Introduction to the study
Introduction
This chapter serves as an introduction to the study, outlining the research background in Section 1.2, where gaps in empirical studies on the efficiency of commercial banks are identified, along with an overview of the Vietnamese banking sector Section 1.3 articulates the research questions and objectives of the study.
In addition, section 1.4 discusses scope and some limitations of the current study
Section 1.5 briefly discusses the general aspects of research methodology such as research types and research design
Section 1.6 provides implications of this study Section 1.7 introduces the structure of the study The structure of chapter 1 is provided in Figure 1.1.
Research background
In January 2010, Vietnam marked three years of membership in the World Trade Organization, leading to significant transformations in its banking industry The ongoing financial liberalization that began in the 1990s has been a driving force behind these changes Over the past two decades, Vietnam's financial system, particularly its banking sector, has evolved from a monopoly to a diversified system, enabling fair and effective competition among all participants.
In recent years, Vietnam's banking system has seen significant growth, marked by an increase in the number of banking institutions, a larger relative size of the banking sector within the economy, a rise in credit availability, and an expansion of various banking services.
The banking system in Vietnam has been impacted by the recent global financial and economic crisis, leading to a rise in bankruptcies among institutions and overall system instability, as evidenced by the liquidation crisis at the end of 2008 and a high ratio of non-performing loans Despite these challenges, our primary focus remains on assessing the effectiveness of Vietnamese commercial banks.
The traditional evaluation of bank performance often relies on financial ratio methods, which provide a straightforward comparison of a bank's financial performance over time However, a significant limitation of this approach is the lack of consensus on the importance of different financial indicators To effectively measure technical efficiency in banks, it is essential to assign appropriate weights to each financial indicator While financial ratios can be useful for firms with a single input or output, banks typically utilize multiple inputs to deliver various services, complicating the selection of relevant ratios A potential solution is to aggregate the averages of all indicators to create a unified measurement.
DEA approach can be employed to solve the issue of weight assignment This approach uses a mathematical programming method to generate a set of weights for each indicator
There is a significant gap in empirical research regarding the efficiency of Vietnam's banking sector over recent decades The confidential nature of Vietnamese banks has hindered access to essential data for both foreign and local researchers, limiting their resources primarily to standard annual reports.
This paper aims to empirically assess the efficiency of the Vietnamese banking system from 2007 to 2009, providing insights into its performance and necessary improvements The author employs a data envelopment analysis (DEA) model, utilizing panel data from twenty-two Vietnamese commercial banks for the research.
This research provides valuable insights for bank managers to assess their banks' efficiency and understand the factors contributing to their successes or failures Additionally, it aids banks in strategic planning and supports policymakers in enhancing the overall efficiency of the banking industry.
1.2.2 Overview of the Vietnamese banking system
Before the Renovation reforms of 1986, Vietnam operated under a centrally planned economy where the government dictated production objectives and methods The banking system was characterized as a "mono-banking system," with the State Bank of Vietnam serving as both the central bank and a commercial bank Although special purpose banks for foreign trade and investment existed prior to 1988, they functioned as sub-units of the State Bank and resembled government organizations rather than commercial entities In March 1988, the banking system transitioned to a two-tier structure, as established by Decree No 53/HDBT from the Prime Minister.
Bank of Vietnam acted as the central bank, overseeing four (state owned) commercial banks (SOCB) which specialize d in different areas of banking activity
In May 1990, a pivotal moment in Vietnam's financial landscape occurred with the announcement of two key Decrees: one concerning the State Bank of Vietnam and the other addressing banks, credit cooperatives, and financial companies These Decrees transformed the Vietnamese financial system from a monopoly, or one-tier system, into a two-tier system, allowing the State Bank of Vietnam to function primarily as a central bank while granting other banks and financial entities the freedom to operate independently, including the ability to establish and close their operations.
Table 1.1 Number of commercial banks in Vietnam, 2007-2009
Source: State Bank of Vietnam, Annual Report, 2007-2009
By the end of 2009, Vietnam's financial and banking system experienced significant growth, with the total number of banking institutions reaching 93 This included five state-owned commercial banks, one social policy bank, 37 joint stock commercial banks, five joint venture banks, 40 foreign bank branches, and five fully foreign-owned banks.
As of now, the total equity of credit institutions in Vietnam reached VND 282,612 billion, representing 9.32% of total assets, with chartered capital at VND 208,873 billion This growth in total equity and chartered capital indicates that credit institutions are leveraging internal resources to enhance their financial capacity They have also expanded total assets through services such as fund mobilization from organizations and individuals, as well as credit extension, including loans and guarantees Compared to 2008, the total assets of the entire system have increased by 36.39% State-owned commercial banks (SOCBs) hold the largest share of assets in the system at 45.5%, down from 51.8% in 2008 Additionally, the ratio of non-performing loans to total loans in the credit institution system is currently 1.99%, a decrease from 2.13% at the end of 2008.
Table 1.2 Some developments of Vietnamese banking system (2007-2009)
Deposit growth rate 36.53 47.64 22.84 29.88 Credits growth rate 25.44 53.89 23.38 37.53
Source: State Bank of Vietnam, Annual Report, 2007-2009
The rapid economic growth in recent years has led to a significant expansion of banking activities, with banking sector assets reaching 120% of GDP by the end of 2008 While the banking system is highly concentrated at the top, it remains fragmented at the bottom.
The four largest State-Owned Commercial Banks (SOCBs), including the partially privatized Vietcombank and Vietinbank, dominate the banking sector, controlling 60% of total assets In 2008, capital mobilized through the banking sector was 22.87%, a decline from nearly 50% in 2007 due to the global financial crisis This shift saw a decrease in capital from SOCBs and an increase from other commercial banks, resulting in domestic credits provided by the banking sector being nearly equivalent to the total GDP.
The global financial crisis of 2007-2009 significantly impacted financial markets worldwide, posing challenges for Vietnam's banking sector To navigate these difficulties, Vietnamese commercial banks must enhance their operational efficiency and boost productivity growth Addressing these critical issues is essential, especially as the country opens its financial market in line with its WTO commitments.
