abnormal loan growth results in a decrease in banks' relative interest income, suggesting that new loans are often issued at rates insufficient to cover the associated default risk.
A study by Rossi (2019) on 100 European banks reveals that unusual loan growth can align with rising bank profitability This phenomenon, known as the "winner's curse," arises from underestimating risk and short-sighted management To mitigate these issues, it is crucial for banks to implement effective provisioning and conduct thorough credit risk assessments Ultimately, controlling abnormal credit growth and ensuring adequate provisions for credit risks are vital policy measures for sustainable banking practices.
Bhowmik (2021) analyzed data from BankFocus, covering 118 commercial banks from 2011 to 2019, revealing that loan growth can elevate bank risk while simultaneously boosting profits The study identified a strong positive correlation between Return on Assets (ROA) and Non-Performing Loans (NPL), suggesting that the rapid development of the banking sector in emerging Asian economies is linked to increased financial resources from industrialization This economic growth has led to a higher standard of living, enabling greater consumer loan usage and expanding lending opportunities, which in turn enhances loan yields.
Fahlenbrach (2018) analyzed a sample of US banks from 1973 to 2014 and discovered that banks in the top quintile for loan growth over three years performed significantly worse in the following three years compared to those in the bottom quarter These rapidly growing banks exhibited lower return on assets and higher provisions for loan losses after periods of expansion Notably, their underperformance could not be attributed to merger activities, suggesting that banks, analysts, and investors tend to overestimate the risks associated with loans during times of high growth.
Paul (2016) investigates the impact of loan portfolio growth on the financial performance of commercial banks in Kenya, revealing that such growth negatively affects their overall performance The study highlights that an expanding loan portfolio leads to a rise in underperforming loans in subsequent years Additionally, it notes that commercial banks often lower their lending rates to boost their loan books, while also becoming more cautious in their lending practices following periods of significant losses attributed to bad loans.
Amador (2013) presents compelling evidence linking abnormal loan growth to increased risk-taking behavior among Colombian banks The study indicates that sustained abnormal credit growth results in heightened bank riskiness, diminished solvency, and a rise in the ratio of underperforming loans to total loans Furthermore, it highlights the critical role of abnormal credit growth in contributing to bank failures during Colombia's late 1990s financial crisis.
Baron và Xiong (2017) analyze 20 developed Asian economies over the period 1920–
In 2012, it was observed that an expansion in bank credit is linked to a heightened risk of bank equity collapse, yet this increased crash risk is associated with lower average bank equity returns over the subsequent one to three years Furthermore, when a country's bank credit exceeds the 95th percentile, the bank equity index experiences an alarming excess return of -37.3% over the next three years.
Le (2020) examines the interrelated factors influencing bank stability, profitability, and loan growth in the Vietnamese banking sector from 2006 to 2017 using the GMM estimator The study reveals a bidirectional relationship among these variables, indicating that sound banks tend to be more profitable but experience lower loan growth While loan growth can enhance bank profitability, it simultaneously decreases bank stability Notably, a quadratic relationship exists, where excessive loan growth negatively impacts both profitability and stability, prompting bank managers to be cautious in pursuing aggressive loan growth strategies Additionally, the U-shaped relationship between bank profitability and loan growth advises managers to diversify into non-traditional activities to sustain profitability Furthermore, the findings highlight a negative correlation between bank size and stability, supporting the "too-big-to-fail" hypothesis, which calls for the State Bank of Vietnam to exercise caution in future bank mergers involving large institutions.
Do (2017) conducts a comprehensive study on the effects of rapid credit expansion on the health of Vietnamese commercial banks, focusing on key factors such as efficiency, profitability, asset quality, and capital structure Utilizing both qualitative analysis and quantitative surveys of big data samples from 2005 to 2013, the study reveals a negative impact of the credit boom on the overall health and efficiency of these banks While rapid credit growth boosts bank profits, as lending remains the primary income source for the Vietnamese banking system, it also poses significant risks to long-term stability.
A study by Dang (2019) investigates the impact of credit growth on bank performance in Vietnam from 2006 to 2017, focusing on credit risk, profitability, and solvency The regression analysis of both static and dynamic panel data models reveals significant evidence that credit growth indicators substantially influence bank performance Notably, an increase in loan growth leads to higher loan risk provisions over the subsequent 2 to 3 years, which subsequently reduces the proportion of bank capital in the following year.
