INTRODUCTION
Problem statements
Banks play a crucial role in economic development by connecting surplus and deficit capital among economic agents Their financial intermediation capabilities have prompted numerous empirical studies aimed at improving bank efficiency and stability Recent changes in the business environment, driven by economic growth, globalization, deregulation, and technological advancements, have created competitive pressures and instability within the banking industry.
Financial deregulation, as outlined by Rosengard & Du (2009), refers to the shift from a closed financial system to a competitive one, particularly within the banking sector This transition involves moving from a monopoly or oligopoly, characterized by restricted competition and limited bank entry, expansion, and diversification, to a more open and competitive banking environment The government aims to establish "a level playing field" among privately owned banks, public sector banks, foreign banks, and domestic banks.
This study investigates the effects of bank deregulation on the efficiency of Vietnamese banks, focusing on improvements in domestic bank performance post-deregulation Rather than contrasting the outcomes of deregulation with those of a regulated environment, we benchmark the efficiency of Vietnamese domestic banks against that of foreign banks Following the findings of Claessens et al (2001), which suggest that foreign banks tend to outperform domestic ones in developing countries like Vietnam, we posit that foreign banks possess superior governance structures, skilled labor, and technological innovations Notably, this research is unique as it utilizes a dataset that includes foreign banks in Vietnam for the first time.
Farrell (1957) introduced the theory of efficiency, defining three types: technical efficiency (TE), allocative efficiency (AE), and economic efficiency (EE) A bank is considered efficient when it maximizes outputs from a set of inputs (technical efficiency) and optimally allocates inputs based on their prices (allocative efficiency) Economic efficiency is the product of allocative and technical efficiency In economic literature, cost and profit efficiency are often viewed as dual aspects of economic efficiency Cost efficiency focuses on minimizing costs to achieve a specific output level, while profit efficiency analyzes how to maximize profits with a given set of inputs, both evaluated through modified cost functions.
In assessing bank efficiency, two primary methodologies are utilized: Data Envelope Analysis (DEA) and Stochastic Frontier Analysis (SFA) DEA, introduced by Charnes, Cooper, and Rhodes in 1978, is a non-parametric approach that employs linear programming to identify best practices Conversely, SFA, developed by Aigner et al., Meeusen, and Van den Broeck in 1977, is a parametric technique that separates residuals from an estimated function into a stochastic error term and efficiency component In this study, we adopt the SFA estimation procedures proposed by Berger et al (2009), Ahn, Lee & Schmidt (2001), and Good & Sickles (1995) instead of the DEA method.
This study, based on Sickles (1984), analyzes the cost and profit efficiency of foreign banks, the big four, and other domestic banks using data from Bankscope (Bureau van Dijk) and annual reports of 37 banks for the post-WTO period from 2009 to 2015 Our findings indicate that while foreign banks are not the most efficient overall, the big four banks demonstrate notable efficiency in both cost and profit metrics Additionally, we explore the underlying reasons for these results.
The evolution of Vietnam’s banking sector
The banking industry faces both opportunities and challenges as it embraces modern technologies to improve market efficiency while navigating intense competition from financial deregulation This shift marks a transition from traditional banking monopolies or oligopolies, characterized by limited competition, to a more open and competitive banking system that encourages market entry, expansion, and diversification.
Before 1945, Vietnam was a feudal-colonial country under the French colonialists’ rule The banking and credit system namely Indo China bank was established to serve the French colonialists.
During the period of 1945 – 1986, the Vietnam’s banking system had functioned as a mono-banking model and had not been market-oriented (Siregar,
In 1999, the State Bank of Vietnam (SBV) served as a crucial instrument for government budgeting and policy implementation, providing fiscal resources to the state and financing for State-Owned Commercial Banks (SOCs) It managed all financial transactions, operating simultaneously as both a central bank and a commercial bank.
Following the Sixth National Congress of the Communist Party in 1986, Vietnam's banking system underwent a significant transformation from a one-tier to a two-tier structure The State Bank of Vietnam (SBV) emerged as the central bank, tasked with regulating monetary policy and overseeing commercial banks Consequently, commercial banks and credit institutions took on the responsibility of conducting banking operations, including currency trading, foreign exchange, payment services, credit, and other banking activities in accordance with the law.
