Tóm tắt Luận án Liquidity positon and operational efficiency of Vietnamese commercial banks

hoice of many studies on measuring effectiveness. There are also many approaches to estimating, measuring and interpreting banking efficiency. For example, there are two measurable approaches to access to structure approach and nonstructural approach Hughes & Mester (2008). 1.2.2.1 The nonstructural approach Accordingly, the nonstructural approach to measure banking efficiency is a traditional and popular way of being used. How to measure through a series of financial indicators such as: ROE (return on equity), ROA (return on assets), ROS (return on sale), C/I (costs ratio) However, this approach has two disadvantages: one is that the market value of the asset, the degree of risk of two, is only appropriate when the bank uses a single input or produces only one output. Follow Wozniewska (2015 2015 ), Hughes and Mester Hughes & Mester (2008 )measuring bank efficiency by the index may not be 4 primary metric groups: - One is, the metric group reflects profitability such as ROE, ROS. - Two is, the indicator group reflects the structure of the balance sheet as the ratio of customers to total assets (DTA), credit balance of credit (LTA). - Three is, the indicator group reflects the quality of assets of banks such as bad debts (NPLs), total debt risk allowance ratio. - Four is, the metric group reflects market value and the risk of assets such as Tobin' s q, Sharpe index, the Capital Adequacy Factor CAR 1.2. 2. 2 Approaches According to Hughles and Mester Hughes & Mester (2008) the structural approach is usually based on the economic cost of the cost minimization or the maximum profit (profit maximization), which is shown through the cost of the cost or the profit function, or may be called the production function. From here, the research on banking efficiency is increasingly diversified and forms a variety of banking evaluation methods and the relationship between banking efficiency and other factors affecting the bank (such as economic growth, equity structure, etc.). Most of the following studies use the structure approach with the main methods to measure bank efficiency, as follows: One is, the parameter approach (parametric) with three main methods: (i) a random method (SFA)); (ii) thick frontier approach (TFA); (iii) and Distribution Free Approach analysis. Two is, the non - parametric approach (non parametric) with two main methods: (i) method for analyzing data (DEA), (ii) and method of free - factor treatment of Hull (FDH) CONCLUSION OF CHAPTER 1 CHAPTER 2: ASSESSMENT OF RESEARCH WORKS ON BANKING EFFICIENCY, FACTORS AFFECTING BANKING OPERATIONS AND RELATIONSHIPS BETWEEN LIQUIDITY AND BANKING OPERATIONS 2. 1 FUNDAMENTAL THEORY 2. 1. 1 Theory of involved parties (Stakeholders Theory) The theory of stakeholders determines that the company is established and operates for the benefit of all stakeholders, so when the decision of the board of management and the operator of the company must consider the benefits of all concerned stakeholders, such as traditional views. It does this by ensuring that the benefits of stakeholders are interested. Therefore, the bank needs to be careful in the state of liquidity regulation, avoiding the profit in order to increase the efficiency of the operation that can protect the interests of the involved parties. H1: The liquidity position is positive to performance in the commercial banks 2.1.2. Dynamic liquidity theory Almeida and associates. (2002) the proposed state of liquidity is based on the assumption that the choices related to liquidity will depend on the access to capital resources and importance of the use of resources in the operations of the banks. Expenses incurred in the absence of a higher liquidity shortage for banks with less efficient operations. Therefore, there will be a positive relationship between banking performance and liquidity position. H2: Active effect on liquidity position at commercial banks 2. 2 RESEARCH STUDIES ON BANKING PERFORMANCE 2. 2. 1 Studies on banking performance in the world commercial banks Berg et al. (1993 ) the study of the banking efficiency in Norway, Sweden and Finland used the data method, both ways to calculate the effectiveness of each territory and compare the three countries. The data from the study of 503 Finnish banks, 150 Norwegian banks and 126 Swedish banks. The author found that Swedish banks had 52 – 63% more effective than Finland and 40 – more effective than the Norwegian bank. At the same time, Sweden' s largest banks were the most effective units in the aggregates, so the group concluded that the Swedish bank was in the best position to expand the bank of Nordic banking. Fecher & Pestieau (1993) )use the SFA parameter method to assess technical efficiency for 11 financial institutions in OECD (Organization for Economic Cooperation and Development). They use total value added tax such as criteria for evaluating the output of each country' s financial services, and the employment in the financial sector and capital is the two inputs. Over there, they found the most effective financial services, and Denmark was the least effective. Most of the research on banking efficiency uses the structural approach (dimensions and dimensions) with the main methods for measuring such as random boundary approach (SFA) ; Thick Frontier Approach (TFA) ; Distribution Free Approach (DFA); data analysis approach (DEA); and method of free - factor treatment of Hull (FDH) 2. 