Symbol Positive effect on CGR Negative effect on CGR
Foos (2010), Fahlenbrach (2018), Hou et al (2015), Hess et al
(2009),Schularick và Taylor, (2012),Paul (2016), Baron và Xiong
Bhowmik (2021),Foos (2010), Hess et al (2009), Laidroo and Mannasoo (2014), Baron và Xiong, (2017),Salas và Saurina
Nguyen Ngoc Diep và Nguyen Minh Kieu
MODEL, RESEARCH METHODS AND DATA RESEARCH
Research models
This study aims to analyze the impact of credit growth on the profitability and credit risk of banks, drawing on Kenton's (1999) fundamental theory of credit supply and demand It examines the relationship between credit growth and profitability, considering the trade-off between risk and expected return Utilizing empirical models from previous research, the thesis develops a dynamic econometric model to investigate how credit growth influences the profitability and credit risk of Vietnamese commercial banks, incorporating lags of the explanatory variables.
Which t and i=[1, 2, N] are year t and bank i respectively, where βs are regression
The coefficients of explanatory variables and the error term εit capture unobservable effects in the analysis of bank performance The dependent variables, Return on Assets (ROA) and Non-Performing Loans (NPL), indicate the banks' profitability and credit risk, respectively, measured through the rate of return on assets and the non-performing loan ratio The lagged values of these dependent variables suggest a significant influence of previous periods on current performance metrics Additionally, the explanatory variable, Customer Loan Growth Rate (CGR), reflects the expansion in banks' customer loan portfolios, as noted in studies by Keeton (1999) and Foos (2016).
This study builds on previous research by examining the effects of loan growth with lags of up to three years, specifically at k = 1, 2, and 3, to identify any lagged impacts Additionally, it incorporates a set of control variables, referred to as Vector Bank, which includes bank-specific factors known to influence the safety and soundness of financial institutions, as highlighted in various studies (Bertay et al., 2013; Cohen & Scatigna, 2016; Kashif et al., 2016).
Macro represents a group of macro factors affecting bank profitability and credit risk
The following table summarizes the calculation formula, sign expectations and prior research of the variables:
Table 2.2 Summarizing the formula of variables and the sign expectation
Expected correlation coefficient Pre-research
ROA Profitability Net Income / Total
Non-performing loans / Total outstanding loans
ROA t-1 Past Profitability Net Income t-1 /
NPL t-1 Past Non- performing loan
Non-performing loans t-1 / Total outstanding loans t-1
CGR t-1 Credit growth before 1 year
(Loan balance t-1 - Loan balance t-2) / Loan balance t-2
(2015), Hess et al (2009),Schularick and Taylor, (2012),Paul
CGR t-2 Credit growth before 2 year
(Loan balance t-2 - Loan balance t-3) / Loan balance t-3
CGR t-3 Credit growth before 3 year
(Loan balance t-3 - Loan balance t-4) / Loan balance t-4
TA Bank size LOG (Total Asset) (+) (+) Dang (2019)
ETA Effective use of capital Equity / Total assets (-) (-)
GDP Economic growth GDP Index-100 (%) (+) (-)
(2011), Sufian and Habibullah (2012), (Jimenez et al
REL Real estate loan growth
(Real estate loan balance t - real estate loan balance t-1)/ real estate loan balance t-1
3.1.1 Model of the impact of credit growth on bank profitability
In this thesis, the author uses the Return on Assets (ROA) as the key variable to measure bank profitability, opting for it over Return on Equity (ROE) to emphasize the bank's perspective ROA is derived from the profit after tax from the income statement and total assets from the balance sheet Given that banks typically operate with high debt ratios, their ROA tends to be lower Additionally, banks utilize significant financial leverage, meaning that a large portion of their assets is financed through mobilized capital from the public According to Athanasoglu et al (2008), banks with lower financial leverage generally exhibit higher ROA but lower ROE, as ROE does not account for the increased risk associated with high leverage or regulatory impacts Consequently, prominent researchers like Athanasoglu et al (2008) and Dietrich and Wanzenried (2011) advocate for the use of ROA as the primary metric for evaluating bank profitability, which justifies its selection in this analysis of commercial banks.
Variable of Return on asset in the past
The author incorporates a lagged variable of Return on Assets (ROA) to illustrate the cyclical effects in banking, paralleling findings by Albertazzi and Gambacorta (2009) that indicate bank performance mirrors broader business cycles, with indicators improving during economic upturns and declining during downturns Additionally, Fahlenbrach et al (2018) highlight that banks in developing countries experience rapid profit growth, attributed to the phenomenon where smaller banks can achieve significant growth rates more easily than their larger counterparts in developed nations Research by Dang (2019) and Foos (2010) further supports that previous profitability notably influences current ROA, particularly during periods of credit expansion Consequently, this study posits that past ROA positively affects current ROA.
