The impact of income diversification on business performance of commercial bank in vietnam in the period 2012 2022

In recent years, bankers, policymakers, andinvestors have become more interested in the research association between incomediversity and bank performance.Studies on income diversificatio

T HE NECESSARY OF THE RESEARCH

The rapid financial integration with international markets has led to significant changes in the Vietnamese banking industry's business model Currently, the sector comprises four Commercial State Banks, 31 Domestic Commercial Bank Shares, nine wholly foreign-owned banks, and two joint-venture banks, all operating within their own systems As competition intensifies, Vietnamese banks are compelled to enhance the quality of their offerings and diversify their products and services to attract customers and ensure profitability Consequently, they are increasingly focusing on income diversification to maintain their market position against foreign competition Additionally, the recent Covid-19 pandemic has driven these banks to explore non-traditional credit activities, reducing their reliance on conventional deposit and lending markets.

Promoting income diversification in banks has become essential in response to recent socio-economic developments Following the restructuring scheme of credit institutions (Decision 254/QD-TTg dated March 1, 2012) by the State Bank of Vietnam, the banking system has actively aligned with strategic goals, significantly contributing to macroeconomic stability Additionally, the recent initiative to transform commercial banks' business models towards non-credit product and service diversification is a key aspect of the "Restructuring the Credit Institution System Associated with Bad Debt Settlement for 2016-2020," approved on July 19, 2017, aimed at achieving targeted profit growth.

To align with competitive development trends and fulfill government initiatives, commercial banks must actively expand traditional interest income and diversify their revenue streams This shift towards income diversification is increasingly recognized as a crucial strategy that impacts bank performance Consequently, there has been a growing interest among bankers, policymakers, and investors in exploring the relationship between income diversity and the overall performance of banks in recent years.

Research indicates varying levels of income diversification among banks in Vietnam, highlighting the need for a deeper understanding of these differences By examining the reasons behind the variation in income diversification, banks can develop effective strategies and business plans to enhance revenue streams and improve operational efficiency Despite the significance of this topic, there is a lack of comprehensive studies on the impact of income diversification on the business performance of Vietnamese commercial banks in recent years Thus, the research titled “Impact of Income Diversification on Business Efficiency in Commercial Banks in Vietnam (2012-2022)” aims to provide valuable insights, enabling Vietnamese commercial banks to recognize the advantages of appropriate income diversification for boosting their overall business efficiency.

R ESEARCH O VERVIEW

L ITERATURE R EVIEW

2.1.1 Literature Review in the world

Research on income diversification in the banking sector primarily examines its relationship with bank performance and the effects on risk This focus is justified, as regulators and supervisors prioritize income stability and risk management in banks.

The relationship between income diversification and bank performance is a subject of debate, with various studies conducted globally and nationally revealing mixed results Research has explored how revenue diversification activities influence the performance of commercial banks, highlighting the complexity of this relationship.

Diversification in banking allows for the simultaneous exchange of labor and technology, enhancing revenue and mitigating risks By leveraging economies of scale across various activities, banks can reduce their exposure to market fluctuations and cyclical profit shifts Empirical studies have shown that effective human resource management significantly contributes to bank efficiency Klein and Saidenberg (1997) highlight that while interest income and human resources may not have a strong correlation, integrating diverse financial operations can enhance profitability This integration stabilizes income, decreases internal management costs, and ultimately boosts a bank's profitability, as supported by research from Elsas et al (2010), Gurbuz et al (2013), Meslier et al (2014), Lee et al (2014), and Moudud-Ul-Huq et al (2018).

Elsas et al (2010) used panel data from nine countries (Australia, Canada, France, Germany, Italy, UK, USA, Spain and Switzerland) over the years 1996–

A study conducted in 2008 analyzed the impact of income diversification on a bank's value, revealing that diversification enhances returns and subsequently boosts market value The findings, derived from regression analysis with fixed effects, indicate that income diversification contributes to increased bank profitability, even amidst the challenges posed by the global financial crisis.

Gurbuz et al (2013) investigated the risk-adjusted business performance and income diversification of 26 Turkish banks from 2005 to 2011 using imbalanced panel data and the GMM method The research reveals that income diversification enhances the risk-adjusted performance of the banks studied, supporting the notion that non-interest income positively influences the profitability of commercial banks.

A study by Meslier et al (2014) examines the impact of revenue diversification on the profitability of emerging nations, utilizing a regression approach with data from the financial statements of 39 Philippine banks from 1999 to 2005 The findings reveal that increasing non-interest activities enhances revenue and return on risk, especially for banks investing in government securities Additionally, the research indicates that diversification benefits international banks more than domestic ones, and is particularly advantageous for banks with less connection to small and medium-sized enterprises (SMEs).

Lee et al (2014) conducted a study on 967 banks across 22 Asian nations from 1995 to 2009, utilizing the GMM approach to examine the impact of non-interest revenue on bank risk and profitability In this research, bank profitability is assessed through the return on assets (ROA) and return on equity (ROE), while risk is measured by the standard deviation of these returns Key independent variables include the ratio of non-interest income to total operating income, bank size, growth rate of total assets, loan-to-asset ratio, deposit-to-asset ratio, capital-adequacy ratio, and provisions for credit risk The findings reveal that non-interest revenue reduces risk for Asian banks but does not enhance profitability.

Moudud-Ul-Huq et al (2018) utilized the GMM approach to examine the influence of asset and income diversification on the performance and risk of banks in emerging East Asian countries, specifically Indonesia, Malaysia, Thailand, the Philippines, and Vietnam, during the period from 2011 to 2015 The study revealed that diversification benefits all banks in the sample by enhancing revenues and reducing risks While the impact of asset diversification varies by country, income diversification consistently shows a significant positive effect on banks' performance and stability.

Income diversification enhances the potential for increased revenue, leading to higher profits Furthermore, it reduces a bank's dependence on traditional lending and deposit activities, particularly during challenging credit conditions.

Opponents of income diversification argue that it negatively impacts bank profitability and heightens risk, a view supported by various studies in the US and EU Research by DeYoung and Roland (2001), DeYoung and Rice (2004), Stiroh (2004), Lepetit et al (2008), and Stiroh and Rumble (2006) indicates that banks face increased switching costs due to income diversification, leading to greater risk and offsetting the potential benefits with negative return volatility.

Research by DeYoung and Roland (2001) on 472 US commercial banks from 1988 to 1995 reveals that an increase in non-interest income necessitates greater investments in technology and human resources, leading to heightened operating leverage and risk Initially, the shift towards non-interest revenue reduced bank risk; however, in later years, it resulted in increased risk and greater return volatility.

A study by Stiroh (2004) examining the diversification of revenue sources in the US banking system from 1984 to 2001 found that non-interest income fluctuated significantly more than interest income, leading to decreased risk-adjusted returns and heightened company risk The analysis concluded that non-interest income adversely impacts bank profitability, suggesting that the benefits of an income diversification strategy for banks are minimal.

Mercieca et al (2007) argue that income diversification does not positively affect small European banks, as high non-traditional returns can lead to negative outcomes, including increased risk and lower risk-adjusted returns Additionally, trading activities introduce potential risks without guaranteeing higher profits Similarly, Lepetit et al (2008) examined the relationship between bank risk and product diversification in European banks, utilizing the OLS regression method on data from 734 banks.

A study conducted between 1996 and 2002 reveals that banks with diversified income sources face greater bankruptcy risks and encounter riskier interest rates compared to those primarily focused on lending Notably, small-sized banks exhibit a significant correlation between non-interest revenue and risk, indicating that they are more vulnerable when diversifying their income streams The authors attribute these findings to the inherent risks associated with non-interest earning activities such as direct investments, security investments, and real estate ventures.

Recent international studies, including those by Nguyen et al (2013), Murharsito (2015), and Meng et al (2017), have explored the impact of revenue diversification on the financial performance of commercial banks Findings indicate that revenue diversification can significantly influence bank performance Additionally, research by Jing Wang and Haowen Zhou (2008) and Chao Xue and Zheng Li (2014) identified a negative correlation between human resources and the performance of commercial banks, suggesting that effective management of human capital is crucial for enhancing financial outcomes in the banking sector.

R ESEARCH GAP

The author identified diverse perspectives on the impact of revenue diversification on the financial performance of commercial banks in Vietnam, based on a review of both international and domestic studies Despite this, there is a scarcity of research specifically examining the effects of income diversification on the profitability and risk of commercial banks, with the overall number of publications on the subject remaining limited Consequently, this study aims to provide new insights and contributions to the field.

A study examining the effects of income diversification on the business efficiency of Commercial Bank Vietnam reveals significant insights into profitability and risk management The research highlights that income diversification positively influences profit metrics, specifically Return on Assets (ROA) and Return on Equity (ROE) Additionally, the analysis utilizes the Z-Score index to assess risk, indicating that increased income diversification contributes to enhanced financial stability Overall, the findings suggest that strategic income diversification can lead to improved performance and reduced risk for commercial banks in Vietnam.

- Studying the impact of non-interest income on the operational performance of Commercial Bank Vietnam.

In conclusion, this article examines the impact of income diversification on business efficiency within the context of Vietnamese commercial banks from 2012 to 2022 By analyzing the positive and negative effects of various factors influencing income diversification, the research aims to address the existing academic gap and provide actionable recommendations to enhance banking performance and operational efficiency.

R ESEARCH O BJECTIVES AND Q UESTIONS

Applying income diversification theory enhances both the efficiency and risk-adjusted performance of banks, supported by theoretical and empirical evidence from 2022 This research identifies key factors influencing income diversification and proposes solutions for Commercial Bank Vietnam to adapt these factors By implementing effective income diversification strategies, the bank can significantly improve its overall business efficiency and performance.

Between 2012 and 2022, Commercial Bank Vietnam has made significant strides in income diversification, showcasing various positive outcomes across different regions However, despite these advancements, certain limitations persist in the diversification process, affecting overall financial stability and growth An assessment of the current status reveals both achievements and challenges, highlighting the need for continued efforts to enhance income sources and address existing barriers.

+ Analyze the impact of income diversification on the business performance of Commercial Bank in Viet Nam.

+ Proposing strategic solutions, and income diversification solutions to help managers improve business efficiency at Commercial Bank Vietnam.

The impact of income diversification on the business performance ofCommercial Bank in Vietnam?

O BJECT AND SCOPE OF RESEARCH

- Research subjects the subject is termites relationship between income diversification and business performance in Commercial Bank Vietnam

Vietnam's banking sector features 28 prominent commercial banks, including ABB, ACB, AGR, BAB, BVB, BIDV, CTG, EIB, HDB, KLB, LPB, MBB, MSB, NAB, NVB, OCB, PGB, SCB, SSB, SGB, SHB, STB, TCB, TPB, VAB, VCB, VIB, and VPB, each playing a vital role in the country's financial landscape.

