Doctor of philosophy in economics thesis summary: The impact of liquidity risk on bank performance efficiency: Empirical evidence from south east Asia countries

The main objective of the thesis is to identify factors influencing liquidity risk and analyze the impact of liquidity risk on bank efficiency performance, study the case of South-East Asia countries in the period of 2004 - 2016.

MINISTRY OF EDUCATION THE STATE BANK OF VIETNAM

BANKING UNIVERSITY OF HO CHI MINH CITY

TRAN THI THANH NGA

THE IMPACT OF LIQUIDITY RISK ON BANK PERFORMANCE EFFICIENCY: EMPIRICAL

EVIDENCE FROM SOUTH EAST ASIA COUNTRIES

DOCTOR OF PHILOSOPHY IN ECONOMICS

THESIS SUMMARY

Major: Finance – Banking Code: 62 34 02 01

Academic advisor: Assoc Prof Dr Tram Thi Xuan Huong

Dr Le Thi Anh Dao

HO CHI MINH CITY - 2018

CHAPTER 1: INTRODUCTION

The relationship between liquidity risk and performance efficiency has made

interested through the approach to hypotheses such as Market Power Hypothesis, Efficient Structure Hypothesis (Diamond and Dybvig, 1983) showed that the impact

of liquidity risk on bank performance efficiency is unclear

Some studies in Africa (Sayedi, 2014; Aburime, 2009; Athanasoglou et al.,2008;

Ajibike & Aremu, 2015; Alshatti, 2015); in Asia (Wasiuzzaman & Tarmizi, 2010; Arif & Nauman Anees, 2012; Shen et al.,2009), in Europe (Bourke,1989; Poposka & Trpkoski, 2013; Goddard, Molyneux & Wilson, 2004; Kosmidou, Tanna & Pasiouras, 2005) found out the positive effect Others in Asia (Lee & Kim, 2013); in Africa (Bassey & Moses, 2015) found out the negative effect A number of studies was also found that there was a very weak relationship or no relationship between liquidity risk and bank performance (Sufian & Chong, 2008; Roman & Sargu, 2015; Alper & Anbar, 2011; Almumani, 2013; Ayaydin & Karakaya, 2014) or depends on economic characteristics and research models (Naceur & Kandil, 2009; Ferrouhi, 2014)

After reviewing studies, the authors found that the majority of the previous studies approaching the impact of liquidity risk on bank performance efficiency (Sufian & Chong, 2008; Sayedi, 2014; Oluwasegun & Samuel, 2015; Lartey, Antwi, & Boadi, 2013; Bourke,1989; Tabari, Ahmadi & Emami, 2013; Arif & Nauman Anees, 2012; Bassey & Moses, 2015; Ferrouhi, 2014; Alshatti, 2015; Aburime,2009; Athanasoglou

et al., 2008; Naceur & Kandil, 2009) Some studies have approached the impact of bank performance efficiency on liquidity risk in different countries (Vodova, 2011;

Abdullah & Khan, 2012; Roman & Sargu, 2015) This shows that the trend of the

impact of the liquidity risk on bank performance efficiency has been attented by the researchers and managers, especially the impact of the financial crisis on bank performance efficiency (Lee và Kim, 2013)

Most empirial researchs approaching to factors affecting to liquidity risk and the impact of liquidity risk on bank performance efficiency have taken in the region of one country only, except study of (Roman & Sargu, 2015) based on European data or (Bordeleau & Graham, 2010) in America, (Shen et al.,2009) both in Europe and America Cross-countries studies on aspect to examine the interlinkage between liquidity risk and bank performance efficiency On the other hand, some recent study claimed that liquidity is endogenously determined and the question on the impact of liquidity risk on bank performance efficiency cannot be studied without controlling for endogeneity