Variable of credit growth in the past
This study analyzes the impact of loan growth over a three-year period, utilizing data from balance sheets and annual audit reports, in line with previous research by Keeton (1999), Foos (2016), and Kashif et al (2016) It explores the "Financial Acceleration Mechanism" proposed by Fisher (1933) and Keynes (1932), which suggests that during credit booms, economic indicators may improve due to rising asset prices and optimistic market expectations Furthermore, Foos (2016) indicates that abnormal loan growth can reduce banks' relative interest income, highlighting the risk that new loans are issued at rates insufficient to cover potential defaults This underscores the necessity for banks to assess whether the additional income from increased lending justifies the heightened risk involved.
The asset size variable (TA) serves as a key indicator of bank size, often measured using the logarithm of total assets due to their large absolute values Research indicates that larger commercial banks tend to be more cost-effective, with studies by Wheelock and Wilson (2012) highlighting economies of scale across various bank sizes Stever (2007) noted that large banks benefit from diversified asset portfolios, unlike smaller banks, which may face challenges in risk management However, some studies, such as Pasiouras and Kosmidou (2007), found a negative correlation between bank size and efficiency, suggesting that smaller banks may have advantages in certain contexts Additional research, including work by Ukuyama (1993) and Sufian (2011), indicates that total asset size may not significantly impact performance Nevertheless, the author posits that bank size positively influences profitability, particularly in Vietnam's lending market, which is dominated by a few large banks.
The ratio of equity to total assets (ETA)
The capital adequacy ratio is essential for evaluating the suitability of capital in banking, as defined by Decision 457 of the State Bank of Vietnam and the Law on Credit Institutions Equity capital, which includes charter capital and reserve funds, serves as the foundation for calculating prudential ratios, encompassing Tier 1 and Tier 2 capital A higher equity ratio is believed to enhance loss tolerance from business risks, particularly credit risks, thereby facilitating credit growth and potentially increasing returns Additionally, a stronger equity position can improve credit ratings and reduce capital costs for commercial banks However, portfolio theory indicates a trade-off between risk and expected returns; as equity ratios increase, overall risk diminishes, which may lead to lower expected returns compared to banks with higher financial leverage Research shows that banks with substantial equity and low leverage often experience lower return on equity (ROE) The relationship between equity ratio and profitability remains inconclusive, varying by research sample and dependent variables This study posits that a higher equity ratio positively influences return on assets (ROA), suggesting that banks with greater equity are more stable and better equipped to handle operational risks, ultimately driving credit growth and profitability.
Loan to deposit ratio (LDR)
A higher loan-to-deposit ratio (LDR) indicates that a bank is effectively fulfilling its intermediary function, as it suggests that the bank's loans exceed its mobilized capital, necessitating additional funding through interbank loans, trust capital, or its own capital Research by Buchory (2006) highlights that the effectiveness of financial intermediation positively influences banking performance and profitability While numerous studies (Nusantara, 2009; Prasanjaya, 2013; Artarina, 2013; Vong, 2009; Widati, 2012; Restiyana, 2011) demonstrate a significant positive relationship between LDR and return on assets (ROA), some studies (Arimi, 2012; Purwoko, 2013) found no such correlation Despite the increased credit risk associated with high lending relative to mobilized capital, it also indicates that banks are effectively utilizing their capital to generate profits, supporting the notion that LDR positively impacts ROA in this analysis.
According to the "Financial Acceleration" theory presented above, GDP growth is usually accompanied by an increase in aggregate demand of the economy and
Vietnamese commercial banks are essential for economic financing, driven by increasing demand for traditional products like credit and capital mobilization, as well as services such as payments and guarantees The author anticipates that GDP will positively influence economic growth, aligning with findings from researchers like Dang (2019) and Sufian and Habibullah (2012) However, some studies present contrasting views, suggesting that heightened competition may lead to instability in the banking system, as indicated by the "Moral hazard" hypothesis (Jimenez et al., 2007).