2. 2 Research studies on banking efficiency in Vietnam . Huong (2002), the need to improve the efficiency of the bank' s business operations through investment activities to raise the competitiveness of commercial banks in the context of international economic integration. Population (2004) has developed an efficient evaluation system of commercial banks through statistical descriptions. Hung (2008) the study conducted the performance of 32 Vietnamese commercial banks in the period 2001 – 2005 through qualitative and quantitative methods. H. T. Vu & Turnell (2010) measuring cost efficiency (cost) by random boundary approach (SFA) in the Bayesian approach of the Vietnamese commercial bank. Research towards a reasonable estimate in estimating marginal costs and using Bayesian. Research results indicate that the effectiveness of the cost of Vietnamese commercial banks is very high, 87%. There is a small and insignificant difference in cost effectiveness between different banks of banks by ownership. However, during the period of research, the banking sector has a slight decline in cost. This is explained by the increase in the cost of managing diversified activities, extending the branch network and upgrading the banking platform. Dang - Thanh (2012 )the study used the analysis of DEA data analysis to analyze the performance of the Vietnamese banking system in 1990 – 2010. H. Vu & Nam (2013) conducting research on the factors affecting the efficiency of the bank of Vietnam in 2000 – 2006. Ngoc Nguyen & Stewart (2013) examined the level of concentration and efficiency of the Vietnamese banking system according to the structure structure (structural) – this is the name of the parameter and the parameter.The study used data of a sample of 48 Vietnamese commercial banks during 1999 – 2009. Empirical results show that the Vietnamese banking system is less concentrated, and the large commercial bank dominates the banking system. Minh et al. (2013 )the study of 32 commercial banks in Vietnam from 2001 – 2005 to identify factors affecting the banking efficiency. Morning (2015) )conducting the study of 48 Vietnamese commercial banks during 1992 – 2013 on banking efficiency and relationship with economic growth through three methods: financial index analysis; analysis of parameters with SFA random approach; method of non - parametric analysis of data (DEA) Through empirical evaluation of banking works. The author found most of the research on banking efficiency using two methods of DEA data and SFA. There are studies like resti (1997), bauer et al ,. (1998), Pelosi (2008), Hung (2008), Vuet al ,. (2010), Ngoc Nguyenet al. (2013) use both methods to measure efficiency in a country. As a result of the efficiency of banking in a country, dea data method is used increasingly common, especially in 2008, such as liang et al ,. (2008), staubet al ,. (2010),et al ,. (2010), yuet al ,. (2013), Replová (2014), Zimková (2014) At the same time, most of the research uses the DEA model, which allocates such as CCR, BCC, SBM, profit efficiency. 2. 2 EVALUATION OF RESEARCH WORKS RELATED TO THE EFFECT OF BANKING OPERATIONS Hoang (2016) is based on the model of the model Williams (2012)research on factors influencing the performance of the Vietnamese commercial bank system of commercial banks - 2011 according to the efficiency measurement method (SFA), and the regression method of regression analysis to analyze the impact factors of the banks. Results show that the effect is affected by 02 main groups: subjective factors (market share, liquidity, foreign investor holdings and bank size and objective factors (gross domestic and inflation). In which the positive factors are the proportion of foreign countries, the size and market share of the bank. Sufian (2009) analysis of factors influencing performance of Malaysian commercial banks in 1995 – 1999 around the crisis of East Asia in 1997. Research using the DEA analysis method to measure performance efficiency and regression analysis to evaluate factors affecting performance of Malaysian commercial banks. Variable the dependence of the model is the performance of the commercial bank by the DEA