The Impact Of Liquidity Risk On Bank Profitability Empirical Evidence From Listed Commercial Banks In Vietnam.pdf

INTRODUCTION

REASONS FOR CHOOSING THE STUDY

Banks hold the most influential position in the financial sector and are essential in driving economic development worldwide through their multifaceted roles as financial intermediaries, funds facilitators, and supporters (Malik, Awais and Khursheed 2016; Mohanty and Mehrotra 2018) The banks are not only the storehouses of the economy’s wealth but also provide funds for resource mobilization for the businesses (Salim and Bilal 2016) Besides, the bank plays a crucial role in engaging in significant activities on both sides of the balance sheet On the asset side, they strengthen the flow of funds by providing essential cash to short-term users through lending Simultaneously, on the liability side, they accumulate liquidity from savers, meaning the bank will both mobilize and receive deposits with an appropriate deposit rate to source capital from various customer bases, thereby playing a role in influencing how the overall financial system operates (Arif and Anees 2012; Diamond and Rajan 2001; Golubeva, Duljic and Keminen 2019) In summary, this indicates that banks play an important role in efficiently transferring capital from surplus economic units to deficit economic units (Tesfaye 2012) and is heightened significant attention from policymakers, leadership, and investors

Due to the multifaceted roles of the banking sector, which involves maintaining the balance between money supply and demand, diversifying its functions, and fulfilling its tasks by providing a wide range of products and services to different entities on various scales, banks may be exposed to various risks such as credit risk, liquidity risk, interest rate risk, market risk, and operational risk, all of which impact the banks' profitability (Arif and Anees 2012; Chen et al 2018) Among these risks, liquidity risk stands out as one of the most influential risks, a significant risk in banking operations and liquidity management that has received considerable attention worldwide, mainly from regulators and policy-makers or most institutions involved in global economic fluctuations (Alalade, Ogbebor and Akwe 2020; Hacini, Boulenfad and Dahou 2021) Effective liquidity risk management ensures the bank has enough cash and resources to meet customer payment and withdrawal demands in all situations Liquidity shortages pose a significant threat not only to individual banks but also to the overall stability of the financial system When a bank experiences a liquidity shortfall, it may be unable to meet its obligations to depositors and other creditors, leading to a domino effect of financial distress This can have far- reaching consequences, including a decline in asset prices, increased market volatility, and even a systemic crisis, impacting the bank's profitability Therefore, banks must prioritize effective liquidity risk management, ensuring they conserve enough liquidity and are prepared to navigate various challenges, potential losses, or vulnerabilities in funding sources (Hlebik and Ghillani 2017)

According to Khan and Ali (2016) and Mohanty and Mehrotra (2018), liquidity and profitability are fundamental concepts in the corporate domain For banks, liquidity holds immense significance because it will impact on their profitability (Charmler et al 2018; Hacini, Boulenfad and Dahou 2021), with liquidity risk serving as a crucial determinant impacting bank's profitability (Almazari 2014; Chen et al 2018) As presented above, the primary function of banks is to gather funds from the public through deposits and then utilize these funds, along with their resources, to promptly meet customers’ demands (Malik, Awais and Khursheed 2016; Mohanty and Mehrotra 2018) To achieve this, banks must strike a delicate balance between adequate liquidity and maximizing profits Profitability is crucial for the sustainability of any commercial entity, and it reflects the relationship between the amount of profit earned and various other factors However, compared to other businesses, banks must emphasize on balancing profitability and liquidity

(Mohanty and Mehrotra 2018) The liquidity and profitability can be likened to opposing forces with different goals, constantly applying pressure that may strain the bank's stability (Malik, Awais and Khursheed 2016) On the one hand, banks need to maintain a sufficiently high level of liquidity to ensure they have enough cash reserves to meet short-term customer needs and payment obligations On the other hand, banks must also enhance profitability by efficiently using assets and optimizing business operations Therefore, the relationship between liquidity and profitability can be seen as two contrasting factors, the balancing of which can sometimes be challenging and create pressure for banks (Malik, Awais and Khursheed 2016; Mohanty and Mehrotra 2018)

In Vietnam, the study of Doan Thanh Ha et al (2022) indicated that the commercial banking system had undergone two system restructuring times from 2012 to 2015, with several banks facing liquidity challenges The commercial banking system decreased by 5 joint-stock commercial banks due to mergers and acquisitions (MandA), including Ficom Bank, TinNghia Bank, Habu Bank, Western Bank, and DaiA Bank Additionally, the SBV acquired 3 commercial banks - VNC Bank, OceanBank, and GPBank - at the price of zero dong This adversely affected the operational efficiency of Vietnam’s entire banking system Moreover, Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan (2022) also indicated that in Vietnam, the banking sector has been experiencing financial crises recently, with several commercial banks collapsing due to non-performing loans Therefore, these commercial banks were acquired and merged by SBV with state-owned commercial banks The SBV has had to step in to address these issues The main reason for these failures has been a lack of liquidity in business operations (State Bank of Vietnam 2018) This has been worsened by commercial banks aggressively expanding credit, especially in real estate investments, putting strain on the banking and financial sectors The COVID-19 pandemic from 2020 to 2021 exacerbated challenges, notably liquidity risks, for commercial banks These risks threatened the safety of individual banks and the overall banking system (Le Ngoc Thuy Trang et al 2021)

In summary, these fluctuations have negatively impacted on the profitability of listed commercial banks in Vietnam

Furthermore, Vietnam must adhere to international banking standards of the Basel Accords of the BCBS Currently, over 20 credit institutions in Vietnam are implementing Basel II by the SBV’s requirements under Circular No 41/2016/TT- NHNN dated December 30, 2016, which stipulates capital adequacy ratios for banks and foreign bank branches The development strategy for the banking sector of Vietnam until 2025, with a vision towards 2030, aims to ensure that bare commercial banks will achieve capital levels compliant with Basel II standards by 2020 By 2025, all commercial banks will adopt Basel II using standardized approaches and implement Basel II on an advanced basis (Quynh Trang 2022) Currently, no regulation mandates Vietnam’s commercial banks to adhere to the Basel III standards However, some commercial banks have voluntarily adopted Basel III due to its need for development This is because the SBV prioritizes credit approval for banks with ample levels of shareholders' equity, high capital adequacy ratios (CAR), and robust risk management capabilities Moreover, this helps banks improve credit ratings and enhance competitiveness in the international market (Nguyen Thi Anh Ngoc and Nguyen Thi Diem 2023) Meeting these standards is challenging, especially in effectively achieving banking regulatory standards related to liquidity These measures aim to promote the growth and stability of Vietnam's banking sector as it matures Therefore, understanding the relationship between liquidity and bank profitability is crucial for navigating economic challenges, especially amid the current global economic uncertainties characterized by complexities and unpredictable developments (Le Ngoc Thuy Trang et al 2021)

There have been numerous relevant empirical studies on the impact of liquidity risk on the profitability of commercial banks, and this is a topic that is gaining attention from various economic stakeholders worldwide, including those in Vietnam However, the research results have been inconsistent The findings indicated that there existed a relationship between liquidity risk and bank profitability, although the relationship varied depending on the specific indicator utilized to measure bank profitability For instance, several studies showed that there was a significant relationship between liquidity risk and profitability (Alalade, Ogbebor and Akwe 2020; Charmler et al 2018; Hacini, Boulenfad and Dahou 2021; Ishari and Fernando 2023; Muriithi and Waweru 2017; Le Ngoc Thuy Trang et al 2021; Nguyen Thanh Phong 2020; Ren 2022; Saleh and Afifa 2020; Salim and Bilal 2016; Tang My Sang 2019; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) The results of each study revealed that the relationship varied depending on the specific indicator utilized to represent bank profitability, so that this relationship could be positive or negative On the other hand, Golubeva, Duljic and Keminen (2019) and Dong (2021) considered that there was an unclear relationship or an insignificant impact between liquidity risk and profitability

From the analysis above, the author has provided an overview of the practical situation regarding liquidity risk issues and their impact on the profitability of listed commercial banks in Vietnam It is acknowledged that liquidity risk is a challenge faced by banks, and striking a balance between liquidity risk and profitability remains a constant challenge Additionally, the banking system has undergone various fluctuations, with the most recent notable event being the COVID-19 pandemic Besides, studies across different areas and timeframes yield provided inconsistent results on the relationship between liquidity risk and bank profitability, both internationally and in Vietnam, because it depended on the measurement of the dependent variable, which led to numerous continuous debates among researchers Therefore, the author sees the need for continuously updated studies to reflect the evolving context and to reexamine the relationship between liquidity risk and profitability of the banking system in the current Vietnam context This is the main reason the author has chosen the topic "The impact of liquidity risk on bank profitability: Empirical evidence from listed commercial banks in Vietnam” as the author’s study Furthermore, in investigating this relationship, the author will also consider the impact of the COVID-19 pandemic during the period of 2020-2021 on the relationship between liquidity risk and profitability of listed commercial banks in Vietnam to provide a broader perspective when Vietnam’s economy had fluctuations.

STUDY OBJECTIVES

The general objective of this study is to analyze the impact of liquidity risk on bank profitability of 26 listed commercial banks in Vietnam, alongside examining the impact of COVID-19 on the relationship between liquidity risk and bank profitability

To address the general objective, this study needs to focus on four specific objectives Firstly, the study will analyze the impact of liquidity risk on the profitability of listed commercial banks in Vietnam Secondly, the study will determine the extent of liquidity risk’s impact on the profitability of listed commercial banks in Vietnam Thirdly, the study determines the extent to which the COVID-19 pandemic has impacted the effect of liquidity risk on bank profitability The final specific objective is to discuss the findings based on the research results regarding the impact of liquidity risk on the profitability of listed commercial banks in Vietnam, both with and without the impact of the COVID-19 pandemic, including a comparison with results from other relevant empirical studies.

STUDY QUESTIONS

To achieve specific objectives, the study needs to address the following questions The first question is whether liquidity risk affects the profitability of listed commercial banks in Vietnam? In response to the second objective, the study has the second question: what is the extent of the impact of liquidity risk on the profitability of listed commercial banks in Vietnam? The third question is, what is the extent of the COVID-19 pandemic on the relationship between liquidity risk and bank profitability of listed commercial banks in Vietnam? The final question is, what are the conclusions regarding the impact of liquidity risk on the profitability of listed commercial banks in Vietnam based on the research results in the context of having and not having the impact of the COVID-19 pandemic? The completion of this study in addressing all the above study questions will fulfill the objectives set for this study topic.

THE SUBJECT AND SCOPE OF STUDY

The study subject of the topic is similar to the author's studies (Alalade, Ogbebor and Akwe 2020; Charmler et al 2018; Golubeva, Duljic and Keminen 2019; Hacini, Boulenfad and Dahou 2021; Muriithi and Waweru 2017; Ren 2022; Saleh and Afifa 2020; Salim and Bilal 2016) internationally or the studies in Vietnam (Le Ngoc Thuy Trang et al 2021; Nguyen Thanh Phong 2020; Tang My Sang 2019; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) These studies of numerous authors conducted their study focusing on the impact of liquidity risk on bank performance or profitability of commercial banks in numerous countries Especially in Vietnam, the author recognizes that the study of Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan (2022) also conducted their study on the impact of liquidity on the profitability of listed commercial banks in Vietnam

Basing on the impacts of inadequate liquidity can be significant, affecting both individual banks and the entire banking system In Vietnam, the banking sector has been experiencing financial crises recently, with several commercial banks collapsing due to non-performing loans and the COVID-19 pandemic Besides that, numerous relevant empirical studies provided inconsistent results on the relationship between liquidity risk and bank profitability The author acknowledges this is a significant issue, so the main factor studied throughout this study is the impact of liquidity risk on the profitability of listed commercial banks in Vietnam This study will clarify the impact of liquidity risk on the profitability of listed commercial banks in Vietnam in a specific direction

About the scope of time, this research data will be collected from 2012 to 2022 (11 years) The banking system in Vietnam experienced structural changes from 2012 to 2015 Specifically, from 2012 to 2015, the commercial banking system had significant structural changes, marked by the reduction of five commercial banks through M&A (Mergers and Acquisitions) such as Ficom Bank (First Joint Stock Commercial Bank), TinNghia Bank (Vietnam Tin Nghia Joint Stock Commercial Bank), Habu Bank (Hanoi Building Commercial Joint Stock Bank), Western Bank, DaiA Bank (DaiA Commercial Joint Stock Bank) At the end of 2015, the SBV acquired three struggling banks at the price of 0 dong, including VNCB Bank (The Vietnam Construction Bank), OceanBank (Ocean Commercial One Member Limited Liability Bank), and GPBank (Global Petro Commercial Joint Stock Bank) (Doan Thanh Ha et al 2022) Moreover, according to Le Ngoc Thuy Trang et al (2021), this period had policy changes that impacted liquidity operations This means that regulatory policy changes during the research period significantly impacted liquidity operations The adoption of the international Basel III agreement, the latest version of Basel standards, aimed to enhance regulations on bank liquidity, improving the banking sector's resilience to financial risks and preventing future economic crises The COVID-19 pandemic happened from 2020 to 2021, which is a crisis period due to the global COVID-19 pandemic, this event also had impact on the profitability of listed commercial banks in Vietnam In other words, under various impactful events, the profitability of listed commercial banks in Vietnam experienced fluctuations Relevant empirical studies such as (Alalade, Ogbebor and Akwe 2020; Muriithi and Waweru 2017; Le Ngoc Thuy Trang et al 2021; Nguyen Thanh Phong 2020), etc have also conducted on a relatively similar or shorter timeframe By combining events related to the impact of liquidity risk on bank profitability and the length of the relevant studies’s period, the author chooses the timeframe (2012-2022) as the official scope of time for this study

About the scope of the space, this study will be conducted at 26 listed commercial banks in Vietnam, which are listed on the HNX, HOSE, and UPCOM exchanges There are a total of 27 listed banks in the Vietnam’s stock market To collect data within the timeframe from 2012 to 2022, data from 26 listed commercial banks have been collected One bank does not meet the criteria, the author excludes Viet Nam Thuong Tin Commercial Joint Stock Bank - Vietbank as this bank lacks sufficient research data for 2012, 2013, and 2014 Besides, relevant empirical studies within the scope of research in Vietnam (Nguyen Thanh Phong 2020; Tang My Sang 2019; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) also conducted research involving 25, 19, and 18 joint-stock commercial banks in Vietnam respectively Given the inconsistency in the number of banks studied, this thesis will be conducted from the author’s standpoint, involving 26 listed commercial banks (as mentioned above) to compare the objectives of this research and those of previous authors Therefore, the official study scope of this study is limited to 26 listed commercial banks to ensure the representativeness of the entire banking system in Vietnam, which is presented in detail in Appendix 1.