Research objectives and questions
A research objective, as defined by Zikmund (1997), represents the researcher's interpretation of a business problem, articulating the purpose of the research in measurable terms and establishing criteria for its success This study aims to address the research problem through specific objectives.
- To investigate to the efficiency of the Vietnamese commercial banks
- To analyze the changes in the productivity and technology of the Vietnamese commercial banks
- To contribute to knowledge of measuring efficiency in banking sector
Research questions involve the research translation of “problem” into the need for inquiry (Zikmund 1997) As discussed above, the current research leads to the following research questions:
Q1 How is the current efficiency of the banking system in Vietnam?
Q2 What effect are the changes in productivity and technology of Vietnamese commercial banks‟ having on efficiency?
Thus, it is important to answer these questions and from that, a clearer view on the Vietnamese banking system will be revealed.
Scope and Limitation
This study focused on 22 Vietnamese commercial banks from 2007 to 2009, highlighting the need for further research involving a broader range of banks to accurately evaluate their scale and efficiency Expanding the research to include the entire banking industry will provide valuable insights for policymakers regarding macroeconomic trends.
Research method
In choosing a research design, Zikmund (1997) discusses three types of business research: exploratory, descriptive and causal research
• Exploratory research is usually conducted to clarify and define the nature of a problem
• Descriptive research is designed to describe characteristics of a population or phenomenon
• Causal research is conducted to identify cause-and-effect relationships among variables where the research problem has already been narrowly defined
The choice of research type is determined by the specific questions the researcher aims to address This study focuses on assessing the efficiency of commercial banks in Vietnam, making exploratory research a suitable approach Additionally, it seeks to uncover cause-and-effect relationships between various factors and bank efficiency, incorporating causal research alongside descriptive methods In conclusion, a blend of exploratory and causal research is selected for this investigation.
After determining the type of research, the next crucial step is selecting the appropriate research design There are four primary research designs to consider: surveys, experiments, observations, and secondary data (Zikmund 1997) The choice of research design should be informed by the pros and cons of each type, as well as the specific context of the research problem In this study, secondary data methods are employed to effectively tackle the identified issue.
This paper employs the non-parametric approach, specifically Data Envelopment Analysis (DEA), to measure efficiency We utilize the Malmquist total factor productivity index to assess productivity changes, distinguishing between technical efficiency and technological changes, in order to analyze the productivity differences among various banks.
Implications of research
The study results bring various practical meanings for the bank managers in this industry, educators as well as researchers in banking and finance as follows:
This research provides valuable insights for bank managers to assess their banks' efficiency and understand the factors contributing to their successes or failures, aiding in effective strategic planning.
Second, the study also helps policy makers in their attempts to improve the overall efficiency of the banking industry
This research enhances the global literature on bank efficiency, serving as a valuable reference for researchers, educators, and Vietnamese students in the fields of banking and finance, particularly within the context of the Vietnamese banking industry.
Finally, the present study will be a reference of research methodology not only in banking field in particular but also the other social sciences.
Structure of the study
This research is structured into 5 chapters:
This chapter introduces the research including research background, research questions, a brief research methodology overview, implications and limitations of research.
Literature Review
Introduction
This chapter presents a comprehensive literary review of the efficiency of commercial banks, starting with a general introduction in section 2.1 Section 2.2 defines commercial banks, outlining their functions and the factors that influence their operations In section 2.3, the concept of efficiency is explored, followed by section 2.4, which discusses methods for measuring efficiency, particularly focusing on previous research utilizing the DEA method Section 2.5 highlights studies related to measuring bank efficiency in Vietnam, and the chapter concludes with a summary in section 2.6.
Commercial banks
According to the Law on Credit Institutions of Vietnam, a commercial bank is defined as a type of credit institution that can perform all banking operations These banks are categorized into commercial banks, policy banks, and cooperative banks based on their characteristics and operational objectives Specifically, a commercial bank is one that engages in all banking activities as well as other profit-oriented business operations.
Law No 47/2010/QH12, enacted on June 16, 2010, by the XII National Assembly of the Socialist Republic of Vietnam during its 7th session, regulates credit institutions in the country.
The law also mentions that banking operations mean the trading in and regular provision of one or some of the following services:
1.)Taking demand deposits, time deposits, savings deposits and deposits of other types
2.) Issuing deposit certificates, promissory notes, treasury bills and bonds to raise capital at home and aboard
3.) Extending credit by: Lending; Discounting and re-discounting negotiable instruments and other valuable papers; Providing bank guarantee; Issuing credit cards;
Domestic factoring; international factoring for banks licensed for international payment; Other forms of credit after obtaining the Slate Bank's approval
4.) Opening payment accounts for clients
The payment services offered include domestic options such as checks, payment orders, authorized payments, collections, letters of credit, bank cards, and various collection and payment services Additionally, international payment services and other payment solutions are available, subject to approval from the State Bank.
Commercial banks are private, profit-driven entities focused on maximizing shareholder wealth by balancing risk and return in their portfolio management Their primary assets consist of loans and securities, with the majority of income generated from interest on loans Meanwhile, deposits represent the main liabilities of banks However, the importance of transaction accounts has decreased as other financial institutions have begun to offer similar accounts, increasing competition with commercial banks.
Banks serve three primary functions: they play a crucial role in the payments system, facilitate the connection between depositors and borrowers through various deposit and loan products, and offer a wide range of financial services, including fiduciary services, investment banking, and managing off-balance sheet risks.
Banks play a crucial role in the payments system, which is essential for settling financial transactions Their involvement is significant for economic stability and growth, highlighting the social importance of an efficient payments system The payment system can be categorized into two main components.
The retail payment system enables individuals to efficiently pay bills and receive funds, while the large-dollar payments system is designed for businesses and governments to manage substantial domestic and international transactions.
Banks serve as financial intermediaries, where deposits are classified as liabilities and loans as assets Their profitability stems from the interest rate differential between borrowing and lending, adjusted for all associated expenses.
2 Benton E.Gup, James W.Kolari (2005), Commercial banking textbook, the
Management of Risk, John Wiley and Sons, Inc
Commercial banks serve as intermediaries, connecting depositors with borrowers They utilize deposits to provide loans, primarily focusing on short-term financing for commercial needs Historically, short-term deposits have been essential in supporting short-term lending activities.
Financial intermediation plays a vital role in economic growth and stability by connecting depositors with borrowers The economy thrives on substantial savings and the efficient allocation of these funds to productive ventures Commercial banks promote savings by providing financial instruments with attractive risk/return profiles, while also ensuring that credit requests are thoroughly screened to direct funds towards socially beneficial and profitable investments.
Banks engage in off-balance sheet activities, including insurance and securities-related services, as well as trust services They utilize financial derivatives such as interest rate swaps, financial futures, and options to mitigate risks associated with interest rates, foreign exchange, and credit defaults Additionally, commercial banks and their affiliates offer a range of life insurance policies, annuities, and related products They also provide brokerage services for buying and selling securities on behalf of customers and may operate as securities dealers, trading for their own accounts.
Finally, they may offer investment banking services such as underwriting securities
Commercial banks often have trust departments that manage clients' funds for a fee, as outlined in a trust agreement These departments generate fee income for the banks.
2.2.3 Major factors affecting commercial banks’ performance
The principal factors that have affected the operations of commercial banks in recent years:
Commercial banks are significantly impacted by inflation and fluctuating interest rates, which have led to intense pressure on the financial system and the failure of numerous institutions Many banks borrowed short-term funds while issuing long-term real estate loans at fixed interest rates As interest rates surged, their borrowing costs rose above the low fixed returns on their assets, resulting in a decline in the market value of these assets Additionally, a substantial number of borrowers defaulted on their loans, exacerbating the situation.
Securitization has significantly transformed the financial services industry by enabling the issuance of debt instruments backed by revenues from a specific pool of loans This process allows banks to unbundle the lending process, facilitating enhanced access to capital markets.
The evolution of technology in financial service delivery, particularly the reduction in delivery costs, has profoundly affected commercial banks Additionally, advancements in computer technology have lowered the expenses associated with screening and monitoring loan portfolios.
The more sophisticated customers making banking markets more efficient and making it, the more difficult for banks to earn an acceptable risk adjusted return
Efficiency and factors affecting efficiency of commercial banks
Efficiency is a key concept across various fields such as economics, technology, and social sciences In the context of economics, a firm is considered efficient when it maximizes output from given inputs, achieving what is known as Pareto optimality.
An economic system is considered more efficient than another if it can deliver a greater quantity of goods and services to society without requiring additional resources (O'Sullivan, 2007) Efficiency, often equated with productivity, is quantified by the ratio of outputs to inputs used in the production process.
P: Productivity (or Efficiency) Out: Output variables (such as quantity, revenues, profit, etc.) In: Input variables (such as wages, cost, expense, etc.)
Efficiency encompasses not just productivity but also economic value, often referred to as value-for-money (SNZ, 2010) Researchers typically define it as economic efficiency, which comprises both technical efficiency and allocative efficiency (Hall & Lieberman, 2006).
A financial institution, often referred to as a decision-making unit (DMU), is considered efficient if it cannot increase its output without a proportional rise in inputs, or if it cannot decrease its inputs without a corresponding reduction in output.
2.3.2 Main factors affecting efficiency of commercial banks
A commercial bank serves as a financial intermediary, connecting regional savings with investments in the economy Consequently, its operations are significantly affected by external factors, including economic, political, and social environments.
In a robust and stable economic growth environment, businesses are prompted to expand their production and operations, leading to a heightened demand for loans This scenario allows commercial banks to easily broaden their credit activities Additionally, a thriving economy contributes to a decrease in bad debts, as the financial strength of enterprises improves.
The evolution of the market economy highlights the crucial role of legal systems in business management An outdated legal framework can significantly obstruct economic progress In contrast to developed nations with robust legal systems, Vietnam faces challenges due to its inadequate legal infrastructure, which poses a barrier to the operations of commercial banks.
Internal factors within Vietnam's commercial banks, including financial capability, management effectiveness, the adoption of advanced technologies, and the quality of labor, significantly impact the efficiency of these banks.
The financial capability of commercial banks is often manifested through the ability to expand sources of equity, because equity represents the financial strength of a bank
The potential of equity significantly influences a bank's operational scope, impacting its ability to mobilize capital and loans, finance investments, and enhance technological capabilities Profitability serves as a key indicator of a bank's financial strength, reflecting the efficiency of its business capital Additionally, a bank's capacity to manage and mitigate risks is crucial; an increase in bad debt reserves indicates a heightened risk level that must be managed to offset potential losses If bad debt reserves rise but remain insufficient to cover financial shortfalls, the bank's ability to address these expenses diminishes.
Capacity management and administration significantly influence bank performance The effectiveness of executive management relies on the organizational structure, workforce quality, and operational efficiency to adapt to market changes Additionally, strong management capacity is demonstrated through the ability to lower operating costs while enhancing the productivity of inputs to achieve optimal output.
The ability to leverage advanced technologies reflects a bank's information technology capacity In the past, before the rise of robust science and technology and its widespread application in society, the banking industry struggled to remain competitive while offering communication services.
Technological capacity of the bank reflect the ability of new technology equipment, including equipment and people, the technology links between banks and the uniqueness of each bank's technology
The success of commercial banks heavily relies on the quality of their workforce, making the human factor a crucial determinant As society evolves, banks must enhance their services and improve the quality of human resources to adapt to market changes Employing professionals with strong ethics and high standards fosters customer loyalty, mitigating business risks and lowering operating costs Furthermore, the development of human resources should align with advancements in technology to ensure continuous growth and adaptation.
The measurement of efficiency of Commercial banks
The efficiency of banks has gained significant attention in recent years, particularly with the rapid growth of financial markets Measuring the efficiency of financial institutions is crucial, as improved operational efficiency can lead to enhanced profitability and increased intermediated funds.