independence includes: (i) the size of the bank measured by natural logarithms of the total deposit, (ii) ratio of credit balance on total assets, (iii) bad on total assets, (iv) total income on total assets, (v) total expenses in addition to the interest on total assets, (vi) ratio of equity on total assets. Follow Williams (2012) use SFA method to analyze factors affecting the efficiency of commercial banks, including the power market (lerner index). ' ownership' and' listed dummy' are included as a false role: ownership receives the value' 1' when the bank has a foreign investor' s holding rate and the value' 0' when the investor' s capital contribution is listed as' 1' and non - listed banks receive the value' 0'. Bringing counterfeit variables into the model to help authors in consideration of positive effects or negative negative effects on the performance of banking systems. Ayadi (2013) analyses factors affecting the performance of commercial banks in Tunisia during 1996 – 2010. Study the application of non - parametric analysis to measure the performance of commercial banks in Tunisia. The model used for the use of the cost - effective variable by DEA is dependent on the variable and the independent variables are the market' s concentration index (hhi), rate of each bank compared with the system, credit balance of credit on the total assets, the ratio of equity on total assets, the size of the measuring bank in the decimal logarithm, and Hausman, shall be used by using the regression method to test the data in the fixed effects, Ayadi (2013) method uses the regression method, according to the fixed effects model and a random effect then using Hausman to test. Alrafadi et al. (2014) measuring performance acts as factors affecting the performance of banking systems in Libya during 2004 – through the data set of 17 Libyan commercial banks. The study applied the DEA analysis method to measure the efficiency of resource use commercial bank and regression Tobit to analyze factors affecting performance of the performance Libyan commercial bank. The research model uses the same technique as the DEA' s dependence and independence including: customer deposits / total assets, owner' s equity on total assets, bank size of the total of the total property, the state of primary liquidity commercial bank . 2.3 EVALUATION OF RESEARCH ON THE RELATIONSHIP BETWEEN LIQUIDITY STATE AND BANKING EFFICIENCY. Some study of liquidity position and performance has been performed by scholars with various aspects. Study of Nkobe (2013 )consider the relationship of the liquidity state, capital safety and performance of commercial banks in Kenya. Study of Sufian et al. (2012 )to examine the relationship of the operation and the factors of the Malaysian banking sector. Research results show that liquidity states have a positive relationship with the bank' s performance. This is consistent with the study of Dang (2011), who discovered the state of liquidity in the performance of the bank' s performance. But it does not fit Ongore & Kusa (2013) when their findings indicate that the relationship between liquidity and banking operations in Kenya is negligible. Bordeleau, Crawford and Graham (2009) reviewed the impact of liquidity position to the performance of 55 American banks and 10 Canadian banks during 1997 to 2009. The study used quantitative measures to assess the impact of liquidity on bank profits. Results from research show that the non - linear relationship, which is effective in improving the bank' s liquidity banks, however, there is a more dominant point than the bank performing effectively to the liquidity state of., bank Eljelly (2004) said that organizations that have high liquidity accounts for most investments in short - term assets, have lower profits than long - term assets. As a result, the high liquidity is expected to work better but the low yield and vice versa. Maintaining liquidity is indicated that money is limited in mobile assets so it is not available for investment purposes to yield higher returns, but it is necessary to meet the capital requirements when it is necessary to help the performance of the banks better. 2. 4 EVALUATIONS FROM PREVIOUS RESEARCH PROJECTS For banking performance: Research works measure the efficiency of banking operations using 2 methods of data (DEA) and randomness (SFA). Some studies use both methods to measure banking performance as Pelosi (2008 ), Resti (1997 ; To (2013) ) For the study of factors affecting the performance of the banking operations, researchers applied different regression methods, including the smallest squares (OLS), regression, for the proposed economic model of the proposed economy. The models are built to assess the movements of the factors in the banking operations in which the explanatory state is the liquidity state Alrafadi et al. (2014 ; Nkobe (2013 ; Sufian et al. (2012 ).The results of the study often show positive impact of liquidity position to banking operations during the study. For the study of the relationship between liquidity state and banking operations Nkobe (2013), Bordeleau & Graham ; Eljelly (2004) towards the analysis of the relationship between liquidity states and performance effectiveness in a multivariate regression model. These studies often use linear analysis techniques through the D-GMM method for the model data model, Granger analysis, Granger Karmar analysis, Tobit censorship regression; minimum squares regression. From theoretical research and experimentation on liquidity position and performance of commercial banks. With the study outside the country, the factors affecting liquidity factors, factors affecting liquidity in this country must not be repeated in other countries. For domestic studies, it is only stopped to identify factors affecting liquidity, operating efficiency without expanding the determination of the relationship between liquidity and banking performance. Studies on banking efficiency in Vietnam are quite a lot of achievement (2008); To (2013) )in both parameters and dimensions. There are also studies that analyze factors affecting banking efficiency, analyzing factors affecting liquidity position such as Hong (2012) ; (2013) but lack of research on the relationship assessment between liquidity and banking efficiency. With the author' s research knowledge evaluation and synthesis, currently no research has been conducted in determining factors affecting the performance of the state - of - the - art operations, as well as the assessment of the relationship between the liquidity state and the effective performance of the state - of - the - art state, in 2007, in 2017 – 2017. In short, when studying the " liquidity and operational position " topics Vietnamese commercial bank will seek (i) empirical evidence for the effect of liquidity position to the efficiency of the banking operation (ii), empirical evidence for linear relations between liquidity states and operational efficiency commercial bank Vietnam period 2007 – 2017. This is the gap that studies solve. CONCLUSION OF CHAPTER 2 CHAPTER 3. RESEARCH MODELS AND METHODS OF RESEARCH RELATIONSHIP BETWEEN LIQUIDITY STATE AND BANKING OPERATION 3.1 MODEL RESEARCH MODEL AND METHODS OF MEASURING BANKING ACTIVITIES 3.1.1. Model background A cost function is represented as follows Ilieva (2003 2003): (1.1) In which: - TC i is the sum of total bank expenses - And y i , p i is the vector that represents the output and input. is error number. Ilieva (2003) assuming that different evaluation techniques and different assumptions for the distribution of errors are results in different models. The optional profit function (alternative profit function) allows the bank to have large rights in the output decision so the function is determined by the input price and output output, as follows: = 1n, (1.2) 3. 1. 2. Model of measuring data efficiency If the assumption is that one DMU uses the input element x to produce the output n elements with the method of coordinating certain inputs and outputs in two different weights, such as v and u (u and v are the set of input variables, the DMU, which is the following, is calculated as the following: i = 1m; j = 1n (2.1) Applying the formula above to compute the efficiency of each DMU and on the theory, each DMU is different about x and y, and u, v, m, n is the same. If the price is not determined, it is possible to assume that a variable in x i or a turn - out i will be assigned to a v i or u i based on the importance of the head input or that output to DMU. However, each DMU will have different evaluation of the importance of each input and output, so each DMU is very different from both u, v, x, and y. so the DEA will intervene and solve the problem. However, for different analysis purposes, researchers often classify the DEA models used in the measure of banking efficiency Kumar & Gulati (2013 ): (i) Non – allocation DEA models; (ii) Allocation DEA models. 3.1.2.1. Non- Allocation DEA models 3.1.2.2. Allocation DEA models 3. 1. 3 Specifies the model and selection of input inputs In this study, commercial bank was considered the financial intermediaries and provide financial services and provide payment services to the economy should be selected with three input variables: the cost of output variables include: earnings from interest (y1) ; revenues from interest (y2) include net income from service activities net income from securities trading, investment and income net from other activities. The cost of the cost is (w1) staff costs, expenditures fees for use of fixed assets (w2) and average interest expense (w3). With the upper approach of theses, input and output variables are selected: - Staff cost (I1) - Fixed assets net (I2) - Customer deposit (I3) - Earnings from interest (O1) - Non - profit income (O2) 3.2/METHOD AND ANALYSIS OF FACTORS TO AFFECT THE PERFORMANCE OF BANKS TRADE. To determine the model and method for analyzing factors to the efficiency of commercial banks, research conducted by the study work - related research works. 3.2.1. Specify the model and analysis of factors affecting banking efficiency Table 3. 1 Detailed description of variables in the tobit regression model Variables Data Mark expectation Research DEA_TE Banking efficiency (results from the dea) / Sufian(2009); Alrafadi and partners (2014) DETA Deposit size (customer deposit/ total assets) + Alrafadi and partners (2014); Kwan (2006) EQTA Capital structure (equity / total assets) + Berger và Mester (1997) ;Sufian(2009);Alrafadi and partners (2014) LATA Liquidity state (property assets / total assets) + Vodová (2011); Aspachs and parters (2005); Alrafadi and partners (2014) LODE Credit density (loan balance / deposit) + Lee and Kim (2013) SIZE The logarithm of the total assets represents the size of the bank + Kwan (2006) Lee and Kim (2013) According to Tobin' s research (1958) and Coelli co - workers (1998), model with the sample, including i bank in a year proposed: In the meanwhile, xi and β are vectors of explanatory variables and parameters needed finding. yi * is dependent variable blocked or cut variable and yi is effective variable of banking activities of bank i by randomly research sample receiving value from 0 to 1; εi is the noise part. Model (2.1) is tobit model standardnizing for cross-sectional however,So as to be suitable to unbalanced data panel in researching and dependent variables on efficient technique, model (2.1) is deployed to be: = + + + + + Hypothesis research With research model and variables as above, hypothesis research is supposed to be: H1: Existing relationship covariencedly between deposit scale and banking efficiency. H2: Existing relationship covariencedly between capital structure and banking efficiency. H3: Existing relationship covariencedly between liquidity position and banking efficiency. H4: Existing relationship covariencedly between credit proportion and banking efficiency. H5: Existing relationship covariencedly between bank scale and banking efficiency 3.3. RESEARCHING MODEL AND RESEARCHING METHOD IN ORDER TO ANALISING THE RELATIONSHIP BETWEEN LIQUIDITY POSITION AND BANKING EFFICIENTCY. Chosen variables are : liquidity position (LATA) and activity efficiency (DEA_TE). Thematic offers researching model as below: ( ) (3.1) ( ) (3.2) There in: n is latency number, is specific bank effect which records systematic difference between banks, is random error which has similiar distribution and independent, DEA_TE is work efficient, LATA is liquidity position In order to do the above statitic, this thesis uses estimating method S – GMM two steps for dynamic panel data models. 3.4 RESEARCHING DATA Researching data of this thesis is taken from Thomson Reuter and Audited financial statements (Separate financial statements) of Việt Nam commercial banks from 2007 to 2017. The period of researching this thesis lasted from 2007 to 2017. This thesis chose the research milestone was 10 years since Việt Nam has joined WTO and the first intergrating branch is banking and finance thus this statement develops very fast until 2017 when banks pilot performing as Basel II standards having went to the route bringing liquidity admistration gradually into standards and stability (Orientation of state bank in performing Basel II route by realizing Archive 1601/NHNN-TTGSNH on 17/3/2014 about performing capital treaty Basel II) The total number of banks in sample research was 32 banks (however, the maximum number of banks for several years was less, because there was merger), including State commercial banks , joint-stock commercial bank. CONCLUSION CHAPTER 3 CHAPTER 4: RESEARCHING RESULT 4.1 OVERVIEW OF BANK SYSTEM. 4.2 RESEARCHING RESULTOF BANK EFFICIENCY Technical effectness as DEA of the lowest commercial bank in 2008 accounted for 80% and the highest in 2007 for 92%. Technical effect according to DEA from the period 2007 to 2017 was 86% on average. Technical inefficiency reflected the deviation of animistration compared to the most effective bank. The result of analising technique according to DEA in appendix 1 showed that AGR,VCB, MBB, TPB, NAB are banks having average highest technique effective (100%) and NAV was the bank having the lowest highest technique effective (56%). The average technique efficiency was maintain at quite high rate show that Việt Nam commercial banks đã focus on increasing administration ability and using effectively their power in order to reach the optimal technique efficiency Commercial banks in sample researc