RESEARCH METHODOLOGY AND DATA SOURCES

With microdata, the data is used to investigate the impact of liquidity risk on bank profitability is secondary data that will be collected from audited financial statements of 26 listed commercial banks in Vietnam from 2012 to 2022 (11 years) All the collected data were taken from the FiinPro-X platform After that, the author verified it against primary financial data from the bank’s published and audited financial statements With macrodata, the author collects macroeconomic data from the official announcement of the General Statistics Office of Vietnam

(https://www.gso.gov.vn/) and the World Bank

(https://www.worldbank.org/vi/country/vietnam)

To solve the study objectives, this study uses a quantitative research method The author conducts the study based on the data collected from 26 listed commercial banks in Vietnam from 2012 to 2022, totaling 286 observations The research data are organized into balanced panel data, and Microsoft Excel 2021 is used for data synthesis and processing Additionally, Stata software version 15 is also utilized to analyze and test regression models, aiming to address the study questions presented at the listed commercial banks in Vietnam and achieve the study objectives The estimation method used for the study model is the GMM method To assess how liquidity risk influences the profitability of listed commercial banks in Vietnam It will also discuss its findings regarding the impact of liquidity risk on the profitability of listed commercial banks in Vietnam Therefore, the regression method (GMM) is an appropriate approach

In the literature review aspect, the author observed that the method which is used in this study (GMM) is also similar to most relevant empirical studies (Muriithi and Waweru 2017; Nguyen Thanh Phong 2020; Saleh and Afifa 2020; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021) when they conducted their study about the relationship between liquidity risk and bank profitability in Vietnam and various countries Additionally, there are several other methods, such as OLS, FEM, and REM which are also used in some other studies (Alalade, Ogbebor and Akwe 2020; Charmler et al 2018; Dong 2021; Golubeva, Duljic and Keminen 2019; Hacini, Boulenfad and Dahou 2021; Ishari and Fernando 2023; Le Ngoc Thuy Trang et al 2021; Ren 2022; Salim and Bilal 2016; Tang My Sang 2019; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022)

However, in this author’s study, the author is going to give lagged effects and interaction variables between the COVID-19 pandemic and liquidity risk in the proposed research model Due to the limitations of the pooled OLS model in estimating panel data, which may introduce bias due to variance, autocorrelation, or endogeneity (Kiviet 1995), FEM and REM estimates are employed to control individual effects However, as FEM and REM fall short in addressing endogeneity issues (Ahn and Schmidt 1995), the GMM method is utilized to overcome these challenges (Arellano and Bond 1991; Hansen 1982; Hansen, Heaton and Yaron 1996) GMM estimators is another technique to address the endogeneity, heteroscedasticity, and serial correlation problem among variables in econometrics introduced by Arellano and Bond (1991) Firstly, this approach rectifies heteroscedasticity, autocorrelation, and endogeneity issues within the context of panel data analysis Secondly, it uses lagged values for dependent variables while addressing concerns regarding instrumental variable incorporation Lastly, this technique offers an estimator capable of capturing correlations among explanatory variables (Arellano and Bond 1991) Anderson and Hsiao (1981) propose using lagged values of dependent variables as instruments, ensuring that these lags are not correlated with the residuals This study’s analysis also utilizes alternative profitability measures as instrumental variables in instrumental variable estimation Additionally, certain exogenous variables serve effectively as instruments in dynamic panel data analysis The research employs dynamic panel data methodology following the framework outlined (Arellano and Bond 1991; Hansen 1982; Hansen, Heaton and Yaron 1996) The GMM technique yields reliable, standardized, and efficient coefficient distributions.

CONTRIBUTIONS OF THE STUDY

The author's study combines with other relevant empirical studies on the topic in Vietnam and internationally to reexamine the extent of the impact of liquidity risk on the profitability of listed commercial banks in Vietnam from 2012 to 2022, considering the impact of the COVID-19 pandemic The thesis provides more detailed and reliable results by utilizing the GMM estimation method, which helps address endogeneity issues In terms of practical application, the study aims to verify the impact of liquidity risk on the profitability of listed commercial banks in Vietnam The study findings will provide further empirical evidence of the extent of this relationship and valuable information for other researchers in the finance and banking sector and various relevant parties Based on these results, recommendations will also be made to enhance banks' liquidity risk management effectiveness.

DISPOSITION OF THE STUDY

Korrapati (2016) proposed a general idea for a quantitative study thesis, suggesting it should consist of five chapters Therefore, this study will be structured according to the five-chapter framework recommended by Korrapati (2016) Moreover, the author observed that the five-chapter structure is commonly used in quantitative studies within the scientific study field

Chapter 1 Introduction This chapter provides readers a comprehensive and concise overview, offering a general perspective on the issues the study aims to address Specifically, chapter 1 presents an overview of the thesis content, the reasons for choosing the topic, study objectives, outlines the study questions, study scope and objectives, and study methodology Additionally, the chapter highlights the contributions of the selected topic and concludes by presenting the overall structure and organization of the study

Chapter 2 Theoretical framework and review of relevant empirical studies Chapter 2 presents concepts related to the study problems Specifically, the study will first offer the theoretical basis of liquidity risk and bank profitability In addition, the theoretical basis and relevant empirical studies from international and Vietnam Therefore, the author identifies the study gaps, which will be a foundation for building the study model in the next chapter

Chapter 3 Research model and methodology This chapter establishes a study model based on the proposal for choosing a model Then, the study will explain the chosen variables within the model, detail the method employed for collecting research data, and outline the sequence for implementing the model

Chapter 4 Research results and discussion Chapter 4 outlines procedures for conducting quantitative analysis on a secondary dataset comprising observed samples obtained from listed commercial banks through econometric methods The results of the data analysis will be discussed, focusing on comparing these results with those obtained in other relevant empirical studies

Chapter 5 Conclusions and recommendations Based on the results obtained in Chapter 4, Chapter 5 will provide some recommendations designed for relevant parties aiming to maintain the liquidity risk but still achieve a stable profitability of listed commercial banks This section will also address the study limitations and propose some future study directions

Chapter 1 of this study highlights the importance of the study subject and clarifies both general and specific objectives that will be addressed through corresponding study questions about the topic “The impact of liquidity risk on bank profitability: Empirical evidence from listed commercial banks in Vietnam” Additionally, the study identifies the subject, scope of the study, and the study methodology and data sources during the 11 years from 2012 to 2022, involving 26 listed commercial banks in Vietnam, building upon the foundation inherited from and extending previous relevant empirical studies, this study also considers the interaction effects of COVID-19 pandemic and liquidity risk on bank profitability of listed commercial banks in Vietnam After affirming the scientific significance and practical contribution of the study topic, the final section of the chapter provides an overview of the structure of the thesis, which will consist of 5 main chapters Therefore, the following Chapter 2 will need to identify relevant theories and relevant empirical studies to make a foundation for this study and address this study objective.

THEORETICAL FRAMEWORK AND REVIEW OF

REVIEW OF COMMERCIAL BANK LIQUIDITY

Regarding assets, liquidity reflects the ability to quickly convert assets into cash at a low cost and vice versa An asset is considered highly liquid if it simultaneously satisfies two characteristics: having a trading market and maintaining a relatively stable price that is not affected by the volume and timing of transactions (Rose 2001)

According to Duttweiler (2011), liquidity represents the ability to meet all payment obligations as they come due, to the fullest extent, and in a specified currency Since liquidity is essentially cash, it is solely related to cash flows Failure to meet these payment obligations will result in insolvency

In summary, liquidity is the ability to quickly and stably convert assets into cash, as well as the capacity to meet payment obligations as they come due.

REVIEW OF COMMERCIAL BANK LIQUIDITY RISK

2.2.1 The definition of liquidity risk

According to Basel Committee on Banking Supervision (2008), liquidity refers to a bank’s ability to fund increases in assets and meet obligations when they mature without suffering unacceptable losses The fundamental role of banks in transforming short-term deposits into long-term loans makes them vulnerable to liquidity shortages, which is precisely liquidity risk This risk can arise from the bank's operations or general market fluctuations

Duttweiler (2011) argued that liquidity risk is the risk that arises when a financial institution is unable to make payments in time or has to mobilize capital at high costs to meet payment demands Besides that, there are also other reasons which impair the ability of the financial institution to make payments This can lead to adverse consequences for the commercial banks Liquidity risk occurs when a bank does not have sufficient financial resources to fulfill its debt obligations at maturity Commercial banks have to use high-cost financial sources, even though the bank can still make payments (Vento and Ganga 2009)

Decker (2000) and Pham Thi Hoang Anh (2015) defined liquidity risk as a type that can be categorized into market liquidity risk and funding liquidity risk Market liquidity risk arises when banks face risks that are not easily settled without market price adjustments due to incomplete market information or disruptions In other words, market liquidity risk is the risk that a bank can not sell its assets in the market quickly and at the lowest cost Funding liquidity risk is a situation where a bank does not have enough capital to meet its requirements for available funds As market liquidity risk and funding liquidity risk interact through the financial market, they can significantly impact credit institutions in the market, especially during periods of financial market volatility To meet their capital needs, commercial banks often conduct the following measures such as borrowing on the interbank money market, selling financial market assets, engaging in foreign exchange swaps, and ultimately resorting to discounting or capital restructuring transactions with other financial institutions A particular bank may still receive support from other banks relatively quickly if financial markets continue to operate normally and temporary liquidity difficulties just happen in a bank However, suppose the financial market is experiencing a period of volatility In that case, the liquidity shortfall of a bank will likely lead to numerous difficulties for other banks due to the tight interconnectedness of banks within the system This situation could result in multiple banks attempting to sell their assets in the financial market, creating a liquidity risk

Thus, liquidity risk is an inherent and inevitable risk commercial banks face Liquidity risk appears when a bank cannot fund its assets or fulfill obligations when they come due at a reasonable cost or when it can't sell assets at a fair price to meet liquidity requirements A bank will be exposed to liquidity risk if it fails to raise enough capital by increasing borrowing or immediately converting assets at a reasonable price

2.2.2 Bank liquidity risk measurement methods

2.2.2.1 Method based on the regulation of the Basel Committee

To measure the liquidity risk of the entire banking system, relevant empirical studies have often employed traditional methods However, since the global financial crisis, banking institutions have increasingly focused on liquidity risk, prompting the Basel Committee to introduce the third version, Basel III agreement In this version, there is a greater emphasis on liquidity risk at both the individual bank level and the overall banking system (Pham Thi Hoang Anh 2015) The studies of Golubeva, Duljic and Keminen (2019) and Muriithi and Waweru (2017) proposed a research method based on Basel III regulations, which include two ratios LCR (Liquidity Coverage Ratio) aims to enhance banks' resilience by ensuring they have high-quality liquidity sources to withstand short-term difficulties and NSFR (Net Stable Funding Ratio) aim to strengthen banks' long-term liquidity resilience by creating incentives for banks to raise funds from more stable sources while ensuring that banks continue their normal operations

According to Basel Committee on Banking Supervision (2010), LCR consists of two components, which are stock of high-quality liquid assets and total net cash outflows over the next 30 calendar days The following formula specifies this

The objective of this ratio is to guarantee that a bank sustains a sufficient amount of unencumbered, high-quality liquid assets that can be transformed into cash This is crucial for addressing its liquidity requirements over a 30-day calendar period, particularly in the face of a notably severe liquidity stress situation outlined by supervisors At the very least, the supply of liquid assets should allow the bank to endure until the 30th day of the stress scenario During this period, it is assumed that suitable corrective measures can be implemented by management and supervisors or that the bank can resolve the issue systematically (Basel Committee on Banking Supervision 2010)

Regarding NSFR, NSFR is specified in the following formula

This ratio was introduced in 2018, the NSFR is the ratio between the available stable funding and the required stable funding over a one-year period (in contrast to the 30-day horizon of the LCR) Basel III requires the NSFR must reach a minimum of 100% Available stable funding refers to capital expected to remain stable over a specific period, typically one year To calculate the NSFR, banks must categorize the book value of all types of capital and debt assets into one of five groups (as defined by the Basel Committee) based on maturity and withdrawal capabilities The required stable funding depends on (i) the liquidity characteristics and remaining term of the assets held by the organization and (ii) the liquidity characteristics and remaining term of the value of off-balance sheet items The NSFR ratio ensures banks maintain a stable capital structure, reducing reliance on short-term funding for long-term assets Together with the LCR ratio, this ratio supports the transformation of the liquidity risk structure of organizations, shifting from short-term capital to more stable, longer-term sources The goal of the NSFR is to limit dependence on short- term funding, promote a better assessment of liquidity risk, and reduce incentives for reserve-financing institutions with short-term capital maturing just beyond the 30- day period This contributes to the overall stability of the financial system by aligning the maturity profiles of the bank’s assets and liabilities (Basel Committee on Banking Supervision 2010)

2.2.2.2 Method based on liquidity ratios

Basel III introduces increased capital, liquidity, and debt standards compared to Basel I and II (Basel Committee on Banking Supervision 2010) Additionally, in response to regulatory inefficiencies observed during the 2007-2009 financial crisis, Basel III tackles business cycles and systemic risk concerns Specifically, banks must maintain capital reserves during economic expansion phases to mitigate potential losses from economic downturns (Basel Committee on Banking Supervision 2010) However, the disadvantage of this method is the lack of data Studies in emerging markets such as Vietnam often lack sufficient data to conduct methods based on the Basel Committee’s regulations

Moreover, there are other methods, such as the sources and uses of funds approach, the structure of funds approach, the liquidity indicator approach, and the market signals (or discipline) approach However, in this study, the author focuses on presenting the liqudity ratios method because (i) it is one of the most common and widely used methods for measuring bank liquidity; (ii) the ratio method has been studied and applied in many relevant empirical studies of Salim and Bilal (2016), Charmler et al (2018), Golubeva, Duljic and Keminen (2019), Alalade, Ogbebor and Akwe (2020), Saleh and Afifa (2020), Hacini, Boulenfad and Dahou (2021), Dong (2021), Ren (2022), Ishari and Fernando (2023), Tang My Sang (2019), Nguyen Thanh Phong (2020), Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), Le Ngoc Thuy Trang et al (2021) or Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan (2022), demonstrating its reliability and accuracy in measurement; (iii) the ratio method is simple and easy to understand, helping to assess the liquidity of banks based on metrics The liquidity ratios method may be simplistic and not very accurate Nevertheless, its strength lies in the availability of data and existing relevant empirical studies which have been studied Many studies advocate for the use of liquidity ratios as an appropriate method of analysis Hence, the primary measurement approach in this study is liquidity ratios

The first ratio is L1, representing the proportion of liquid assets to total assets This metric assesses liquidity by dividing total liquid assets by total assets, demonstrating the percentage of a bank's assets readily convertible into cash This ratio has been used in several studies (Alalade, Ogbebor and Akwe 2020; Charmler et al 2018; Saleh and Afifa 2020; Salim and Bilal 2016; Tang My Sang 2019; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022)

(2.3) Commercial banks maintain liquidity as a solution to mitigate the risk of bankruptcy by reducing costs and increasing profits However, increasing holdings of numerous liquid assets may decrease the ROE and reduce profitability While holding liquid assets can make banks more resilient to liquidity shocks, having too much can incur significant costs in conditions of reduced business efficiency High- liquidity assets do not generate high profits, so it is necessary to reconsider asset ownership and optimize asset management to improve liquidity while simultaneously optimizing profitability In summary, a higher ratio would indicate an increased capacity to absorb shocks (AlAli 2020; Salim and Bilal 2016)

(2.4) L2 ratio is used in several studies (Charmler et al 2018; Ren 2022; Salim and Bilal 2016; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) The L2 ratio measures liquidity risk in dealing with high demands for short-term liquidity When the L2 ratio is higher, the bank possesses sufficient liquid assets to deal with short- term liquidity Therefore, the bank has liquidity in the short term (Salim and Bilal 2016)

(2.5) L3 ratio is used in studies (Salim and Bilal 2016; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) The L3 ratio compares a bank's liquid assets (including cash and assets easily convertible to cash) to the amount of customer deposits When this ratio is high, it means the bank has more liquid assets compared to the amount of customer deposits This indicates that the bank is better equipped to manage long- term liquidity risk (AlAli, 2020; Salim and Bilal, 2016) It signifies that the bank can meet customers' withdrawal demands and other payment obligations without relying heavily on borrowing from other banks Therefore, a high L3 ratio indicates the bank’s liquidity is good

(2.6) The L4 liquidity ratio of loans to total assets This ratio is used in studies (Salim and Bilal 2016; Tang My Sang 2019; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) A low L4 ratio indicates that the bank has more liquid assets in its total assets, indicating good liquidity Conversely, a high L4 ratio means that the total loans of the bank account for a high proportion of total assets In summary, the higher this ratio, the lower the liquidity of the bank, and vice versa (AlAli 2020; Salim and Bilal 2016)

REVIEW OF COMMERCIAL BANK PROFITABILITY

2.3.1 The definition of bank profitability

According to Rose (2001), bank profitability is the difference between the interest rates on deposits and loans plus the profit from other investment activities minus associated costs These costs are the expenses incurred corresponding to the revenue generated by the bank

The study by Mendoza and Rivera (2017) has presented that profitability is the ability to earn money for a business entity This is one of the crucial factors in creating value for the bank and is a significant step in maximizing shareholders’ wealth

According to the Xu, Hu and Das (2019), bank profitability refers to generating profit or income from the bank's activities and investments This is an essential measure of the financial efficiency of the bank and is calculated by comparing the bank's revenue to its costs, losses, and other expenses A profitable bank will generate enough income to cover operating expenses, loan losses, taxes, and other costs while still providing profits to shareholders Various factors, such as interest rates, economic conditions, competition, and the quality of the loan portfolio can influence bank profitability