Registered capital: No less than 3000 billion VND forRegistered capital no less than 3000 billion VND-
At least 10 billion US dollars in total assets in previous year The capital adequacy ratio must exceed 8%.
Requirements for establishment all commercial banks -
Figure 1: Structure of banking system in Vietnam
The State bank of Vietnam
Joint-stock commercial banks Foreign banks
(Source: the State Bank of Vietnam, 2016)
The primary goal of this reform is to diversify ownership, strengthen private oversight, enhance stability, and maximize competition Specifically, it involves the establishment of Joint Stock Commercial Banks (JSCBs) to achieve these objectives.
1990s and foreign banks have been licensed to participate in a type of branches or joint ventures with domestic banks.
However, the government conducted some laws which were further restrict foreign banks entry in term of products, a possibility of operations and registered capital.
Table 1: Laws of domestic banks and foreign banks
The ratio of NPLs is lower than 3% at the previous year. Positive profit at least 3 years previous.
The loan-deposit ratio - The
(LDR) is less than 90% (Big 90% four); 80% (other domestic less than
No.36/2014/TT-No.36/2014/TT-NHNN)
The ratio of loans to a single borrower is less than 15% of its total equity
The ratio of loans to a single borrower is less than 15% of its total equity
9% (Government Circular No.36/2014/TT-NHNN) - 9% (Government Circular
Make loans, take deposits, business credit card, insurance agency, settlement of accounts, bill acceptance.
Meet the legal deposit reserve requirement set by SBV
After WTO entry in 2007, Vietnam conducted a series of financial deregulation with the purpose of enhancing the efficiency and stability of the banking system.
In line with its WTO commitments, the Vietnamese government issued Decree No 22/2006/NĐ-CP, permitting foreign institutional investors to own up to 15% of shares in domestic banks, with a total cap of 30% for all foreign holdings Strategic investors can increase their stake to 20% Additionally, from April 1, 2007, Vietnam allowed the establishment of wholly foreign-owned banks, with HSBC Vietnam being the first to set up in early 2009 By 2017, Vietnam had welcomed seven fully foreign-invested banks, including HSBC (Hong Kong), ANZ (Australia), Shinhan (South Korea), Standard Chartered (UK), Hong Leong (Malaysia), Public Bank (Malaysia), and Woori Bank (South Korea).
The upcoming deregulation focuses on bank ownership diversification, highlighting strong governmental support for privatization The central bank advocates for the transition from state to private ownership, as evidenced by Government Notice No 03/TB-VPCP, which reported a 49% reduction in state ownership by 2010.
The competition between banks has a dramatic improvement since 1990s As the number of joint stock commercial banks increase from 4 in 1991 to 51 in 1997.
The rapid changes in the banking sector have led to increased instability, including cross-ownership and a rise in non-performing loans (NPLs) To enhance the stability of the banking industry, the State Bank of Vietnam (SBV) has implemented new laws promoting mergers and acquisitions, as well as bank restructuring As a result, the number of domestic banks decreased from 42 to 34 between 2006 and 2015.
Table 2: A snapshot of Vietnamese Banks in 2015 (bil VND)
The research objective
Investigating the economic efficiency of the Vietnamese banking system under the impact of Vietnam’s financial deregulation since 2007.
Organization of the thesis
This study includes five chapters, which can be structured as follows
Chapter 01 introduce the problem statements.
Chapter 02 provides a brief review of previous studies
Chapter 03 describes methodology and data.
Chapter 04 will discuss the results
LITERATURE REVIEW
The theoretical literature
2.1.1 Theory of productivity and efficiency
Productivity and efficiency are often confused, but they have distinct meanings Productivity, as defined by Vincent (1968), refers to the ratio of output to input It encompasses various forms, including total factor productivity (TFP), which considers all production factors Common metrics for measuring productivity in firms include labor productivity, fuel productivity, and land productivity.
Calculating the productivity of a firm using a single input to produce one output is straightforward However, when multiple inputs and outputs are involved, researchers often refer to this concept as efficiency (Grosskopf and Lovell, 1994; Siems and Barr, 1988).
Efficiency is defined as the optimal relationship between input and output, representing the production frontier It signifies a performance level in a process that maximizes outputs while minimizing inputs.