To estimate banks‟ efficiency, we can use different methods These methods can be classified in various ways:
• The traditional method of financial indices based on balance sheet analysis,
• Parametric methods based on the knowledge of production function,
• Non-parametric methods that do not re quire such knowledge
2.4.1 Efficiency analysis by the traditional method of financial indicators
Financial indicators remain a crucial tool for analyzing and comparing bank performance among owners and potential customers However, a significant limitation of ratio analysis is the lack of consensus on the importance of different indicators While the financial ratio method is effective for firms with a single input or output, banks, which utilize multiple inputs to deliver various services, face challenges in selecting the appropriate ratios Consequently, evaluators must navigate a multitude of financial indicators, which can be categorized into four main groups: profitability rates, margin rates, weighted result rates, and employment efficiency rates.
The first group of the indicators is profitability rates The most common ones in this group are:
ROE (the rate of Return On Equity) is a ratio of financial result to a bank‟s own fund;
ROA ( the rate of Return On Asset) is a ratio which measures the ability of management to utilize the actual financial resources of the bank to generate returns
ROA is commonly used to evaluate bank management;
(ROS) (the rate of Return On Sale) is a ratio of financial result to a bank‟s income;
(C/I) (Costs ratio)is a ratio of costs to incomes.
The second group of efficiency indicators are margin rates Two basic rates of this group are based on interest margin:
Net interest margin is a key financial metric that measures the ratio of interest income to total assets It reflects the difference between the average interest earned on interest-bearing assets and the average interest paid on interest-bearing liabilities.
Weighted result rates are a key financial measure in efficiency analysis This metric reflects the difference between the accumulation and dissolution of reserves, alongside the achieved results Additionally, the result rate related to operating costs indicates the ratio of operating expenses to the overall result If the result rate associated with reserves is positive, meaning a bank is accumulating more reserves than it is dissolving, it suggests that this reserve buildup negatively impacts the bank's results, effectively lowering its performance level.
The last group of measures constitutes the employment efficiency rates The most frequently used ones are:
• The rate presented as a ratio of assets to the number of employees (job positions);
• The rate presented as a ratio of a result to the number of employees
Analyzing financial indicators is the most widely used method for assessing efficiency in banks While numerous financial indicators can be utilized, interpreting the results can become challenging Conversely, relying on a single indicator often yields insufficient information to determine the accuracy of a given value.
Evaluating a bank's financial ratio often involves comparing it with peer group banks, which can provide valuable insights Tracking these ratios over time, even without peer comparisons, can reveal important trends in the bank's performance However, a limitation of financial ratio analysis is that it assumes other factors remain constant To gain a more comprehensive understanding of a bank's financial health, it is essential to calculate multiple financial ratios.
2.4.2 Efficiency analysis by parametric frontier efficiency approach
In addition to traditional methods, researchers worldwide now employ frontier efficiency analysis to assess bank performance This approach is categorized into two groups: parametric and non-parametric methods The parametric approach necessitates the definition of a specific functional form for frontier efficiency and allows for the specification of inefficiency distribution or random errors.
But if the selected function is incorrect, calculation results will affect the opposite direction of the efficiency index
A significant challenge for both parametric and non-parametric approaches lies in differentiating random errors caused by accounting practices or other inefficiencies While parametric methods employ various strategies to address random errors, non-parametric methods typically overlook them.
The parametric approach emphasizes the production or cost functions of banks, allowing for the estimation of an optimal banking system function through regression models (Banker & Maindiratta, 1988) This method enables the calculation of a bank's efficiency by comparing its production or cost levels to the optimal benchmark The parametric estimates, grounded in regression analysis with defined confidence intervals and deviations, are statistically validated According to Berger and Humphrey (1997), over 52 percent of researchers favored the parametric approach for measuring the efficiency of financial institutions between 1992 and 1997.
Key studies in the field include Mester (1993), Berger and DeYoung (1996), and Peristiani (1997) However, the assumptions underlying these estimations are frequently unreliable, particularly with small sample sizes Consequently, a nonparametric approach is often favored in such cases.
Numerous studies, including those by Bell and Murphy (1967), Longbrake and Johnson (1975), and Kolari and Zardkoohi (1987), have aimed to estimate cost function characteristics and assess economies of scale and scope under the assumption of efficient bank operations Banker and Maindiratla (1988) contended that the estimated cost function reflected the average behaviors of banks within the sample, suggesting that regression methods could be adjusted to focus estimates towards the efficient frontier From 1992 to 1997, efficient cost frontier methodologies were employed in this context.
116 out of 130 studies related to financial institution frontier efficiency across 21 countries (Berger and Humphrey, 1997)
2.4.3 Efficiency analysis by Non-parametric approach (DEA)
This study employs Data Envelopment Analysis (DEA), a non-parametric method first introduced by Charnes, Cooper, and Rhodes in 1978 Initially based on constant returns to scale, DEA has since been expanded by Banker, Charnes, and others.
Cooper (1984) developed a model that accommodates variable returns to scale using Data Envelopment Analysis (DEA), a linear programming technique DEA constructs the efficiency frontier piecewise from the best practice observations, which are considered 100% efficient This method does not impose a specific functional form on the data, allowing the weights for inputs and outputs to be derived directly from the data itself.
Data Envelopment Analysis (DEA) evaluates the efficiency of financial institutions by comparing them to a reference set of efficient peers Each inefficient institution is benchmarked against this group, allowing for a clear identification of performance gaps This operational approach not only highlights inefficiencies but also offers actionable insights for improvement, making DEA a valuable tool for enhancing efficiency in the financial sector.
This approach distinguishes between technical and allocative efficiencies in cost efficiency, further breaking down technical efficiency into pure technical and scale efficiencies The Malmquist index is a key tool for evaluating changes in bank productivity, often separating these changes into technical efficiency and technological progress The primary goal of Data Envelopment Analysis (DEA) is to assess the relative efficiency of similar units that utilize the same technology and resources to achieve comparable outputs Efficiency scores for Decision-Making Units (DMUs) range from zero to one, with fully efficient banks achieving a score of one.