Thus, the profitability of a bank refers to its ability to generate profit or income from its activities and investments This is a crucial metric for assessing the financial performance of the bank and is calculated by comparing the bank's revenue with its expenses, losses, and other costs A profitable bank will generate enough income to cover its operating expenses, loan losses, taxes, and other costs while still providing profits to its shareholders Profitability is one of the key metrics used to determine financial efficiency and provide early warnings for investors

Many indicators can be used to measure the profitability of a bank According to Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), the analysis of profitability of banks usually use three main financial ratios which are ROA, ROE, and NIM

Studies (Alalade, Ogbebor and Akwe 2020; Charmler et al 2018; Dong 2021; Golubeva, Duljic and Keminen 2019; Ishari and Fernando 2023; Le Ngoc Thuy Trang et al 2021; Nguyen Thanh Phong 2020; Ren 2022; Saleh and Afifa 2020; Salim and

Bilal 2016; Tang My Sang 2019; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) use ROA ratios to measure profitability of banks The ROA is viewed as an essential indicator of the profitability of an entity and its overall total assets

ROA measures a bank’s management effectiveness, illustrating its capacity to transform assets into net income (Rose and Hudgins 2012) ROA provides analysts insight into the management's efficiency in utilizing its assets to generate earnings (Do Hoai Linh, Ngo Thanh Xuan and Phung Quoc Anh 2020) The higher the ROA, the greater the bank’s profitability and the flexibility in managing assets in response to economic fluctuations This means that the bank is earning more on less spending Nevertheless, a high ROA doesn’t always indicate that a bank knows how to use its assets effectively, it could result from insufficient investment in assets, leading to a decline in asset values and affecting the long-term business performance of the bank When comparing this ratio with historical data and other enterprises in the same industry, it reveals whether the enterprise’s ability to generate profit from its assets is favorable or not When comparing two enterprises in the same sector, if one enterprise has a larger asset base, it is likely to generate higher revenue and profit than the remaining enterprise (economies of scale is explained that the cost advantage of this business increases) Moody’s financial strength assessment standards for the banking industry consider ROA ≥1% as indicative of good profitability CAMEL shows banks are most efficient when ROA is greater than or equal to 1.5% (Rozzani and Rahman 2013)

ROE is much similar to the ROA and is widely used in numerous relevent empirical studies (Alalade, Ogbebor and Akwe 2020; Charmler et al 2018; Golubeva, Duljic and Keminen 2019; Hacini, Boulenfad and Dahou 2021; Ishari and

Fernando 2023; Muriithi and Waweru 2017; Le Ngoc Thuy Trang et al 2021; Nguyen Thanh Phong 2020; Ren 2022; Saleh and Afifa 2020; Salim and Bilal 2016; Tang My Sang 2019; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) This ratio also represents the bank profitability

ROE is a gauge of evaluating the profitability generated per dollar of equity

By looking at the ROE ratio, investors can know how much profit the bank earns with the amount of money that shareholders have invested The ROE is the most helpful ratio when comparing the profitability of two or more firms within the same industry (Do Hoai Linh, Ngo Thanh Xuan and Phung Quoc Anh 2020) The higher the ROE, the greater the effective use of equity by the bank ROE also provides a signal of financial success because it indicates whether the entity is gaining profit without pouring new equity capital into the business and is used by fund managers to estimate the level of growth that a business can achieve in the future (Do Hoai Linh, Ngo Thanh Xuan and Phung Quoc Anh 2020) However, an increased ROE may not solely be attributed to the efficient use of equity by the bank, it could result from the bank reducing the equity proportion and increasing the proportion of loan capital, which leads to liquidity risk When comparing ROE, it is essential to conduct horizontal and vertical comparisons, comparing with other enterprise in the same industry to get the most comprehensive view of the enterprises situation According to Moody’s financial strength assessment standards for the banking industry, an ROE ranging from 12% to 15% indicates good profitability Moreover, according to CAMEL, banks operate optimally when ROE is greater than or equal to 22% (Rozzani and Rahman 2013)

In addition, NIM can be also used as a metric to assess the bank’s profitability of a commercial bank

NIM is an intermediary that contributes to collecting savings and providing loans (Do Hoai Linh, Ngo Thanh Xuan and Phung Quoc Anh 2020) NIM illustrates how effectively a bank manages the performance of mobilization and lending by controlling profitable assets and finding sources of capital at low cost A high NIM can stem from a combination of low deposit interest rates and high-interest rates on loans This situation reduces the incentive for saving, increases the cost of borrowing for potential borrowers, and consequently leads to a decreased investment However, a low NIM should not be considered a good indicator This complexity has resulted in fewer studies incorporating the NIM rate than the other two above profitability indicators (Do Hoai Linh, Ngo Thanh Xuan and Phung Quoc Anh 2020) However, study by Salim and Bilal (2016), Saleh and Afifa (2020), Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan (2022), Le Ngoc Thuy Trang et al (2021) has still demonstrated that NIM remains a reliable indicator of profitability.

REVIEW OF RELEVANT THEORIES ABOUT THE IMPACT OF

Referring to relevant empirical studies of Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), and Le Ngoc Thuy Trang et al (2021) when researching the impact of liquidity risk on the profitability of commercial banks, it typically relies on two fundamental theories: the Market – Power theory and the Efficiency - Structure theory

The MP theory also has two main approaches: SCP (Structure - Conduct - Performance) and RMP (Relative market power) SCP theory was developed by Edward Mason and his doctoral student Joe S Bain (Lee 2007), this theory analyzes and provides a general overview of the relationship between market structure, market behavior, and market performance According to SCP, market structure determines the behavior of firms, which in turn determines their performance Thus, the environment dictates the existence and operations of businesses, with their primary mission being to respond and adapt Overall, banking markets become more concentrated, loan interest rates typically increase, while deposit interest rates tend to decrease An increase in loan interest rates can result in higher income for banks, while a decrease in deposit interest rates helps reduce the costs banks have to pay for customers Therefore, the spread between loan and deposit rates may increase, creating favorable conditions for banks to enhance profitability Furthermore, as the market becomes more concentrated, competition among banks decreases This can make it difficult for other entities to compete, as they may lack the opportunity or resources to compete with large banks that have a significant market share and influence This also increases the opportunity for large banks to control the market and increase profits This contributes to an improvement in bank profitability as competition among banks diminishes and it becomes difficult for other units to compete

Beyond SCP, there is also the RMP theory, a theory of relative market power This theory suggests that firms with a large market share and differentiated products can execute market power and earn non-competitive profits (Berger 1995) Banks with brand, product quality, and scale advantages may increase prices for their products and services, leading to higher profits To sum up, each aspect of the theory focuses on a different aspect of market power, with SCP concentrating on market structure and RMP focusing on market control through market share and unique product, service characteristics, and quality

Demsetz (1973) studied the ES theory, indicating that banks achieving profitability results from improved management efficiency Banks that can effectively manage their operations will have a competitive advantage over others Additionally, author Anyanwaokoro (1996) argued that profitability is crucial in persuading stakeholders, such as investors, borrowing customers, etc., to deposit money in the bank This ES theory has two main approaches: X-efficiency and Scale_efficiency Concerning X-efficiency, banks with high efficiency typically gain larger market shares, higher profits, and more customers because they can minimize production costs at any input level (Al-Muharrami and Matthews 2009) Regarding Scale_efficiency, large banks may have lower production costs through cost-cutting, resulting in higher profits due to economies of scale (Olweny and Shipho 2011) Moreover, large banks operate with lower costs than small ones because large organizations often operate in a more competitive environment thanks to economies of scale, technological efficiency, better financial management and risk management capabilities, etc

Several foundational theories have attempted to analyze the impact of liquidity risk on bank profitability This theory about the relationship between risk and return was first formalized by Hary Markowitz in 1952 According to this theory, all organizations have a trade-off between liquidity and profitability In the context of a bank, the two objectives of achieving good profits and maintaining liquidity cannot be followed simultaneously without one affecting the other (Akinwumi, Micheal and Raymond 2017) This means that banks should aim at maintaining an optimum level of liquidity to balance between the benefit of holding cash in the form of saving transaction costs to raise funds and the cost of holding cash in the form of tax disadvantage and liquidity premium (Edem 2017) The theory is relevant for the study as it examines the relationship between liquidity and bank profitability

According to the trade-off theory, banks can mitigate their risk by maintaining higher levels of liquidity This means they might not have to offer investors as much money to convince them to invest because there's less chance they'll lose it all (Osborne, Fuertes, and Milne, 2012) Moreover, the optimal liquidity level of banks tends to fluctuate throughout the business cycle, typically increasing during periods of higher expected financial distress Consequently, the relationship between liquidity and profitability can vary significantly, often showing more favorable outcomes during distress This suggests that banks aiming to bolster their liquidity positions may also see improvements in their profitability As a result, the relationship between liquidity and profitability in the short term can be either positive or negative, depending on the bank's current liquidity status relative to its optimal level In summary, a successful banker must balance these competing objectives by investing in a well-diversified portfolio mix

According to Osborne, Fuertes and Milne (2012), the author addresses the relationship between liquidity and profitability of banks in a banking industry crisis

In a distressed banking market, the risk of bankruptcy for banks increases Consequently, banks seek to maintain a higher level of liquidity to mitigate heightened risks Enhanced liquidity can assist banks in navigating through these challenging periods without resorting to selling assets at distressed prices or seeking costly supplementary capital However, under normal conditions, banks may not consistently achieve their optimal liquidity levels This could be due to an overly cautious approach leading to the retention of higher liquidity levels than necessary, potentially sub-optimizing profitability metrics Thus, in a normal environment, the relationship between liquidity and profitability may not be robust or may lack clarity Furthermore, reducing the liquidity asset structure of a bank can increase liquidity risk, but it may also augment profitability through the acceptance of higher levels of risk Therefore, according to this hypothesis, a positive correlation between risk and profitability may exist within the banking sector In summary, the crisis environment drives banks to bolster liquidity to minimize the risk of bankruptcy, whereas, under normal circumstances, the relationship between liquidity and profitability may be indeterminate Additionally, accepting higher levels of risk could be associated with increased profitability within the banking sector.

A REVIEW OF RELEVANT EMPIRICAL STUDIES

Salim and Bilal (2016) conducted a study to examine the relationship between liquidity risk and the performance of Omani banks and the relative impact of liquidity position on the financial performance of these banks with the eventual objective of advice policies in order to improve the management of liquidity risk in Omani banks The data of this study were collected from the annual reports of four Omani banks operating in the Sultanate of Oman, which are listed on the Muscat Securities Market (MSM) during the period of five years from 2010 to 2014 The study used the SPSS software and Multiple regresion analysis and ANOVA – Oneway Analysis to test the study hypothesis with the original model comprising independent variables which are namely L1 (Ability to absorb liquidity shocks), L2 (Ability to cope with a high demand of short term liquidity), L3 (Liquidity in the case of inability to borrow from other banks), L4 (Percentage of assets related to illiquid loans), L5 (Relation of illiquid assets and liquid liabilities), and L6 (Liquidity risk exposure) representing liquidity ratios variable and ROA, ROE, ROAA, and NIM which are represented for bank performance The study results are that with the ROA model, there are negative associations between L1 and L2 with ROA, and there is a positive relationship between L3, L4, L5, and L6 with ROA With the ROE model, there is a positive association between L3, L5, and L6 on the financial performance (ROE) of Omani Banks, while the results indicate that there are negative association between L1, L2, and L4 with ROE The ROAA model shows the positive relationship between L3, L4, L5, L6 and ROAA while the results indicate negative associations between L1 and L2 with ROAA Moreover, no variable in the model affects NIM

Muriithi and Waweru (2017) used two variables for liquidity risk measurement variables based on the Basel agreement The authors conducted a study to examine the impact of liquidity risk on the profitability of commercial banks in Kenya, based on data from audited financial statements and annual reports of commercial banks from 43 commercial banks in Kenya over the period 2005-2014 The study used Panel data techniques of REM and GMM after testing the model using the Wald test and F-Test The proposed model included two independent variables, including Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NSFR), and dependent variable was bank performance, which was measured by ROE After the analysis, the study found that NSFR negatively impacts ROE (both short-term and long-term), while LCR does not significantly impact on ROE This demonstrates that liquidity risk has a negative effect on bank performance

Charmler et al (2018) conducted a study to examine the level of bank liquidity, the trend of banks liquidity, and the impact of liquidity on profitability of commercial banks The data of this study was collected from annual reports of 23 Universal banks in Ghana from 2010 to 2016, with data arranged in the form of a panel Data were analyzed using descriptive statistics, correlation analysis, and regression analysis, the study is a quantitative study relying mainly on panel regression analysis The original model used profitability as the dependent variable of the model, including ROA and ROE Regarding independent variables, the study used two measures of liquidity for banks sampled for this study The first one is the ratio of liquid assets to total assets (LIQD1), which indicates the easily ability to convert bank assets to cash The second measure of liquidity is LIQD2, which measures the average number of times that liquid assets cover total interest-bearing liabilities Moreover, this study also used the NIM, capital adequacy ratio (CAR), foreign ownership (OWN), and bank size (SIZE) as the firm-specific variables The results of the study revealed that in the ROA model, there was a positive relationship between the two bank liquidity ratios However, LIQD1 also had a positive influence on ROE while LIQD2 had a negative impact on ROE, but both of them were statistically insignificant Furthermore, the findings on the control variables indicated that NIM, ETA, OWN, and SIZE are positively associated with bank profitability In summary, the study revealed a positive relationship between bank liquidity and profitability

Golubeva, Duljic and Keminen (2019) conducted a study on the impact of liquidity risk on profitability, applying the Basel III framework in the case of European countries The data were collected from annual reports of the banks, which were taken from the Thomson Reuters Eikon database of 45 European banks (the banks with the highest amounts of assets) with 180 observations from 2014 to 2017 and 37 observations in 2018 The analysis method was the OLS method The authors used variables such as ROA, ROE, net profit margin (NPM), and earnings before interest, taxes, depreciation, and amortization (EBTDA) representing bank profitability Additionally, the study included independent variables representing liquidity, such as loan cover ratio (LCR), loan to deposit ratio (LTD), and financing gap ratio (FGR) Control variables included deposits, equity, size, non-performing loan ratio (NPLL), and gross loans to deposits (GLS) Macroeconomic variables such as GDP and inflation were also considered in the study The study results revealed that LCR and deposits was required for further investigation because of insignificant contributors LTD had a positive relationship with EBITDA and ROE but no statistics with ROA and NPM models The authors demonstrated that FGR and ROA had a negative correlation and positive correlation with EBTDA Besides that, equity of the regressions model showed that equity had a negative relationship with ROE but a positive one on NPM and ROA Bank size and NPLL had a significant negative impact on bank profitability for the period 2014-2017 The impact of variable GLS is positive on EBTDA but negative on the other return ratio, ROE and NPM Regarding GDP, GDP growth had a negative relationship with ROE, and NPM ratio and it had no statistically significant impact on EBTDA and ROA ratios The final variable is inflation which was an insignificant contribution to bank profitability The study results revealed that the impact of liquidity risk on bank profitability is diverse, leading to the conclusion that specific relationships between liquidity risk and business performance could not be definitively determined

Alalade, Ogbebor and Akwe (2020) conducted a study examining the effect of liquidity risk on the financial performance of listed Deposit Money Banks in Nigeria The data of this study were secondary data, which were collected from published financial statements of fourteen listed banks on the Nigerian Stock Exchange for ten years, from 2009 to 2018 Data were analysed by using panel regression analysis with emphasis on pooled OLS, FEM, and REM models with the original model including the dependent variables are ROE, ROA, and EPS Moreover, the independent variables are net debt to total asset (NDTA), loan to deposit ratio (LTD) and liquid asset ratio (LAR) This study provides strong evidence supporting the existence of a positive correlation between LTD, LAR, and banking profitability in terms of ROA, ROE, and EPS Conversely, the NDTA ratio showed an insignificant inverse effect on bank profitability Hence, the importance of effective liquidity management in the banking sector cannot be overstated, especially for maximizing profit levels while maintaining liquidity The study concluded that proficient liquidity management can notably impact banks' EPS and positively affect their profitability and overall stability

Saleh and Afifa (2020) conducted a study to investigate the influence of credit risk, liquidity risk, and bank capital on the profitability of thirteen banks in the context of Jordanian commercial banks The study focused on data collected from the Amman Stock Exchange of thirteen commercial banks in Jordan during the financial crisis, covering the period from 2010 to 2018 because of the changes in the Basel regulations gathered in Basel III, which occurred in 2010, with a total of 117 observations The study used panel data and the GMM method as the analytical method The initial proposed model included profitability (dependent variable), which was measured by ROAA, ROAE, NIM, and six independent variables, including credit risk, liquidity risk, bank capital, bank size, loan growth, and efficiency (cost) After the analysis, the study results revealed that credit risk had a negative relationship with ROAA,

NIM, and no effect on ROEA model Liquidity risk had a negative effect on bank profitability when using ROAA and ROAE to examine The study demonstrated that bank capital positively affects on profitability (ROAA, ROAE, NIM) Bank size had a negative influence on ROAA and NIM no impact on ROAE The loan growth had a positive affect on ROAA, ROAE and no impact on NIM Furthermore, when using ROAA and NIM as a profitability variable, the cost had a negative influence on bank profitability In summary, credit risk, liquidity, and bank capital significantly impact bank profitability, with liquidity risk having a notable inverse effect on the bank’s financial performance

Hacini, Boulenfad and Dahou (2021) conducted study on the impact of liquidity risk management on the business performance of Saudi Arabian banks The data were collected from the annual financial reports of banks in Saudi Arabia during the period 2002-2019, and the analysis was performed using methods which are OLS, FEM, and REM The authors used the ROE as a variable to represent the business performance of banks, while the independent variables measuring liquidity risk included the loan to deposit ratio (LTD) and cash to deposit ratio (CTD) Additionally, the study also included a control variable named equity to total asset ratio (ETA) The results indicated that both LTD and CTD, as well as ETA, negatively impact the business performance of the banks This demonstrates that liquidity risk has an adverse effect on the business performance of the banks

Dong (2021) conducted a study to examine the impact of COVID-19 on the financial performance of U.S and Chinese banks in 2018-2020 The study used data from quarterly frequency and annual reports of 10 listed banks (5 US banks and 5 Chinese banks) through 2 countries for 12 quarters, which began from the first quarter of 2018 to the fourth quarter of 2020, with a total of 120 observations The study used the OLS method to test hypotheses in analyzing the impact of the COVID-19 pandemic on the financial performance of U.S and Chinese banks in multiple linear regression The initial model had ROA, the dependent variable to measure the profitability of U.S and Chinese banks The independent variables were the capital adequacy ratio (CAR), nonperforming loan ratio (NPL), loan to deposit ratio (LDR), and efficiency ratio (ER) Moreover, a moderation analysis was employed to examine whether the association between the dependent and independent variables varies based on the moderator CASES This analysis aims to investigate whether COVID-

19 influenced the profitability of US and Chinese banks, this variable received the value of 1 when referring to the COVID-19 pandemic period (2020) in both Chinese and US cases and 0 otherwise The US variable was also a dummy variable that was assigned a value of 1 when referring to banks located in the United States and 0 otherwise The results indicated that CAR positively correlated with ROA on US and Chinese banks during the COVID-19 pandemic The ER ratio significantly negatively impacted the bank performance (ROA) of Chinese and US banks during the crisis NPL and LDR had insignificant impacts on US and Chinese banks during the COVID-19 pandemic when using ROE, which represented bank performance

Ren (2022) conducted a study with to investigate the direction of the impact of credit risk and liquidity risk on the profitability of Chinese commercial banks during COVID-19 The study focused on financial data and financial indicators which were collected from the wind database and CSMAR and additional information from the China Banking Regulatory Commission and annual reports of 5 state-controlled commercial banks, 10 joint-stock commercial banks, and 17 city commercial banks, with a total of 32 listed commercial banks in China from 2009 to 2021 Moreover, macroeconomic data were collected from the official website of the World Bank The study used panel data regressions for all banks and three separate bank characteristics categories The study conducted descriptive statistics to determine numerous information, then used correlation analysis to check the existence of multicollinearity among the variables After that, the study determined the appropriate model by using the Hausman test, then selected a FEM model to analyze the results The study had two objectives The first one is to investigate the impact of credit risk and liquidity risk on bank profitability which the initial proposed model included profitability variable which was measured by ROA, ROE, and independent variables, including non-performing loan ratio (NPL), current ratio (CR) which representing credit risk and liquidity risk The study also had control variables, which were bank size (SIZE) and cost to income ratio (CIR), as well as control macro variables (GDP growth rate and inflation rate) COVID was a dummy variable as well as moderating variable) It depended on the model being used (1 if the period is during the COVID-19 pandemic (2020-2021), and 0 otherwise) The second one is to investigate the impact of COVID-19 on the correlation of these risks and bank profitability of listed commercial banks in China by using interaction terms (credit risk and COVID - NPLCOVID, liquidity risk and COVID - CRCOVID) After the analysis, the study results revealed that in the model without interaction terms, NPL and CR, which represented credit risk and liquidity risk, has a negative relationship on ROE The model with interaction term between risk and COVID-19 to examine the impact of COVID-19 on the relationship between credit risk, liquidity risk, and profitability, the results indicated that NPLCOVID had a negative coefficient, CRCOVID had a positive coefficient, proving a stronger positive relationship between liquidity risk and bank profitability in the presence of COVID-19 pandemic In summary, there was a positive correlation between liquidity risk and commercial banks' profitability in the context of COVID-19 In the context of credit risk, the opposite outcome is provided with liquidity risk

Ishari and Fernando (2023) conducted a study to examine the impact of the COVID-19 pandemic on the profitability of commercial banks in the South Asian region The study focused on quarterly financial statements of respective banks of all the licensed commercial banks in Sri Lanka, Bangladesh, and Pakistan from 2018 quarter 1 to 2021 quarter 3 (15 quarters) The study selected a sample consisting of

12 commercial banks from Sri Lanka and Bangladesh and 10 commercial banks from Pakistan The study was conducted to analyze the panel data by using pooled OLS, FEM, and REM methods After that, the study continuously determined the appropriate model by using the Hausman test, then selected the REM method to test the hypotheses and analyze the results of the Sri Lankan and Bangladesh banking sector While the study used the FEM model to test the hypotheses for the Pakistan banking sector The initial proposed model included profitability variables which were measured by ROA, ROE, and independent variables, including capital adequacy ratio (CAR), loan to deposit ratio (LDR), non-performning loan ratio (NPLR), and COVID (1 if the period is during the COVID-19 pandemic, and 0 otherwise) Control variables and moderating variables, are bank size and ownership (1 if the bank is a private sector bank and otherwise) After the analysis, the study results revealed that the CAR variable had a significant positive impact on the profitability (ROE) of Pakistan’s banking sector while an insignificant positive impact on the profitability of Sri Lanka and Bangladesh’s commercial banks The LDR variable had a significant positive impact on the profitability of Pakistan, and it had a negative impact on the profitability of Sri Lanka and Bangladesh The NPLR variable had a significant negative impact on the profitability (ROA) of Bangladesh’s banking sector and Sri Lankan banking sector and it had an insignificant positive impact on profitability of Pakistan banking sector The last moderating variable, called bank size (SIZE) had a negative impact on the profitability of Sri Lankan and a positive impact on profitability of the Sri Lankan and Pakistani banking sectors The COVID-19 pandemic had a significant impact on the profitability of commercial banks in the Sri Lanka and Pakistan banking sector Specifically, COVID (dummy variable) had a positive impact on Sri Lanka banking sector while a negative impact on the Pakistan banking sector However, the COVID pandemic had an insignificant impact on the profitability of commercial banks in Bangladesh Finally, ownership of banks (dummy variable) also demonstrated a notable positive influence on the financial performance of commercial banks in Pakistan and exhibited a moderating effect on the correlation between profitability and the period affected by COVID-19

2.5.2 Review of studies in Vietnam

Tang My Sang (2019) conducted a study to test the impact of liquidity management on the profitability of the commercial banking system in Vietnam The study utilized a dataset comprising 1026 observations from the annual financial reports of 19 commercial banks in Vietnam from 2010 to 2018 The entire dataset was extracted from financial reports available on the websites of these 16 banks, and from data on the websites www.cafef.vn and www.vietstock.vn As presented in the author’s study, the total assets of these 19 banks, as of 2017, account for over 86% of the whole banking system, making it representative of the whole Vietnamese banking system The study employed panel data, so the research model is estimated using Pooled OLS, FEM, REM, and GLS estimation techniques In which the GLS method was used to address the issue of autocorrelation The proposed model included two variables, which was ROA and ROE, to measure the efficiency of using assets and equity to generate profits The independent variables assessing liquidity included IR stands for the investment ratio (Loans/ Customer’s deposits), NCFTA standed for the lending ratio (Loans/ Total assets), CR standed for the capital adequacy ratio (CAR),

LR standed for the liquidity ratio (Short-term assets/ Total assets), and QR was quick ratio ((Short-term assets – Inventory)/ Total assets) The analysis revealed the following results For the ROA model, variables IR, CR, and QR had significant positive impacts, while LR and NCFTA did not significantly affected the profitability of Vietnamese commercial banks Among the factors with significant consequences,

RESEARCH GAP

Empirical studies about the impact of liquidity risk on bank profitability have been conducted both internationally and in Vietnam using quantitative research methods These studies vary in perspectives, databases, and methodological limitations, resulting in differences in the variables used in different research models After conducting a review of previous relevant empirical studies, the author identified several research gaps regarding the impact of liquidity risk on bank profitability:

Firstly, despite considerable attention, the relationship between liquidity risk and bank profitability remains inconclusive While some studies suggest a significant impact of liquidity risk on profitability, others argue that there is no significant impact Moreover, variations in the measured profitability indicators yield differing results among models even within the same study

Secondly, previous experimental studies have not extended their analysis to include data up to 2022 from listed commercial banks in Vietnam Specifically, for international studies, the research period extends until 2021, while for studies conducted in Vietnam, the research period extends until 2020

Thirdly, in Vietnam, particularly during the heavily COVID-19 affected period of 2020-2021, there is a lack of comprehensive research analyzing the impact of the COVID-19 pandemic on the relationship between liquidity risk and bank profitability Research is crucial to understanding the changes in this relationship influenced by the COVID-19 pandemic The author noted that the impact of COVID-19 has only been studied in foreign markets, as detailed in the literature review section For studies conducted in Vietnam, the author has not found research conducted on interaction variables between liquidity risk and the COVID-19 pandemic Instead, the author has only found studies conducted on interaction variables between liquidity risk and financial crises

In summary, the author recognizes the need for more detailed research to address these above research gaps identified in experimental studies related to the impact of liquidity risk on bank profitability Specifically, the author conducted a study focusing on 26 listed commercial banks in Vietnam from 2012 to 2022 using the SGMM method to enhance the robustness and accuracy of the research model and determine the impact of liquidity risk on bank profitability in a specific direction The author also considers the impact of the COVID-19 pandemic on this relationship during the period 2020-2021

Chapter 2 presents the theoretical framework regarding liquidity risk, bank profitability, and related theories on the impact of liquidity risk on the profitability of listed commercial banks in Vietnam This impact also indicates that there have been numerous empirical studies, both internationally and in Vietnam on the impact of liquidity risk on bank profitability Through these empirical evidences, despite employing various research methodologies, the scope of the study, etc., the findings of studies consistently demonstrate the impact of liquidity risk on profitability Overall, the results of relevant empirical studies are diverse and lack uniformity among different studies

These findings serve as the foundational basis for the following Chapter 3 of this study In the next chapter, the author will propose a research model corresponding to the issues addressed in the topic from the perspective of learning from previous relevant empirical studies on liquidity risk and its impact on bank profitability and extend to incorporate the involvement of COVID-19 pandemic although the limitations of relevant studies on this relationship in Vietnam.

RESEARCH MODEL AND METHODOLODY

IMPLEMENTATION PROCESS

To analyze the impact of liquidity risk on the profitability of 26 listed commercial banks in Vietnam during the period 2012 - 2022, the following study process was utilized:

The study process conducted by the author consists of the following steps, and the specific content of each step is presented in the following sequence:

Step 1 The author will conduct a comprehensive review of relevant theoretical frameworks about liquidity risk and bank profitability In addition, the author will review relevant empirical studies published internationally and in Vietnam about the impact of liquidity risk on bank profitability while also considering additional studies concerning the impact of the COVID-19 pandemic on this relationship to determine the research gaps Besides that, based on a theoretical foundation and a comprehensive review of relevant empirical studies, the author identifies the variables (independent, dependent, control, and interaction variables) for the research model

Step 2 From theoretical foundations and an overview of relevant empirical studies, The author explains and determines the measurement methods for each variable and constructs appropriate study hypotheses for the independent variables After that, the research constructs an appropriate research model regarding the impact of liquidity risk on bank profitability and the impact of COVID-19 pandemic on this relationship Besides that, the author also employs appropriate research methodology to analyze the impact of liquidity risk on bank profitability and the impact of the COVID-19 pandemic on this relationship in the next steps

Step 3 From the proposed research model, based on the constructed variables, the data are collected and processed The research data are secondary data obtained from the audited and published financial reports of 26 listed commercial banks in Vietnam from 2012 to 2022 from the FiinPro X- platform, and macroeconomic data are collected from the General Statistics Office of Vietnam and the World Bank From the collected data, the author processes the relevant data according to the measurable requirements of the variables After that, all data are processed using Stata 15 software

Regression analysis using the GMM method The author conducts regression analysis of the proposed model using the Generalized Method of Moments (GMM) estimation method Compared to other methods such as OLS, FEM, and REM, GMM helps to address issues of autocorrelation, varying error variances, endogeneity issues, and lagged variables of the dependent variable that other methods cannot handle, which could bias the results if used Additionally, GMM has two estimation forms, including Dif-GMM and Sys-GMM This study utilizes the Sys-GMM (SGMM) estimation

Step 4 To ensure the reliability of the research findings, the author conducts relevant tests such as autocorrelation test, multicollinearity test, heteroscedasticity test, and endogeneity test Moreover, the research also tests the appropriation of the SGMM model The author checks the reliability and validity of instrumental variables used in the model, with key tests including the Sargan test (also known as the Hansen test) and the Arellano-Bond AR(2) test

Step 5 The author will discuss the research results The study presents regression results on the impact of liquidity risk on bank profitability of listed commercial banks in Vietnam and the impact of the COVID-19 pandemic on this relationship, followed by a discussion and comparison of results with relevant empirical studies

Step 6 Following the discussion of the research results, the author infers the impact of liquidity risk on the profitability of listed commercial banks in Vietnam and concludes on the impact of the COVID-19 pandemic on this relationship Based on these findings, the author also provides recommendations for helping listed commercial banks in Vietnam effectively control liquidity risk to achieve expected profitability.

DEVELOPING RESEARCH HYPOTHESES AND PROPOSED

3.2.1 The variables in the model and the research hypotheses

ROE reflects the efficiency of the bank in managing shareholders' equity Additionally, the ratio indicates how much profit the bank will earn from shareholders' equity, demonstrating whether the bank operates effectively based on its own resources or not (Hacini, Boulenfad and Dahou 2021)

From the literature review perspective, the author utilizes ROE as the primary measure for the profitability of listed commercial banks in Vietnam This is because ROE is a crucial indicator widely used in numerous relevant empirical studies both in Vietnam and internationally to measure the profit-generating capability of banks in different regions, including international studies (Alalade, Ogbebor and Akwe 2020; Charmler et al 2018; Golubeva, Duljic and Keminen 2019; Hacini, Boulenfad and Dahou 2021; Ishari and Fernando 2023; Muriithi and Waweru 2017; Ren 2022; Saleh and Afifa 2020; Salim and Bilal 2016), studies in Vietnam (Le Ngoc Thuy Trang et al 2021; Nguyen Thanh Phong 2020; Tang My Sang 2019; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021; Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan 2022) Particularly, studies of Muriithi and Waweru (2017) and Hacini, Boulenfad and Dahou (2021) solely employed this indicator to measure the profitability of banks in their study The above authors measured this indicator by calculating net income over shareholder’s equity

The research of Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021) about the impact of liquidity risk on bank profitability indicated that lag of bank profitability has a positive impact on bank profitability This means that the bank profitability in this year will depend on its profitability in previous years

High profitability in the previous year will positively impact the bank's business activities Therefore, in my research, the author expects that the lag of bank profitability positively affects bank profitability

Hypothesis 1: There is a positive relationship between the lag of bank profitability and bank profitability

LTD (Loan to deposit) ratio

LTD is the ratio of loan to deposit This ratio indicates how much a bank has lent based on the total capital it has mobilized from customer deposits, or in other words, how much capital the bank has mobilized to serve its lending activities This ratio is crucial for assessing the safety level of banks A higher LTD ratio signifies that the bank has lent out more than it has available through customer deposits While this may increase the opportunity for higher loan profits, it also implies higher liquidity risk In situations when many customers demand withdrawals simultaneously, the bank may struggle to return funds to customers, thus facing liquidity risk Conversely, a low LTD ratio indicates that the bank has lent out a smaller portion compared to the total deposits it has collected This suggests that the bank has higher liquidity, with greater flexibility to return deposits to customers in various scenarios However, lower LTD ratios may result in reduced profitability opportunities as the bank is lending out less In summary, the lower the LTD ratio, the higher the liquidity, and the lower liquidity risk; therefore, bank profitability will also reduce According to Lee and Hsieh (2013), this ratio should ideally fluctuate between 70% and 90%, as exceeding this range may pose risks to bank operations There are several studies found a positive relationship between LTD ratio and bank profitability (Alalade, Ogbebor and Akwe 2020; Golubeva, Duljic and Keminen 2019; Ishari and Fernando 2023; Salim and Bilal 2016; Tang My Sang 2019) On the contrary, Hacini, Boulenfad and Dahou (2021), Ishari and Fernando (2023) when researching in Sri Lanka, Le Ngoc Thuy Trang et al (2021) and Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021) reported that the LTD ratio has an inverse relationship with bank profitability Additionally, research by

Dong (2021) and Tran Quoc Thinh, Le Xuan Thuy and Dang Anh Tuan (2022) suggested that this ratio and profitability have a statistically insignificant impact on each other In my research, the author expects that LTD ratio positively affects bank profitability

Hypothesis 2: There is a positive relationship between the loan to deposit ratio (LTD) and bank profitability

ETA (Equity to total assets)

The ETA ratio serves as a measure of a bank's financial strength within the market A higher equity ratio indicates reduced reliance on external funding sources, leading to lower capital costs when seeking financing This reduces cost pressures and increases profitability for the bank Moreover, elevated capital levels correspond to lower leverage and risk for the bank Abundant capital facilitates flexibility in adjusting policies and products in response to market dynamics, thereby enhancing operational efficiency Additionally, abundant capital resources enable banks to proactively mitigate unexpected liquidity risks, fostering stability in their operations This variable is used in studies by authors Charmler et al (2018), Golubeva, Duljic and Keminen (2019), Tang My Sang (2019), Saleh and Afifa (2020), Nguyen Thanh Phong (2020), Hacini, Boulenfad and Dahou (2021), Dong (2021), Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), Le Ngoc Thuy Trang et al (2021), Ishari and Fernando (2023) The research results of Charmler et al (2018), Saleh and Afifa (2020), Nguyen Thanh Phong (2020), Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), Le Ngoc Thuy Trang et al (2021) found a positive relationship Conversely, studies by Golubeva, Duljic and Keminen (2019), Tang My Sang (2019), Hacini, Boulenfad and Dahou (2021) found an inverse relationship between ETA and bank profitability The study of Ishari and Fernando (2023) indicated that ROE had both an insignificant and a positive impact on profitability, depending on each country that the author studied Therefore, in this study, the author expects ETA to impact bank profitability positively

Hypothesis 3: There is a positive relationship between equity to total assets (ETA) and bank profitability

Bank size is calculated by taking the natural logarithm of total assets and is considered one of the characteristic variables of banks Through the market power theory presented above, banks with larger size are expected to provide higher profitability by potentially reducing costs Bank size has been used in various studies by numerous studies of authors Charmler et al (2018), Golubeva, Duljic and Keminen (2019), Saleh and Afifa (2020), and Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), Le Ngoc Thuy Trang et al (2021), Ren (2022), and Ishari and Fernando (2023) The studies by Charmler et al (2018), Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), Le Ngoc Thuy Trang et al (2021), and Ishari and Fernando (2023) have supported this argument However, Golubeva, Duljic and Keminen (2019) and Ishari and Fernando (2023) found an inverse relationship between bank size and profitability Saleh and Afifa (2020) found different results from the aforementioned studies, suggesting that bank size did not significantly impact bank profitability Therefore, in this study, the author expects bank size to affect bank profitability positively

Hypothesis 4: There is a positive relationship between bank size (SIZE) and bank profitability

This is a relative index defining the total value added that an economy can produce within a certain period, and this ratio can be used to compare the scale of economies and the growth rates among countries GDP growth will impact bank profitability Increasing this ratio is a positive signal, indicating increasing borrowing demand because during periods of growth, banks tend to lower interest rates to attract higher borrowing demand, thus increasing bank profitability GDP growth will positively impact on bank profitability (Nguyen Thanh Phong 2020; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021) However, the study of

Golubeva, Duljic and Keminen (2019) found an inverse correlation between GDP growth and bank profitability Le Ngoc Thuy Trang et al (2021) found different results from the aforementioned studies, suggesting that GDP did not significantly impact bank profitability Therefore, in this study, the author expects that GDP growth rate has a positive relationship with bank profitability

Hypothesis 5: There is a positive relationship between GDP growth (GDP) and bank profitability

The inflation rate (INF) is used to measure the level of price changes for goods and services and provides information about the inflation or deflation status in the economy This rate is calculated by the change in the Consumer Price Index (CPI) for each country for each year The inflation rate directly and indirectly impacts bank profitability (Ferrouhi 2014) Directly, banks have to bear costs for inputs such as labor and equipment when prices rise Inflation occurs, and prices of raw materials, services, and labor may increase, leading banks to incur higher expenses This tends to reduce bank profitability Indirectly, through changes in interest rates, exchange rates, etc, as inflation rises, monetary authorities seeking to curb inflation may raise interest rates, tighten monetary policies to reduce inflation Higher interest rates may reduce customer borrowing demand, thereby affecting bank lending activities and bank profitability The study by Golubeva, Duljic and Keminen (2019) suggests that the inflation rate has no significant impact on bank profitability Nguyen Thanh Phong (2020) has not found a correlation between the two variables On the other hand, the study by Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021) indicates an inverse relationship between the inflation rate and profitability or Le Ngoc Thuy Trang et al (2021) found the positive relationship between INF and bank profitability In summary, in this study, the author expects that the inflation rate has an inverse relationship with bank profitability

Hypothesis 6: There is a negative relationship between inflation (INF) and bank profitability

The impact of COVID-19 on the relationship between liquidity risk and bank profitability (LTDCOVID)

Despite being only two years old, COVID-19 has already shown its significant economic impact, affecting nearly all sectors, including banking operations According to Ozili and Arun (2023), the pandemic hampers economic activity from two perspectives Firstly, by closing financial markets, jobs, and activities; secondly, by instilling caution among consumers and investors due to uncertainty about the outbreak's progression and duration Therefore, maintaining an appropriate liquidity ratio in banking operations becomes crucial during an epidemic Liquidity crises compel banks to borrow at higher rates from the market, leading to decreased earnings Moreover, COVID-19 markedly influences banks' savings rates (a primary source of mobilization) and customers' loan repayment punctuality According to the study by Ren (2022) that the COVID-19 boosted the positive impact of liquidity risk on bank profitability On the contrary, the study of Dong (2021) indicated that COVID-19 had negative impacts on the profitability of commercial banks in their study country The study of Ishari and Fernando (2023) found the relationship between COVID-19 and the profitability of commercial banks is diverse, depending on each research’s country, which will provide different results (details of the results can be found in the Table 2.1) In my study, the author expects that COVID-19 will positively impact on the relationship between liquidity risk (LTDCOVID) and bank profitability

Hypothesis 7: COVID-19 has positive impact on the relationship liquidity risk and bank profitability

Figure 3.2 Synthesize the research hypotheses within the model

Firstly, in the first model, the study builds upon the approach of authors Salim and Bilal (2016), Hacini, Boulenfad and Dahou (2021) and Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021) while also making adjustments to better suit the operational context of listed commercial banks in Vietnam The authors design a research model to investigate the impact of liquidity risk on bank profitability The proposed multivariate regression model 3.1 is as follows:

ROE i,t = 𝜷 0 + 𝜷 1 ROE i,t-1 + 𝜷 2 LTD i,t +𝜷 3 ETA i,t + 𝜷 4 SIZE i,t + 𝜷 5 GDP ,t + 𝜷 6

(3.1) Secondly, in the second model, the study analyzes the influence of the COVID-19 pandemic on the correlation between liquidity risk and profitability of listed commercial banks in Vietnam With this model, the study will refer to the model used by Ren (2022) in his research when using the interaction variables for the relationship between credit risk, profitability (NPLCOVID), and liquidity risk, profitability (CRCOVID) The study of Dong (2021) also used interaction variables about COVID-19 with independent variables representing liquidity risk named LDR to consider whether the COVID-19 pandemic affected the profitability of US and Chinese banks The author also recognizes that the study of Ishari and Fernando (2023) also investigated the impact of the COVID-19 on the profitability of commercial banks in the South Asian region with the interaction variables between bank size and the COVID-19 pandemic Therefore, in my thesis, the author decides to add an interaction variable (LTDCOVID) for liquidity risk and COVID-19 to model 3.1 and the research will analyze the coefficients of the interaction variable to assess the influence of this pandemic Consequently, the study establishes the model 3.2 presented below:

ROE i,t = 𝜷 0 + 𝜷 1 ROE i,t-1 + 𝜷 2 LTD i,t + 𝜷 3 LTD i,t * COVID + 𝜷 4 ETA i,t + 𝜷 5

The model uses one indicator (ROEi,t) to measure the listed commercial bank’s profitability as an explanatory variable Independent variables include the lagged dependent variable (ROEi,t-1 ), which indicates that bank profitability impacts each other over time Liquidity risk is measured by loan to deposit ratio - LTDi,t (Loan/ Deposit) Bank-specific control variables include the ratio of total equity to total assets - ETAi,t (total equity/total assets), bank size - SIZEit (natural logarithm of total assets) The study also uses two macro-control variables include GDPit (GDP growth rate) and INFit (Inflation rate) Besides that, the dummy variable in the study is COVID which represents the COVID-19 pandemic to asses the differrential impact of liquidity risk on bank profitability in the case of the economy being affected by COVID-19 in the years 2020 and 2021 The author gives the value of 1 to the period (2020 – 2021), and 0 otherwise In model 3.2, LTDi,t*COVID is the interaction variable between loan to total deposit ratio and COVID (LTD is a variable of liquidity risk) With i (bank), t (year), u (error).

THE RESEARCH DATA

3.3.1 Description of data collection source

The research data are selected through a data screening method, specifically filtering the scope of the space to be listed commercial banks in Vietnam Therefore, other financial institutions were excluded Additionally, the author's research data consists of balance sheet data, so listed commercial banks in Vietnam without sufficient data within the author's study timeframe were excluded Specifically, the author excluded one listed commercial bank in Vietnam, Vietnam Thuong Tin Commercial Joint Stock Bank – Vietbank, due to insufficient data to meet the author's study requirements within the study period Moreover, the author used descriptive statistics to fully understand the variables, including mean, maximum, minimum values, and standard deviation These values were calculated and presented using Stata 15 software

The data source for the dependent variable (bank profitability, which is measured by ROE) and independent variables belonging to commercial banks (LTD, ETA, SIZE) will be collected from the financial reports of listed banks in Vietnam that have been audited and fully disclosed on the FiinPro-X platform The entire dataset was re-verified by the author through the official websites of the banks to ensure the accuracy of the study The data source for the remaining variables (macroeconomic variables: GDP, INF) was collected by the author from the General Statistics Office of Vietnam (https://www.gso.gov.vn/) and the World Bank (https://www.worldbank.org/vi/country/vietnam)

3.3.2 Characteristics of the research sample

The research sample in this study consists of 26 listed commercial banks in Vietnam over an 11-year study period from 2012 to 2022, totaling 286 observations The author selected this study sample because these commercial banks provide comprehensive information about the data conditions and are suitable for the study topic conducted by the author from 2012 to 2022

3.3.3 Descriptive statistics of research sample

This study conducts a descriptive statistical analysis to provide a comprehensive overview of the various changes in the sample data, thereby offering the researcher and readers a general view of the research data and the reflective measures of the study objective The author focuses on describing the statistical measures such as mean, standard deviation, minimum value, and maximum value of balanced panel data of 286 observations of the 26 listed commercial banks in Vietnam from 2012 to 2022 (11 years), based on the audited financial report data from banks from FiinPro-X platform and macroeconomic data from the General Statistics Office of Vietnam The study findings are presented in Table 3.2

Table 3.1 Statistical results of variables used in the research model

Variable Obs Mean Std.Dev Min Max

Source: Analysis results from STATA 15.0 software

Regarding the dependent variable representing the profitability of banks, represented by the ROE ratio, the descriptive statistics show an average ROE value of approximately 10.78%, with a standard deviation of 7.63%, suggesting that the average ROE value ranges from 3.15% to 18.41% The minimum ROE value is 0.00016% (National Citizen Commercial Joint Stock Bank – NVB in 2022), and the maximum value is 3.033% (Vietnam International Commercial Joint Stock Bank – VIB in 2021)

Regarding Figure 1 in Appendix 7, from 2012 to 2015, the ROE values of the banks showed a downward trend, which can be attributed to the post-economic crisis recovery period, with a slowdown in the economy's demand and weak business prospects, causing a decline in the economy's absorption capacity Additionally, the financial situation of the businesses was weak, leading to a reduction in the value of collateral assets, making banks more cautious in lending to minimize risk However, from 2016 onwards, the profitability of banks started to recover, with higher economic growth and stronger lending activity, contributing to a reduction in the difficulties faced by the businesses Although there was a slight slowdown in 2019-

2020, one of the typical reasons is that banks are affected by the complex COVID-19 pandemic The profitability of banks started to rebound in 2021 and 2022, with a favorable lending environment In summary, this variable proves that listed commercial banks in Vietnam gradually uses equity more efficiently, leading to improved and enhanced bank profitability

The variable LTD representing the LTD ratio of 26 listed commercial banks in Vietnam has an average value of approximately 88.68% (This ratio falls within the recommended range set forth by Lee and Hsieh (2013)), indicating that banks have used deposits to fund approximately 88.68% of their customer loans The minimum value is 37.19% (The Maritime Commercial Joint Stock Bank - MSB in 2014), and the maximum is 146.91% (Vietnam Prosperity Joint Stock Commercial Bank - VPB in 2021) This LTD variable has a standard deviation of 17.07%, suggesting that the average LTD value ranges from 71.61% to 105.75% Circular 22/2019/TT-NHNN stipulates the limits and safety ratios for banking operations, including branches of foreign banks, effective from January 1, 2020, the maximum LTD is set at 85% This is a change from the previous regulation outlined in Circular 36/2014/TT-NHNN, where the maximum LTD for commercial banks was 90%, for joint-stock commercial banks, joint venture banks, and banks with 100% foreign-owned capital, the maximum LTD is 80% Analysis reveals that several listed commercial banks in Vietnam, such as BID, CTG, OCB, VIB, and VPB, tend to concentrate their capital on lending activities This strategic focus is driven by the fact that the primary source of income for these banks derives from lending operations, as evidenced by their LTD ratios consistently exceeding 1 (100%) However, despite this trend, several other banks have yet to leverage their capital to maximize profitability fully According to Figure 2 in Appendix 7, the LTD ratio demonstrates a decreasing trend before 2014, followed by a gradual increase in subsequent years This upward trend can be attributed to proactive capital restructuring initiatives undertaken by these listed banks aimed at maintaining the LTD ratio following SBV's signals for tighter credit controls Notably, in 2021, the ratio experienced significant growth, rebounding strongly after a slight decline in 2020 This resurgence underscores the increasing strength of banks' lending activities, indicating robust development in this activity

Regarding the control variables for the banks, the ETA variable represents equity to total assets ratio, with an average value of 8.99% and a standard deviation of 3.52%, suggesting that the average ETA value ranges from 5.47% to 12.51% The minimum value is 4.06%, corresponding to Joint Stock Commercial Bank for Investment and Development of Vietnam - BID in 2017, and the maximum value is 23.84%, corresponding to Saigon Bank for Industry and Trade - SGB in 2013 Looking at Figure 3 in Appendix 7, a decreasing trend of ETA can be observed from

2012 to 2017, and since 2018, listed commercial banks in Vietnam have shown stable growth Specifically, the ETA ratio decreased from 11.83% in 2012 to 7.69% in 2017 Despite many listed commercial banks in Vietnam organizing capital increases to meet competitive conditions from 2012 to 2017, the increasing in equity was still smaller than the increase in total assets (equity to total assets ratio) From 2018 to

2022, the ETA ratio showed a stable increase among listed commercial banks in Vietnam This is a result of capital adjustment to maintain capital adequacy according to SBV regulations when implementing the Basel Accord in Vietnam (Circulars 13/2010/TT-NHNN, 19/2010/TT-NHNN, and 36/2014/TT-NHNN setting the capital adequacy ratio at 9%) Additionally, to enhance capital sources to meet the demands of integration and long-term financial market development, the SBV has implemented various policies to support commercial banks in balancing their funds This is achieved through compliance with general regulations and the developing of strategies for increasing equity to improve ETA

Bank size (SIZE) with an average value of 20.1109, the lowest is 16.5023 (Saigon Bank for Industry and Trade - SGB in 2013), and the highest is 410.5826 (Vietnam Prosperity Joint Stock Commercial Bank - VPB in 2022) This variable has a standard deviation of 11.44%, suggesting that the average SIZE value ranges from 3.6086 to 36.6132

Regarding the macroeconomic control variable (Figure 4 in Appendix 7), the Vietnam GDP growth rate has an average value of 5.87%, with the highest being 8.82% in 2022 and the lowest being 2.58% in 2021 This GDP variable has a standard deviation of 5.87%, suggesting that the average GDP value ranges from 4.13% to 7.61% The GDP growth rate showed an overall increasing trend from 2012 to 2019, with a high growth rate of over 6% in 2015-2019 This indicates a stable development of the Vietnam’s economy and demonstrates the efforts to drive our country's economy during the international integration process, addressing challenging issues effectively to create favorable conditions for business and production Despite the low GDP growth in 2020 and 2021 due to the negative impact of the COVID-19 pandemic, leading to prolonged social distancing measures in major economic centers This led to a decline in GDP In 2022, thanks to the government's economic recovery policy, such as low-interest loan support packages which assisted the recovery of businesses facing bankruptcy due to COVID-19 and created conditions for the emergence of new enterprises, Vietnam’s economy is poised for a gradual rebound, leading to an increase in the GDP ratio to 8.82%, after experiencing the COVID-19 pandemic This demonstrates the resilience and effective utilization of opportunities to drive the economy with Vietnam's high level of effort

According to Figure 4 in Appendix 7, the inflation rate (INF) has an average value is 3.75% with a standard deviation of 2.23%, it means it will fluctuate from 1.52% to 5.98% The lowest value of 0.63% is among the 26 listed commercial banks in Vietnam within the research period in 2015, and the highest value of 9.21% occurred in 2012 The main reason for the high inflation rate in 2012 was the impact of the global economic crisis before 2012, with issues such as slow growth, imbalance, and constant volatility Vietnam’s economy, which was linked to the low growth rate and high inflation rate (To Ngoc Hung and Nguyen Duc Trung 2011), particularly in 2007-2009, experienced a sharp increase in inflation, with a 12-month inflation rate of 2.9% in December 2007 In the last six months of 2008, inflation showed significant signs but remained high, and by the end of 2009, inflation showed a downward trend In 2010-2011, after the decline in 2009, inflation continued to increase in September 2010, leading to economic challenges in 2011, with high commodity prices in the world market and high domestic prices of most goods According to Nguyen Tri Khiem and Nguyen Cong Khanh (2022), up to 2012, the government tightened credit for real estate and some non-manufacturing sectors, aiming to control inflation and stabilize the macroeconomy as the top priority From

2012 to 2015, Vietnam's inflation rate decreased, and inflation dropped significantly, reaching its lowest point of 0,63% in 2015 From 2016 to 2020, the inflation rate fluctuated between 2.66% and 3.23%, decreasing to 1.84% in 2021 and increasing to 3.15% in 2022.

RESEARCH METHODOLOGY

From the proposed research model above, the author identifies several technical issues that, if not addressed, could lead to ineffective estimations The first issue is that the research model may encounter endogeneity when introducing interaction variables The second issue is the presence of an independent variable as the lag of the dependent variable (bank profitability), as there may be suspicion that bank profitability is correlated with its lag, leading to an increase in standard errors and a bias in estimating the lag coefficients of the dependent variable (Nickell 1981) Finally, the dataset comprises 26 listed commercial banks in Vietnam over a short period, and economic cycle effects may persist for more than one year (data inertia) Thus, the model is prone to experiencing strong correlation in errors

Given these model issues, conducting model diagnostics using methods such as OLS, FEM, and REM would lead to biased estimates and incorrect results The GMM is the most effective method to address these issues GMM is a method that helps address endogeneity and other issues in panel data (Arellano and Bond 1991) GMM has two estimation forms: Dif-GMM and Sys-GMM This study employs the SGMM method because it improves of the Arellano and Bond (1991) method, incorporating additional assumptions

The SGMM addresses the endogeneity issue of some explanatory variables through a weighting matrix of instrumental variables The efficiency of SGMM estimation depends on the appropriateness of instrumental variables The author employs two testing techniques proposed by Arellano and Bond (1991) to tackle this issue First, the Sargan test, also known as the Hansen test for overidentification, allows for testing the validity of instrumental variables This test determines whether there is a correlation between instrumental variables and residuals in the model Theoretically, the Hansen test in the two-step estimation is considered more effective than the Sargan test in the one-step estimation (Roodman 2009b) Another important test in dynamic panel data is the AR(2) test for second-order autocorrelation of residuals in the model

Multicollinearity occurs when independent variables in a linear regression model are highly correlated with each other The strong linear correlation among these independent variables can increase the standard errors of regression coefficients and decrease their statistical significance, rendering the regression coefficients unreliable and affecting the study results To identify and conclude on multicollinearity, we can use the variance inflation factor (VIF) When a variable has a VIF greater than a certain threshold (VIF > 10), it indicates that the variable may cause multicollinearity According to Gujarati (2022), the VIF values for the variables in the model range from 1.10 to 1.85, indicating the absence of multicollinearity To mitigate this issue, the variable can be removed from the model to minimize the impact of multicollinearity

Autocorrelation is a phenomenon where the error term at time t is related to the error term at time t–1 or at any other point in the past When a model exhibits autocorrelation, it leads to an artificially high R-squared estimate compared to reality, and the estimates become biased, affecting the regression model The Wooldridge test is conducted with the following hypotheses, including H0: The model does not have autocorrelation, and H1: The model has autocorrelation When the Prob < 5%, it allows rejecting the hypothesis H0, concluding that the model exhibits autocorrelation (Wooldridge 2002)

If the phenomenon of heteroscedasticity occurs in the regression model, the estimates obtained through the ordinary least squares regression method will become inaccurate, and the regression coefficient tests will no longer be reliable This outcome may lead to the misconception that the independent variables in the study model are significant In this case, both the regression coefficient tests and R-squared lose their validity for use With the Wooldridge test, the hypotheses are as follows: H0: There is no heteroscedasticity in the model H1: There is heteroscedasticity in the model When Prob < 5%, it allows rejecting the hypothesis H0, concluding that the model exhibits heteroscedasticity (Greene 2000)

Wooldridge (2002) defines endogenous variables as those that are determined within the model, while factors outside of the model determine exogenous variables Arellano and Bond (1991) proposed that efficiency can be enhanced by using lagged values of the dependent variable and lagged values of exogenous regressors as instrumental variables The Hausman test is used to select exogenous variables as instrumental variables with appropriate lagged values Furthermore, assuming no autocorrelation, lagged dependent variables are deemed appropriate as instrumental variables

There are two hypotheses: H0, which is the exogenous variable, and H1, which is the endogenous variable When P-value < 5%, it allows rejecting the hypothesis H0, concluding that the variable is an endogenous variable On the contrary, if hypothesis H0 is accepted, concluding that the variable is an exogenous

3.4.3 GMM method (Generalized Method of Moments)

This study uses SGMM, which is the method introduced by Arellano and Bover (1995) and Blundell and Bond (1998) to test the the relationship between liquidity risk and bank profitability of listed commercial banks in Vietnam SGMM approach has five advantages in comparison to other panel models (OLS, FEM, REM, dif-GMM)

Firstly, the conventional models of panel analysis, including pooled OLS regression, REM model, and FEM model, produce parameter estimates that suffer from bias and inconsistency because of the presence of lagged-dependent variables (lagged bank profitability) or potential endogeneity issues arising from explanatory variables (Harris and Matyas 2004; Nickell 1981) The pooled OLS regression model tends to overestimate the coefficient on the lagged dependent variable due to its positive association with the effects in dynamic panel models (Hsiao 2014), whereas the FEM estimator tends to underestimate it because of unobserved bank-specific effects (Nickell 1981) Consequently, the SGMM method aims to address the bias inherent in standard panel models, such as pooled OLS and FEM models

Secondly, the SGMM estimation provides efficient and reliable estimates even in cases where explanatory variables are not strictly exogenous and when there is heteroskedasticity and autocorrelation within individual observations

Thirdly, the SGMM technique addresses issues of endogeneity and fixed effects, and eliminates dynamic panel bias (Nickell 1981) The problem of endogeneity is tackled by employing instruments such as lagged-dependent variables and endogenous variables with their respective lags, including lagged two, and so forth This approach is commonly known as the first-difference GMM method The dif-GMM model is further classified into one-step and two-step GMM estimators However, the estimates produced by the difference GMM estimator may suffer from bias and inconsistency due to the omission of potential information that likely captures the relationship between levels and first differences (Ahn and Schmidt 1995)

Fourthly, the SGMM estimation demonstrates greater efficiency compared to the dif-GMM estimation due to its utilization of a system that combines regressions involving both levels and first differences Consequently, it is referred to as the SGMM estimator The SGMM approach permits incorporating of a larger set of instruments by employing lagged first differences of variables in the original equation at the level, while lagged levels of variables are utilized as instruments in the first differenced equation Additionally, first differences are assumed to serve as instruments but remain uncorrelated with the fixed effects (Roodman 2009b) Furthermore, SGMM estimation yields superior outcomes when dealing with unbalanced panel data compared to dif-GMM, as dif-GMM tends to accentuate discrepancies (Roodman 2009a)

Finally, the SGMM estimator is more efficient when the number of time periods is small as well as the persistence in dependent variable highly correlates with the autoregressive term which is close to unity (Blundell and Bond 1998)

This study includes 26 listed commercial banks in Vietnam over 11 years from

2012 to 2022 Additionally, the study model incorporates lagged variables of the dependent variable (bank profitability - ROE) and contains endogenous variables by introducing instrumental variables into the model Therefore, the author opts for the SGMM method to estimate the model effectively and accurately, aiming to examine the impact of liquidity risk on the profitability of listed commercial banks in Vietnam

3.4.4 Testing the appropriation of the GMM method

Based on the dynamic models initially chosen, the study conducts regressions utilizing the SGMM estimator to address endogeneity issues (Arellano and Bond 1991; Arellano and Bover 1995) To ensure the trustworthiness of the GMM estimator, the study performs the Arellano-Bond test to check for residual autocorrelation and the Hansen test to assess the appropriateness of instrumental variables

RESEARCH RESULTS AND DISCUSSION

CORRELATION MATRIX ANALYSIS

The purpose of analyzing the correlation between variables is to examine the relationship between independent and dependent variables The results obtained from the correlation matrix analysis serve as a basis for assessing the model and considering the phenomenon of multiple linear regression where independent variables are related to each other

Table 4.1 Correlation coefficients between variables of research model

ROE LTD ETA SIZE GDP INF

Source: Analysis results from STATA 15.0 software

According to Gujarati (2004), if the correlation coefficient between independent variables exceeds 0.8, there is a possibility of multicollinearity in the regression model Table 4.1 above shows that, in general, the correlation coefficients between independent variables are quite low Table 4.1 above shows that most of the correlation coefficients between variables are less than 0.8, indicating that there is no strong correlation between variables in the regression model Therefore, the author can fully use multiple independent variables simultaneously to explain the impact of liquidity risk on the profitability of listed commercial banks in Vietnam without causing multicollinearity In this model, LTD, SIZE are variables that are positively related to the dependent variable ROE On the other hand, variables such as ETA, GDP, and INF are negatively related to the dependent variable ROE

However, multicollinearity is not dependent on a high or low correlation coefficient but rather on the consequences of multicollinearity The impact of multicollinearity can cause the regression coefficient to change signs and lead to inaccurate results in the regression model Therefore, the author will continue test for multicollinearity in the model through the variance inflation factor (VIF) coefficient

Table 4.2 Multicollinearity test of model 3.1 and model 3.2

Variable VIF 1/ VIF VIF 1/ VIF

Source: Analysis results from STATA 15.0 software

According to the discussion in Chapter 3, if the VIF coefficient of independent variables is less than 10, it indicates that the model does not have multicollinearity

In this research model 3.1, the VIF values range from 1.01 to 1.13 (all of which are less than 10) This suggests that there is no multicollinearity in the model Therefore, all independent variables can be used to explain the impact of liquidity risk on the profitability of listed commercial banks in Vietnam without causing multicollinearity in model 3.1 Similarly, when implementing model 3.2, the VIF of all variables in model 3.2 is smaller than 10, indicating that the model does not suffer from multicollinearity

Thus, for the study conducts on listed commercial banks in Vietnam, both model 3.1 and model 3.2 exhibit no multicollinearity In other words, the independent variables can explain the impact of liquidity risk on bank profitability without causing multicollinearity issues in the model.

REGRESSION DEFECT TESTING OF RESEARCH MODEL

The Wooldridge test can be used to test for autocorrelation This test examines two following hypotheses, including H0: The model does not have autocorrelation, and H1: The model has autocorrelation

If the Prob of the Wooldridge test is less than the significance level (5%), then the null hypothesis is rejected, and it is concluded that the model has autocorrelation

If the Prob of the Wooldridge test is greater than the significance level (5%), then the null hypothesis is not rejected and it is concluded that the model does not have autocorrelation

Table 4.3 Wooldridge test of model 3.1 and model 3.2

Prob > F = 0.0000 The research model has autocorrelation

Prob > F = 0.0000 The research model has autocorrelation

Source: Analysis results from STATA 15.0 software

From Table 4.4, with a 5% significance level, the autocorrelation test result of Wooldridge test shows that Prob > F = 0.0000 < 5%, so the null hypothesis (H0) is rejected and it is concluded that the model 3.1 exhibits autocorrelation Similarly, when implementing model 3.2, the autocorrelation test result of the Wooldridge test shows that Prob > F = 0.0000 < 5%, so the null hypothesis (H0) is also rejected like model 3.1 and it is concluded that model 3.2 has autocorrelation In summary, both model 3.1 and model 3.2 have autocorrelation

To check for heteroscedasticity, the White test can be used This test checks the following hypotheses that H0: The model does not have heteroscedasticity and H1: The model has heteroscedasticity If the Prob of the test is less than 0.05, then the author rejects the null hypothesis (H0) and concludes that the model has heteroscedasticity

Table 4.4 Modified Wald test of model 3.1 and model 3.2

Prob>chi2 = 0.0006 The research model has heteroscedasticity

Prob>chi2 = 0.0053 The research model has heteroscedasticity

Source: Analysis results from STATA 15.0 software

The result of the test shows that Prob of the model 3.1 is 0.0006, which is less than the 5% significance level (Table 4.3) Therefore, the author rejects the null hypothesis (H0), meaning the research model has heteroscedasticity Similarly, when implementing model 3.2, the Modified Wald test’s result shows that Prob > chi2 0.0053 < 5%, so the null hypothesis (H0) is also rejected like model 3.1, and it is concluded that model 3.2 has heteroscedasticity In summary, the conclusion is that both model 3.1 and model 3.2, which use the ROE variable, have the problem of heteroscedasticity

4.2.3 Endogenous and exogenous variables testing

The author conducts to test the phenomenon of endogenous variables with the hypothesis H0: Endogenous variables and H1: Exogenous variables

Table 4.5 Endogenous and exogenous variables in model 3.1

Variables Wu-Hausman Endogenous variables

Source: Analysis results from STATA 15.0 software

Table 4.6 Endogenous and exogenous variables in model 3.2

Variables Wu-Hausman Endogenous variables

Source: Analysis results from STATA 15.0 software

From Table 4.5, it can be seen that endogenous variables include equity to total assets (ETA), bank size (SIZE), GDP growth date (GDP), and inflation rate (INF) Exogenous variables include the variables loan to deposit (LTD) Besides that, from

Table 4.6, it can be seen that endogenous variables include equity to total assets (ETA), bank size (SIZE), GDP growth date (GDP), and inflation rate (INF) Exogenous variables include loan to deposit (LTD) and interaction variable (LTDCOVID).

THE RESULT OF THE SGMM REGRESSION ANALYSIS

Addressing the shortcomings of the model, this thesis employs the SGMM method to estimate the regression model Notably, a strength of the SGMM method lies in its ability to address the issue of endogeneity within the model

4.3.1 The impact liquidity risk on profitability of listed commercial banks in Vietnam

Table 4.7 The impact liquidity risk on profitability of listed commercial banks in Vietnam

Variables Coef Std Err P-value

Arellano-Bond test for AR(1)

Arellano-Bond test for AR(2)

Note: The symbols (***), (**), (*) indicate the level of statistical significance respectively 1%, 5%, 10%

Source: Analysis results from STATA 15.0 software

From the results in Table 4.7, the Sargan test in model 3.1 is used to check the over-identifying properties of the instrumental variables, and the Prob is 0.465, which is greater than 5% Therefore, the hypothesis (H0) is accepted that the instrumental variables are exogenous This indicates that the instrument variables, which are the lag of bank profitability, satisfy the over-identifying condition, and instrument variables are exogenous variables, and model 3.1 does not have an endogenous phenomenon Besides that, the number of instruments in model 3.1 is smaller than the number of groups (25 < 26) The Arellano-Bond test for AR(2) test result also showed that the Prob was 0.218, which is greater than 5%, hypothesis H0 is accepted that the model does not have a series autocorrelation problem

Therefore, all the instrumental variables used in the SGMM model meet the above criteria By including the lagged value of the dependent variable (Bank profitability – ROE L1.) as an instrumental variable, the autocorrelation problem in the model was addressed, resulting in a more robust and valid model Consequently, the SGMM regression method is used as a conclusion in this research to examine the impact of liquidity risk on the profitability of listed commercial banks in Vietnam in the period 2012 - 2022, as shown in the following equation according to the proposed research model 3.1:

4.3.2 The impact of COVID-19 on the relationship between liquidity risk and bank profitability of listed commercial banks in Vietnam

Table 4.8 The impact of COVID-19 on the relationship between liquidity risk and bank profitability of listed commercial banks in Vietnam

Variables Coef Std Err P-value

Arellano-Bond test for AR(1)

Arellano-Bond test for AR(2)

Note: The symbols (***), (**), (*) indicate the level of statistical significance respectively 1%, 5%, 10%

Source: Analysis results from STATA 15.0 software

Similarly with model 3.1, the Sargan test in model 3.2 is performed in Table 4.8 to check for over-identification of instrumental variables From the results in Table 4.8, the Sargan test in model 3.2 is used to check the over-identifying properties of the instrumental variables, and the Prob is 0.355, which is greater than 5% Therefore, the hypothesis (H0) is accepted that the instrumental variables are exogenous This indicates that the instrument variables, which are the lagged value of the profitability (ROE L1.), satisfy the over-identifying condition and instrument variables are exogenous variables and model 3.2 does not have an endogenous phenomenon Besides that, the number of instruments in the model 3.1 is smaller than the number of groups (25 < 26) The Arellano-Bond test for AR(2) test result also showed that the Prob was 0.078, which is greater than 5%, hypothesis H0 is accepted that the model does not have a series autocorrelation problem Therefore, all the instrumental variables used in the SGMM model meet the above criteria By including the lag of the dependent variable (Bank profitability) as an instrumental variable, the autocorrelation problem in the model was addressed, resulting in a more robust and valid model Consequently, the SGMM regression method is used as a conclusion in this research to examine the impact of COVID-19 on the relationship between liquidity risk and bank profitability of listed commercial banks in Vietnam in the period 2012 - 2022, as shown in the following equation according to the proposed research model 3.2:

ROE i,t = -0.0344 + 0.5454 ROE i,t-1 + 0.0544 LTD i,t + 0.0315 LTD i,t * COVID + 0.2658 ETA i,t + 0.0002 SIZE i,t + 0.7075 GDP i,t - 0.8809 INF i,t

SUMMARIZING AND DISCUSSING RESEARCH RESULTS

The SGMM model has been applied in this study to investigate the impact of liquidity risk on the profitability of listed commercial banks in Vietnam Additionally, the study also examines the effect of COVID-19 on this relationship The research findings are presented in Table 4.9, with model 3.1 being the model without the interaction variable between COVID and liquidity risk and model 3.2 being the model with the interaction variable between COVID and liquidity risk

Result Coef P-value Result Coef P-value

Source: Analysis results from STATA 15.0 software

Table 4.9 reveals that the impact of liquidity risk on the profitability of listed commercial banks in Vietnam in both model 3.1 and model 3.2 aligns with scientific predictions

From the results of model 3.1 and model 3.2, which are shown in Table 4.9, about the lag variable of bank profitability (ROE), the lag of bank profitability in both model 3.1 and model 3.2 has a positive relationship with the bank profitability at a 1% level of significance, indicating that a higher profitability in the previous year has a positive impact on the profitability of current year With model 3.1, under unchanged conditions of other factors, a 1% increase in the lag variable of bank profitability will cause the bank profitability coefficient to increase by 46.26% About model 3.2, the profitability will increase by 54.54% if increasing last year’s profitability by 1% It means that in the absence of COVID-19, the impact of the lagged profitability variable on ROE is 46.26%, but when having the impact of COVID-19, the impact of the lagged profitability variable on ROE increases by 54.54% The increased influence of the lagged profitability variable on ROE during the COVID-19 pandemic can be achieved from banks' enhanced risk management efforts aimed at maintaining stability and resilience amid the pandemic Consequently, profitability is bolstered as banks adjust their strategies and allocate capital prudently to mitigate the pandemic's impact, thereby enhancing their ability to generate returns This result is completely similar to the author’s initial expectation in hypothesis 1 and relevant empirical studies of Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021) This suggests that there is a positive relationship between bank profitability and its interaction across periods Explaining this positive relationship is that high profits in the previous year enable banks to have additional capital to enhance investment in their business activities, thus positively impacting the profitability of the banks in the current year Moreover, by demonstrating stability and efficiency in past operations, banks can enhance investor confidence in the bank’s future growth potential As a result, new investors and customers are attracted, leading to further enhancements in bank profitability

The research findings reveal a significant positive impact of the LTD ratio on bank profitability in both model 3.1 and model 3.2, indicating that liquidity risk strongly influences the profitability of listed banks in Vietnam from 2012 to 2022 Specifically, under unchanged conditions of other factors, a 1% increase in the LTD ratio leads to a 6.03% increase in bank profitability in model 3.1, and a 5.44% increase in model 3.2 It means that in the absence of COVID-19, the impact of the LTD variable on ROE is 6.03%, but when the impact of COVID-19, the impact of LTD on bank profitability, which is measured by ROE ratio also increases to 5.44% However, the impact is less pronounced compared to a situation without COVID-19 This aligns with hypothesis 2 and is consistent with relevant empirical studies (Alalade, Ogbebor and Akwe 2020; Golubeva, Duljic and Keminen 2019; Ishari and Fernando 2023; Salim and Bilal 2016; Tang My Sang 2019) This suggests that increasing the LTD ratio can indeed enhance the profitability of banks A higher ratio signifies that banks can leverage customer funds to provide loans, increasing loan revenue and interest income However, in this situation, banks may face high liquidity risk if many customers suddenly withdraw their deposits To offset this cash shortfall, banks may need to compete by raising deposit interest rates to attract additional customer capital, leading to an increasing liquidity risk This perfectly aligns with the risk-return trade-off theory, indicating that while bank profitability rises, liquidity risk also increases

Next, the author adds an interaction variable between liquidity risk and COVID-19 (LTDCOVID) to investigate the impact of the COVID-19 pandemic on the correlation between liquidity risk and bank profitability This interaction occurs when the impact of LTD on the bank's profitability is influenced by the presence of the pandemic The regression coefficient of LTD is positive, and the regression coefficients of the liquidity when having the interaction term LTDCOVID are also positive with bank profitability at a significance level of 5% The results can be used to compare with findings from previous studies and provide practical recommendations The results show that this interaction has a significant and positive effect on the bank's profitability during the pandemic This can be understood as during the pandemic, the bank's lending activities may have a significantly positive impact on its profitability This positive relationship has been proven that in the presence of the COVID-19 pandemic, the relationship between interaction variable and profitability is still positive This result is similar to the initial expectation in hypothesis 7 of the study and relevant empirical study of Ren (2022) The results can be utilized to conduct comparisons with previous study findings and provide practical recommendations The empirical results provide evidence that the profitability of listed commercial banks in Vietnam is not significantly affected by the COVID-19 pandemic, unlike the general trend observed in banks worldwide during a pandemic, which tends to experience a slight increase Under the impact of the COVID-19 pandemic, in 2020, the banking system was tasked with supporting businesses and individual customers in restoring production and business activities through debt restructuring and interest rate reduction Even amid the COVID-19pandemic, there remains a demand for loans from businesses and individuals, although possibly at different levels compared to normal situations Therefore, maintaining a high LTD ratio allows banks to continue earning interest income, positively impacting their profitability Moreover, although there may be an increase in non-performing loans due to economic disruptions caused by COVID-19, banks remain cautious in lending They often employ various risk management strategies to mitigate risks; thereby the LTD ratio still positively influences profitability

The profitability of listed commercial banks in Vietnam is not influenced solely by liquidity risk but also by other control variables The research results indicate that the ETA ratio positively impacts on profitability in both models, although it is statistically significant only in model 3.1 In model 3.1, the ETA variable has a positive coefficient at a 1% significance level This result is similar to the initial expectation in hypothesis 3 of the study and market power theory (increasing capital reflects the strength and position of the bank in the financial market) and structure-behavior-efficiency The results are also similar to several relevant empirical studies of Charmler et al (2018), Saleh and Afifa (2020), Nguyen Thanh Phong (2020), Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021), Le Ngoc Thuy Trang et al (2021) Under unchanged conditions of other factors, a 1% increase in the ETA ratio will cause the bank profitability coefficient to increase by 42.71% In model 3.2, although the ETA variable still has a positive relationship with bank profitability but there is no statistical significance As mentioned earlier, equity is a crucial factor in explaining the operations of financial institutions Bank capital indicates the bank's abundant self-capital and less reliance on external funding sources Therefore, banks with high ETA ratios can minimize default risks because they can better settle their loans, even in challenging financial situations, ensuring safety and stability Banks with high ETA ratios can also enhance risk absorption capacity With substantial self-capital, banks can easily mobilize additional capital to cope with unexpected risk situations Moreover, banks can enhance their reputation and competitiveness with a high ETA ratio, as it reflects the financial health of the bank, thereby attracting more depositors All these analyses demonstrate that with a high ETA ratio, bank profitability tends to increase Therefore, based on the regression model results, it is recommended that listed commercial banks in Vietnam improve their ETA to enhance bank profitability

Regarding the impact of bank size on profitability, in both model 3.1 and model 3.2, SIZE has positive coefficients like the initial expectation, but there is no statistical significance This result is still similar to the relevant empirical studies of Saleh and Afifa (2020) Explaining the positive impact of bank size on bank profitability, larger banks have an advantage in diversifying their products and services to enhance profitability and capture more market share Diversifying products and services helps banks mitigate and limit risks during operations Moreover, larger scale enables banks to access more funding sources compared to smaller banks, as they can competitively attract deposits and offer customers services at lower fees (consistent with the market power theory) On the other hand, bank operations entail numerous risks and uncontrollable factors When a bank's size exceeds a certain threshold, it may negatively affect bank profitability Initially, increasing size may have a positive effect on bank profitability, but reaching a certain point of scale could lead to inefficiency A larger scale allows banks to diversify, engage in risky investments, or rely on government intervention in cases of liquidity shortages, rising costs, and impacts on bank profits Therefore, while increasing bank size may create conditions for enhancing bank profitability, excessive expansion may have a counterproductive effect on profitability Thus, the SIZE variable in this study does not exhibit a clear impact on profitability, as it lacks statistical significance during the research period

The macroeconomic variable (GDP) has a positive relationship with ROE in both models 3.1 and 3.2 at a significance level of 1% With model 3.1, under unchanged conditions of other factors, a 1% increase in GDP rate will cause the bank profitability coefficient to increase by 12.51% About model 3.2, the profitability will increase by 70.75% if increasing GDP 1% This implies that in the absence of COVID-19, the effect of the GDP variable on ROE is 12.51% However, when impacted by COVID-19, the influence of GDP on bank profitability, measured by the ROE ratio, rises to 70.75%, which is greater than the impact without COVID, which is 58.24% Since the first outbreak of COVID-19 in Vietnam, the Prime Minister promptly issued resolutions and directives with the motto "fighting the pandemic as fighting the enemy" and a determination to achieve the "dual goal" of both effectively preventing and controlling the pandemic and focusing on restoring socio-economic development Controlling the epidemic has been given top priority, and the government has also timely directed the implementation of support packages, monetary and fiscal policies, and social security measures to assist businesses and people in overcoming the difficulties caused by the COVID-19 pandemic The timely management by the government reflected in the recovery and development support packages and ensuring social security towards sustainable development, has helped Vietnam's economy grow positively despite the pandemic This result is similar to the initial expectation in hypothesis 5 of the research and relevant empirical studies

(Nguyen Thanh Phong 2020; Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh 2021) about a positive relationship between GDP and bank profitability This indicates that economic growth positively impacts investment portfolios, increasing the value of assets and cash flows for banks, resulting in higher profitability Conversely, during economic recession, the likelihood of borrowers defaulting increases, contributing to an increase in credit risk, which in turn affects bank profits As outlined in the hypothesis section of Chapter 3, increasing this ratio is a positive sign, signaling rising borrowing demand During periods of growth, banks tend to decrease interest rates to attract more borrowers, thereby enhancing bank profitability

The second macroeconomic variable (INF) has a negative relationship with ROE in both model 3.1 and model 3.2 at a significance level of 5% With model 3.1, under unchanged conditions of other factors, a 1% increase in the INF rate will cause the bank profitability coefficient to decrease by 76.15% and vice versa About model 3.2, the profitability will decrease by 88.09% if increasing INF 1% and vice versa In the context of the COVID-19 pandemic, inflation leads to a greater reduction in bank profitability due to several adverse effects Firstly, it devalues the currency, forcing banks to pay higher interest rates on deposits, thus reducing profits Secondly, inflation reduces the purchasing power of customers as income sources become uncertain during this period, leading to decreased deposits and increased non- performing loans Thirdly, increased operating costs due to rising prices of goods and services during the pandemic further decrease profitability Finally, inflation creates an unstable and unpredictable business environment, making risk management and business strategy challenging, ultimately negatively impacting bank profits These impacts can become more severe during the COVID-19 pandemic due to the unstable and uncertain economic conditions This result is similar to the initial expectation in hypothesis 6 of the study and relevant empirical studies of Tram Thi Xuan Huong, Tran Thi Thanh Nga and Tran Thi Kim Oanh (2021) Explaining this relationship is that in an economy experiencing inflation, the purchasing power of currency diminishes, leading to increased prices of goods and consequently raising production costs, affecting the circulation of goods Furthermore, as inflation rises, individuals tend to withdraw their bank deposits to invest in alternative channels to avoid currency devaluation risks Therefore, banks will need a significant amount of cash to meet this increased demand for withdrawals

In this chapter, the author analyzes descriptive statistics of the variables and performs tests for heteroscedasticity, autocorrelation, and multicollinearity from the results of Stata 15.0 software and Microsoft Excel The author then estimates the model using the SGMM estimation method and conducts Sargan and AR(2) tests to ensure the reliability and appropriateness of the instrumental variables, as well as addressing the issue of endogeneity and autocorrelation in the model The estimation results show that liquidity risk has a significant positive impact on the profitability of listed commercial banks in Vietnam in normal conditions, represented by loan to deposit (LTD) ratio, which impacts on ROE These variables representing liquidity risk show a significant positive relationship with bank profitability in the model Besides that, bank profitability in the previous year also had a strongly positive impact on the bank profitability of the current year Additionally, control variables such as equity to total assets (ETA), and control macroeconomic variables including GDP growth rate (GDP), and inflation rate (INF) have significant effects on bank profitability However, the bank size variable (SIZE) still positively impacts ROE but has no statistical significance Besides, the study also examines the impact of independent variables on bank profitability while assessing the impact of COVID-19 The model has interaction terms, called LTDCOVID, which also have a positive impact on the profitability of listed commercial banks in Vietnam

Based on the regression analysis results of Chapter 4, the author will present conclusions for the four research questions presented in Chapter 1 and simultaneously suggest recommendations to assist listed commercial banks in Vietnam in controlling liquidity risk appropriately and having the desired profitability Moreover, the study also proposes future study directions related to the impact of liquidity risk on the profitability of listed commercial banks in Vietnam, which is presented in Chapter 5 of this study.

CONCLUSIONS AND RECOMMENDATIONS

CONCLUSIONS

The study about the impact of liquidity risk on the profitability of listed commercial banks in Vietnam uses data from 26 listed commercial banks in Vietnam over 11 year period from 2012 to 2022, utilizing balance sheet data collected from audited financial reports via the FiinPro-X platform The study employs the SGMM study method to estimate the regression model using Stata 15 software The author conducted a review of theoretical foundations and relevant empirical studies published internationally and in Vietnam to propose 2 suitable research models for the thesis The first proposed research model includes the dependent variable of bank profitability measured by the ROE ratio Additionally, the author uses the independent variable of liquidity risk through loan to deposit ratio (LTD) Furthermore, the study also incorporates control variables into the model Bank characteristic control variables include equity to total assets (ETA), bank size (SIZE), and 2 macro-control variables GDP growth rate (GDP) and inflation rate (INF) The second proposed research model is similar to the first proposed research model, but it considers additional interaction variable to examine the impact of COVID-19 on the relationship between liquidity risk and bank profitability (LTDCOVID)

Overall, the test results regarding the impact of liquidity risk on bank profitability (ROE) through the model before and after the impact of the COVID-19 pandemic are completely similar to the extent of impact and consistent with the initial author’s expectation The research results indicate that high profitability banks will have high liquidity risk This is similar to the risk-return trade-off theory The author compares the regression coefficients of LTD in model 3.1 and model 3.2, and the author finds that the positive impact of liquidity risk on bank profitability has a significance level of 1% This can be specifically explained as the increase in this ratio indicates that listed commercial banks in Vietnam are lending more than the capital they raise from customer deposits Consequently, while liquidity risk may increase, it is accompanied by an increase in bank profitability because lending is one of the primary activities of banks

Furthermore, the control variables used in the model, such as ETA, positively impact bank profitability with a significance level of 1% in model 3.1 However, in model 3.2, this relationship is not statistically significant, although this variable still has a positive relationship with ROE This finding is consistent with the structure - behavior - efficiency and market power theory A rise in capital reflects the bank's strength and standing in the financial market It's believed that alterations in market structure or the level of bank concentration impact bank profitability Generally, larger banks tend to achieve higher profitability Bank size (SIZE) also has a positive impact on ROE in both model 3.1 and model 3.2, but there is no statistical significance The macroeconomic variable GDP has a positive relationship with ROE in both models, with a significance level of 1% Similarly, the variable INF also has a significance level of 1% in both models, but the relationship is inversely related Furthermore, when considering the impact of COVID-19 on the relationship between liquidity risk and bank profitability (LTDCOVID) of listed commercial banks in Vietnam, we observe that higher liquidity risk during the COVID-19 pandemic period tends to lead to higher profitability for banks in the case of empirical research conducted on listed commercial banks in Vietnam

In summary, this study provides empirical evidence demonstrating the impact of liquidity risk on the profitability of listed banks in Vietnam The results conclude that liquidity risk has a positive relationship with bank profitability Most listed banks in Vietnam with high profitability have encountered high liquidity risk, which aligns perfectly with the risk-return trade-off theory Moreover, the evidence also shows that when there is a COVID-19 pandemic, this event will boost the positive impact of liquidity risk on bank profitability.

RECOMMENDATIONS

From the study results on the impact of liquidity risk on the bank profitability of listed commercial banks in Vietnam in the period 2012 -2022, the author has some suggestions for listed commercial banks and state management agencies as follows:

For bank managers, ensuring sustained and stable profitability growth is crucial, as the research findings indicate that past profits significantly influence current profits, thereby contributing to future bank profitability In other words, higher profits in previous years act as a driving force to increase profits in subsequent periods This suggestion can be explained by the fact that high profits in the previous year provide banks with additional capital to invest in business activities, thus enhancing the bank's profitability During the author's study period from 2012 to

2022, it can be observed that banks in the industry are currently competing at a relatively high level When banks aim to increase profits, they need to strongly promote business activities, diversify products beyond traditional lending strategies, and improve service quality to enhance their brand, thereby increasing competitiveness and profitability in the market

The research findings indicate that increasing the LTD ratio enhances the profit-making potential of banks Consequently, listed commercial banks in Vietnam must implement robust measures to improve risk management, financial capacity, and credit conditions for efficient credit allocation However, intensified competition among these banks through easy credit provision to customers could negatively impact long-term banking safety Therefore, when listed banks in Vietnam raise the

LTD ratio, it should be accompanied by controls and enhancements in credit quality to effectively manage and minimize non-performing loans This ensures that the mobilized capital is used by the bank safely and efficiently, contributing significantly to bolstering overall banking operations and the economy The decision to increase the LTD ratio aligns with the SBV's Circular 22/2019/TT-NHNN, which sets the maximum LTD ratio at 85% from January 1, 2020, up from the previous limit of 80% outlined in Circular 36/2014/TT-NHNN This relaxation from 80% to 85% eases pressure on banks, allowing for more lending against the capital they have mobilized and reduced capital mobilization competition in the market While higher lending implies increased liquidity risk, amid global monetary easing trends, Vietnam plans to adjust its policies accordingly based on domestic realities In summary, banking management should exercise caution in pursuing aggressive lending growth strategies, and policymakers should closely monitor the Vietnamese banking market when lending expansion accelerates too rapidly

Based on the positive impact of LTDCOVID, as revealed by the research findings, banks can leverage the business opportunities arising from the COVID-19 pandemic Consequently, the study offers recommendations for banks to diversify their sources of income beyond traditional lending activities This diversification strategy can help banks maintain profitability and reduce reliance on interest income, thus ensuring a more secure financial plan while enhancing profitability, but reducing liquidity risk Additionally, banks must collaborate with regulatory authorities to navigate through uncertain economic conditions

Research has confirmed that increasing the scale of shareholder equity will positively impact the profit potential of listed banks in Vietnam Therefore, banks need to enhance their resources and maintain a robust capital structure to withstand losses and minimize liquidity risks A solid capital structure will empower banks to navigate market fluctuations and reinforce customer trust and safety To achieve this, banks can increase internal capital from retained earnings, raise external capital through issuing common and preferred shares in the market, or augment shareholder equity through mergers and acquisitions This will contribute to creating financially capable banks, adequately positioned to compete with domestic and international counterparts Furthermore, banks should improve access to and diversify funding sources, especially shareholder funds, while assessing and maintaining relationships with shareholders Moreover, banks need to establish balanced policies in distributing financial results, allocating dividends to shareholders, and retaining profits to supplement shareholder equity After increasing capital, listed banks in Vietnam also need to enhance the efficiency of capital utilization to avoid capital idleness Additionally, banks should ensure that increasing scale aligns with government and SBV policies to safeguard capital in line with Basel agreements, effectively controlling banking risks Listed commercial banks continue to increase their equity Increasing the scale of equity is a prerequisite for these banks to expand their loan portfolios and increase the scale of profitable assets, thereby boosting income from interest This is also a crucial objective for these banks to ensure the minimum capital adequacy ratio (CAR) compliance as per SBV’s circular.

Despite the SIZE variable lacks statistical significance in the research model, the authors do not have sufficient basis or evidence to demonstrate its impact However, according to the relevant literature reviews mentioned above, attention should be paid to the issue in subsequent studies and should be consistently maintained In cases where scale becomes significant, similar to the research scenarios of most other authors, a large bank asset size will contribute to improving bank’s profitability Expanding the scale of the bank is also necessary to enhance the profit-making ability of listed banks in Vietnam When the bank's scale is sufficiently large, it will have the ability to exploit and utilize economic advantages regarding scale, thereby increasing the bank's profitability Additionally, having a large asset size also helps enhance the bank's brand image and credibility, thereby attracting customers more efficiently Expanding the scale of the bank's network through establishing new branches and transaction points will help increase coverage and attract more new customers, better meeting their financial product and service needs in various areas However, this does not mean that the larger the bank's asset size, the higher the profitability for the bank because increasing the scale requires a reasonable balance between the bank's scale and the management capacity of bank managers If the bank's scale increases excessively, it will have the opposite effect, meaning it will decrease the bank's profitability because the management capacity and level of expertise of the banks do not increase in line with the speed of scale expansion, leading to difficulties in management and cost control due to the significant increase in scale, thereby affecting the bank's profitability Therefore, listed commercial bank managers should carefully manage their bank size expansion plans to effectively improve bank profitability, as excessive expansion may not be effective and could have adverse effects

From the bank's perspective, regarding macroeconomic management, banks need to establish an early forecasting system for macroeconomic conditions and develop specific plans to cope with unusual risks that may affect liquidity risk This has been evidenced by study results showing that GDP and inflation both impact bank profitability Therefore, having early macroeconomic forecasting plans will enable listed commercial banks in Vietnam to react promptly, make appropriate policy adjustments to minimize risks, and ensure bank profitability To enhance proactive forecasting and response to macroeconomic information, banks can establish expert groups and build dedicated study models to develop suitable scenarios under different macroeconomic conditions Additionally, leveraging information technology through the use of artificial intelligence (AI) applications can help identify early potential risks and provide support in management and operations to improve performance, achieve profit targets, or at least minimize risks and potential decreases

Based on Vietnam's economic situation and banking activities, appropriate macroeconomic policies regarding liquidity should be formulated by the regulatory authorities body For instance, considerations should be concentrated and boost the liquidity ratio or loan ratio If banks overly focus on liquidity according to strict monetary regulations, it may negatively impact the profitability of the listed commercial banking system in Vietnam However, loosening monetary policies can also put pressure on the liquidity situation of the system Therefore, it is crucial for regulatory authorities to grasp the economic situation and implement timely and appropriate policies to ensure the safe and sustainable development of the banking system In response to the above issue, regulatory authorities need to enhance regular inspections to have plans and timely corrective measures, ensuring the safe, stable, and efficient operation of the banking system, contributing to the overall development of Vietnam country

Market economy activities heavily depend on the issue of inflation Increasing inflation will have significant impacts on all aspects of the economy High inflation leads to easier debt repayment for customers due to the reduced real value of loans However, increasing inflation may also reduce borrowers' ability to repay debts because higher inflation decreases in customers' real incomes, increasing the risk of bad debts and thereby increasing liquidity risk and negatively affecting bank profitability Therefore, regarding the issue of inflation, the SBV needs to first collaborate with relevant regulatory agencies to control inflation to a suitable level for the economy This should be done gradually to combat inflation issues Thus, all relevant parties must have appropriate adjustment plans concerning the relationship between various indicators, such as inflation and exchange rates, etc, while aligning with the state regulations.

LIMITATIONS OF THIS THESIS AND FUTURE RESEARCH

Despite achieving certain results as presented above, the study about the impact of liquidity risk on the profitability of listed commercial banks in Vietnam still faces several limitations due to constraints related to time, and experience of the author

Firstly, regarding the limitations of the research data, this study only uses data from 26 listed commercial banks in Vietnam during the period 2012-2022, which does not cover all commercial banks in Vietnam Incomplete data disclosure by some banks also leads to asymmetrical data, which may affect the regression results in the model Moreover, the results obtained do not reflect a high level of generality, nor do they fully demonstrate the real impact of liquidity risk on the profitability of the Vietnamese banking system

Secondly, the study only applies the financial indicator ROE as a proxy for the profitability of listed commercial banks in Vietnam to measure liquidity risk, while other financial indicators such as ROA and NIM have not been considered by the authors Furthermore, the study solely focuses on evaluating the impact of liquidity risk on bank profitability, while there are still many other factors influencing bank profitability, such as credit risk, non-interest income, and non-interest expenses, which have not been incorporated into this study due to limitations in data collection and study capabilities of the authors

Thirdly, regarding macroeconomic factors, there are many other macroeconomic variables that have not been considered by the authors in the model, such as the annual unemployment rate (UEP), money supply (M2), etc These variables should be considered in future study

Fourthly, this quantitative study only provides experimental evidence of the positive impact of liquidity risk on bank profitability However, the study results have not examined the impact of liquidity risk on profitability among different groups of banks (large, medium, and small banks) to provide a more detailed and appropriate evaluation of the consistency among entities within the same bank group to increase the reliability of the study results

Lastly, the study only tests one-way regarding the impact of liquidity risk on profitability It has not tested two-way causality in the relationship between liquidity risk and bank profitability

Based on the limitations mentioned above, the authors propose several suggestions for future studies to overcome the constraints of this study and provide more comprehensive contributions to experimental study

Firstly, future studies could increase the number of observations by expanding the scope of the study, such as conducting a study with all commercial banks (according to the list of commercial banks in Vietnam by the SBV as of September

30, 2023, which includes 31 joint-stock commercial banks and 4 state-owned commercial banks) Additionally, future studies could be conducted over a longer period to provide a comprehensive long-term assessment and minimize deviations in study results

Secondly, future studies could use more dependent variables representing bank profitability or combine all three financial indicators (ROA, ROE, NIM) or other new variables Moreover, the study should innovate representative variables for the liquidity risk issue of banks, replacing them with other variables such as LLR (liquid assets/total loans), FGAP (financing gap), etc

Thirdly, the new study needs to include more control and macroeconomic variables such as SIZE 2 , Money supply (M2), annual unemployment rate (UEP), etc Additionally, regarding dummy variables, studies can add a combination of two dummy variables for financial crises and COVID-19 to the relationship between liquidity risk and bank profitability instead of just considering COVID-19 as in this study

As solutions for limitations four and five, future studies should consider and assess the impact of liquidity risk on bank profitability between large, medium, and small banks and simultaneously compare the model of the impact of bank profitability on liquidity risk to understand the relationship between these two factors further

In general, this study has achieved the initial research objectives set out However, the liquidity risk measurement method used by the author of this study relies on traditional liquidity ratios Therefore, future research should consider indicators based on Basel III standards, as mentioned in the theoretical section When Vietnam implements Basel III in the near future, sufficient data sources will be available for further research

Chapter 5 presents a summary of the study results on the impact of liquidity risk on bank profitability of listed commercial banks in Vietnam, based on the regression results of 26 listed commercial banks during the study period from 2012 to 2022 Based on the study findings, recommendations are provided for liquidity risk management to achieve the profitability goals set by banks Additionally, these study findings are expected to serve as an additional reference for other research students and bank managers as supplementary materials This chapter also discusses some limitations of the study that the author could not address and suggests some new research directions for future authors to contribute to the completion of the author's study topic on the impact of liquidity risk on bank profitability

Profitability plays a crucial role in the operations of economic entities, including the banking sector However, banking activities inherently involve various risks If these risks are not well controlled, they can diminish the profitability of banks Building upon theoretical frameworks and practical contexts, aiming to fill study gaps, the dissertation has chosen to conduct the study "The impact of liquidity risk on bank profitability: Empirical evidence from listed commercial banks in Vietnam" to contribute additional empirical evidence and useful information regarding the impact of liquidity risk on bank profitability

The general study objective of this thesis is to analyze the impact of liquidity risk on the profitability of listed commercial banks in Vietnam from 2012 to 2022 Furthermore, the study also considers to the impact of the COVID-19 pandemic on the relationship between liquidity risk and bank profitability