In 1957, Farrell introduced the theory of efficiency, identifying three types: technical efficiency (TE), allocative efficiency (AE), and economic efficiency (EE) Technical efficiency is defined as the ratio of inputs used by a fully efficient firm to those used by the firm in question, aiming to minimize resource waste—such as materials, labor, energy, and time—while enhancing management quality Firms operating on the production frontier are considered technically efficient, while those below it demonstrate varying degrees of inefficiency When information on input and output prices is accessible, and profit maximization and cost minimization assumptions hold, allocative efficiency can be assessed Allocative efficiency refers to the optimal use of inputs for cost minimization, and a firm must achieve both technical and allocative efficiency to be considered fully efficient The product of allocative and technical efficiency yields the overall economic efficiency.
EE = TE * AE Economic efficiency may be obtained from two approaches, input and output.
Cost reduction involves minimizing the inputs required to produce a specific output quantity In this context, a firm utilizes two inputs, x1 and x2, to generate output y The isoquant YY' represents the efficient frontier, showcasing the minimum input combinations necessary for a given output level, indicating fully efficient firms Conversely, the iso-cost line CC’ illustrates the optimal input ratio for achieving the lowest cost If a firm operates at point P on this line, the technical inefficiency can be measured by the distance QP, suggesting that the firm can reduce inefficiency without decreasing output Technical efficiency (TE) is quantified by the ratio OQ/OP, with values ranging from 0 to 1, reflecting the firm's efficiency in utilizing its inputs.
1 If the value equals 1, the firm may be seen as fully technically efficient For instance, since the point Q and Q’ lie on the isoquant line, it gains maximum technical efficiency While allocative efficiency (AE) may be calculated by the rate of OR/OQ Then two measures above are grouped together to form an economic efficiency.
According to Colie (2005), overall cost efficiency equates to economic efficiency when comprehensive input price information is available The research question (RQ) highlights the potential reduction in production costs necessary to achieve both technical and allocative efficiency at point Q', as opposed to merely reaching technical efficiency with allocative inefficiency at point Q Ariff and Can (2008) further define cost efficiency as the firm's endeavor to minimize input costs while maintaining a consistent level of output.
Figure 2: Economic efficiency following input approach x2/y
Input-oriented measures assess the extent to which input levels can be reduced without affecting output, while output-oriented measures evaluate how much output can be increased without altering input levels This concept is illustrated by the production frontier, which indicates the maximum output achievable with a given input In our scenario, a firm utilizes a single unit x to produce two outputs, y1 and y2, represented by the production possibility curve ZZ’ An inefficient firm is identified at point A, with technical efficiency calculated as OA/OB The distance AB highlights the technical inefficiency, suggesting that output can be increased without additional input to achieve both technical and allocative efficiency at point B’, rather than remaining at point B, which reflects technical efficiency but allocative inefficiency.
Figure 3: Economic efficiency following output approach
According to Colie (2005), overall revenue efficiency can be equated to economic efficiency when complete price information for both inputs and outputs is available With this comprehensive price data, the iso-revenue DD’ line can be established, allowing for the calculation of allocative efficiency using the formula AE = OB/OC.
Finally, the economic efficiency is estimated by overall technical and allocative efficiency.
• Following the previous research, there are two main approaches to assess the efficiency of banks, data envelope analysis (DEA) and stochastic frontier analysis (SFA)
2.1.2 The method of efficiency measurement
In 1978, Charnes, Cooper, and Rhodes introduced Data Envelopment Analysis (DEA), a non-parametric linear programming method designed to identify optimal frontier observations This model evaluates n banks (i = 1, …, n) that utilize a vector of m inputs, xi = (xi1, …, xim), incurring costs represented by prices ci = (ci1, …, cim) to generate a vector of s outputs, yi = (yi1, …, yis), which are sold at prices ri = (ri1, …, ris) The DEA technique effectively estimates cost and profit efficiency.
The cost efficiency of bank j may be described in linear programming as follows:
The idea behind this model is the optimal inputs demand vector
Bank j's cost efficiency (CEj) is assessed by comparing its optimal production costs to its actual costs, highlighting the importance of minimizing expenses while considering the current input prices.