Literature review on measuring efficiency of banking system in Vietnam
Vietnamese researchers have extensively explored the liberalization of the financial system and banking sector (Le, 2006; Ngo, 2004, 2009a), assessed the efficiency of commercial banks (Ngo, 2010b; V H Nguyen, 2007), and applied bootstrapping techniques to enhance the Malmquist productivity index for these banks (X Q Nguyen & DeBorger, 2008).
In his 2006 discussion paper, Le highlighted that numerous banking regulations were introduced post-2005 to enhance the stability of Vietnam's banking system and to privatize major state-owned commercial banks (SOCBs) However, he criticized the poor sequencing of these reforms, stating that "Vietnam's banking system is quantitatively and qualitatively inadequate." Consequently, Le concluded that the efficiency of the Vietnamese banking system was low as of 2006, a viewpoint that was further corroborated by Nguyen in 2007.
In 2007, Nguyen conducted a study on the efficiency performance of 13 commercial banks in Vietnam, analyzing data from 2001 to 2003 The research specifically examined aspects such as efficiency change, productivity growth, and technological advancements within these banks.
The author concluded that Vietnamese banks exhibited inefficiencies in both allocative and technical aspects, with technical inefficiency being the more critical issue This indicates that enhancing the efficiency of input utilization in these banks is more crucial than optimizing the selection of the appropriate input mix (V.H Nguyen, 2007).
A discussion paper presented at the Asia-Pacific Productivity Conference 2008 analyzed the efficiency and productivity of 15 commercial banks in Vietnam, including 4 out of 5 state-owned commercial banks (SOCBs), authored by Nguyen and DeBorger.
In their 2008 study, X Q Nguyen and DeBorger claimed to be the pioneers in analyzing efficiency and productivity indices for commercial banks in Vietnam, revealing a concerning trend of declining productivity within these financial institutions.
2008) However, as they concluded, the bootstrapping result proved that this trend is not significant, and therefore, more detailed studies are needed as well.
Summary
This chapter examined the efficiency and measurement of commercial banks in Vietnam, highlighting the work of various researchers in the field However, there remains a significant need for further research on the Vietnamese banking system, particularly regarding efficiency and performance Enhancing efficiency is crucial for the Vietnamese banking sector to compete effectively with foreign banks during the integration process and to fulfill commitments made upon joining the WTO.
Research methodology
Introduction
The literary review in the previous chapter introduced and discussed the theoretical model constructed for this research
This chapter aims to outline and justify the research methodology employed in the study, detailing the research design, data collection, and analysis methods used to address the research questions presented in Chapter 1 It is organized into five sections, with the next section focusing on the research design, while Section 3.3 will describe the data and variables involved.
Next, the study shows the data analysis techniques The final sector is the summary.
Research design
The research design is crucial for addressing theories and hypotheses, guiding subsequent research steps Business research methods can be categorized by function or technique (Zikmund 1997) Functionally, research types include exploratory, descriptive, and causal studies Technically, methods encompass experiments, surveys, observational studies, and secondary research According to Zikmund (1997), exploratory research seeks to clarify and define problems by focusing on "how" and "what," while causal studies investigate the "why" behind the influence of one variable on another This study integrates both exploratory and causal approaches, as reflected in its research questions.
This study employs secondary research techniques to evaluate the research questions, utilizing data sourced from the annual reports of Vietnamese commercial banks The analysis is conducted using DEAP 2.1 software to process the collected data.
Figure 3.1 Research process Literature review
Colleting financial statement of 22 Vietnam commercial banks
• Calculating the data set from the financial statement
• Encode and input the data set
Descriptions of Data Sample and Variables
In Vietnam, over 90 banks exhibit significant variations in capital, labor force, and location, leading to differing levels of availability This research aims to analyze the top twenty-two commercial banks in the country.
In 2009, the VNR-500 identified the 50 largest enterprises in the banking, financial, and securities industry, which included a sample of 22 Vietnamese commercial banks comprising four state-owned banks and eighteen private or joint-stock banks However, data for many joint-venture and smaller banks were not available.
Table 3.1 provides the list of these 22 banks
Table 3.1 Sample of Vietnamese banks
Bank ID Abbreviation Used Full name
1 ABB An Binh Commercial Joint Stock Bank
Vietnam Bank for Agriculture and Rural Development
Bank for Investment and Development of Vietnam
Vietnam Export Import Commercial Joint Stock Bank
6 HBB Hanoi Building Commercial Joint Stock Bank
The VNR500 ranks the top 500 largest enterprises in Vietnam by revenue, akin to the Fortune 500 model For more information, visit the official ranking table at vnr500.com.vn.
Housing Development Commercial Joint Stock Bank
8 MB Military Commercial Joint Stock Bank
9 MHB Housing Bank of Mekong Delta
10 MSB Maritime Commercial Joint Stock Bank
11 OCB Orient Commercial Joint Stock Bank
12 SEAB Southeast Asia Commercial Joint Stock Bank
13 SGB Saigon bank for Industry & Trade
14 SHB Saigon-Hanoi Commercial Joint Stock Bank
15 PNB Southern Commercial Joint Stock Bank
Saigon Thuong Tin Commercial Joint Stock Bank
Vietnam Technological and Commercial Joint Stock Bank
18 VAB Viet A Commercial Joint Stock Bank
20 VCB Bank for Foreign Trade of Vietnam
21 ICB Vietnam Bank for Industry and Trade
Vietnam Joint Stock Comercial bank for Private Enterprises
Measuring the technical efficiency and productivity of banks presents challenges, particularly in defining the inputs and outputs of banking activities According to the banking theory literature, notably the work of A N Berger and D B Humphrey (1997), there are two primary approaches to assessing the services provided by financial institutions: the production approach and the intermediation approach.
The production approach defines a financial institution as a service provider for account holders, focusing on transaction management for deposit accounts and document processing for loans This perspective is particularly useful for assessing the efficiency of financial institution branches.
The intermediation approach posits that financial firms serve as intermediaries between savers and investors, making it a suitable method for evaluating entire financial institutions This approach takes into account interest expenses, which can represent a significant portion of a bank's total costs, often ranging from one-half to two-thirds Additionally, the intermediation approach is advantageous for assessing the frontier efficiency of financial institutions, as it emphasizes the minimization of total costs—beyond just production costs—to maximize profits.
The approach of input and output definition used in this study is a variation of the intermediation approach, which was originally developed by Sealey and Lindley
(1977) The intermediation approach posits total loans and securities as outputs, whereas deposits along with labor and physical capital are defined as inputs
The study adopts the intermediation or asset approach to define bank inputs and outputs, following the works of Drake (2003), Sathye (2001), and Fukuyama (1993, 1995) The primary targets for bank outputs are interest income and non-interest income, as identified in the research conducted by Cevdet A Denizer and Mustafa Dinc (2000), Matthews and Tripe (2002), and Richard S Barr and Kory A.
Killgo, and Thomas F Siems (1999) [87], Thomas, 100 F Siems and Richard, S Barr
In this study, outputs are categorized into interest income and non-interest income, which includes fees, commissions, foreign currency and gold dealings, and investment income, all of which are vital earning assets for commercial banks The author identifies three key inputs for banks: labor, measured by the number of employees; fixed assets, representing the book value of these assets on balance sheets; and customer deposits, which are crucial from an asset perspective Additionally, intermediate materials serve as a proxy for other operating expenses, excluding labor costs and depreciation.
Accordingly, three inputs and two outputs would be used consisting of: y1: Interest income y2: Non-interest income x1: Personnel expenses (Labor) x2: Fixed assets (Physical Capital) x3: Savings deposits (Deposits)
Figure 3.2 Relationship between input-output factors and efficiency of commercial banks
Generally speaking, the product of inputs and outputs in a DEA application should optimally be less than the sample size in order to effectively distinguish the banks
Therefore, the author uses three inputs (labor, capital, and deposits) and two outputs (interest income and non-interest income).
Data analysis techniques
After completing data collection, descriptive statistics were performed to summarize the sample The discussion starts with Farrell's original concepts, which were represented in input/input space, emphasizing an input-reducing approach known as input-oriented measures.
Farrell demonstrated his concepts through a straightforward example where firms utilize two inputs (x1 and x2) to generate a single output (y), assuming constant returns to scale Understanding the unit isoquant of a fully efficient firm, depicted by SS’ in Figure 3.3, allows for the assessment of technical efficiency Typically, the technical efficiency (TE) of a firm is quantified by the ratio.
TE I = 0Q/0P, (1) which is equal to one minus QP/0P
Allocative efficiency (AE) can be determined when the input price ratio, indicated by line AA’ in Figure 3.3, is known For a firm operating at price P, allocative efficiency is defined by the ratio \$AEI = \frac{0R}{0Q}\$.
Total economic efficiency (EE) is expressed as the ratio \$EEI = \frac{0R}{0P}\$, where the distance \$RP\$ can also signify a reduction in costs It is important to note that the combination of technical and allocative efficiency yields the overall economic efficiency.
Figure 3.3: Technical and Allocative Efficiencies
TEI x AEI= (0Q/0P) x (0R/0Q) = (0R/0P) = EEI (4) Note that all three measures are bounded by zero and one
Efficiency measures rely on the assumption that a firm's production function is fully known; however, this is often not the case in practice To address this, the efficient isoquant must be estimated from sample data Farrell proposed two methods for this estimation: (a) a non-parametric piecewise-linear convex isoquant that ensures no observed data points fall to the left or below it, and (b) a parametric function, such as the Cobb-Douglas form, also fitted to the data with the same constraint He demonstrated these methods using agricultural data from the 48 continental states of the U.S.
Output-oriented measures can be analyzed by examining a scenario with two outputs (y₁ and y₂) and a single input (x₁) Assuming constant returns to scale, the technology can be illustrated using a unit production possibility curve in two dimensions In this context, the line ZZ′ represents the unit production possibility curve, while point A indicates an inefficient firm, as it lies below the curve, demonstrating that ZZ′ marks the upper limit of production capabilities.
Figure 3.4: Technical and Allocative Efficiencies from an Output Orientation
In Figure 3.4, the distance AB illustrates technical inefficiency, indicating the potential increase in outputs without additional inputs Therefore, the output-oriented technical efficiency can be quantified using the ratio \$TEO = \frac{0A}{0B}\$ (5).
If we have price information then we can draw the is revenue line DD, and define the allocative efficiency to be AEO= 0B/0C (6)
Overall economic efficiency can be understood as the combination of revenue-increasing and cost-reducing interpretations, similar to the concept of allocative inefficiency in the input-oriented context.
EEO= (0A/0C) = (0A/0B)(0B/0C) = TEOxAEO (7) Again, these three measures are bounded by zero and one
This study builds on Coelli (1996) by exploring various assumptions, including the variable returns to scale (VRS) model proposed by Banker, Charnes, and Cooper (1984) The discussion of Data Envelopment Analysis (DEA) starts with an overview of the input-oriented constant returns to scale (CRS) model in section 3.4.1, as it was the first model to gain widespread application.
3.4.3.1 The Constant Returns to Scale Model (CRS)
The introduction of Data Envelopment Analysis (DEA) is effectively achieved through its ratio form, which measures the output-to-input ratio for each Decision Making Unit (DMU) This is represented mathematically as \$\frac{u^T y_i}{v^T x_i}\$, where \$u\$ is an \$M \times 1\$ vector of output weights and \$v\$ is a \$K \times 1\$ vector of input weights To determine the optimal weights, we formulate a mathematical programming problem: maximize \$\frac{u^T y_i}{v^T x_i}\$ subject to the constraint \$\frac{u^T y_j}{v^T x_j} \leq 1\$ for \$j = 1, 2, \ldots, N\$ with \$u, v \geq 0\$.
To maximize the efficiency measure of the i-th Decision Making Unit (DMU), it is essential to identify values for u and v while ensuring that all efficiency measures remain at or below one However, a significant challenge with this ratio formulation is the presence of an infinite number of potential solutions.
To avoid this issue, one can impose the constraint \$\nu' x_i = 1\$, leading to the maximization problem: \$\max_{\alpha, \nu} (\alpha' v_i)\$ subject to \$\nu' x_1 = 1\$ and \$\alpha' y_j - \nu x_j \leq 0\$ for \$j = 1, 2, \ldots, N\$, with \$\alpha, \nu \geq 0\$ The change in notation from \$u\$ and \$v\$ to \$\alpha\$ and \$\nu\$ signifies a transformation, and this formulation is recognized as the multiplier form of the linear programming problem.
By applying duality in linear programming, one can formulate an equivalent envelopment problem defined as minimizing \$\theta\lambda\$ subject to the constraints \(-y_i + Y\lambda \geq 0\) and \(\theta x_i - X\lambda \geq 0\) with \(\lambda \geq 0\) In this formulation, \$\theta\$ represents a scalar, while \$\lambda\$ is an N×1 vector of constants This envelopment form is advantageous as it has fewer constraints compared to the multiplier form, making it the preferred approach for solving such problems.
The efficiency score for the i-th Decision-Making Unit (DMU) is represented by the value of \$\theta\$, which must satisfy \$\theta \leq 1\$ A score of 1 indicates that the DMU is technically efficient and lies on the frontier, as defined by Farrell (1957) It is important to note that the linear programming problem needs to be solved N times, once for each DMU in the sample.
The issue of efficiency measurement in Data Envelopment Analysis (DEA) stems from the piecewise linear nature of the non-parametric frontier This challenge arises because the segments of the piecewise linear frontier, which align parallel to the axes, are not typically found in most parametric functions For example, in Figure 3.6, the Decision-Making Units (DMUs) utilizing input combinations C and D represent the two efficient DMUs that establish the frontier, while DMUs A and B are identified as inefficient.
The Farrell (1957) measure of technical efficiency gives the efficiency of DMU’s
Conclusion
This chapter outlines the research methodology and procedures utilized in this study, justifying the chosen approach It emphasizes the selection of the sample and the analytical methods used to evaluate the propositions and address the research questions.
This chapter also described the statistical methods employed for data analysis, including labor expenses, physical capital, and savings deposit The next chapter reports the results of the data analysis.
Empirical results of the research
Introduction
This chapter presents the results of the main study, following the research methodology discussed in Chapter 2 Section 4.2 provides descriptive statistics of the sample, focusing on an analysis of 22 Vietnamese commercial banks using Data Envelopment Analysis (DEA) to assess their efficiency The findings of the efficiency estimates are detailed in Section 4.3, which also includes the Malmquist index to indicate productivity improvements The data analysis was conducted using the software DEAP 2.1, and the chapter concludes with a summary of the findings.
Descriptive Statistics
The study identifies the optimal number of inputs and outputs based on available data, measuring labor (L) through total labor expenses, capital (K) by the value of physical capital, and deposits (D) as the sum of savings and other deposits Additionally, the analysis includes input prices to assess cost efficiency, with the unit price of labor calculated as the total cost of bank employees divided by the total number of employees The unit price of capital is determined by physical capital expenses relative to the value of physical capital, while the price of deposits is derived from total interest expenses on deposits divided by the total of savings and other deposits.
The below table summarizes relevant variables and their definitions
Table 4.1: Summary of Variables on Vietnamse commercial banks, 2007-2009
Output Input input prices y1 y2 x1 x2 x3 w1 w2 w3 interest income
Non- interest income labor capital deposits price of labor price of physica l capital the price of deposit s
Definition Operating income total labor expense s
Fixed assets saving deposits and other deposits labor expense s/ number of employ ees
Fixed assets expense s/x2 total interest expense s/x3
Average 7218020 946314 651943 543030 49448490 93.720 3.282 0.144 max 45021387 6567211 5111540 3176455 299954030 174.143 6.742 0.273 min 1031749 38627 68380 64178 4336883 48.573 0.655 0.047 Std.dev 10323824 1463376 1215696 678665 73844114 30.210 1.838 0.055
Source: Author’s estimates based on 22 Vietnamese commercial banks Annual reports
Table 4.1 summarizes the statistics for the maximum, minimum, average, and standard deviation of the variables utilized as inputs and outputs in the models for estimating efficiency measures The data clearly indicates that commercial banks in Vietnam exhibit significant diversity in both size and activity.
Input variables as Labor, Capital, Deposit quantity tends to increase over time, but Deposit increased strongly in 2009 This may explain by
1.) the trend of increasing savings in the population, 2.) the easing policy of State bank in credit operations
3.) and importantly, it is due to innovation and growth of the commercial banking system, especially the Vietnam Joint Stock Commercial Bank in 2007
Figure 4.1 Trend of interest income and non-interest incomes, 2007-2009
Source: Author’s estimates based on 22 Vietnamese commercial banks Annual reports
Figure 4.1 illustrates the trends in interest and non-interest income among 22 banks in the research sample, revealing that interest income remains the primary revenue source Although non-interest income has shown a gradual increase over time, it still constitutes a minor portion of total income This indicates that the banks primarily rely on credit provision, which may expose them to systemic risks.
Analysis of Efficiency Estimates
The empirical strategy involves estimating efficiency scores and Malmquist indexes, followed by deriving statistical inference from these components The author utilized the DEAP 2.1 software, developed by Tim Coelli in 1996, to obtain detailed information on distances, efficiency scores, and Malmquist indexes Table 4.2 presents a summary of the estimated efficiency scores throughout the study period, while Table 4.3 outlines a summary of five efficiency scores Additionally, Table 4.4 provides a comprehensive presentation of the estimated five efficiency scores for each bank.
Table 4.2: Technical efficiency of commercial banks, 2007-2009
Source: Author’s estimates based on DEA result
Table 4.2 reveals the technical efficiency (TE) scores of commercial banks from 2007 to 2009, indicating a significant increase in average TE in the last two years, with the peak score of 0.864 in 2009 Private banks (JSCB) demonstrated higher efficiency than state-owned commercial banks (SOCB), achieving average TE scores of 74.9% compared to 74.2% This suggests that JSCBs utilized their resources more effectively during this period, likely due to advantages such as better risk management, reduced financial crisis pressures, increased customer trust, and enhanced competitiveness in fundraising and expansion efforts.