2

The empirial researchs showed that Vietnamese is one of the countries with lower average income population of Southeast Asia and there are too many banks but lacked

a main banking to competitive with other regional economies (Nguyen Cong Tam &

Nguyen Minh Ha, 2012) Thus, this study used Bankscope data of 171 banks during

the period 2004–2016 and the system Generalized Method of Moments (SGMM) method to analyze the impact of liquidity risk on bank efficiency performance in South-East Asia countries to estimate the impact of liquidity risk on banks financial performance in South-East Asia countries case Studies in different spaces and periods will give unequal results

To fill the gap research, thesis combining research approach to factors influencing liquidity risk and the impact of liquidity risk on bank efficiency performance in South-East Asia countries are extremely important and valuable

So, the author selected the topic "The Impact of Liquidity Risk on Bank Fficiency Performance : A Case Study in South-East Asia Countries" as a thesis In addition, the study combines a case study of South - East Asia and Vietnamese to propose policy suggestions for Vietnam This study will contribute to empirical evidence and provide some useful information on the factors affecting liquidity risk and the impact of liquidity risk on bank efficiency performance

1.2 Research objectives

1.2.1 Research objectives

The main objective of the thesis is to identify factors influencing liquidity risk and analyze the impact of liquidity risk on bank efficiency performance, study the case of South-East Asia countries in the period of 2004 - 2016

1.2.2 Specific objectives:

Based on that, the specific objectives of the project are defined as:

The firstly, to analyze factors influencing liquidity risk, study the case of South-East

Asia countries and Vietnamese

The secondly, to analyze the impact of liquidity risk on bank efficiency performance,

study the case of South-East Asia countries and Vietnam

Thirdly, to suggested policies on liquidity risk management and Effective

management of banking in Vietnam

1.2.3 Research questions

1) What factors influence the liquidity risk and the level impact of factors on liquidity risk the case of South-East Asia countries?

2) Are there any differences in the results study the factors affecting the liquidity risk in the case of South-East Asia countries and Vietnam?

3) What is the impact of liquidity risk on bank efficiency performance, in the case

of of South-East Asia countries?

4) Are there any differences in the results study impact of liquidity risk on bank efficiency performance, in the case of South-East Asia countries and Vietnam?

5) What are the policy implications of liquidity risk management and Effective management of banking in Vietnam

1.3 Research Object and research scope:

Research Object: The study Object of the thesis are liquidity risk and bank

efficiency performance, in the case of South-East Asia countries

Research scope: The scope of the study was extended to for 11 countries in

South-East Asia (Brunei, Cambodia, EasiTimor, Indonesia, Laos, Myanmar,

Malaysia, Philippines, Singapore, Thai Land, Viet Nam) from 2004 to 2016

The database was collected from 2 sources: (i) bank-level data from Bankscope, (ii) macroeconomic information data from Asian Development Bank (ADB)

1.4 Recearch methodology:

The research has combined the approach of (Ferrouhi & Lahadiri, 2014; Trenca, Petria & Corovei, 2015) to analyze impact of the factors on liquidity risk and its approach (Growe và cộng sự, 2014; Ferrouhi, 2014) to analyze the impact of liquidity risk on bank efficiency performance, in the case of South-East Asia countries

On the other hand, comparative study of the result in the case of South-East Asia countries and Viet Nam to proposed policy implication for Vietnam

1.5 Thesis structure:

The structure of the thesis consists of 5 chapters:

Chapter 1: Introduction

Chapter 2: Theory basis and literature review

Chapter 3: Research Methodology

Chapter 4: Research results

Chapter 5: Conclusions and Policy Implications

4

CHAPTER 2 THEORY AND LITERATURE REVIEW

2.1 Liquidity Risk in Commercial Banks

2.1.1 Theoretical Framework Underlying the Study

2.1.1.1 Commercial Loan Theory and Liquidity

2.1.1.2 The Shiftability Theory of Liquidity

2.1.1.3 Anticipated Income Theory of Liquidity

2.1.2 The concept of liquidity risk

The Basel Committee on Banking Supervision (2003) contends that Liquidity Risk

is a risk that a bank's inability to accommodate decreases in liabilities or to fund increases in assets

Rudolf Duttweiler1, contends that Liquidity represents the ability to payment all payment obligations upon maturity The inability of banks to raise liquidity can be attributed to a funding liquidity risk that is caused either by the maturity mismatch between inflows and outflows and/or the sudden and unexpected liquidity needs arising from contingency conditions Lack of liquidity will lead to liquidity risk

According Bonfim and Kim (2014), the complexity of the functions of banks gives rise to an intrinsic risk that lies deep in their core function; their unique intermediation role Banks use a limited amount of their own resources in granting loans to entrepreneurs and consumers and thus provide them with the liquidity to finance their investment and consumption demands Much of these resources used by these banks are normally associated with liabilities to third parties traditionally in the form of deposits For profit purposes, this transformation of liquid liabilities (deposits) into risky liquid (illiquid) assets in the form of loans capitalizing on their maturity

mismatch expose them to liquidity risk (Diamond and Dybvig, 1983; Jekinson, 2008)

In order to lessen the maturity gap between assets and liabilities or the inherent illiquidity, banks can adequately manage the liquidity risk underlying their balance sheet structure by holding a buffer of liquid assets However, aside the high opportunity cost of holding a buffer of liquid assets as compared to the higher returns associated with illiquid assets, it manifests a degree of inefficiency on the part of management as it limits banks' ability to provide liquidity to entrepreneurs and

1

Rudolf Duttweiler: "Liquidity Management in Banking: Top-down Approaches", Ho Chi Minh City General Publisher, p.23

consumers Hence, even though banks have some incentives to hold a fraction of liquid assets (in the form of cash, short term assets or government bonds), these buffers will hardly ever be sufficient to fully insure against a bank run or liquidity risk (Bonfim and Kim, 2014)

2.1.3 Liquidity risk measurement methods

2.1.3.1 Approaches to the guaranteed ratio is regulated by the the Basel Monitoring Committee

2.1.3.2 Approaches to liquidity indicators

2.1.4 Empirical literatures on determinants of liquidity risk

Though liquidity risk has always been considered in literature as a major determinant of bank performance, only a few of studies have gone further to take into consideration the various determinants of liquidity risk in the daily operations of a bank Work done by some few researchers show varied determinants in different banking environments basically categorized under bank specific and macro-economic factors

The determinants affected liquidity risk are focused on the following factors:

The bank size: previous studies found that a negative relationship between

bank size and liquidity risk (Lucchetta, 2007; Munteanu, 2012; Abdullah & Khan, 2012; Delécha et al., 2012; Bonfim & Kim, 2014) While other studies suggested that the relationship between bank size and liquidity risk may be nonlinear or ambiguous (Vodova, 2011; Shen et al., 2009; Aspachs & cộng sự, 2005; Truong Quang Thong, 2013)

Asset quality: A key liquidity ratio is the liquid assets ratio (Liquid

assets/Total assets) Previous studies (Bonfim và Kim, 2014; Bunda và Desquilbet,

2008; Delécha và cộng sự, 2012; Lucchetta, 2007; Munteanu, 2012; Vodova, 2011) was found that lower liquidity means higher risk The portfolio theory suggests higher risk leads to higher profitability In addition, some studies (Lucchetta, 2007; Bunda &

Desquilbet, 2008; Vodova, 2011; Delécha et al., 2012) is used be the ratio of liquid

assets/total deposits Liquidity has also been measured by liquid assets to total deposits (Liquid assets/deposits) and some studies measured by liquid assets to short-term deposits (Bunda & Desquilbet, 2008;Vodova, 2011; Cucinelli, 2013; Delécha et

al., 2012) Hence, higher values of this ratio denote less liquidity The higher the

liquidity structure, the lower the liquidity risk

Capital: indicators measure the strength of the bank’s capital position,

including its ability to withstand and recover from economic shocks Theoretical expectations, as well as empirical results (Lucchetta, 2007; Bunda & Desquilbet,

6

2008; Cucinelli, 2013; Munteanu, 2012; Bonfim & Kim, 2014; Trương Quang Thông, 2013), for the equity to assets ratio (Total equity/Total assets) suggest that the ratio will be positively related to liquidity risk In addition, studies across industries have

found that the actual relationship between Capital and liquidity risk was negative

(Delécha et al., 2012; Berger & Bouwman, 2013) The implication is that the level of equity in a bank’s capital structure should be negatively related to liquidity risk

Credit risk: is the risk that a portion of interest or both interest and principal

of a loan will not be repaid as committed The competitiveness of the bank depends largely on its ability to manage credit risk (Bonfim & Kim, 2014) Previous studies (Delécha et al., 2012; Cucinelli, 2013; Bonfim & Kim, 2014; Trenca, Petria & Corovei, 2015) using Loan Loss Provision/Total Loans is assessed by credit risk While other studies (Bonfim & Kim, 2014; Cucinelli, 2013; Delécha et al., 2012)

suggested that higher lending ratios, lower liquidity It means, bank have more

vulnerable capital structures, liquidity risk is higher

Interest income:a key ratio is the efficiency or cost to income ratio (interest expense/Total income) Several studies have found that high values on it higher liquidity risk (Abdullah & Khan, 2012; Bonfim & Kim, 2014; Delécha et al., 2012;

Munteanu, 2012)

The macro factors include: The GDP growth variable is assessed by the

year’s real change in gross domestic product (GDP) for the nation the bank is located

in, sometimes on a per capita basis GDP growth was related to liquidity risk in a number of studies (Aspachs et al., 2005; Bonfim & Kim, 2014; Bunda & Desquilbet, 2008; Cucinelli, 2013; Delécha et al., 2012; Munteanu, 2012; Growe et al., 2014;

Trương Quang Thông, 2013; Vodova, 2011) The inflation rate is assessed by

entering in the CPI change rate for the nation and the year GDP as a measure of total economic activity in an economy, higher economic growth encourages banks to lend more and permits them to charge higher margins, and improve the quality of their assets as suggested by previous studies (Aspachs et al., 2005; Bonfim & Kim, 2014; Bunda & Desquilbet, 2008; Cucinelli, 2013; Delécha et al., 2012; Munteanu, 2012;

Growe et al., 2014; Trương Quang Thông, 2013; Vodova, 2011) Financial crisis

(Bunda & Desquilbet, 2008; Delécha et al., 2012; Lucchetta, 2007; Munteanu, 2012;

Growe et al., 2014; Shen et al., 2009; Skully & Perera, 2012; Vodova, 2011) was found that is one of the factors affecting the liquidity risk

As, the findings of previous studies are quite consistent with the realities in the

financial markets The empirical studies continue to assess the determinants of the impact on the liquidity risk in banks through the specific factors (bank size, asset

quality, capital, credit risk, interest income,…) and macro factors (Real GDP

growth rate, fluctuations of inflation, financial crisis, )

2.2 Bank Performance Efficiency

2.2.1 Theories on Bank Performance Efficiency

Bank Performance Efficiency is often measured by profitability Studies of Bank Performance Efficiency or profitability is used basing on two theories: market power theory (MP – market power) and structural efficiency theory (ES - efficient structure)

2.2.1.1 The theory of market power

2.2.1.2 The theory of efficient structure

2.2.2 The concept of Performance Efficiency in bank

When evaluating the Performance Efficiency in business, it can be based on

two indicators that are absolute efficiency and relative efficiency

Absolute efficiency: Measured by business results minus cost to achieve results

This ratio reflects the scale, volume and profits gained in specific conditions, time and place

Relative efficiency: based on comparative ratio between inputs and outputs

Relative efficiency is defined as: Efficiency = output / input or Efficiency = input / output This assessment is very convenient on comparing different organizations from sizes, space and time

2.2.3 Methods of measuring bank’s Performance Efficiency

In methodology, previous studies used three different approaches to measuring business performance: measure Performance Efficiency by the ratio method, measure the Performance Efficiency from market data and measure Performance Efficiency from profit margin The results of applying different measurements can lead to different results No research has shown which measurement method is the best For the selection methods, the authors combine the assessment, analysis of advantages and disadvantages of each method, then approach methods of measuring bank’s Performance Efficiency in accordance with the scope and objects of research

2.3 The impact of liquidity risk on bank’s Performance Efficiency

2.3.1 Theory of the relationship between liquidity risk and bank’s Performance Efficiency

2.3.1.1 The theory of risk for profit

2.3.1.2 The theory of Banking Specificities Theory

Firstly, Gap on the approach There are many studies on liquidity risk, but studies

focus on the causes of liquidity risk (Ahmed, Ahmed & Naqvi, 2011; Angora & Roulet, 2011; Bonfim & Kim, 2014; Bunda & Desquilbet, 2008; Gibilaro, Giannotti & Mattarocci, 2010; Horváth et al., 2012; Rauch et al., 2010; Skully & Perera, 2012; Vodova, 2013) and studies on liquidity risk management to stabilize banks such as (Acharya & Naqvi, 2012; Scannella, 2016; Wagner, 2007) Researchs on liquidity risk

is also considered one of the types of bank risk such as credit risk

(Bissoondoyal-Bheenick & Treepongkaruna, 2011) or one of the factors affecting bank’s Performance Efficiency (Athanasoglou et al., 2008; Shen et al., 2009) However, only

a few studies that combine an analysis of factors affecting liquidity risk and the impact

of liquidity risk on bank’s Performance Efficiency across multiple countries

Secondly, Gap on the spaces and periods researchs Most empirial researchs

have taken in the region of one country only, except study of (Roman & Sargu, 2015) based on European data or (Bordeleau & Graham, 2010) in America, (Shen et al., 2009) both in Europe and America Cross-countries studies on aspect to examine the interlinkage between liquidity risk and bank’s performance efficiency According to the author, these are only three well-known empirical studies of liquidity risk and bank’s performance efficiency across multiple countries and are published in highly reliable journals In the case of South East Asia countries, there is no separate study

on the impact of liquidity risk on bank’s performance efficiency across multiple countries Different spaces and periods researchs, will be result in dissimilar results on the relationship between liquidity risk and bank’s performance efficiency

Thirdly, Gap on the measurement elements Other empirical studies also showed

that there are many factors affecting the bank’s performance efficiency such as: lending chanel through the ratio of loan of total assets (Nguyen Viet Hung, 2008; Gul

et al., 2011; Trinh Quoc Trung & Nguyen Minh Sang, 2013…); Banking capital mobilization and operation using bank capita used be the ratio of total mobilized capital of total loan (Nguyen Viet Hung, 2008; Nguyen Thi Loan & Tran Thi Ngoc Hanh, 2013 ); the size of equity (Nguyen Viet Hung, 2008; Gul et al., 2011; Nguyen Thi Loan & Tran Thi Ngoc Hanh, 2013; Ongore & Kusa, 2013 …); the size of asset (Nguyen Viet Hu, 2008; Gul et al., 2011; Ongore & Kusa, 2013; Ayadi, 2014 …), the economic growth rate (Gul et al., 2011; Ongore & Kusa, 2013;…), the inflation rate

(Gul et al., 2011; Ongore & Kusa, 2013;…) Particularly, the factors influencing liquidity risk have used but rarely Some studies had just used liquidity ratios to measure liquidity risk but Poorman and Blake (2005)2 indicated that it was not enough

to measure liquidity just using liquidity ratios and it was not the best solution In this study, the combination of financial gap method and liquidity ratios to measure liquidity risk in banking business

Fourthly, Gaps in character:

Liteliture review showed that the trend of studying the impact of liquidity risk on bank’s performance efficiency is mainly (Sufian & Chong, 2008; Sayedi, 2014; Oluwasegun & Samuel, 2015; Lartey, Antwi & Boadi, 2013; Bourke,1989; Tabari, Ahmadi & Emami, 2013; Arif & Nauman Anees, 2012; Bassey & Moses, 2015; Ferrouhi, 2014; Alshatti, 2015; Aburime, 2009; Athanasoglou et al., 2008; Naceur & Kandil, 2009; Wasiuzzaman & Tarmizi, 2010; Lee & Kim, 2013) Other studies have approached the impact of bank’s performance efficiency on liquidity risk across multiple countries (Abdullah & Khan, 2012; Roman & Sargu, 2015) So, the recent trend, scientists and managers are very interested in the impact of liquidity risk on bank’s performance efficiency Especially, the impact of the financial crisis on bank’s performance efficiency (Lee & Kim, 2013)

However, it have rarely researched combined analysis of factors affecting liquidity risk with the impact of liquidity risk on bank’s performance efficiency across multiple countries In general, previous studies identified that it is necessary to study the impact of liquidity risk on bank’s performance efficiency

In this study, we use each bank’s financing gap ratio (FGAP) as the independent variable and financial crisis variable to compare the case studies of South East Asia and Vietnamese countries From which to propose policy suggestions for Vietnam Besides, research thesis the case of Vietnam to to compare the different impact of liquidity risk on bank’s performance efficiency with the case of South East Asia countries

2

Diamond and Dybvig (1983) developed a model to explain why banks choose to issue deposits that are more liquid than their assets

10

CHAPTER 3: RESEARCH METHODOLOGY

3.1 Research Methods

Due to the limitations of the Pool OLS model in panel data estimation (Kiviet, 1995),

so that FEM and REM model can be used to analyse individual effects However, since FEM and REM do not handle endogenous phenomena (Ahn & Schmidt, 1995), so the SGMM estimation technique is used to solve the problems mentioned above (Arellano & Bond, 1991; Hansen, 1982; Hansen, Heaton, & Yaron, 1996) The the system Generalized Method of Moments (SGMM) estimators applied to panel data models address the problem of the potential endogeneity of all explanatory variables, measurement errors and omitted variables Stata software version 12 was used to for all the estimations to determine the results of this study

3.2 Research models of factors influencing liquidity risk

3.2.1 Research models

The research has combined the approach of (Ferrouhi & Lahadiri, 2014) and (Trenca, Petria & Corovei, 2015) with the addition of lag liquidity variables and credit risk variables to examine factors influencing liquidity risk in South-East Asia case Econometric models are thus presented as follows:

Models (1): LIQUIDITYRISK t = f(α, LIQUIDITYRISK t-1, SIZE it , SIZE it ^2, LIA it , LLR it, LADS it, ETA it , LLP it , NIM it GDP it , INF it , M2 it , D_CRIS t , u)

include liquidity risk (FGAPit – Bank’s loans – customer deposits/ total assets; NLTAit - Loans /total Assets, NLSTit - Loans/deposits + Short term liabilities); The independent variables include lag liquidity risk variables; Bank size, Natural logarithm of total assets (SIZEit); Natural logarithm of total assets squared (SIZEit^2); liquid assets/ total asset(LIAit); liquid assets / total Loans (LLRit); liquid assets / short term liabilities (LADSit); the ratio of equity to total assets (ETAit); Loan loss reserves/Total loans (LLPit); Net interest income (NIMit) Macroeconomic variables include change in GDP (GDPit); change in inflation (INFit); change in money supply (M2it); Dummy variables the impact of crisis on banking performance efficiency (D_CRIS)

With α (the constant), i (bank), t (year), u (the error)

Table 3.1: Relationship between dependent and independent variable in the model of factors influencing bank liquidity risk

SUMMARY VARIABLES TO BE USED IN THE MODEL

(Factors affecting liquidity risk in South East Asia countries and Vietnam)

d

Data Source The dependent variables

Bank’s loans – customer deposits/ total assets Ferrouhi & Lahadiri (2014), Shen et al., (2009);

Saunders & Cornett (2006), Arif & Nauman Anees (2012)

BankScope

NLTA

Loans /total Assets Ferrouhi & Lahadiri (2014); Lucchetta (2007);

Vodova (2011), Roman & Sargu (2015);

Munteanu (2012)

BankScope

NLST

Loans/deposits + Short term liabilities Ferrouhi & Lahadiri (2014); Vodova (2011),

Saunders & Cornett (2006), Shen et al., (2009);

Munteanu (2012)

BankScope

The independent variables

d

Data Source Lag

Liquidity

risk

There is substantial evidence that

bank Liquidity levels persist from

one year to the next

According to the economic theory

of scale, The larger the bank size,

The lower the liquidity risk

Log (total Assets) Delécha et al., (2012); Lucchetta (2007); Truong

Quang Thong, (2013); Vodova (2011), Trenca, Petria & Corovei (2015), Ferrouhi & Lahadiri (2014), Bonfim & Kim (2014); Horvàth et al., (2012)

(-) BankScope

SIZE ^2

Hypothesis “too big too fall” the

larger the bank size, the greater the

risk

Log (total Assets)^2 Shen et al., (2009); Truong Quang Thong,

(2013); Lee & Kim (2013); Ferrouhi &

Lahadiri (2014)

(+/-) BankScope

LIA

Components of liquid assets may

vary across countries, but generally

include cash, government securities,

interbank deposits, and short-term

marketable securities Lower

liquidity means higher risk

liquid assets/ total asset Delécha et al., (2012), Vodova (2011), Ferrouhi

& Lahadiri (2014), Ahmed et al., (2011)

(-) BankScope

12

LADS

Liquid assets /short term liabilities Vodova (2011), Cucinelli (2013), Delécha et al.,

(2012), Ferrouhi & Lahadiri (2014)

(-) BankScope

al., (2012), Ferrouhi & Lahadiri (2014)

(-) BankScope

ETA

Access to fragile financial structure

and dominance deposit structure

effects suggests that capital and

Liquidity risk are positively

correlated

the ratio of equity to total assets Delécha et al., (2012), Lucchetta (2007);

Cucinelli (2013), Munteanu (2012), Ferrouhi &

Lahadiri (2014)

(+) BankScope

LLP Banks with a higher proportion

of reserves may be those with more

aggressive lending strategies

Capital Structure is vulnerable, the

higher the risk

Loan loss reserves/Total loans Delécha et al., (2012), Cucinelli (2013), Trenca,

Petria & Corovei (2015), Arif & Nauman Anees (2012)

(+) BankScope

NIM The higher the interest from the loan

portfolio, The higher the liquidity

risk

(Interest income - Interest expense) /

Average asset

Delécha et al., (2012), Munteanu (2012);

Trenca, Petria & Corovei (2015), Roman &

INF Inflation has changed a lot,

affecting liquidity risk

Consumer Price Index Ferrouhi & Lahadiri (2014); Bonfim & Kim

(2014); Vodova ( 2011); Cucinelli (2013)

Notes: (-) negative correlation, (+) positive correlation, (- / +) nonlinearity Source: Self-synthesis of the author

3.3 Research models the impact of liquidity risk on Performance Efficiency

3.3.1 Research models

The research has combined the approach of (Growe et al., 2014) with the addition of crisis variables to examine the impact of liquidity risk on Performance Efficiency In addition, this study is based on the model (Ferrouhi, 2014) to supplement other variables to examine the impact of liquidity risk on Performance Efficiency in case of South-East Asia and Vietnam Econometric models are thus presented as follows:

Models (2): P t = f(α, P t-1, LIQUIDITY RISK it , CONTROL it , u)

From the equations above, the bank specific variables include liquidity risk (FGAPit – Bank’s loans – customer deposits/ total assets; NLTAit - Loans /total Assets, NLSTit - Loans/deposits+Short term liabilities); Control variables include bank size, