Commercial Banks Adequacy Ratio

(1)

Commercial Banks Adequacy Ratio

Nguyen, Hong Yen¹ - Le, Ngoc Minh Chau²

1 Banking Academy of Vietnam; ² KIS Vietnam Securities Corporation

Ngày nhận: 08/05/2021 Ngày nhận bản sửa: 26/05/2021 Ngày duyệt đăng: 09/06/2021

Abstract: The purpose of this research is to findout the impact of liquidity transformation on capital adequacy ratio (CAR) of Vietnamese commercial banks.

By using Generalized Least Square regression model for 16 Vietnamese banks in the period 2012-2020 with dependent variable ‘capital adequacy ratio CAR’, independent variable ‘lag liquidity transformation LTG t-1’ and some additional control variables (namely: the lag capital adequacy ratio CARt-1, return on equity ROE, credit risk CRSK, gross domestic product GDP, inflation rate INFL), this study finds that liquidity transformation (LTGt-1) has negative effect on capital adequacy ratio (CAR), while the variables lag capital adequacy ratio (CARt-1) and credit risk (CRSK) are positively related to CAR, while, ROE, GDP and INFL have

Tác động của khả năng chuyển đổi thanh khoản đến tỉ lệ an toàn vốn tối thiểu của các ngân hàng thương mại Việt Nam

Tóm tắt: Bài viết thực hiện đánh giá tác động của khả năng chuyển đổi thanh khoản đến tỉ lệ an toàn vốn tối thiểu của các Ngân hàng thương mại Việt Nam. Sử dụng phương pháp hồi quy bình phương tối thiểu tổng quát để kiểm định cho 16 ngân hàng ở Việt Nam trong giai đoạn từ năm 2012 - 2020 với biến độc lập là CAR, biến phụ thuộc là biến trễ của LTGt-1 và một số biến kiểm soát là nhân tố bên trong là tỉ lệ vốn tối thiểu năm trước (CARt-1), khả năng sinh lời (ROE) và rủi ro tín dụng (CRSK) và các biến số vĩ mô có thể có tác động tới tỉ lệ vốn tối thiểu là: tổng sản phẩm quốc nội (GDP) và tỉ lệ lạm phát (INFL). Kết quả của nghiên cứu đã chỉ ra rằng biến LTGt-1 có tác động ngược chiều tới biến CAR trong khi đó các biến CARt-1 và CRSK có tác động cùng chiều đến tỉ lệ an toàn vốn tối thiểu CAR.

Ngược lại, các biến ROE, GDP và INFL không có ý nghĩa kinh tế trong quan hệ với tỉ số CAR.

Từ khóa: Ngân hàng thương mại Việt Nam, Tỉ lệ an toàn vốn tối thiểu, Khả năng chuyển đổi thanh khoản, Phương pháp bình phương tối thiểu tổng quát

Nguyễn Hồng Yến Email: yennh@hvnh.edu.vn Học viện Ngân hàng Lê Ngọc Minh Châu

Email: minhchau1812@gmail.com

Công ty cổ phần chứng khoán KIS Việt Nam

(2)

insignificant effect on CAR.

Keywords: Capital adequacy ratio, Liquidity transformation, generalized Least Square, Vietnam

1. Introduction

The modern theory of financial intermediation suggest that banks perform two basic functions: risk transitions and liquidity transitions (Berger & Bouwman, 2009). It is evident that to fulfill these two transformational roles, maintaining capital adequacy is always extremely important for banks, hence why much concern is being expressed in the relationship between these two transition fuctions and bank’s capital. However, the relationship between bank captital and bank risk-taking has been mentioned in many studies, there is not much attention paid to the liquidity transformation - the use of highly liquid sources to finance low-liquidity bank assets. People often approach bank liquidity from the perspective of liquidity risk management, but the empirical studies on liquidity transformation are quite rare (Horvath, Seidler & Weill 2012). So far, the studies on the two issues CAR and liquidity are quite detached. The relationship between liquidity transformation and CAR is not interested in researches (Novokmet & Marinovic, 2016). Alhassan (2017) is the only one which studied the effect of liquidity convertibility on CAR of 20 banks in Ghana. This study concludes that the liquidity of banks in Ghana is positively correlated with CAR.

In Vietnam, the implementation of the Basel II Capital Agreement in particular as well as compliance with the capital adequacy ratio (CAR) for commercial banks in general is extremely urgent. Studies on the factors affecting the minimum capital ade-

quacy ratio of commercial banks have been conducted by many authors from different angles. In which, there are only two studies on factors affecting capital adequacy from a liquidity perspective, which are the studies of Vu Huu Thanh et al. (2016) and Le Tu (2018). Both of these studies talk about the correlation between liquidity conversion and bank capital, not about CAR. These two studies conclude that liquidity conversion and bank capital are negatively related.

Up to date, in Vietnam, there has been no research on the impact of liquidity transformation on the CAR of banks.

With these research gaps, in order to provide other perspective study on the factor affect the bank CAR, based on the research of Alhassan (2017), the study is the first attempt to investigate the impact of liquidity transformation on Vietnam- ese commercial banks’ CAR. The study considers 16 commercial banks (which accounting for nearly 80% of Vietnames bank total asset) as of December 2020 it the spans of 9 years. The variables used include lagged liquidity transformation gap, return on average equity, credit risk, gross domestic product, and inflation. The main finding of the study is that liquidity transformation has a statistically negative effect on bank’s capital adequacy in Viet- nam, the Financial Fragility Hypothesis and Crowding Out Hypothesis are reasonable for Vietnam during the research period. It will be a theoretical as well as practical suggestion for the management of sufficient capital through management of banks’ ability to convert liquidity.

The rest of the paper is organized as fol-

(3)

lows. Section 2 reviews the theoretical and empirical literature of the relationship between liability transformantion and bank’s captial. Hypotheses developed in this section, relating in impact of liquidtity transformation to captital adiquacy ratio is tested in Section 3. After analyzing the regression result in Section 4, the Section 5 concludes the paper with some policy implications and directions for future research.

2. Literature review 2.1. Theoretical review

There are two opposite theories about the relationship between the liquidity transformation and the bank’s capital. On the one hand “Risk Absorption Hypothesis” argues that bank capital increases banks’ ability to bear risk since capital absorbs risk (e.g.

Von Thadden, 2004; Repullo, 2004; Coval

& Thakor, 2005; Bhattacharya & Thakor, 1993) and the risk from liquidity transformation is no exception. Specifically, more liquidity created from bank also means that the more illiquid assets, the higher the level of loss for handling illiquid assets. To avoid bankruptcy when not meeting the liquidity demands of customers, the bank needs to find sources to compensate when the liquid liabilities has expired but still has not collected money from investing in illiquid assets. At this time, in order not to passively borrow on the interbank market at a large cost, banks will actively increase their capital in proportion to the increase in liquidity created. Therefore, the “risk absorption hypothesis” predicts a positive association between liquidity transformation and bank capital adequacy. The more liquidity banks transform, the higher capital level the bank will maitain to avoid bank insolvency.

On the other hand, “Financial Fragil- ity- Crowding Out Hypothesis” predicts an inverse relationship between liquidity transformation and the bank capital. Ac- cording to the Crowding-out hypothesis, the increasing amount of deposits forced banks to perform credit operations to earn profits to offset the costs payable to depositors. The increase in deposits from large numbers of retail customers with short maturities- liquid liabilities will increase the bank’s liquidity transformation for credits with fewer customers, large credit value and long credit terms- illiquid assets, and liquidity risk accordingly also increased. The theory of financial fragility indicates that the increase in equity is much more passive than the increase in depositors. The bank shareholders are not at liberty to provide funds to the bank at their chosen time, while the depositors can flexibly send money to the bank at any time. In this situation, deposits are said to have “crowded- out” capital, thereby com- pelling banks to fund long- term loans and investments with deposits. This situation increases the bank’s dependence on deposits, banks will be less likely to seek to increase equity. Over time it will reduce the bank’s capital ratio in the capital structure.

Therefore, liquidity transformation has a negative relationship with bank capital.

In both theories, an increase in liquidity conversion is accompanied by an increase in less liquid assets (fewer liquid assets are often associated with higher risk), the difference between these two theories is just the relationship between the liquidity transformation and the incentive to increase the bank’s equity. Therefore, it can be seen that both theories show that liquidity transitions have an impact on a bank’s CAR through risky assets and bank capital, but the effect is in opposite directions.

(4)

2.2. Empirical Review

There are many empirical literatures on liquidity risk or liquidity transformation but few of them assess matter associated relationship between bank’s liquidity transformation and its capital (Berger and Bouwman, 2009).

Among the empirical studies on liquidity transitions, two are well-known, the researches of Deep & Schaefer (2004) and Berger & Bouwman (2009). These two studies provide the method to measure liquidity transformation and become foun- dation for other studies to develop.

Deep and Schaefer (2004) gives a measure bank’s liquidity transformation, using the concept of “Liquidity Transition Gap - LTG”. They applied this on a sample of 200 large banks in the United States and indicated that the banking industry created liquidity of up to 20% of its total assets, banks do not seem to create liquidity.

Berger & Bouwman (2009) use the concept of “Liquidity Creation- LC” to measure bank’s liquidity transformation, with the four measures: ‘cat fat’, ‘mat fat’, ‘cat nonfat’, and ‘mat nonfat’ which are based on category, maturity, category with off balance sheet activities, and maturity with off balance sheet activities, then assign weighting for asset types to calculate total liquidity creation. The study found that banks in the United States created nearly double the amount of liquidity during the period from 1993 to 2003. In 2003, the banking system generated approximately

$4.56 of liquidity per $1 of capital.

Based on the measurement method of Deep & Schaefer (2004) and Berger &

Bouwman (2009), Alhassan (2017) used GMM regression method to determine the impact of liquidity transformation to the capital adequacy ratio CAR of 20 banks

in Ghana in the period of 2006- 2015.

This study concluded that banks’ liquidity transformation in Ghana is positively correlated with CAR, suggesting that banks often increase their capital reserves with increasing liquidity, in order to absorb liquidity risks from this activity and ensure solvency.

In Vietnam, there are only two studies mentioning the ability to liquidity transformation. These studies were Vu Huu Thanh et al. (2016) and Le Tu (2018), both of these studies are about the correlation between liquidity creation and banking capital. Based on Berger & Bouwman’s (2009) study, GMM and three-stage least square regression methods, the two studies conclude that liquidity transformation and bank capital have a negative relationship. Specifically, bank capital affects 30%

of liquidity transformation but liquidity transformation only explains 4% of the variation of bank capital. Le Tu (2018) also concluded that big banks generated 92% of liquidity in the banking system over the period 2007- 2015.

The research conducted by Vu Hung Phu- ong and Dang Ngoc Duc (2020) identi- fied the factors affect the capital adequacy ratio (CAR) of Vietnamese commercial banks for the period from 2011 to 2018.

However, the variables that are hypoth- esized to affect the capital adequacy ratio of commercial banks in Vietnam are bank size (SIZE), deposit (DEP), loan (LOA), loan loss reserves (LLR), liquidity (LIQ), return on assets (ROA), return on capital (ROE), net interest margin (NIM), non- performing loans (NPL) and leverage (LEV). There is no examination of the impact of liquidity transformation to CAR in this research.

2.3. Research Gap

(5)

Research on the relationship between liquidity transformation and bank capital has been available in Vietnam. However, research on the impact of liquidity transformation on capital adequacy ratio CAR has not been done yet in Vietnam while the bank’s liquidity transformation not only affects bank capital but also affects the bank’s risk assets. Moreover, liquidity transformation is one of the two core activities of the bank in addition to the risk transformation, so it will partly affect the capital adequacy ratio CAR- the ratio measures the safety in banking operations.

Therefore, a study on the impact of liquidity transformations on the banks’ CAR is needed.

3. Research Methodology 3.1. Sample

Our study uses the data of 16 commercial banks (Appendix 1) that had assets account- ed for nearly 80% of the total assets of the Vietnamese commercial banking system by the end of 2020. This sample also includes 10 commercial banks selected by the State Bank of Vietnam to implement the capital and risk management method according to the Basel II Capital Accord.

Among the selected sample banks, Sai Gon- Ha Noi Bank (SHB) was merged with HBB in August 2012. Therefore, to ensure consistency, the analytical data is based on these banks’ Financial Statement from 2012 to the end of 2020. Finally, our number of observations is 144 (include 16 banks in the period of 2012-2020).

Financial data were collected from Finan- cial Statemenets and Annual Reports which are available at the official websites of these commercial banks over the period 2012- 2020 and financial data platform FiinPro.

Macroeconomic data are collected from the official websites of The World Bank.

3.2. Research Model

Our research model is based on the original model of Alhassan (2017) on the impact of liquidity transformation to capital adequacy ratio CAR of commercial banks in Ghana over the period 2006 - 2015.

Of which, the dependent variable CAR represents the bank solvency. In fact, banks are more likely to fail if they have a sufficiently low capital adequacy ratio.

The capital adequacy ratios (CAR) in this study were calculated according to the current regulation, Circular No.36/2014/

TT-NHNN (equivalent to the calculation in Basel I).

The independent variable is the lagged liquidity transformation gap (LTG)- represents for bank’s ability to convert liquidity. To calculate the LTG, our research firstly reclassified the banks’ assets and liabilities into liquid, semi-liquid and illiquid groups by pointing out some limitations in the research of Vu Huu Thanh et al. (2016) (as shown in Apendix 2). Then, by combining two methods of Deep & Schaefer (2004)- liquidity transformation gap and Berger & Bouwman (2009)- liquidity creation- we calculates the LTG for Vietnamese banks. The LTG, therefore, is calculated by the following formula:

Liquidity Transformation Gap = (Liquid Liabilities - Liquid Assets) ÷ Illiquid Assets (1)Based on this, higher values of LTG imply higher liquidity risk or the more liquidity banks transform, the higher the liquidity risk they are exposed to. Because they have to transform redundent liquid liabilities to finance illiquid assets.

(6)

Control variables are included in the mod- els as they have effect on bank solvency.

Of which, bank internal factors are profitability (Return on Average Equity - ROE) and loan losses (Credit Risk - CRSK) and external factors variables are economic growth (Gross Domestic Product - GDP) and price instability (Inflation- INFL).

Basically, our study based on the original study of Alhassan (2017) after removing the two variables ROA and SIZE. The reasons for this elimination are: (i) For the ROA variable: there is autocorrelation between ROA and ROE (as shown in Appendix 3) and after variable filtering we choose ROE variable to represent for bank’s profitability; (ii) For the SIZE variable: we found that in Vietnam there is an unreasonable correlation between the size of the bank and the CAR. Specifically, as shown in Appendix 2, the actual CAR data of Vietnamese banks shows that large- scale commercial banks such as BIDV and VietinBank have low capital adequacy

ratios, only approximately 9%- the minimum requirement. Meanwhile, small-sized commercial banks like Maritimebank or VIB always maintain a high capital adequacy ratio, up to 18% or over 24% in some years. However, a high CAR may not indicate that small banks are performing well, but it can show that the bank has difficulty in lending or attracting deposits from customers (Than Thi Thu Thuy and Nguyen Kim Chi, 2015). Subsequently, the SIZE variable is removed from our regression model.

Therefore, the following research model is used in our study:

CAR_i,t = β₀ + β₁CAR_i,t-1 + β₂LTG_i,t-1 +

β₃ROE_i,t + β₄CRSK_i,t + β₅GDP_t + β₆INFL_t + ε_i,t

i = 1÷16 and t = 2012÷2020 3.3. Specification of variables

Based on the second correlation matrix table (Appendix 4) analyzing the correla-

Table 1. Expected Impact of Explanatory Variables on Capital Adequacy

Variable Symbol Measure Expected sign

Capital Adequacy Ratio CAR Tier 1 Capital + Tier 2 Capital Risk Weighted Assets Capital Adequacy Ratio of

the previous year CARt-1 Tier 1 Capital + Tier 2 Capital Risk Weighted Assets ⁺ Liquidity Transformation of

the previous year LTGt-1 Liquid Liabilities - Liquid Assets Illiquid Assets ^+/- Return on Equity ROE Net Profit after Tax

Average Assets ⁺

Credit Risk CRSK Loan Loss Provision

Gross Loans ^+/-

Gross Domestic Product GDP Annual rate of growth of Gross

Domestic Product -

Inflation Rate INFL Annual Inflation Rate +/-

Source: Summarized by the authors

(7)

tion among the variable in the model, we found that, the correlation between the variables is relatively low (less than 50%), so it can be considered that the multicol- linearity model does not occur.

And the table 2 shows the average varian inflation factor (VIF) magnification coefficient for the model is less than 2, so the model does not have a multi-collinearity phenomenon.

The statistic of the variables can be sum-

marized in the following Table 3.

It can be seen that the value of capital adequacy ratio CAR of Vietnamese commercial banks over the period 2012- 2020 does not have a specific volatility trend, ranging from 8.34% to 40.15%, the average value is 13.12% with the standard deviation at 4.24%. This is a relatively high number compared to the 9% required by the SBV during this period in the Cir- cular 36/2014/TT-NHNN. These figures show a quite healthy way in operating of the sample commercial banks. However, when deeply looking in these banks’ An- nual Reports we found the CAR of big commercial banks such as BIDV (with

minimum value of the all observations at 8.34% in 2020), Vietinbank and Mili- tary Bank tend to be low, only around the level of the SBV’s regulations; while the small commercial banks’ CAR is higher, such as Maritime Bank’s CAR reaches the highest level at 24.53 in 2015 or CAR of TPBank has been up to 40% in 2012, however this ratio of TPBank has been decreasing at around 9-10% in the recent years. As shown in research by Reynolds et al. (2000) on the financial structure and performance of banks in Asian countries during 1987- 1997 with the conclusion that capital safety has a negative relationship with scale, large banks have lower

Table 2. Variance Inflation Factor – VIF

Variable VIF

CARt-1 1.24

LTGt-1 1.38

ROE 1.10

CRSK 1.14

GDP 1.10

INFL 1.29

Mean VIF 1.21

Source: Authors’ calculations from STATA software Table 3. Descriptive Statistic

Variable Obs Mean Std. Dev. Min Max

CAR 144 0.1312507 0.0424494 0.0834 0.04015

CARt-1 144 0.1327424 0.0421736 0.0877 0.04015

LTGt-1 144 -0.1964005 0.4154136 -2.35584 0.8246939

ROE 144 -0.1128303 0.0757943 0.0002829 0.2956575

CRSK 144 0.0134357 0.0047878 0.0063098 0.0475195

GDP 144 0.0592889 0.0124131 0.0291 0.0708

INFL 144 0.0404178 0.0236525 0.0063 0.0921

Source: Authors’ calculations from STATA software Note: Obs: Number of Observations (which derived from 16 commercial banks over the period of 9 years from 2012 to 2020); Mean: Mean value; Std.Dev.: Standard Deviation; Min: the minimum value; Max: the maximum value

(8)

capital adequacy ratio than small banks, the CAR value of banks in Vietnam is perfectly suited.

Similar to capital adequacy ratio (CAR), the value of lagged capital adequacy ratio (CARt-1) of commercial banks ranges from 8.8% to 40.15%, the average value is 13.27% with standard deviation at 4.22%.

LTG_t-1 has an average value of 0.1964 indicating that sample banks are creating liquidity, but at a extremely low level.

Although using liquid liabilities to fund illiquid assets, it is still at a safe and low risk level. With a standard deviation of up to 41.54%, the difference in LTG_t-1 between banks is quite large. The smallest LTG_t-1 value is -2.35584 of TP Bank in 2012 and the largest one is +0.8245 of Vietcombank in 2018. That means Vietcombank illiquid assets are financed almost by liquid liabilities in that year- an extremely dangerous level. Besides VCB and ACB that had positive LTG almost during the study period, in recent years, BIDV, MBB, SHB, STB and KLB always maintained a positive LTG ratio. Base on the LTG-calculated fomular (1) above, the positive LTGs in these cases mean in these banks liquid liabilities are bigger than liquid assets. In other words, these banks have to use liquid liabilities to finance illiquid assets in the recent years.

Return on equity ROE has an average value of 11.28% with a standard deviation of 7.58%, maximum and minimum values are 29.57% (of VIB in 2020) and 0.03%

(of NCB in 2020) respectively. In particular, low value of ROE often appears in banks with small capitalization in the post- crisis period of 2007- 2008 and the period of low credit growth in 2013- 2014.

The average value of CRSK in this period is 1.34% with a standard deviation of 0.48%, the largest and smallest values

are 4.75% (of TPB in 2012) and 0.63%

(of EIB also in 2012) respectively. With a relatively low number of CRSK in general, these commercial banks show good quality credit balance and they don’t need to make high provision or vice versa. This may indicate that the provisioning is not taken seriously by commercial banks during the study period.

The GDP of Vietnam’s economy over the period 2012- 2020 averaged at 5.93%

with the smallest value is 2.91% in 2020 (because of covid pandamic) and the largest value is 7.08% in 2018. The inflation rate was quite low during the study period consistently around 2-3%. Particularly, only 2012 was at the highest level of more than 9% and the lowest level was in 2015 at only more than 0.6%.

4. The Regression Results and Analysis By using the mathematical statistics tool with the support of the STATA software to run the model. At first, we used the FEM, REM regressions method to run the model and after check for defects in the model. We see that the model has the heteroskedasticity and autocorrelation er- rors. According to Wooldrige (2002), the Generalized Least Squares (GLS) method can be used to ensure that the obtained estimates are stable and efficients. Subse- quently, the study uses the GLS regression model with some addition options namely panels (heteroskedastic), corr (ar1) and rhotype (dw) in order to minimize serial heteroscedasticity and autocorrelation problem. The results are summarized for all three methods in Table 4.

The results of the GLS method is chossen, then the sample regression model is shown below:

CAR_i,t = 0.0462 + 0.2979CAR_i,t-1

(9)

- 0.0172LTG_i,t-1 - 0.0142ROE_i,t +

2.3143CRSK_i,t + 0.1407GDP_t + 0.0516IN- FL_t + ε_i,t

As expected, the lag value of CAR (CAR_t-

1) has a positive and 1% statistically significant impact on the CAR of the current year, reflecting the consistency in the attitude of maintaining the bank’s capital adequacy ratio. When the capital adequacy ratio of the previous year increased by 1%, the CAR of the following year will

increase by 0.2979%. This finding entirely consistent with the conclusion of Alhas- san’s research (2017).

Table 3 shows that the coefficient of LTG_t-

1 is significant and negative, suggesting that the greater the bank liquidity transformation reduce the CAR. With a extremely high level of significant, a 1% increase in the previous liquidity transformation gap decreases the CAR of the current year by 0.0172%. This finding does not support Table 4: The Impact of Liquidity Transformantion to the Banks’ Capital Adequacy Ratio

Dependent Variable: CAR

Variables Expected Sign FEM REM GLS

CARt-1 + 0.240478 0.4463106 0.2979427

(0.001)*** (0.000)*** (0.000)***

LTGt-1 +/- - 0.0224561 - 0.0186645 - 0.0172463

(0.025)** (0.012)** (0.003)***

ROE + 0.0500589 - 0.0427185 - 0.0141882

(0.326) (0.242) (0.581)