• Prof� it effic iency (PE)mod el
Suggested by Fare and Grosskopf (1997) and developed by Fare (2004), the profit efficiency of bank j may be described in linear programming as follows:
The solution to this model is the connection of the ideal outputs supply vector � ∗ ∗ � ∗ ∗
(� �1 , … , � �� ) and the potential input demand vector
The profit efficiency of bank j (PEj) is determined by the ratio of its actual profit to the optimal profit, which maximizes its profitability given the current prices of inputs and outputs.
Meeusen and van den Broeck (1977) and Aigner et al (1977) pioneered the stochastic frontier analysis method, which has since gained popularity for estimating firm efficiency Numerous studies have built upon their original models, leading to the creation of various formulas and extensions The methodology evolved from a general stochastic frontier cross-sectional model to more advanced panel-sectional models, with significant contributions from researchers like Pitt and Lee (1981).
Panel data offers a more comprehensive information set, leading to highly accurate assessments of inefficiencies The initial generalization of the panel model was introduced by Pitt and Lee in 1981, utilizing maximum likelihood estimation of the half-normal distribution This approach assumes that inefficiencies are time-invariant while varying across different banks, indicating that the inefficiency term can be defined accordingly.
Schmidt and Sickles (1984) introduced a model tailored for the Truncated-normal case, which allows for the application of a stochastic frontier model with time-invariant efficiency by modifying traditional fixed effect estimation This approach suggests that inefficiency can be correlated with the frontier of independent variables, relaxing certain assumptions regarding the inefficiency term, u i Additionally, Battese and Coelli (1988) contributed to this field with further advancements.
Empirical studies
Research on the Vietnamese banking system is limited due to data availability, resulting in few published studies Most scholars focus on the deregulation process of Vietnam's financial system and the factors influencing bank performance Notably, there has been a lack of research examining the impact of foreign bank entry on the efficiency of domestic banks, whether positive or negative.
Nguyen (2007) conducted an evaluation of cost efficiency among 13 commercial banks in Vietnam from 2001 to 2003 using the Malmquist DEA approach to analyze productivity, efficiency, and technical changes The findings indicated that neither technical nor allocative efficiency was achieved during this period However, the study faced limitations due to time constraints and the limited number of banks analyzed.
A study by Nguyen (2011) assessed the efficiency of 20 commercial banks in Vietnam over a three-year period from 2007 to 2010, evaluating various efficiency determinants The analysis considered inputs such as labor, fixed assets, and deposits, while outputs included interest and non-interest income The findings revealed that state-owned banks exhibited lower efficiency scores compared to their joint-stock commercial bank counterparts.
In their 2012 study, Nguyen and colleagues analyzed 32 commercial banks from 2001 to 2005 using the slacks-based model of Data Envelopment Analysis (DEA), concluding that there is significant potential for enhancing the efficiency of these banks.
Nguyen (2011) and Lieu and Vo (2012) assessed the efficiency of commercial banks in Vietnam from 2008 to 2012, utilizing labor, total deposits, and various expenses as inputs to evaluate total loans as outputs Their findings revealed that banks with larger asset sizes demonstrated efficiency levels 11 times greater than those of smaller banks.
Vu and Turnell (2010) employed a Bayesian stochastic frontier approach to assess the cost efficiency of the Vietnamese banking system, which was found to be relatively high at approximately 0.87, although it experienced a slight decline from 2000 to 2006 Additionally, their findings indicated that there is no significant difference in cost efficiency between state-owned banks and joint stock banks.
In 2010, Ngo utilized the Data Envelopment Analysis (DEA) method to assess the efficiency of 22 Vietnamese commercial banks from 2008, all of which were listed among the top 500 largest enterprises by revenue according to VNR500 in 2009 Despite an impressive average efficiency score of 0.946, the findings indicated that these banks still had potential for further efficiency improvements.
Nguyen, Nghiem, Roca, and Sharma (2016) conducted a study on the cost efficiency of 32 Vietnamese banks from 2000 to 2014, utilizing the translog stochastic cost frontier method They analyzed input prices, including financial capital, physical capital, and labor, against outputs such as net loans and other earning assets The findings revealed that Vietnamese banks demonstrated strong cost management, with an average cost efficiency of 0.93, and state-owned commercial banks (SOCBs) exhibited greater efficiency compared to joint-stock commercial banks (JSCBs) at a 5% significance level Despite previous research highlighting limitations in understanding the banking system's efficiency in Vietnam, particularly before 2010, a slight upward trend emerged post-2010 Building on existing literature, this study aims to further investigate efficiency scores and gaps among three categories of banks: the Big Four, foreign banks, and other domestic banks.
2.2.2 International research of the efficiency effects of deregulation
Efficiency improvement of the financial system is a key objective of deregulation.
The entry of foreign banks is anticipated to enhance overall bank performance and create spillover effects, as increased competition pressures all banks to minimize costs while still boosting profits This may lead to careful research into new technologies and resource allocation However, empirical studies suggest that the entry of foreign banks can have negative impacts on the relative efficiency of domestic banks Claessens (2001) analyzed panel data from 7,900 banks across 80 countries between 1988 and 1995, revealing that foreign banks in developing countries achieved higher interest margins, returns, and tax payments compared to domestic banks, while adverse effects were noted in developed countries.
In developing countries, foreign banks are generally as efficient as or more efficient than domestic banks However, domestic banks can be categorized into private and state-owned institutions State-owned banks often receive favorable treatment and subsidies from the government, which can result in them being more efficient than foreign banks in certain developing nations.
One of the earliest research on the efficiency of Indian banking system conducted in
Between 1986 and 1991, 70 banks emerged during a period of liberalization According to Bhattacharyya (1997), in this deregulated environment, public sector banks demonstrated the highest efficiency, followed by foreign and private banks.
Research by Choi and Hasan (2005) and Berger (2009) indicates that the entry of foreign banks influences bank performance Conversely, Jiang (2013) argues that while there may be a short-term negative impact, the long-term effects are positive Initially, foreign acquisitions can lead to a decline in bank performance due to the challenges of addressing previous inefficiencies.
– performing assets and rectifying the fraudulent accounting numbers by the foreign strategic investors.
Hsiao, Shen, and Bian (2015) analyzed the cost and profit efficiency of 107 Chinese banks from 2007 to 2012 to assess the impact of financial reforms since 1978 Their findings revealed that foreign banks exhibited higher cost efficiency compared to domestic banks during this timeframe, although domestic banks rapidly improved their performance under competitive pressure, narrowing the efficiency gap to 1 Interestingly, while the big four domestic banks demonstrated the highest profit efficiency, foreign banks lagged behind in this aspect, and the disparity between foreign and domestic banks has widened over time The Chinese government's financial reforms, particularly following its WTO accession in 2001, have aimed to foster a competitive banking environment.
Numerous studies have highlighted the advantages of financial deregulation, with Fry (1988), Fischer (1993), and Lozano (1997) suggesting that reforms in the financial sector can enhance the level of financial savings Research conducted by Maudos and Pastor (2003) analyzed the costs and profits across a sample of 14 European Union countries, as well as Japan and the US during the 1990s Their findings indicated that increased competition resulted in profit efficiency in Europe and the US, while Japan did not experience the same benefits.
Some literature highlights the limitations of financial deregulation, with Killick and Martin (1989) and Harris (1996) noting that such reforms in developing countries resulted in high nominal lending rates Early studies on technical efficiency and productivity, including research by Lovell and Leightner (1998) on Thai banks, Wilson and Gilbert (1998) on Korean banks, and Shyu (1998) on Taiwanese banks, demonstrated that banks adapted to deregulation by changing their input and output mix, which ultimately led to increased productivity.
DATA AND METHODOLOGY
Research methodology
Stochastic Frontier Analysis (SFA) is a parametric technique developed by Aigner et al., Meeusen, and Van den Broeck in 1977 This method effectively separates residuals into two components: inefficiency (u) and a stochastic error term (v) The stochastic error term accounts for random influences such as accidents, weather conditions, and chance events.
Let � �� ′ �� +be some observed performance measure for bank i in period t.= �− � ′ � + � (1)
Where xit is the factors that may affect the performance of banks The “stochastic
The one-sided error � �� ≥ 0 represents for the inefficiency If there are no inefficiency, meaning uit = 0, the frontier of outputs should be: ∗
In this analysis, we consider two firms, A and B, where Firm A utilizes input xA to generate output yA, and Firm B uses input xB for output yB The output frontier demonstrates diminishing returns to scale If Firm A's potential output exceeds the deterministic frontier, it indicates the presence of positive statistical noise (vA > 0) Conversely, Firm B experiences negative statistical noise, affecting its output.
(vB