As of December 31, 2022, the State Bank of Vietnam reported that the commercial banking sector comprises four state-owned commercial banks and 31 domestic joint-stock commercial banks Notably, three banks—DongA Bank, Pvcombank, and Vietbank—were excluded from the data due to their acquisition for zero dong and their special control status, as well as the inability to gather sufficient financial statement information during the research period.

- Research period: From 2012 to 2022, the period since the Prime Minister approved Decision 254/2012 of the State Bank of Vietnam on the Scheme of restructuring credit institutions.

R ESEARCH M ETHODS

This study employs a mixed-methods approach, integrating both qualitative and quantitative research techniques to comprehensively analyze the factors influencing income diversification It aims to assess the degree of income diversification and its effects on the performance of commercial banks.

By method of qualitative research, the thesis contains secondary data from macro theories about factors affecting income diversification and its influence on business performance, especially in the period 2012-2022.

The author developed a primary data table utilizing a quantitative research approach, analyzing financial statements from 28 Vietnamese commercial banks spanning 2012 to 2022 By employing various quantitative methodologies, including pooled OLS regression, fixed effect regression (FEM), random effects regression (REM), and System GMM (SGMM) estimation, the author effectively examines the research data.

C ONTRIBUTION OF RESEARCH

This thesis aims to enhance the existing literature by exploring how income diversification impacts the business performance of commercial banks, utilizing both qualitative and quantitative research methods.

- This study examines the impact of income diversification on business performance taking into account the risk factor of Vietnamese commercial banks.

- Income diversification is measured in two ways, respectively, through the diversification index (DIV) and the ratio of non-interest income (NON).

This study provides updated insights for Vietnamese commercial banks on the beneficial impacts of income diversification on banking performance It also proposes strategic policies for effective income diversification to enhance banking efficiency Furthermore, the findings serve as a valuable resource for researchers in the banking sector, facilitating further exploration of related topics.

R ESEARCH TOPIC STRUCTURE

The research project consists of four chapters:

This article provides an overview of the theoretical foundations and concepts related to income measurement and diversification, highlighting its significance in banking performance It discusses various theories associated with income diversification and summarizes key findings from previous studies in this area, emphasizing the impact of diversified income streams on financial stability and growth within the banking sector.

This chapter gives a research model, discusses how to measure variables, and formulates research hypotheses after presenting the data source and estimating technique of the research model.

Chapter 3: Research Results and Discussion

This article presents an analysis of research findings on the impact of revenue diversification in Commercial Bank Vietnam The study utilizes two key metrics: the diversity index (DIV) and the percentage of non-interest income (NON), to effectively measure income diversification within the bank's operational model.

Chapter 4: Conclusion and policy suggestions

This article outlines key conclusions from research findings, proposing solutions and policies aimed at enhancing income diversification and improving operational efficiency for Vietnamese commercial banks Additionally, it addresses the study's limitations and suggests future research directions related to this topic.

THEORETICAL OVERVIEW

T HEORETICAL BASIS OF BANK INCOME DIVERSIFICATION

1.1 The definition of bank income diversification

- The definition of bank income

A bank's income is derived from its output services, which encompass revenue from capital usage and other operations, as defined by Rose et al (2006) Traditional banking primarily generates interest income through borrowing arrangements, where banks accept customer deposits, pay interest on those deposits, and then reinvest the funds by lending them at higher interest rates (Bailey, 2010; Young & Rice, 2003).

Non-interest income refers to revenue generated from activities outside of lending Stiroh (2004) categorizes this income into four primary components: trust income, service fees, fees, and other income, highlighting the diverse nature of non-interest income sources.

According to the Vietnam Commercial Bank's financial report, non-interest income encompasses earnings from various sources beyond credit activities, including service activities, forex and gold trading, and securities trading The significance of revenue from service-related activities is increasing as client needs diversify, highlighting the growing importance of payment services, brokerage, consultation, insurance, guarantee activities, letters of credit, and treasury intermediaries.

Diversification, as defined by Ansoff (1957), refers to a shift in a company's product line or market, distinguishing it from strategies like market penetration, market development, or product development It signifies a transformation in the product structure of a business, highlighting the importance of adapting to new markets or product offerings for overall growth and sustainability.

Businesses diversify to operate in multiple fields simultaneously In the banking sector, this diversification happens when banks broaden their primary focus on interest income and traditional revenue streams by entering additional product and service markets.

Diversification is an effective investment strategy aimed at minimizing risk by integrating a variety of investments into a portfolio This approach results in a multi-directional portfolio, ensuring that not all investments will fluctuate in the same direction simultaneously.

According to Mercieca, Schaeck, and Wolfe (2007), diversification in the banking sector encompasses three key dimensions: the diversification of financial products and services, geographical diversification, and a combination of both geographic and business diversification.

Banks can adopt various diversification strategies to enhance their business performance, aiming to boost revenue, reduce costs, attract more customers, and ensure sustainable growth A key indicator of successful diversification is the increase in bank income, making income diversification a crucial measure of the effectiveness of these strategies rather than merely a tactic within the banking process.

Bank income diversification, as noted by Rose and Hudgins (2008), is indicated by the changing ratio of non-interest income to total income A bank is considered concentrated if it relies solely on net interest income; however, if it balances this with non-interest income, it achieves diversification This shift primarily emphasizes commercial operations and activities that generate fees and commissions Consequently, banks are moving away from traditional credit activities and are increasingly embracing non-traditional services and business ventures to enhance their income streams.

According to Elsas et al (2010), commercial banks are increasingly diversifying their income sources by transitioning from traditional activities, such as collecting interest from deposits and loans, to generating fee-based income This shift allows banks to stabilize their fee income, which then enables them to explore non-traditional activities, including investment operations, to enhance the share of non-interest income within their overall revenue Consequently, the diversification of bank income reflects a strategic move away from a sole focus on traditional business practices towards a balanced approach that incorporates both interest and non-interest income streams.

Diversification of bank income refers to the shift from traditional banking activities, where banks primarily relied on interest income, to a more balanced approach that includes both interest and non-interest income sources This strategy enables banks to enhance their overall revenue streams and reduce dependence on any single income source.

1.2 Indicators to measure the diversification of bank income

Asif and Akhter (2019) highlight that income diversification in banks is primarily assessed using the non-interest income ratio and the Herfindahl Hirschman index, with 44% of studies employing the former and 29% the latter Furthermore, these studies also examine the components of non-interest income alongside the non-interest income ratio.

1.2.1 Measure through net non-interest income ratio

Bank income is categorized into two primary types: interest income and non-interest income Interest income refers to earnings from loans and other interest-bearing assets, while non-interest income encompasses various revenue streams, including commission revenue, business income, investment revenue, and additional sources Specifically, non-interest income includes earnings from service fees, business activities, investments, and other miscellaneous income.

In the financial statements of banks, investment activities, securities trading, and foreign exchange trading are reported on a net basis, leading to the calculation of bank income also on a net basis Total net income comprises net interest income and net non-interest income, where net interest income is derived from the difference between interest income and interest expenses Additionally, net non-interest income is calculated by subtracting non-interest expenses from net interest income.

Net interest income = Interest income – Interest expense

= Non-interest income – Non-interest expenses

= Net income from services + Net income from investment business + Other net income

Total net income = Net interest income + Net non-interest income

The financial statements of Commercial Bank Vietnam reveal that the bank's total net income comprises various components, including net interest income, income from service activities, foreign exchange operations, trading securities, investment securities, and other activities, as well as income from capital contributions and share purchases Consequently, the net non-interest income for Commercial Bank Vietnam is derived from these diverse revenue streams.

T HE THEORETICAL BASIS OF BANKING BUSINESS EFFICIENCY

2.1 The concept of banking business performance

Business efficiency is a widely used term in all economic, technical, and social fields Efficiency can be approached from many different angles.

Business efficiency measures the relationship between the results achieved and the total costs incurred, highlighting the quality of economic activity A greater difference between these two factors indicates higher efficiency In a market economy, a key criterion for assessing the success of enterprises is their level of business efficiency.

Business efficiency in economics, as defined by Nguyen Khac Minh (2004), refers to the relationship between the inputs of limited resources and the outputs of goods and services It evaluates how effectively markets allocate resources Therefore, business efficiency can be interpreted as the extent to which banks successfully manage and distribute their resources to achieve specific objectives.

Banking efficiency refers to a bank's ability to optimize its input resources to achieve maximum output, typically measured by its profit targets.

Business efficiency of commercial banks, according to Berger and Mester

In 1997, efficiency is defined by the relationship between revenue and resource utilization costs, highlighting the ability to effectively transform input resources into outputs An organization or bank is considered efficient when it maximizes output while optimally utilizing available inputs.

Commercial banks, as highlighted by Rose (2004), are structured as business entities focused on maximizing profits while maintaining an acceptable level of risk This profit-oriented approach not only helps preserve capital but also enhances market share expansion and attracts investment capital, which are essential for the bank's sustainability and future growth.

Business performance in financial institutions, as noted by Rose and Hudgin (2008), hinges on fulfilling the needs of shareholders, employees, depositors, borrowers, and other stakeholders The effectiveness of these institutions in meeting diverse expectations is revealed through financial statements, primarily focusing on profit, equity, assets, and stock value Consequently, profit serves as a key indicator of both business performance and the sustainable development of a bank.

According to Nguyen Viet Hung (2008), a bank's business efficiency is indicative of how effectively resources are utilized to meet objectives This efficiency illustrates the relationship between inputs and outputs, highlighting the importance of cost reduction in enhancing competitiveness against other financial institutions.

Banking efficiency is considered the level of success that a bank achieves in optimizing input resources to achieve maximum output results, expressed through profit targets.

This study evaluates risk-adjusted business performance in banks, emphasizing earnings volatility as a key risk factor Earnings volatility is defined as the deviation between actual and expected returns, reflecting the variability in profitability Specifically, bank risk is quantified by the standard deviation of its rate of return Consequently, risk-adjusted business performance is determined by calculating the rate of return relative to the standard deviation of the bank's risk, providing a comprehensive view of performance that incorporates risk factors.

Numerous studies have evaluated risk-adjusted business performance by analyzing the rate of return in relation to its standard deviation Notable research in this area includes works by Stiroh (2004), Stiroh and Rumble (2006), Mercieca et al (2007), Chiorazzo et al (2008), Lepetit et al (2008), Sanya and Wolfe (2011), Gurbuz et al (2013), Batten and Vo (2016), and Nguyen Quang Khai (2016), along with contributions from Vo Xuan Vinh and Tran Thi Phuong Mai (2015).

In summary, business performance in a bank is defined by its success in effectively utilizing resources to maximize outputs, ultimately reflected in profit targets The primary operational goal of a commercial bank is to achieve maximum profits while maintaining an acceptable level of risk.