CRSK +/- 4.601281 2.363713 2.314321

(0.000)*** (0.000)*** (0.000)***

GDP - 0.2349408 0.1715062 0.140682

(0.242) (0.443) (0.226)

INFL +/- 0.0123279 0.0814163 0.0516033

(0.917) (0.520) (0.510)

_cons 0.0130213 0.0279433 0.0462359

(0.519) (0.168) (0.001)***

t statistics in parentheses; * p < 0.1, ** p < 0.05, *** p < 0.01

N 144 144 144

R-sq 0.4850 0.4087

F test F(15,122) = 3.35

Prob>F = 0.000 Hausman with sigmamore Chi2(4) = 38.14

Prob > Chi2 = 0.000

Wooldridge test F (1,15) = 29.115

Prob > F = 0.0001

Modified Wald test Chi2 (16) = 986.29

Prob > Chi2 = 0.0000

Source: Authors’ calculations from STATA software

(10)

the finding of our based research Alhas- san (2017). But it is in line with those of Vu Huu Thanh et al (2016) and Le Tu (2018) work on Vietnamese banks for the earlier period of time. While the effect found by Le Tu (2018) is quite high, our finding is more consistent with Vu Huu Thanh (2016) of relatively small effect.

This negative effect confirm the Financial Fragility-Crowding Out Hypothesis is ap- propriate in Vietnam. Accordingly, with the underdeveloped capital market in Viet- nam, both Vietnamese banks and enterprises find it difficult to raise fund through this market. On the one hand, with only a handful of investment chanels, people tend to deposit their idle money into bank. And banks, of course, have better access to deposit facilities while have lower ability to access capital markets. On the other hand, enterprises have to raise fund by borrow- ing more money with a longer term at the bank, making the risk-weighted assets of the bank increase. Both of these trends cause CAR (the ratio of bank’s capital over risk-weighted assets) to decline.

The Nextremely high and positve coefficient of CRSK (+2.3143) with p-value<

1% represents a significant effect on CAR.

This finding is perfectly consistent when facing bigger risk banks have to keep higher capital adequacy ratios because they are required to set aside more capital as a buffer against losses. This conclusion is consistent with the research results of Mili et al (2014), Masood and Ansari (2016) on the banking system at Pakistan and Alhassan’s research (2017) on the Ghana banks. However, the conclusion is different from the results of empirical evidence of banks in Jordan (Al-Sabbagh, 2004) and research of Le Thanh Tam et al. (2017) on the determinants of capital adequacy ratios of 26 Vietnamese com-

mercial banks in the period of 2009- 2015.

The variables Gross Domestic Product (GDP), Return on Equity (ROE) and Infla- tion rate (INFL) have statistically non- significant effect on CAR at all (as shown in Table 4).

5. Conclusion and Recommendation The authors successfully completed the goals of the research, which is empirical study impact of liquidity transformation on the Capital Adequacy Ratio (CAR) of the bank and it showed that the impact is inversely proportional in Vietnam. This impact clearly showed that liquidity transformation is a signal to show in which direction the CAR is developing in. Because of this, it is one of the most important statistic for banks to follow in managing their business in general and in control- ling capital adequacy ratio to be specific.

Based on the results of the research, the authors came up with some following recommendations:

5.1. Recommendations for commercial banks Banks need to build a large database in order to accurately identify their liquidity convertibility (represented by liquidity transformation gap) to manage their capital adequacy ratio. As explained in the research, liquidity transformation gap is calculated by determining the actual duration of bank assets and liabilities. This determination is in fact based not only on the terms of assets and liabilities but also on the behavior of the customer which must be based on a big database to help the bank clearly identify. Based on the big database, banks can identify the factors affecting customer behavior in early deposit with- drawal or early loan repayment. There for,

(11)

building the database system is a problem that banks should put an eye on, especially in the Fourth Industrial Revolution.

Reduce the amount of liquidity created.

The result of the research show that bank’s liquidity transformation gap has an inverse impact on capital adequacy ratio. There- fore, the overuse of short-term funds for medium and long-term loans will not only affect bank liquidity risk as known prior but it’ll also reduce the bank’s capital adequacy ratio. Because of this, banks need to reduce their liquidity transformation if they want to improve their capital adequacy ratio. There are two requirements in order to perform this. Firstly, increasing long term mobilized funds by creating big gap between short-term and long-term deposit interest rates to encourage long-term deposits. Besides, banks need to increase mobilization through issuing valuable papers, which shows a more stable term compared to deposit. Secondly, bank need to use funds with duration corresponding to mobilized funds. In fact, the impact from liabilities is more difficult, since depositing in the bank is solely a cli- ent’s decision. The impact from the asset side is needed. Therefore, in addition to extending the term of mobilized capital, banks need to have strategies to develop short-term credit products or to invest in corporate bonds (when the corporate bond market has developed).