Variable loan balance in the real estate market (REL)
Monetary policy, particularly the M2 money supply policy, significantly influences credit growth, especially in real estate, impacting the housing market's development The growth rate of real estate loans is crucial, as banks assess collateral values when extending credit This relationship between asset prices and loan growth cycles suggests that rising collateral prices enable banks to finance more borrowers Research by Davis and Zhu (2009) indicates that increasing property prices lead to expanded lending, reduced risk premiums, and enhanced bank profitability Conversely, declining property prices can restrict loan growth, widen loan margins, and decrease bank profits, potentially resulting in credit constraints In Vietnam, real estate loans constitute a substantial portion of the banking system's total loan balance, meaning fluctuations in the housing market can significantly affect banks An upturn in the real estate market typically drives robust credit growth, particularly in mortgage loans, while a high concentration of business and real estate loans heightens credit risk during downturns Ultimately, rising real estate prices are anticipated to boost customer loans for property purchases, thereby enhancing bank profits.
3.1.2 Model of the impact of credit growth on credit risk
In this thesis, the dependent variable representing credit risk is the non-performing loan (NPL) ratio, which is calculated as the percentage of non-performing loans relative to the total outstanding balance of each bank According to Vietnam's debt classification guidelines, non-performing loans are categorized as those in groups 3 to 5 on banks' balance sheets The relevant debt items for these groups are sourced from the notes to the financial statements and annual income statements, while the total outstanding balance is obtained from the banks' balance sheets The NPL ratio is computed annually for each bank using this formula.
Variable of Non-performing loan in the past
The past non-performing loan (NPL) ratio significantly influences the current NPL ratio, indicating a persistent impact of credit risk over time Research by Jimenez and Saurina (2006) supports the notion that higher past NPLs correlate with a bank's diminished capacity to manage lending risks, resulting in elevated current NPL levels Empirical evidence from studies by Salas and Sauria (2002), Jimenez and Saurina (2006), and Klein (2013) reinforces this relationship, suggesting that the non-performing loan variable exhibits cyclical tendencies within the economic landscape.
Research on credit growth, referencing Kenton (1999) and studies by Gambacorta & Marques-Ibanez (2011) and Kashif et al (2016), indicates that credit growth typically experiences a lag of 1-4 years, influenced by economic conditions and regional factors Empirical studies, including those from Vietnam, consistently show that credit growth has a strong positive correlation with non-performing loans, as banks often relax lending standards to increase market share and boost interest income However, some conflicting studies suggest an inverse relationship between credit growth and non-performing loans.
Weinberg's (1995) theory suggests that lending risk rises during economic growth due to improved expected returns on investment projects, leading banks to loosen underwriting standards despite the need for tighter credit criteria This trend results in an increase in bad loans, particularly when banks extend credit to lower-quality borrowers In Vietnam, the lending market is dominated by state-owned banks, with Vietinbank, Vietcombank, and BIDV collectively holding approximately 34% of the market share as of late 2020 This concentration challenges smaller banks striving to expand their lending market share, often compelling them to relax policies that ensure loan quality The author anticipates that credit growth will align with the trends of non-performing loans, consistent with previous studies conducted in Vietnam.
Research Methods
This thesis employs the Generalized Method of Moments (GMM) as the primary estimation technique due to two significant issues concerning error composition in panel data models: the correlation between explanatory variables and individual effects, and the correlation between explanatory variables and the noise error component The presence of either or both of these issues can lead to biased or inefficient Ordinary Least Squares (OLS) estimates To address these challenges, Arellano and Bond (1991) introduced the GMM method.
In Generalized Method of Moments (GMM) analysis, as highlighted by Arellano and Bond (1991), the number of moment conditions increases with the inclusion of additional representative variables To assess the validity of these conditions, the Sargan test is conducted to identify any excessive binding The Hansen test results for all models show a p-value greater than 0.1, indicating acceptance of the null hypothesis (H0), which posits that the instrumental variable is exogenous and uncorrelated with the model's error term Additionally, the AR(2) test also yields a p-value exceeding 0.1, leading to the acceptance of its null hypothesis (H0), which asserts the absence of second-order autocorrelation within the model.
This thesis employs the panel threshold regression (PTR) method introduced by Hansen in 1999, which is valued for its straightforwardness and clear policy implications The analysis aims to identify the credit growth threshold at which the effects of credit growth on return on assets (ROA) or non-performing loans (NPL) change direction.
Research data
The study's data set comprises annual reports and financial statements from 20 commercial banks over a time series from 2012 to 2020 This sample includes 4 state-owned commercial banks and 16 joint stock commercial banks, representing a significant portion of the 35 total joint stock commercial banks in Vietnam Thus, the sample is deemed representative of the joint stock commercial banking sector in the country.