2.2 Factors Affecting the Bank's business performance

Income diversification, as supported by modern portfolio theory, enhances bank profitability while mitigating risks Research conducted by Meslier et al (2014), Lee et al (2014), and Vo Xuan Vinh and Tran Thi Phuong Mai (2015), along with studies by Le Van Hau and Pham Xuan Quynh, underscores the importance of diversifying income streams for banks to achieve these financial benefits.

Research indicates a complex relationship between income diversification and bank performance While studies by Moudud-Ul-Huq et al (2016) and Moudud-Ul-Huq et al (2018) suggest that greater income diversification correlates with improved business performance, contrary findings from DeYoung & Rice (2004), Stiroh (2004a, 2004b), and Mercieca et al (2007) argue that increased diversification may lead to decreased efficiency However, the author contends that income diversification can mitigate risks and enhance profitability by spreading risk and optimizing financial resources, ultimately boosting service offerings, increasing bank income, and improving both overall and risk-adjusted business efficiency.

Larger banks often exhibit enhanced stability due to their ability to manage individual risks more effectively, as they possess superior risk management capabilities and can invest in advanced technologies (Meslier et al., 2014) These institutions benefit from better diversification opportunities and reduced income volatility when entering new markets (Sanya & Wolfe, 2011) While larger banks can better manage risk and diversify their operations, smaller banks tend to be more flexible (Sanya and Wolfe 2011; Chiorazzo et al 2008; DeYoung and Rice 2004) Research indicates that large-scale banks enhance business efficiency and mitigate risks (Lee et al., 2014; Vo Xuan Vinh and Tran Thi Phuong Mai, 2015) However, Meslier et al (2014) found a conflicting relationship between bank size and performance Overall, the consensus in literature supports that larger banks possess superior business expansion capabilities and risk management, leading to increased efficiency and reduced risks.

The capital adequacy ratio (CAR) reflects a bank's financial leverage and stability in the market, showcasing its ability to withstand economic shocks A strong CAR indicates that a bank can absorb significant losses, thereby reducing bankruptcy risk and enhancing operational efficiency Banks with higher capital adequacy ratios require less external funding and incur lower capital costs, positioning them for greater resilience and success in fluctuating economic conditions.

High capital levels correlate with lower leverage and risk, leading to the argument that shareholder returns should increase as ownership decreases and risk rises Research by Goddard et al (2004), Stiroh & Rumble (2006), and Le Long Hau and Pham Xuan Quynh (2017) indicates a positive relationship between equity and total assets in banking performance However, studies by Almumani (2013) and Anbar and Alper (2011) found no such correlation This variable remains prevalent in recent income diversification literature (Sanya and Wolfe 2011; Chiorazzo et al 2008; Stiroh, 2004b) Overall, both theoretical and empirical research supports the notion that a high capital adequacy ratio is linked to reduced risks and enhanced business efficiency.

T HEORIES ON THE IMPACT OF INCOME DIVERSIFICATION ON BANKING

The theoretical model of portfolio diversification, established by Markowitz and James in 1970, emphasizes the importance of minimizing portfolio risk through strategic asset allocation Effective diversification can enhance portfolio performance, but its risk-reducing benefits largely depend on the correlation between the assets involved When investments exhibit low or negative correlation during periods of risk, diversification can significantly mitigate or even eliminate potential portfolio losses.

Diversification is an investment strategy aimed at reducing risk by incorporating a variety of investments, ensuring that not all assets move in the same direction (Sanya & Wolfe, 2011) In the banking sector, income diversification manifests through three key trends: the expansion of financial products and services, geographical diversity, and a blend of both geographical and business diversification (Mercieca et al., 2007) This approach enhances non-interest income, contributing positively to the overall net income of commercial banks.

Commercial banks are increasingly diversifying their income by transitioning from traditional activities, such as collecting interest from deposits and loans, to generating revenue through fee collection and other non-traditional activities This shift allows banks to enhance their non-interest income, which includes service fees, commissions, and investment income, thereby raising the overall percentage of non-interest income in their total operating revenue According to Elsas et al (2010) and Rose & Hudgins, this strategic move signifies a significant change in the banking landscape, as institutions seek to stabilize their income sources and promote growth beyond conventional credit activities.

Income diversification in banks is assessed using the Herfindahl-Hirschman Index (HHI), as outlined by Mercieca et al (2007) This index evaluates the concentration of a bank's core activities, providing a comprehensive measure of diversification.

퐻퐻 푅� = ( 푁�푁

With푁�푇� = 푁�푁+푁�푇

NON refers to non-credit income, whereas net interest income is represented by NET and NETOP, indicating net operating income The relationship can be summarized as: a rise in the Herfindahl-Hirschman Index (HHI) signifies greater revenue concentration and reduced diversification.

The Bank's income diversification index is assessed using the Herfindahl-Hirschman Index (HHI), as outlined in research by Stiroh & Rumble (2006) and Chiorazzo, Milani & Salvini This method effectively estimates the level of income diversification within banks.

(2008) The degree of diversification is calculated according to the following formula:

퐷 = 1−( 푁푇 2 +푁�푁 2 )

INT: Ratio of net interest income to total operating income

NON: Ratio of net non-interest income to total operating income

The diversification formula can also be rewritten as:

Net non-interest income (NOI) encompasses various revenue streams, including income generated from service fees, earnings from securities trading, profits from foreign exchange trading, and revenue from additional activities.

NETOP: The bank's total net operating income includes net interest income and net non-interest income.

푁�푇� = 푁�푇+푁�

If net non-interest income is negative, the study assigns a value of zero to the ratio of net non-interest income, indicating that revenue from non-interest activities does not enhance net income (Ho Thi Hong Minh and Nguyen Thi Canh, 2015).

This study explores the effect of income diversification on banking performance by utilizing the non-interest income to total income ratio (NON) as a measurement method This approach aligns with previous research conducted by Lepetit et al (2008), Lee et al (2014), Meslier et al (2014), and Batten & Vo (2016), highlighting the significance of non-interest income in assessing a bank's income diversification.

3.2 The impact of income diversification on the business performance of Commercial Bank

Diversifying income sources in a bank enhances business performance and profitability by expanding markets and customer bases By offering a variety of products and services, banks can identify and exploit market gaps, thereby increasing market share and revenue This approach allows banks to fully utilize their capital, technical resources, and human capital, leading to reduced management and operating costs while maximizing profits Additionally, product diversification fosters the development of complementary businesses.

Global researchers remain divided on the effects of income diversification on bank profitability One viewpoint suggests that income diversification enhances banks' profitability.

Smith et al (2003) show that when banks increase non-credit income- generating activities, it will help stabilize and increase bank profits Chiorazzo et al

Research indicates that increased diversification in banks can lead to higher profitability through non-credit income sources, a finding supported by studies from Baele et al (2007), Carlson (2004), and Gurbuz et al (2013) However, contrasting empirical evidence exists, suggesting that diversifying income sources may not yield positive outcomes for banks, as highlighted by DeYoung and Roland (2001) and Stiroh and Rumble (2006).

Numerous empirical studies indicate that income diversification may not enhance, and could potentially diminish, bank profitability Research conducted by DeYoung and Roland (2001) on 472 US commercial banks from 1988 to 1995 reveals that revenues from traditional lending activities tend to be more stable over time This stability is attributed to the conversion and information costs incurred by both borrowers and lenders when switching banks, leading customers to prefer maintaining established credit relationships.

A study by Stiroh (2004a) reveals a correlation between the growth of net interest income and non-credit income during the 1990s; however, non-credit income exhibits greater volatility, which negatively impacts business performance and overall profitability of commercial banks As the number of banks increases, competition intensifies, further affecting the stability of these institutions Consequently, it is essential for commercial banks to carefully develop strategies to diversify their product offerings, fee structures, and investment portfolios to enhance sustainable competition and improve operational efficiency in a globalized environment.

3.3 The impact of income diversification on operational risk Commercial Bank

Income diversification is essential for banks as it mitigates risks and enhances profitability beyond traditional lending activities By offering a range of complementary products and services, banks can stabilize their revenue streams and maintain consistent profits, even during market fluctuations This strategic approach not only helps in risk management but also contributes to the overall financial stability of the institution.

RESEARCH METHODOLOGY AND DATA

R ESEARCH P ROCESS

The following are the primary measures that were taken in order to carry out this thesis:

Step 1: Identify the research problem Determine the requirement for work study diversification of bank income and business efficiency of commercial banks.

Step 2: Overview of the theoretical basis and proposed research model The author examines the theoretical underpinnings of the research problem in step 1 and the empirical studies conducted globally and in Vietnam to determine the impact of bank income diversification on bank performance An overview of prior research will be used to suggest a theoretical research model.

Step 3: Collect and evaluate research data From the theory in the model, collect and calculate the necessary data for running the research model Internal data for Commercial Bank Vietnam is gathered from audited financial accounts, and macroeconomic data is gathered from the websites of the World Bank (WB) and the International Monetary Fund (IMF).

Step 4: Run and test the research model On Stata 14.0, the regression model is tested using the System GMM estimate To determine whether multicollinearity exists, examine the correlation coefficients between each of the independent variables separately Simultaneously, when running the regression model with GMM estimation on Stata software, the tests on autocorrelation of order

2 (AR2) and Hansen test are performed to ensure that the results of the research model are appropriate.

Step 5: Present and discuss the research results Presenting research findings on how revenue diversification affects Commercial Bank Vietnam's business success, while also discussing and contrasting them with findings from relevant previous studies.

Step 6: Conclusion and policy recommendations Based on the research results, the thesis makes conclusions and policy suggestions to increase the level of

The research process is summarized as follows:

R ESEARCH MODEL

This article explores the development of a research model to assess the impact of income diversification on business performance at Vietnam Commercial Bank Drawing on the findings of Lee et al (2014) and other prior research, key factors influencing banking performance—such as bank size, growth rate, customer loan ratio, deposit ratio, capital adequacy ratio, operating efficiency, economic growth rate, and inflation—are identified alongside income diversification Consequently, a model is proposed to analyze how income diversification affects the business performance of Vietnam Commercial Bank.

Yit= β0 +β1Yit-1 +β2DIVit+ βxXit +�i +�it (1)

In there: β 0, …., β j: Estimated parameters λi: Unobserved effect ԑit: Error

- The dependent variable measures the business performance of commercial banks (ROA, ROE, SDROA, SDROE).

- The independent variable is: the bank's income diversification index (DIV).

Key control variables in the analysis include firm size (SIZE), capital structure (ETA), asset growth rate (GROW), credit operations size (LTA), loan provision (LLP), deposit size (DTA), management efficiency (OTR), economic growth rate (GDP), inflation rate (INF), and variable FO.

Y: That dependent variable is the bank's business performance, expressed in profit measured in turn by:

ROA (return on average assets)

When evaluating a bank's risk factor, Y represents the risk-adjusted business performance through the SDROA and SDROE metrics Higher values of these ratios indicate lower risk and signify superior business performance for the bank.

SDROAi,t=ROAi,t/σROAi ; SDROEi,t=ROEi,t/σROEi

N is the number of years of observation (11 years, from 2012 to 2022) xt is ROA (or ROE) at year t μ is the mean value of ROA (or ROE) over the observation period

The bank's income diversification is assessed using the Herfindahl-Hirschman Index (HHI), which quantifies the extent of diversification in bank income streams The formula employed in this study effectively measures the level of income diversification within the banking sector.

푫� =� − 푯푯�=� −(�� � +��� � ) INT: Ratio of net interest income / Total operating income

NON: Ratio of net non-interest income / Total operating income

Another way to rewrite the diversification formula is as follows:

NOI: Net non-interest income is comprised of net service fee income, net investment income from trading in securities, net foreign exchange income, and pure other income.

NETOP: The bank's overall net operating profit, which includes net interest and net non-interest revenue.

If the net non-interest income is negative, the study will set the net income ratio for non-interest activities to zero, indicating that these activities do not contribute to net income.

Size of the bank SIZE Logarithm of the bank's total assets Capital adequacy ratio ETA The ratio of equity / Total assets.

Bank growth rate GROW Growth rate of total assets

(GROW1 - GROW0)) / GROW0 Bank loan rate LTA Loan balance/total assets ratio

Bankruptcy risk ZSC Measured by logarit of Z-Score.

풐� �풒풖 � 풐� 풔풔 �풔

Payment risk LTD Measured by the ratio of customer loans to total customer deposits

Bank credit risk LLP Measured by the ratio of provision for loans to customers / Total assets

Customer deposit rate DTA Measured by the ratio of customer deposits / Total assets.

OTR Measured by the ratio of operating expenses / Total operating income of the bank.

Source:Source: Synthesis of the author

GDP: Annual economic growth rate, measured by the annual growth rate of gross domestic product (GDP)

INF: inflation rate, measured by the annual growth rate of the consumer price index

CPI: To examine the impact of diversification on business performance in the context of increasing financial openness in Vietnam, the study adds the control variable for inflation.

FO : Financial Openness, Foreign Direct Investment/Gross Domestic Product (FDI/GDP)

⇒Equation (1) can be rewritten as:

Yit = β0 + β1Yit-1 + β2DIVit + β3SIZEit + β4ETAit + β5GROWit + β6LTAit + β7LLPit + β8DTAit + β9OTRit + β10GDPit + β11INFit +β12 FOit + λi + ԑit

The author uses the data source in the financial statements of 28 Commercial Bank Vietnam in Graduation Thesis period 2012-2022 The author uses data from

28 Vietnamese commercial banks' financial statements in this Graduation Thesis.

The State Bank of Vietnam compiles essential information, including macroeconomic indicators such as GDP, inflation (INF), and foreign investment (FO), sourced from the International Monetary Fund (IMF) and the General Statistics Office of Vietnam.

R ESEARCH HYPOTHESIS

Based on an analysis of prior studies and statistical insights, the author proposes several research hypotheses that predict both positive and negative impacts of independent variables on the diversification of bank income.

- Business performance has a favorable effect on the diversity of bank income.

- The income diversification of the bank is positively impacted by the bank's size.

- Diversification of a bank's income is positively impacted by capital adequacy ratio.

- The bank's ability to diversify its sources of income is negatively impacted by the ratio of loans to total assets.

- Bankruptcy risk has a positive impact on bank income diversification

(Because the bankruptcy risk is higher when the Z-score is lower and vice versa, the above hypothesis corresponds to the negative sign expectation)

- Diversification of income is positively impacted by liquidity risk.

- Diversifying income has a beneficial effect on credit risk.

- Income diversification benefits from the bank's operational expense ratio.

- Diversification of income is negatively impacted by economic expansion.

- Diversification of income is negatively impacted by inflation.

- Financial openness has a positive effect on income diversification

Table 2.2 Summary of variables in the research model of factors affecting bank income diversification

Expected sign on ROA, ROE

Expected sign on SDROA, SDROE

Stiroh & Rumble (2006), Chiorazzo et al

Stiroh & Rumble (2006), Lepetit et al

Wolfe (2011), Gurbuz et al (2013), Meslier et al (2014), Ho Thi Hong Minh and Nguyen Thi Canh (2015), Batten and Vo

Stiroh & Rumble (2006), Lepetit et al

(2008), Sanya & Wolfe (2011), Gurbuz et al

(2013), Lee et al (2014), Meslier et al

(2014), Ho Thi Hong Minh and Nguyen Thi Canh (2015), Nguyen Quang Khai (2016)

Stiroh & Rumble (2006), Lepetit et al

(2008), Sanya & Wolfe (2011), Gurbuz et al

(2013), Lee et al (2014), Meslier et al.

(2014), Vo Xuan Vinh and Tran Thi Phuong Mai (2015)

Lee et al (2014) , Ho Thi Hong Minh and Nguyen Thi Canh (2015), Nguyen Quang Khai (2016)

Stiroh and Rumble (2006), Lepetit et al

(2008), Sanya and Wolfe (2011), Gurbuz et al (2013), Lee et al (2014), Vo Xuan Vinh

& Tran Thi Phuong Mai (2015), Nguyen Quang Khai (2016)

Lepetit et al (2008), Lee et al (2014), Vo Xuan Vinh & Tran Thi Phuong Mai (2015) + +

Ho Thi Hong Minh, Nguyen Thi Canh

GDP worldbank Ho Thi Hong Minh and Nguyen Thi Canh

INF IMF Ho Thi Hong Minh and Nguyen Thi Canh

FO IMF Simplice Anutechia (2010), Hanh (2010),

Source: Synthesis of the author

R ESEARCH DATA

The audited financial accounts of 28 Commercial Banks in Vietnam from 2012 to 2022 provided essential secondary data for this study, including banks such as ABB, ACB, AGR, BVB, BIDV, CTG, EIB, HDB, KLB, LPB, MBB, and MSB.

NAB, NVB, OCB, PGB, SCB, SSB, SGB, SHB, STB, TCB, TPB, VAB, VCB, VIB,

VCA, and VPB are among the Commercial Bank Vietnam in the sample

The information used in the study was gathered from 28 Vietnamese commercial banks' audited financial statements and annual reports for the years

Between 2012 and 2022, a comprehensive analysis was conducted to identify banks with complete financial statements, including balance sheets, income statements, cash flow statements, and accompanying notes To ensure precision and reliability, data on external macroeconomic factors were sourced from reputable organizations such as the International Monetary Fund (IMF), World Bank (WB), State Bank, the websites of the commercial banks under study, the General Statistics Office, and the Ministry of Finance.

E STIMATION METHOD

Data is collected and initially stored in an Excel file, where it undergoes editing and encryption The next phase involves data cleaning, which includes identifying and correcting errors, filling empty cells, and verifying the accuracy of the information to complete the dataset Subsequently, data is processed and analyzed using Stata 14 software, applying various models such as Ordinary Least Squares (OLS), Fixed Effects Model (FEM), Random Effects Model (REM), and System Generalized Method of Moments (SGMM) to estimate regression parameters and derive the most effective equation that illustrates the relationships among the influencing factors.

According to Gujarati (2004), significant multicollinearity in a model is indicated by a correlation coefficient exceeding 0.8 between independent variables, which can alter the sign of regression coefficients and distort research findings To address this, it is essential to verify the correlation among independent variables prior to regression analysis and to assess multicollinearity using the Variance Inflation Factor (VIF) If the VIF coefficients are below 10, multicollinearity is not expected to substantially impact the model's estimation results.

This study employs the two-step SGMM (System GMM) methodology developed by Arellano & Bover (1995) and Blundell & Bond (1998) to execute the regression model and achieve precise estimations The inclusion of the lagged dependent variable introduces endogeneity issues, complicating the model due to the simultaneous relationship between income diversification and company performance Consequently, the estimates may be biased as past and present firm performance can influence diversification decisions, creating a reciprocal effect that necessitates careful consideration in the analysis.

In Chapter 2, the author outlines the scope and objectives of the research, detailing the construction of research models and methods, as well as the synthesis of research hypotheses The study analyzes financial and income statement data collected over an 11-year period from 2012 to 2022 across 28 commercial banks in Vietnam Utilizing a multivariate regression model, the research employs calculated data processed by state-run systems to derive its findings.

This study examines the impact of income diversification on banking performance by analyzing 14 software variables, including independent variables such as DIV, SIZE, GROW, ETA, LTA, LLP, DTA, ORT, GDP, INF, and FO, alongside dependent variables like ROA, ROE, SROA, and SROE Utilizing a two-step System Generalized Method of Moments (SGMM) approach, the findings are presented comprehensively in Chapter 3.

RESEARCH RESULTS AND DISCUSSION

D ESCRIPTIVE STATISTICS OF RESEARCH DATA

This study analyzes a sample of 28 commercial banks in Vietnam from 2012 to 2022, utilizing audited financial statements and macroeconomic data sourced from the World Bank and the IMF The dataset consists of unbalanced panel data with a total of 308 observations, with detailed statistics presented in Table 3.1.

Table 3.1: Statistics of the study sample

Source: Source: Calculated from Stata

The analysis of Vietnamese commercial banks reveals an average Return on Assets (ROA) of 0.88%, as shown in Table 3.1 Among these banks, the highest ROA recorded is 3.65%, while the lowest, representing a loss-making institution, stands at -59.93%.

The average Return on Equity (ROE) stands at 9.46%, showcasing a substantial gap between banks, with the highest ROE reaching 30.33% and the lowest plummeting to -56.42% This disparity highlights the varying financial performance among banks in the industry.

The analysis reveals minimal differences in risk-adjusted business performance among banks, indicated by the mean values of SDROA at 1.9874 and SDROE at 2.077, along with their standard deviations of 1.3872 and 1.3869, respectively.

The average diversification index (DIV) of Vietnamese commercial banks stands at 0.2928, reflecting a low and moderate level of diversification compared to a maximum of 0.5 The banks in the sample exhibit minimal variation in diversification over time, with a standard deviation of 0.1244 Some banks show a DIV of 0, indicating they are nearly undiversified, while others approach full diversification with a DIV of 0.4999.

The average non-interest income ratio for Vietnamese commercial banks stands at 20.9%, indicating a moderate level of revenue diversification According to Table 3.1, banks with a DIV index of 0.5 are considered completely undiversified, showcasing a non-interest income ratio that ranges from zero to 100%.

The SIZE variable, indicating bank size, ranges from 16.5023 to 21.475, highlighting minimal size disparity among the sampled banks With a mean of 18.85197 and a standard deviation of 1.184, the data reflects a relatively uniform distribution in bank sizes within the sample.

The growth rates of banks, known as GROW, exhibit a substantial range, with a maximum of 141% and a minimum of 39% Typically, these rates fluctuate around an average of 15.47%.

The capital adequacy ratios among the sampled banks show significant variation, with the lowest ratio at 2.62% and the highest reaching 23.84% The data reflects a standard deviation of 0.36 and an average ratio of 8.7%.

The average deposit-to-total-assets ratio (DTA) for banks stands at a significant 67.88%, highlighting their strong dependence on customer deposits Notably, there is a stark contrast between banks, with the highest DTA reaching 90.95% and the lowest at just 40.49%, illustrating a more than twofold disparity in reliance on deposits.

The Loans to Total Assets (LTA) ratio averages 59.32%, indicating that a significant portion of the bank's assets is comprised of customer loans The LTA ratio varies among banks, with a maximum of 80.06% and a minimum of 22.24% Notably, the low standard deviation of 0.1141 suggests minimal variation in this ratio across different banks.

The study on bank bankruptcy risk among Vietnamese commercial banks reveals an average Z-Score of 3.1102, indicating a moderate level of financial stability The Z-Score ranges from a low of 1.4291 to a high of 6.3955 across different banks Notably, there has been minimal variation in risk levels over time, as evidenced by a standard deviation of 0.6773.

Vietnamese commercial banks exhibit an average Loans to Deposits (LTD) ratio of 88.32%, with significant variability ranging from a high of 146.91% to a low of 37.19% Despite this variation, the LTD's standard deviation of just 0.1684 suggests stability in the banks' performance over time.

Over the years, the loan loss provision (LLP) ratio to total assets of banks has remained relatively stable, averaging 0.99% This ratio has varied significantly, peaking at 4.9% and dropping to as low as 0.0001% Overall, the asset quality of these banks is robust.

The Operating Expenses/Earnings Ratio (OTR) for banks stands at 52.14%, with a standard deviation of 14.65% Historically, the ratio has peaked at 92.74% and dipped to a low of 4.54%, indicating that some banks face elevated operational costs due to subpar management and ineffective practices.

R ESEARCH R ESULTS

2.1 The research results on the impact of income diversification measured by DIV index on bank performance

According to Gujarati (2004), significant multicollinearity in a model is indicated by a correlation coefficient exceeding 0.8 between independent variables This issue can lead to changes in the sign of the model's regression coefficients, potentially resulting in biased research outcomes.

After detect fix multicollinearity in stata, the author removing the autocorrelation variables ( ZRC, LTD) - VIF more than 10 (Appendix 2)

The correlation matrix presented in Table 3.5 indicates that all correlation coefficients between the independent variable pairs are below 0.8 This suggests that multicollinearity does not significantly affect the model's estimates.

The study re-evaluated the multicollinearity phenomenon using variance exaggeration techniques The results indicate an absence of multicollinearity within the model, with VIF coefficients averaging 2.08 and ranging from 1.14 to 3.11, suggesting minimal correlation among the variables.

Table 3.2: Correlation coefficients between pairs of independent variables in the model of the impact of income diversification on banking performance

| DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO

Source: Source: Calculated from Stata

Tables 3.3 and 3.4 exhibit the regression findings of the SGMM estimation of the factors influencing the business performance and risk-adjusted business performance of Vietnamese commercial banks.

The impact of bank income diversification, evaluated through the DIV index, on business success, indicated by return on assets (ROA) and return on equity (ROE), is illustrated in models (1) and (2) without considering the control variable of openness.

The impact of bank income diversification, evaluated using the DIV index, on business performance, indicated by Return on Assets (ROA) and Return on Equity (ROE), is illustrated in models (3) and (4) while accounting for the control variable.

This study investigates the impact of income diversification on business performance, specifically through Return on Equity (ROE) and Return on Assets (ROA), as well as on risk-adjusted performance indicated by the standard deviations of ROE and ROA (SDROE and SDROA) It explores how bank income diversification, measured by the DIV index, affects risk-adjusted performance when controlling for other variables, excluding financial openness.

The impact of bank income diversification, assessed through the DIV index, on risk-adjusted business performance, indicated by SDROA and SDROE, is illustrated in models (7) and (8), with financial openness as a controlled variable.

The model's p-value coefficients exceed 0.05, suggesting that the instrumental variable is over-identified, as confirmed by Hansen's test Furthermore, the second-order autocorrelation test (AR2) yields p-values greater than 0.05, indicating an absence of second-order autocorrelation in the model's residuals With all tests meeting the necessary criteria for SGMM estimates, the model's results are both stable and fully interpretable.

Table 3.3: Regression results of impact of DIV on ROA, ROE

Variables ROA ROE ROA ROE

Table 3.4 Regression results of impact of DIV on SDROA, SDROE

Variables SDROA SDROE SDROA SDROE

Research indicates that income diversification significantly enhances the financial performance of commercial banks in Vietnam, as evidenced by improved Return on Assets (ROA) and Return on Equity (ROE), along with better risk-adjusted performance metrics (SDROA, SDROE) from 2012 to 2022 By diversifying their income sources, Vietnamese banks not only boost profitability and operational efficiency but also mitigate risks These findings align with contemporary portfolio diversification theory, suggesting that banks can achieve higher profits and reduced risk through varied revenue streams.

Research by Vo Xuan Vinh and Tran Thi Phuong Mai (2015), along with Le Van Hau and Pham Xuan Quynh (2016), indicates that income diversification positively impacts the efficiency of Vietnamese commercial banks The study reveals that by expanding their income sources, these banks enhance profitability and overall business performance.

The findings align with international studies, such as those by Meslier et al (2014) and Moudud-Ul-Huq et al (2018), but contrast with certain U.S banking research, including DeYoung & Rice (2004) and Stiroh (2004), as well as European studies by Mercieca et al (2007) Some research suggests that income diversification can hinder bank efficiency, possibly due to customer mobility in the absence of competitive pricing or quality Additionally, diversification often necessitates significant investment, which may diminish profits However, the operational efficiency of Vietnam's commercial banks has improved over the years, despite increased investments in non-interest activities and high associated costs, supporting the notion that income diversification positively impacts their performance.

Diversifying income through expanded fee income from securities investment, foreign exchange trading, and other activities allows Vietnamese Commercial Banks to mitigate risks while enhancing their risk-adjusted business efficiency The stability of Vietnam's stock market and minimal changes in customer behavior contribute to a less volatile non-interest income stream Consequently, as income diversification grows, the risk-adjusted business performance of Commercial Banks in Vietnam improves.

This study investigates the effect of income diversification on the business performance of commercial banks, considering financial openness as a control variable The analysis is based on the regression results from models (3), (4), (7), and (8), highlighting significant findings from models (3) and (4).

(7), (8) show the impact of financial openness on business performance and risk- adjusted business performance of companies.Commercial Bank Vietnam:

The positive and statistically significant regression coefficient for the variable FO at the 1% level indicates that financial openness positively impacts bank performance This study contributes to existing research, such as that by Simplice Anutechia (2010), demonstrating the influence of macro-environmental factors on business performance As Vietnam's economy becomes increasingly open to international investment, financial openness fosters overall economic growth and enhances the expansion of banks, resulting in improved operational efficiency The findings suggest that as Vietnam's financial sector continues to grow, Vietnamese commercial banks are likely to enhance their business efficiency in the future.

The lagged banking performance variable significantly influences banking performance positively at a 1% significance level, indicating a strong correlation between different time periods and the operational success of banks This finding aligns with the research by Moudud-Ul-Huq et al (2018) Additionally, the lagged variable of risk-adjusted business performance positively impacts the risk-adjusted business performance of Vietnamese commercial banks, highlighting that the continuous nature of banking operations means that the outcomes from previous quarters directly affect current performance metrics.

CONCLUSIONS AND POLICY RECOMMENDATIONS

C ONCLUSIONS

This study examines the impact of income diversification on the operating and risk-adjusted performance of Vietnamese commercial banks, utilizing a two-step SGMM estimation method for optimal results By analyzing audited financial statements from 28 banks between 2012 and 2022, the report identifies key variables influencing revenue diversification and offers actionable recommendations for enhancing business efficiency and performance The findings underscore the significance of effective income diversification strategies in improving the overall business outcomes of Vietnam Commercial Bank.

Between 2012 and 2022, Vietnamese commercial banks experienced improved business performance and risk-adjusted returns due to revenue diversification This aligns with contemporary investment portfolio theory, which suggests that banks can enhance earnings and mitigate risks by diversifying their revenue streams across various products and services Consequently, both overall business efficiency and risk-adjusted efficiency increase with greater income diversification.

The lagged variable of business performance significantly influences a bank's current performance, indicating that annual results are closely tied to the previous year's outcomes As banks enhance their efficiency, this relationship fosters continuous improvement in risk-adjusted business performance over time.

Bank size, growth rate, capital adequacy ratio, and customer loan ratio positively influence banking performance To improve financial outcomes, banks should focus on increasing their size and enhancing their equity ratio Additionally, prioritizing lending while diversifying their portfolios is essential for boosting client lending rates, ensuring loan quality, and effectively managing risks.

The bank's business performance suffers due to high credit risk, evidenced by its elevated non-performing loan (NPL) ratio This significant credit risk necessitates substantial risk provisions, which, in turn, have led to a decline in the bank's profits as it complies with State Bank standards.

The operational expenses of a bank are inversely related to its performance, indicating that poor management of these costs can lead to decreased business efficiency.

The positive influence of inflation on a bank's financial performance highlights the importance of effective management strategies Banks often anticipate inflation trends and adjust interest rates accordingly to ensure that income grows at a faster pace than costs This proactive approach not only mitigates the adverse effects of inflation on commercial outcomes but also enhances the bank's operational efficiency and overall performance.

Evaluating factors that affect business success requires consideration of new macroeconomic elements, particularly financial openness This aspect significantly enhances the performance of banking operations As the Vietnamese economy and financial system continue to open up, the overall business performance of banks is expected to improve steadily.

- The deposit ratio, economic growth rate with banking performance did not show any correlation in the study.

S OLUTION

Vietnamese commercial banks need to diversify their income streams beyond traditional credit activities, which currently dominate their revenue To enhance operational effectiveness, banks should adopt a strategy focused on product and service diversification, investing in the growth of non-interest activities This approach will not only broaden their revenue sources but also increase the overall income from non-interest activities Implementing this strategy requires the development of targeted solutions to support these initiatives.

To enhance efficiency, Vietnamese commercial banks need to diversify their income sources This requires a strategic focus on product and service diversification, alongside investments in non-interest activities to create additional revenue streams and increase overall non-interest income.

To enhance the business efficiency of commercial banks in Vietnam, it is essential to improve management skills By minimizing investments in physical assets and focusing on technology development and high-quality human resources, banks can reduce operational expenses This strategic shift will enable them to utilize personnel and assets more effectively, leading to diversified income streams and increased banking efficiency while ensuring that the benefits outweigh the costs.

To enhance company productivity and reduce credit risk, Vietnamese commercial banks need to improve credit quality, as highlighted by the study's findings A higher loan ratio coupled with low credit risk positively influences productivity, while the research also reveals that provisions for credit risks adversely affect the operational performance of banks.

Vietnamese commercial banks need to prioritize effective liquidity risk management, as research indicates that liquidity risk adversely affects the ratio of non-interest income To mitigate this risk, banks should focus on managing their liquid assets—those that can be swiftly and cost-effectively converted into cash to meet liquidity needs Furthermore, enhancing relationships with shareholders and key capital providers is essential for improving capital access and ensuring a diversified capital base.

To mitigate the risk of bank failure, Vietnamese commercial banks need to enhance their risk management capabilities Introducing new products and services can inherently carry risks, making it essential for banks to thoroughly assess these potential threats By developing a robust risk monitoring, prevention, and warning system, banks can effectively manage risks and minimize the likelihood of insolvency.

Vietnamese commercial banks can leverage economic downturns to enhance income diversity and improve business performance Research shows a negative correlation between income diversification and economic growth, suggesting that during weak economic periods, banks should focus on developing non-interest operations to bolster their income streams and overall performance.

During periods of low economic growth, Commercial Bank Vietnam can leverage this challenge to enhance income diversification, ultimately boosting its business performance By prioritizing the development of non-interest activities, the bank can effectively increase its revenue streams and strengthen overall financial results.

To enhance banking efficiency amid high inflation, the Commercial Bank Vietnam must prioritize accurate inflation forecasting By improving the quality of these forecasts, the bank can develop effective business strategies that align with economic conditions, ultimately boosting its overall performance.

The Commercial Bank Vietnam should leverage financial openings to innovate and diversify its products and services, enhancing income and operational efficiency A robust long-term strategy focused on training and attracting skilled talent is essential, as both traditional and non-traditional transactions require a tech-savvy workforce with new competencies Additionally, to boost productivity and service revenues, commercial banks must refine their incentive and reward systems.

R ECOMMENDATIONS

Commercial banks can enhance their income streams by providing investment services, such as mutual funds, stocks, and bonds, to their customers This strategy allows banks to generate fees and commissions from these transactions while also attracting new clients who prefer a one-stop solution for their investment needs instead of seeking separate investment firms.

- Developing New Products and Services

Commercial banks can enhance their income diversification by creating innovative products and services, such as mobile banking apps, online bill payment, and financial planning assistance By providing these unique offerings, banks can set themselves apart from competitors, draw in new customers, and generate additional revenue through fees and commissions, ultimately strengthening their financial performance.

Commercial banks can enhance their revenue streams by geographically diversifying, which includes establishing new branches in different cities or countries This expansion allows banks to access new markets and attract a broader customer base Additionally, banks can achieve economies of scale through this growth, leading to reduced costs and improved profitability.

Commercial banks can enhance their income streams by collaborating with fintech companies, which focus on creating innovative financial products and services through technology This partnership allows banks to provide customers with access to advanced financial solutions while leveraging the expertise of fintech firms to maintain a competitive edge in innovation.

Diversifying income streams is essential for commercial banks aiming to stay competitive and profitable By providing investment services, creating innovative products, expanding into new markets, and collaborating with fintech companies, banks can establish multiple revenue sources It is crucial to assess each strategy carefully to align with the bank's specific needs and objectives This thoughtful approach enables banks to effectively diversify their income and succeed in the fast-evolving financial landscape.

Solutions that government can implement to increase income diversification for commercial banks:

To enhance income diversification, commercial banks should be encouraged to provide investment banking services, including underwriting, mergers and acquisitions, and securities trading By doing so, they can generate additional fees and commissions, thus broadening their income sources Governments can facilitate this shift by offering tax incentives and lowering regulatory hurdles, which may motivate more banks to participate in the investment banking sector This increased participation can foster competition and lead to improved pricing for consumers.

To enhance income diversification, commercial banks should focus on promoting digital banking, as many customers now prefer online and mobile banking solutions By investing in digital infrastructure, banks can lower operational costs and boost revenue through transaction fees and various digital services Governments can support this shift by offering tax incentives to banks that invest in digital technologies and by collaborating with them to establish regulations that facilitate the widespread adoption of digital banking services.

Microfinance institutions play a crucial role in offering financial services to low-income individuals and small businesses lacking access to traditional banking By supporting these institutions, governments can enhance income diversification for commercial banks Through funding and regulatory assistance, governments enable microfinance institutions to collaborate with commercial banks, allowing them to provide comprehensive financial services to their clients This partnership benefits both parties by facilitating resource sharing and leveraging expertise.

Foreign investment serves as a vital means of income diversification for commercial banks, enabling them to enhance their revenue streams By fostering a favorable business environment and providing incentives for foreign investors, governments can stimulate foreign investment Commercial banks can capitalize on this by offering tailored financial services to foreign investors and facilitating cross-border transactions, ultimately leading to increased profitability and a more diverse income portfolio.

In conclusion, enhancing income diversification for commercial banks is essential for economic stability and growth Governments can facilitate this by promoting investment banking services, advancing digital banking, supporting microfinance institutions, and encouraging foreign investment, ultimately helping banks to diversify their income sources and mitigate risk.

Collaboration between governments and commercial banks is essential for implementing solutions that can significantly enhance the financial sector By fostering this partnership, governments can cultivate a more dynamic and resilient financial environment that effectively meets the diverse needs of both businesses and individuals.

3.3 Recommendations to the the State Bank of Vietnam

Income diversification is essential for the State Bank of Vietnam to reduce risks and improve financial stability To achieve this, the central bank should seek new revenue sources beyond conventional banking operations, exploring various innovative income streams.

To promote income diversification in Vietnam's commercial banks, the State Bank should offer incentives like tax breaks and regulatory relief for those that effectively expand their revenue streams.

The State Bank should enhance its recommendations to the Government to expedite the equitization of state-owned commercial banks, establishing a framework for their safer and more efficient operation Additionally, it is essential for the bank to offer guidance and support on managing risks related to income diversification, which may involve implementing training programs and best practice guidelines for effective portfolio management.

To enhance the efficiency of management and operations at the State Bank, it is crucial to develop scenarios that ensure effective consulting and direction across all units and departments This includes securing management and transaction activities in both domestic and international currency markets, as well as the payment system and information technology Additionally, attracting local and international investment capital is essential to meet future economic development needs Reforming the institutional framework and implementing technological solutions will also help achieve the necessary stock market ranking criteria.

L IMITATIONS OF THE STUDY AND DIRECTIONS FOR FURTHER RESEARCH

This study evaluates bank income diversification by employing the Herfindahl-Hirschman Index (HHI) and the ratio of non-interest income The HHI is calculated using the sum of the squares of each type of non-interest income ratio, providing a comprehensive measure of income diversification.

To enhance the bank's business efficiency, it is essential to assess financial indicators and marginal efficiency This study focuses on evaluating business performance through these financial metrics Future research will aim to measure income diversification by employing the adjusted Herfindahl-Hirschman index, which calculates the sum of the squares of various non-interest income ratios, while also incorporating marginal efficiency to achieve more comprehensive study outcomes.

This study focuses exclusively on income diversification and its effect, along with non-interest income, on the profitability of Vietnamese commercial banks Unlike similar research, which also considers the risks associated with income diversification, this analysis emphasizes the direct relationship between income diversification and profitability in banking operations.

The study overlooked the influence of various non-interest income sources on bank profitability Future research should focus on analyzing how each source of non-interest income affects bank profitability while taking risk factors into account.

This study focused on 28 Vietnamese commercial banks, excluding joint ventures and foreign banks, due to insufficient research data To better understand the impact of revenue diversification on business performance, future research should incorporate joint venture and international banks operating in Vietnam Furthermore, the limited research period of 11 years, from 2012 to 2022, restricts the comprehensiveness of the data analyzed.

Diversification income involves investing in various income sources to reduce risk and enhance financial stability Utilizing SGMM stata allows investors to access a range of options, including stocks, bonds, and mutual funds, which can lead to higher returns by capitalizing on diverse market conditions This strategy mitigates the impact of poor performance in any single asset, fostering a more resilient portfolio However, challenges such as higher fees and the complexity of managing multiple investments must be addressed Regular portfolio reviews and adjustments are crucial to maintaining alignment with financial goals Ultimately, with careful planning, diversification income can be a powerful method for building long-term wealth.

Research indicates that diversification positively influences the business performance of Vietnamese commercial banks, while improved business performance also enhances income diversification Key factors contributing to business performance include lagged efficiency variables, bank size, growth rate, capital adequacy ratio, customer loan ratio, inflation, and financial openness Conversely, negative influences stem from the credit risk provision ratio and the operating expenses to operating income ratio When measuring income diversification through non-interest income ratios, additional factors such as bankruptcy risk, liquidity risk, interest rates, and operating expenses also play a role Regression analysis reveals multicollinearity between bankruptcy risk and customer payment risk To boost diversification and efficiency in Vietnamese commercial banks, several solutions have been proposed for the Government and State Bank However, the study did not utilize the adjusted Herfindahl-Hirschman index for measuring income diversification and business performance Future research should adopt this method and consider including joint venture and foreign banks in Vietnam to comprehensively evaluate the impact of income diversification on the efficiency of the country's commercial banking system.

Prove� ≤ 퐃� ≤ � � and the highest DIV means the bank is fully diversified when INT = NON

The Herfindahl Hirschman Index is widely used to measure the concentration of a bank's income The component of bank income consists of two basic parts: interest income - non-interest income.

HHI = INT 2 + NON 2 DIV = 1 − HHI = 1 − (INT 2 + NON 2 )

INT: Ratio of net interest income to total net income

NON: Ratio of net non-interest income to total net income

Total net income = Net interest income + Net non-interest income

The Herfindahl-Hirschman Index (HHI) is a key metric for assessing the concentration of income sources in banks, with values ranging from 0.5 to 1 when net earnings (both INT and NON) are positive An HHI of 0.5 indicates a minimum level of income concentration, reflecting optimal income diversification within the bank Conversely, an HHI of 1 signifies the lowest degree of income diversification, highlighting a lack of variety in income sources.

Banks exhibit low income diversity when the Diversity Index Value (DIV) is 0, indicating that their income relies solely on one source Conversely, a bank achieves full diversification at a DIV of 0.5 when interest income equals non-interest income, demonstrating a balanced revenue stream.

Table 1: Statistics of the study sample.

Table 2: Correlation coefficients between pairs of independent variables in the model of the impact of income diversification on banking performance (DIV,

| DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO

| NON SIZE GROW ETA DTA LTA ZSC LTD OTR GDP INF FO

FO | 3.11 0.322047 SIZE | 2.97 0.337119 GDP | 2.28 0.438445 ETA | 2.09 0.478829 LTA | 2.04 0.489254 OTR | 1.97 0.508439 DTA | 1.79 0.559297 LLP | 1.72 0.581339 INF | 1.67 0.600437 GROW | 1.14 0.874195 -+ - Mean VIF | 2.08

Table 4: DIV-ROA-without FO

xtabond2 ROA l.ROA DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF, gmm(l.ROA,collapse) iv (DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF) twostep

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 21 Obs per group: min = 8

ROA | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 1.013583 0793185 12.78 0.000 8581214 1.169044 DIV | | 0046535 0010073 4.62 0.000 0026794 0066277 SIZE | -.0008207 0003345 -2.45 0.014 -.0014763 -.0001651 GROW | -.0001378 0023905 -0.06 0.954 -.0048231 0045474 ETA | -.0285724 0104849 -2.73 0.006 -.0491225 -.0080223 DTA | -.0033001 0027866 -1.18 0.236 -.0087617 0021614 LTA | 0010327 0017306 0.60 0.551 -.0023593 0044246 LLP | -.0063679 0284796 -0.22 0.823 -.0621868 049451 OTR | -.0096821 0024472 -3.96 0.000 -.0144784 -.0048857 GDP | 0174162 0075932 2.29 0.022 0025338 0322986 INF | -.0141182 0104912 -1.35 0.178 -.0346805 0064441 _cons | 0233479 0062068 3.76 0.000 0111828 035513 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.ROA collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.ROA collapsed

The Arellano-Bond test results indicate significant autocorrelation in first differences for AR(1) with a z-value of -2.71 and a p-value of 0.007, while the test for AR(2) shows no evidence of autocorrelation, with a z-value of -0.07 and a p-value of 0.945 Additionally, the Sargan test for overidentifying restrictions yields a chi-squared value of 21.24 with a probability of 0.012, suggesting that the model is not robust but remains unaffected by the presence of multiple instruments.

Hansen test of overid restrictions: chi2(9) = 14.29 Prob > chi2 = 0.112 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 7.36 Prob > chi2 = 0.498Difference (null H = exogenous): chi2(1) = 6.93 Prob > chi2 = 0.008

Table 5: DIV-ROE without FO

xtabond2 ROE l.ROE DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF, gmm(l.ROE,collapse) iv (DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF) twostep

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 21 Obs per group: min = 8

ROE | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 14.12176 1.533879 9.21 0.000 11.11541 17.1281 DIV | | 0638203 0181209 3.52 0.000 0283041 0993366 SIZE | -.0114175 0047819 -2.39 0.017 -.0207899 -.0020451 GROW | -.0137868 0319787 -0.43 0.666 -.0764638 0488903 ETA | -1.387211 2222876 -6.24 0.000 -1.822887 -.9515353 DTA | 0303303 0437063 0.69 0.488 -.0553325 1159932 LTA | 0078649 0414516 0.19 0.850 -.0733787 0891085 LLP | -.2900274 4729383 -0.61 0.540 -1.216969 6369146 OTR | -.0869139 0356496 -2.44 0.015 -.1567858 -.017042 GDP | 2377063 1118389 2.13 0.034 018506 4569065 INF | -.3310335 1587566 -2.09 0.037 -.6421907 -.0198762 _cons | 3296047 1074493 3.07 0.002 119008 5402014 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.ROE collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.ROE collapsed

The Arellano-Bond test results indicate significant autocorrelation of order one (AR(1)) in first differences, with a z-value of -2.76 and a p-value of 0.006, while the test for order two (AR(2)) shows no evidence of autocorrelation, with a z-value of -0.64 and a p-value of 0.521 Additionally, the Sargan test for overidentifying restrictions yields a chi-squared value of 11.84 and a p-value of 0.223, suggesting that the model's instruments are not overly correlated, though the robustness of this result is not guaranteed.

Hansen test of overid restrictions: chi2(9) = 11.26 Prob > chi2 = 0.259 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 5.47 Prob > chi2 = 0.706Difference (null H = exogenous): chi2(1) = 5.78 Prob > chi2 = 0.016

Table 6: DIV- SDROA-without FO

xtabond2 SDROA l.SDROA DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF, gmm(l.SDROA,collapse) iv (DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF) twostep

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 21 Obs per group: min = 8

SDROA | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 4824425 0999847 4.83 0.000 2864761 6784088 DIV | -.1415957 | 4145273 -0.34 0.733 -.9540544 6708629 SIZE | 1172524 0679662 1.73 0.084 -.015959 2504638 GROW | 4522012 349306 1.29 0.195 -.232426 1.136828 ETA | -1.952222 1.429647 -1.37 0.172 -4.754279 8498351 DTA | -1.233557 5620899 -2.19 0.028 -2.335233 -.1318806 LTA | 1.808232 348134 5.19 0.000 1.125902 2.490562 LLP | 19.74335 7.837102 2.52 0.012 4.38291 35.10379 OTR | -2.298556 5048839 -4.55 0.000 -3.28811 -1.309001 GDP | 2.262355 1.793476 1.26 0.207 -1.252793 5.777504 INF | 3.847654 2.527247 1.52 0.128 -1.105658 8.800966 _cons | -.5393153 1.196036 -0.45 0.652 -2.883503 1.804872 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.SDROA collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.SDROA collapsed

The Arellano-Bond test for AR(1) in first differences yielded a z-value of -2.93 with a p-value of 0.003, indicating significant autocorrelation In contrast, the test for AR(2) showed a z-value of -0.44 and a p-value of 0.661, suggesting no evidence of second-order autocorrelation Additionally, the Sargan test for over-identifying restrictions resulted in a chi-squared value of 22.41 with a probability of 0.008, indicating that the model is not robust but remains unaffected by an excessive number of instruments.

Hansen test of overid restrictions: chi2(9) = 12.14 Prob > chi2 = 0.206 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 7.56 Prob > chi2 = 0.478Difference (null H = exogenous): chi2(1) = 4.58 Prob > chi2 = 0.032

Table 7 presents the results of the DIV-SDROE analysis using the xtabond2 method, highlighting the relationship between l.SDROE and various independent variables, including DIV, SIZE, GROW, ETA, DTA, LTA, LLP, OTR, GDP, and INF The GMM estimation is conducted in a two-step procedure, indicating a robust evaluation of the dynamic panel data model.

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 21 Obs per group: min = 8

SDROE | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 6658508 1001512 6.65 0.000 469558 8621436 DIV | | 0782217 4176487 0.19 0.851 -.7403546 8967981 SIZE | -.0574092 0495398 -1.16 0.247 -.1545055 039687 GROW | 5877485 3402967 1.73 0.084 -.0792208 1.254718 ETA | -7.13381 1.546795 -4.61 0.000 -10.16547 -4.102148 DTA | -.7432295 4995492 -1.49 0.137 -1.722328 2358689 LTA | 1.076459 4640805 2.32 0.020 1668781 1.98604 LLP | 4.017224 8.007573 0.50 0.616 -11.67733 19.71178 OTR | -2.450113 5516526 -4.44 0.000 -3.531333 -1.368894 GDP | 2.046265 2.288824 0.89 0.371 -2.439748 6.532278 INF | -3.500107 2.222856 -1.57 0.115 -7.856825 8566116 _cons | 3.423877 7888307 4.34 0.000 1.877797 4.969956 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.SDROE collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.SDROE collapsed

The Arellano-Bond test results indicate a significant first-order autocorrelation (AR(1)) in first differences, with a z-value of -3.00 and a p-value of 0.003 Conversely, the second-order autocorrelation (AR(2)) test shows no significant evidence, with a z-value of -0.68 and a p-value of 0.499 Additionally, the Sargan test for over-identifying restrictions reveals a chi-squared value of 19.04 with a probability of 0.025, suggesting some robustness, though the results are not overly reliant on a large number of instruments.

Hansen test of overid restrictions: chi2(9) = 14.73 Prob > chi2 = 0.099 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 10.82 Prob > chi2 = 0.212Difference (null H = exogenous): chi2(1) = 3.91 Prob > chi2 = 0.048

Table 8 presents the results of a GMM analysis examining the relationship between dividend payouts (DIV) and return on assets (ROA), utilizing the xtabond2 method The model incorporates various control variables, including size (SIZE), growth (GROW), earnings to total assets (ETA), debt to total assets (DTA), long-term assets (LTA), loan loss provisions (LLP), operating tax rate (OTR), gross domestic product (GDP), inflation (INF), and foreign ownership (FO) The analysis employs a two-step estimation approach to ensure robust results.

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 22 Obs per group: min = 8

ROA | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 1.000959 0805944 12.42 0.000 8429965 1.158921 DIV | | 0045039 0010245 4.40 0.000 002496 0065118 SIZE | -.0007867 0003288 -2.39 0.017 -.0014311 -.0001424 GROW | -.0002985 0023686 -0.13 0.900 -.0049409 004344 ETA | -.0262721 0105328 -2.49 0.013 -.0469159 -.0056283 DTA | -.0032378 0028133 -1.15 0.250 -.0087518 0022763 LTA | 0007044 0019073 0.37 0.712 -.0030337 0044426 LLP | -.005119 0286558 -0.18 0.858 -.0612834 0510454 OTR | -.0096738 0025474 -3.80 0.000 -.0146666 -.004681 GDP | 0104487 0114948 0.91 0.363 -.0120807 0329782 INF | -.0097782 0104947 -0.93 0.351 -.0303474 0107911

FO | 0462823 0435514 1.06 0.288 -.0390769 1316415 _cons | 0203788 0061649 3.31 0.001 0082959 0324618 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.ROA collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.ROA collapsed

The Arellano-Bond test results indicate significant first-order autocorrelation (AR(1)) in the first differences with a z-value of -2.78 and a p-value of 0.005, while the second-order autocorrelation (AR(2)) shows no evidence of autocorrelation, with a z-value of -0.03 and a p-value of 0.979 Additionally, the Sargan test for overidentifying restrictions yields a chi-squared value of 21.14 with a probability of 0.012, suggesting the model is not robust but remains unaffected by an excessive number of instruments.

Hansen test of overid restrictions: chi2(9) = 14.50 Prob > chi2 = 0.106 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 7.97 Prob > chi2 = 0.437Difference (null H = exogenous): chi2(1) = 6.54 Prob > chi2 = 0.011

Table 9 presents the GMM analysis of Return on Equity (ROE) using the xtabond2 method, incorporating variables such as dividends (DIV), size (SIZE), growth (GROW), earnings to assets (ETA), debt to assets (DTA), long-term assets (LTA), loan loss provisions (LLP), other revenues (OTR), GDP, inflation (INF), and foreign ownership (FO) The model employs a two-step estimation approach to examine the relationship between lagged ROE and the specified independent variables.

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 22 Obs per group: min = 8

ROE | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 14.83916 1.449383 10.24 0.000 11.99842 17.6799 DIV | | 0737901 0160693 4.59 0.000 0422948 1052853 SIZE | -.0129415 0047171 -2.74 0.006 -.0221868 -.0036962 GROW | -.0174496 0321417 -0.54 0.587 -.0804461 045547 ETA | -1.428382 2249822 -6.35 0.000 -1.869339 -.9874247 DTA | 028932 04398 0.66 0.511 -.0572672 1151312 LTA | 0136423 047101 0.29 0.772 -.0786739 1059586 LLP | -.340126 469232 -0.72 0.469 -1.259804 5795518 OTR | -.0752856 0346262 -2.17 0.030 -.1431517 -.0074194 GDP | 387136 1816317 2.13 0.033 0311443 7431277 INF | -.3937675 1630354 -2.42 0.016 -.7133111 -.074224

FO | -.8852903 6722568 -1.32 0.188 -2.202889 4323088 _cons | 3908853 1057122 3.70 0.000 1836932 5980775 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.ROE collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.ROE collapsed

The Arellano-Bond test results indicate a significant presence of first-order autocorrelation (AR(1)) with a z-value of -2.71 and a p-value of 0.007, while the second-order autocorrelation (AR(2)) shows no evidence of autocorrelation, with a z-value of -0.66 and a p-value of 0.509 Additionally, the Sargan test for over-identifying restrictions yields a chi-squared statistic of 11.71 with a probability value of 0.230, suggesting that the model's instruments are not overly restricted, although the results are not robust.

Hansen test of overid restrictions: chi2(9) = 12.04 Prob > chi2 = 0.211 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 4.46 Prob > chi2 = 0.813Difference (null H = exogenous): chi2(1) = 7.58 Prob > chi2 = 0.006

Table 10 presents the findings from the DIV-SDROA analysis using the FO xtabond2 method, highlighting the influence of various factors such as l.SDROA, DIV, SIZE, GROW, ETA, DTA, LTA, LLP, OTR, GDP, INF, and FO The results were obtained through a two-step GMM estimation process, focusing on the relationship between lagged SDROA and the selected independent variables.

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 22 Obs per group: min = 8

SDROA | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 4656725 0974671 4.78 0.000 2746405 6567046 DIV | -.1046989 | 4022046 -0.26 0.795 -.8930055 6836076 SIZE | 1366896 0660375 2.07 0.038 0072585 2661206 GROW | 5615944 3211366 1.75 0.080 -.0678217 1.191011 ETA | -1.94707 1.42994 -1.36 0.173 -4.749701 8555609 DTA | -1.341953 5865781 -2.29 0.022 -2.491625 -.1922807 LTA | 2.120503 3834349 5.53 0.000 1.368984 2.872021 LLP | 24.39203 8.855904 2.75 0.006 7.034772 41.74928 OTR | -2.333736 5026256 -4.64 0.000 -3.318864 -1.348608 GDP | 5.965727 2.578687 2.31 0.021 9115943 11.01986 INF | 1.99197 2.42739 0.82 0.412 -2.765626 6.749567

FO | -23.82721 10.87282 -2.19 0.028 -45.13754 -2.516876 _cons | 2012096 1.251773 0.16 0.872 -2.252221 2.65464 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.SDROA collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.SDROA collapsed

The Arellano-Bond test results indicate a significant presence of first-order autocorrelation (AR(1)) in first differences, with a z-value of -2.87 and a p-value of 0.004 Conversely, the test for second-order autocorrelation (AR(2)) shows no significant evidence, yielding a z-value of -0.51 and a p-value of 0.608 Additionally, the Sargan test for overidentifying restrictions reveals a chi-squared statistic of 25.37 with a p-value of 0.003, suggesting that while the results are not robust, they are not adversely affected by the use of multiple instruments.

Hansen test of overid restrictions: chi2(9) = 11.79 Prob > chi2 = 0.226 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 5.69 Prob > chi2 = 0.682Difference (null H = exogenous): chi2(1) = 6.09 Prob > chi2 = 0.014

Table 11: DIV SDROE - with FO

The xtabond2 command is utilized to analyze the relationship between the dependent variable, SDROE, and several independent variables, including DIV, SIZE, GROW, ETA, DTA, LTA, LLP, OTR, GDP, INF, and FO This analysis employs a two-step Generalized Method of Moments (GMM) approach, ensuring robust estimation of the effects of these variables on SDROE while addressing potential endogeneity issues.

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 22 Obs per group: min = 8

SDROE | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 6384196 0982612 6.50 0.000 4458312 831008 DIV | | 0390901 415387 0.09 0.925 -.7750535 8532337 SIZE | -.0467926 0470367 -0.99 0.320 -.1389829 0453977 GROW | 6813775 3315216 2.06 0.040 0316071 1.331148 ETA | -7.047322 1.484557 -4.75 0.000 -9.957 -4.137644 DTA | -.7530288 5170425 -1.46 0.145 -1.766414 2603559 LTA | 1.248416 4935407 2.53 0.011 2810941 2.215738 LLP | 3.578668 9.074036 0.39 0.693 -14.20612 21.36345 OTR | -2.612303 5525253 -4.73 0.000 -3.695233 -1.529374 GDP | 4.197381 3.051321 1.38 0.169 -1.783099 10.17786 INF | -4.264278 2.332479 -1.83 0.068 -8.835853 3072966

FO | -14.40523 10.26678 -1.40 0.161 -34.52775 5.717293 _cons | 4.008191 802089 5.00 0.000 2.436125 5.580256 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.SDROE collapsed

DIV SIZE GROW ETA DTA LTA LLP OTR GDP INF FO

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.SDROE collapsed

The Arellano-Bond test results indicate an AR(1) in first differences with a z-value of -2.92 and a p-value of 0.004, suggesting significant autocorrelation Conversely, the AR(2) test shows a z-value of -0.76 and a p-value of 0.448, indicating no significant autocorrelation at this lag Additionally, the Sargan test for overidentifying restrictions yields a chi-squared value of 23.88 with a p-value of 0.004, which, while not robust, is not adversely affected by the presence of numerous instruments.

Hansen test of overid restrictions: chi2(9) = 16.54 Prob > chi2 = 0.056 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 11.50 Prob > chi2 = 0.175Difference (null H = exogenous): chi2(1) = 5.04 Prob > chi2 = 0.025

Table 12 presents the findings from the xtabond2 analysis, focusing on the relationship between non-return on assets (NON-ROA) and various economic indicators such as size, growth, earnings, debt-to-asset ratio (DTA), long-term assets (LTA), loan loss provisions (LLP), operating tax rate (OTR), gross domestic product (GDP), and inflation (INF) The two-step generalized method of moments (GMM) estimation is employed to evaluate the impact of these variables on the lagged return on assets (l.ROA) The results highlight the significance of these factors in understanding the dynamics of non-ROA performance.

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 21 Obs per group: min = 8

ROA | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 9677299 0812469 11.91 0.000 8084889 1.126971 NON | | 0066461 0015021 4.42 0.000 0037019 0095902 SIZE | -.0007827 0003067 -2.55 0.011 -.0013838 -.0001816 GROW | 0001305 0024718 0.05 0.958 -.0047141 0049751 ETA | -.0258212 0101492 -2.54 0.011 -.0457133 -.0059291 DTA | -.0041949 0026034 -1.61 0.107 -.0092975 0009076 LTA | 002293 0017982 1.28 0.202 -.0012313 0058174 LLP | 0036709 028683 0.13 0.898 -.0525468 0598885 OTR | -.0099447 002527 -3.94 0.000 -.0148975 -.0049919 GDP | 0155368 0071921 2.16 0.031 0014406 0296331 INF | -.0122545 0109175 -1.12 0.262 -.0336524 0091435 _cons | 0226397 005943 3.81 0.000 0109915 0342878 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(NON SIZE GROW ETA DTA LTA LLP OTR GDP INF)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.ROA collapsed

NON SIZE GROW ETA DTA LTA LLP OTR GDP INF

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.ROA collapsed

The Arellano-Bond test results indicate a significant first-order autocorrelation (AR(1)) in first differences, with a z-value of -2.80 and a p-value of 0.005 However, the second-order autocorrelation (AR(2)) shows no significant evidence, as reflected by a z-value of -0.04 and a p-value of 0.964 Additionally, the Sargan test for overidentifying restrictions yields a chi-squared value of 20.86 with a probability of 0.013, suggesting that while the results are not robust, they are not adversely affected by the use of multiple instruments.

Hansen test of overid restrictions: chi2(9) = 13.69 Prob > chi2 = 0.134 (Robust, but weakened by many instruments.)

Difference-in-Hansen tests of exogeneity of instrument subsets:

Hansen test excluding group: chi2(8) = 6.15 Prob > chi2 = 0.630Difference (null H = exogenous): chi2(1) = 7.54 Prob > chi2 = 0.006

Table 13: NON-ROE without FO

xtabond2 ROE l.ROE NON SIZE GROW ETA DTA LTA LLP OTR GDP INF, gmm(l.ROE,collapse) iv (NON SIZE GROW ETA DTA LTA LLP OTR GDP INF) twostep

Dynamic panel-data estimation, two-step system GMM

Group variable: Bank Number of obs = 278

Time variable : YEAR Number of groups = 28

Number of instruments = 21 Obs per group: min = 8

ROE | Coef Std Err z P>|z| [95% Conf Interval] -+ -

L1 | 13.12541 1.619983 8.10 0.000 9.950298 16.30052 NON | | 0915563 0234453 3.91 0.000 0456044 1375082 SIZE | -.0097056 0046613 -2.08 0.037 -.0188416 -.0005697 GROW | -.0126999 0325427 -0.39 0.696 -.0764825 0510826 ETA | -1.28769 2265466 -5.68 0.000 -1.731714 -.8436673 DTA | 0076787 0427754 0.18 0.858 -.0761596 0915169 LTA | 0222321 0390006 0.57 0.569 -.0542076 0986718 LLP | -.1747065 4681894 -0.37 0.709 -1.092341 7429278 OTR | -.0922165 035639 -2.59 0.010 -.1620677 -.0223652 GDP | 2056861 1090857 1.89 0.059 -.008118 4194903 INF | -.3234242 1508908 -2.14 0.032 -.6191646 -.0276837 _cons | 3069613 1062408 2.89 0.004 0987332 5151894 - Warning: Uncorrected two-step standard errors are unreliable.

Instruments for first differences equation

D.(NON SIZE GROW ETA DTA LTA LLP OTR GDP INF)

GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/10).L.ROE collapsed

NON SIZE GROW ETA DTA LTA LLP OTR GDP INF

GMM-type (missing=0, separate instruments for each period unless collapsed) D.L.ROE collapsed