Among the State-Owned Commercial Banks (SOCB), Vietcombank (VCB) and BIDV demonstrated higher technical efficiency (TE) compared to the Joint-Stock Commercial Banks (JOCB), indicating their strong performance Conversely, Vietinbank (ICB) recorded the lowest TE score, particularly in 2009, due to poor outcomes during the financial crisis Overall, the average technical efficiency for the 22 commercial banks in the study was 0.748, suggesting that these banks operated at only 74.8% efficiency, resulting in a resource wastage rate of 33.7%.
Table 4.3: Summary of Estimated Efficiency Measures, 2007-2009 Year ALL OB S Mean Std.Dev Maximum Minimum Obs
6 The relat ionship between technical effic iency (TE) and inefficiency (ITE) as TE = 1 / (1 + ITE)
Note: CE = cost efficiency; AE = allocative efficiency; TE = technical efficiency;
PE = pure technical efficiency; and SE = scale efficiency Source: Author’s estimates based on DEA result
During the study period, the efficiency scores of the banks showed an upward trend, with the mean score of cost efficiency (CE) recorded at 52.7% in 2007, 56.6% in 2008, and 51.9% in 2009 Notably, the mean technical efficiency (TE) was 0.748, surpassing the mean allocative efficiency of 0.722 This indicates that the primary source of cost inefficiencies in Vietnamese banks is likely due to regulatory issues rather than the managerial capabilities of the banks involved in the study.
The mean score of the scale efficiency (SE) for Vietnamese banks was 0.862, which is slightly lower than the pure technical efficiency (PE) score of 0.868 during the observed period Additionally, the average scale inefficiency was 0.160, while the pure technical inefficiency was 0.152 These findings indicate that the observed technical inefficiency is likely due to scale inefficiency rather than pure technical inefficiency.
Table 4.4: Estimated Efficiency Scores for Individual Banks, 2007-2009
Bank ID AE CE TE PE SE Scale type
7 The relat ionship between effic iency (E) and ineffic iency (IE) as E = 1 / (1 + IE)
Note: CE = cost efficiency; AE = allocative efficiency; TE = technical efficiency;
PE = pure technical efficiency; and SE = scale efficiency Source: Author’s estimates based on DEA result
The model estimates presented in Table 4.4 indicate that three out of four state-owned commercial banks are experiencing decreasing returns to scale (DRS) To enhance their operational efficiency, these banks should prioritize the development of new products rather than expanding existing ones In contrast, most joint-stock commercial banks are positioned to continue market expansion, as they operate under increasing returns to scale (IRS).
Table 4.5 Number of banks with DRS, IRS, and Cons, 2007 -2009
Source: Author’s estimates based on DEA result
Table 4.5 presents the model estimates for the operational scale of commercial banks in Vietnam, revealing that among the 22 banks analyzed, nine exhibit increasing returns to scale, nine operate on the efficient frontier, and four experience decreasing returns to scale This indicates that 18.2% of the banks are currently underperforming, while 41% are exceeding optimal size The findings suggest that several banks have maintained constant returns to scale over the years, implying that further scaling up their performance could enhance overall efficiency.
Productivity Improvement with Malmquist index
Table 4.6 summarizes the annual Malmquist index, with all indices relative to the previous year, using 2007 as the base year and starting output from 2008 Additionally, Table 5 illustrates the productivity changes for each bank in the sample (further details can be found in Appendix 3).
Table 4.6: Malmquist Index summary of Annual Means Year Type of banks obs effch techch pech sech tfpch
Note: effch = technical efficiency change; techch = technical or technology change; pech = pure technical efficiency change; sech = scale efficiency change; and tfpch = total factor productivity change
According to Coelli et al (1998), productivity changes are influenced by both technological progress and technical efficiency The findings indicate that the average Malmquist index (tfpch) was 0.975, signifying a 2.5% decline in total factor productivity (TFP) during the study period, despite a mean technical efficiency index (effch) of 1.141 The primary factor contributing to the decrease in TFP was a technological change index (techch) of only 0.855, reflecting a 14.5% decline Furthermore, the technical efficiency change index (effch) rose from 1.043.
In 2009, the technical efficiency index improved to 1.249, reflecting a 20.6% increase from 2008 However, the technological change index saw a significant decline, dropping from 1.201 in 2008 to 0.608 in 2009, which represents a 59.3% decrease Consequently, the total factor productivity (TFP) index experienced a net reduction of 49.2%, falling from 1.252 in 2008 to 0.760 in 2009.
The study indicates that technological progress has not significantly boosted total factor productivity (TFP), primarily due to a 14.5 percent decline in the rate of mean technological progress during the period analyzed The findings reveal that changes in total factor productivity are more closely linked to improvements in technical efficiency (effch) rather than technological change (techch) This trend suggests that technological advancements were not fully realized, with many banks continuing to rely heavily on labor.
Considering different types of banks, the average of factor productivity change of SOCBs is 0.977 (or a decline of 2.3%), while the aveage of this index of JOCBs is 1.084 (or improvement of 8.4%)
Table 4.7 Malmquist Index Summary of Individual Commercial Bank Means,
2007-2009 Bank ID effch techch pech sech tfpch
Among state-owned commercial banks, VCB demonstrates the highest technical efficiency with a total factor productivity change (tfpch) of 1.26, as shown in Table 4.7 Additionally, several joint-stock commercial banks, including ACB, STB, TCB, VIB, and EIB, also exhibit technical efficiency levels exceeding 1.
MB, PNB, and VPB This indicates that JOCB group have increasing trend in productivity.
Conclusion
This chapter reveals the empirical findings of the study, highlighting that Vietnamese banks have significant potential to enhance their performance without additional resource investment The average efficiency improvements for the entire sample are 31.7%, 30.4%, and 13.6% for the years 2007, 2008, and 2009, respectively.
This study estimates non-parametric Malmquist TFP indexes and its components for Vietnamese commercial banks in the sample period 2007-2009 We found that there is a decreasing trend in productivity.