Improve bank’s credit risk management capacity. Overdue debts will on one hand lengthen the actual loan maturities more than the maturity in loan contact. This will unexpectedly increase liquidity transformation ratio and then reduce capital adequacy ratio. On the other hand, these overdue debts are factors that increase the denominator (risk-weighted assets) leading to reduce capital adequacy ratio.

Therefore, banks need to control overdue debts.

5.2. Recommendations for management agencies

Management agencies need to be aware of the importance of the liquidity transformation of the banking system.

Supervising authorities need to see the importance of managing liquidity transformation of commercial banks. From that, they need to establish a reasonable range boundary for liquidity transformation for commercial banks in order to balance between profits and risks of the banks, which will increase their capital adequacy ratio. For the liquidity transformation gap measurement used in research, authors suggest the range for liquidity transformation gap to be at a range from 0.1 to 0.25 because of the fact that banks in this range had ROE at above 10% and CAR at above 9% during the study period. Therefore, if LTG is maintained at this range, the bank will not only be safe but also ensure good profitability.

Besides reporting about the status of liquidity, supervising authorities need to ask commercial banks to report about liquidity transformation gap by day, changes of duration of each assets and liabilities and not just focusing mainly on managing high liquid asset or non-term deposits as current rules. As mentioned above, statistics about liquidity that are being managed by supervising authorities have never been the whole chain but rather showed only in particular banks, supervising the transformed liquidity will help authorities understand the liquidity transformation of the whole banking system. This will solve liquidity problems between banks can be solve in a more connective way in the

(12)

future, instead of the disjointed solutions being used now.

5.3. Recommendations for further research Our research employs all the banks in one sample, therefore, we can not see the detail impact of liquidity transformantion to banks capital adequacy ratio in different size of banks. The researchers can expand the scope of research both in space and time and break up research space into specific banking groups. Because the difference in liquidity status between large and small commercial banks can be noticed, the impact of liquidity conversion on bank groups may be different or even contra-

dictory according to Berger & Bouwman (2009) research on United State commercial banks.

In addition, research on other classifica- tions of liquidity levels of assets and liabilities, and refer to the liquidity being created off-balance sheet should be conducted.

Research on the impact of liquidity transformation on each factor that calculates capital adequacy ratio CAR including equity and risky assets will help to clarify the extent and direction of the impact of liquidity transformation to each component of CAR, thereby more detailed policies can be recommended. ■

References

Ahmad, R., Ariff, M. & Skully, M. (2008), ‘The Determinants of Bank Capital Ratios in a Developing Economy’, Asia- Pacific Finan Markets, 15. pp. 255–272.

Aktas, R., Acikalin S., Bakin B., Celik G., 2015, ‘The Determinants of Banks’ Capital Adequacy Ratio: Some Evidence from South Eastern European Countries’, Journal of Economics and Behavioral Studies, Vol. 7, No. 1, p. 79-88.

Al-Sabbagh, N. (2004). Determinants of Capital Adequacy Ratio in Jordanian Banks. Master thesis. Yarmouk University. Irbid, Jordan.

Alhassan A. S., 2017, ‘Capital adequacy of banks in Ghana: Does liquidity transformation matter?’, Thesis, University of Ghana.

Bateni L., Vakilifard H. & Asghari F., 2014, ‘The Influential Factors on Capital Adequacy Ratio in Iranian Banks’, International Journal of Economics and Finance, Vol. 6, No. 11; 2014

Berger, A. N. & Bouwman, C. H., 2009, ‘Bank liquidity creation’, Review of Financial Studies, No. 22(9), pp. 3779- 3837.

BIS, (1999) ‘Basel Committee on Banking Supervision: A new capital adequacy framework’, online, available from:

https://www.bis.org/publ/bcbs50.pdf (Accessed 12 March 2017), [online]. Available from: https://www.bis.org/

publ/bcbs50.pdf (Accessed 12 March 2017).

Bhattacharya, S., & Thakor, A. V. (1993). Contemporary Banking Theory. Journal of Financial Intermediation, 3(1), 2-50.

Coval & Thakor, 2005, Financial intermediation as a beliefs-bridge between optimists and pessimists, Journal of Financial Economics 75(3):535-569

Deep, A., & Schaefer, G. (2004), Are Banks Liquidity Transformers? Harvard University, John F. Kennedy School of Government.

Diamond D. W., 2007, ‘Banks and liquidity creation: a simple exposition of the Diamond-Dybvig model’, FRB Richmond Economic Quarterly, Vol.93(2), pp. 189-200.

FiinPro, http://fiinpro.com/Feature

Horvath, R., Seidler, J., & Weill, L. (2012). Bank Capital and Liquidity Creation. Social Science Research Network.

Le Thanh Tam and Nguyen Dieu Linh, 2017, “Cac yeu to quyet dinh toi ty le an toan von cua ngan hang: bang chung thuc nghiem tu Viet Nam” (Factors affect bank’s capital adequacy ratio: an empirical study in Vietnam), NEU publisher, Hanoi, pp. 84-107.

Le Tu (2018), The interrelationship between liquidity creation and bank capital in Vietnamese banking, Managerial Finance, Vol. 45 Issue: 2, pp.331-347

Masood, U., & Ansari, S. (2016), ‘Determinants of Capital Adequacy Ratio - A Perspective from Pakistani Banking Sector’, International Journal of Economics, Commerce and Management, 4(17), 247-273.

(13)

APPENDICES

Mili, M., Sahut, J. M., & Trimeche, H. (2014), Determinants of the Capital Adequacy Ratio of a Foreign Bank’s Subsidiaries: The Role of the Interbank Market and Regulation of Multinational Banks (No. 2014-366).

Novokmet & Marinović, 2016, Solvency and Liquidity Level Trade-off: Does it Exist in Croatian Banking Sector?, Scientific Annals of Economics and Business 63(3):429-440

Repullo, R. (2004). Capital Requirements, Market Power, and Risk-taking in Banking. Journal of Financial Intermediation, 13(2), 156-182.

Than Thi Thu Thuy and Nguyen Kim Chi, 2015, ‘Nghien cuu cac yeu to anh huong den he so an toan von tai cac NHTM CP Viet Nam’ (Studying factors affect capital adequacy ratio of Vietnamese joint stock commercial banks), Banking Review, No. 11 June, 2015.

The State Bank of Vietnam (2014). Circular No. 36/2014/TT-NHNN dated November 20, 2014 on stipulating minimum safety limits & ratios for transactions performed by credit institutions & branches of foreign banks [online].

Available from: http://vbpl.vn/nganhangnhanuoc/Pages/ivbpq-toanvan.aspx?ItemID=38214&Keyword=36/2014 (Accessed 1 March 2017).

The World Bank, Data/Vietnam, https://data.worldbank.org/country/vietnam?view=chart

Tran, V. T., Lin, C. T., & Nguyen, H. (2016), ‘Liquidity Creation, Regulatory Capital, and Bank Profitability’, International Review of Financial Analysis, 48, pp. 98-109.

Von Thadden, E. L. (2004), ‘Bank Capital Adequacy Regulation under the New Basel Accord’, Journal of Financial Intermediation, 13(2), pp. 90-95.

Vu Hung Phuong and Dang Ngoc Duc (2020), Determinants influencing capital adequacy ratio of Vietnamese commercial banks, Accounting, 6 (2020), p. 871–878

Vu Huu Thanh et. al, 2016, ‘Von ngan hang, su tao thanh khoan va hieu qua cua ngan hang’ (Bank capital, liquidity creation and bank efficiency), Science Magazine, Open University of Ho Chi Minh City, No. 4(49)2016.

Wooldridge, J. (2002), Econometric Analysis of Cross Section and Panel Data, MIT Press.

Yuanjuan, L. &Shishun, X., 2012, ‘Effectiveness of China’s Commercial Banks’ Capital Adequacy Ratio Regulation, A Case Study of The Listed Banks’, Inter disciplinary Journal of Contemporary research in busines, 4 (1) p. ijcrb.

webs.com.

Appendix 1. List of Sample Banks

Acronyms Name of Banks

1 ACB Asia Commercial Joint Stock Bank

2 BID JSC Bank for Investment and Development of Vietnam 3 CTG Vietnam Joint Stock Commercial Bank for Industry and Trade 4 EIB Vietnam Joint Stock Commercial Export Import Bank

5 HDB Ho Chi Minh City Housing Development Bank 6 KLB Kien Long Commercial Joint Stock Bank 7 MBB Military Commercial Joint Stock Bank

8 MSB Vietnam Maritime Joint – Stock Commercial Bank 9 NCB National Citizen Commercial Joint Stock Bank 10 SHB Saigon – Hanoi Commercial Joint Stock Bank 11 STB Sai Gon Thuong Tin Commercial Joint Stock Bank

12 TCB Viet Nam Technological and Commercial Joint Stock Bank 13 TPB Tien Phong Commercial Joint Stock Bank

14 VCB JSC Bank for Foreign Trade of Vietnam

15 VIB Vietnam International and Commercial Joint Stock Bank 16 VPB Vietnam Prosperity Joint Stock Commercial Bank

Source: Summarized by the authors

(14)

Appendix 2. Reclassified Assets and Liabilities Liquidity Catagories

(15)

Source: Reclassified by the authors Appendix 3: The first correlation matrix between variables

Source: Authors’ canculations from Stata software Appendix 4: The second correlation matrix between variables

Source: Authors’ canculations from Stata software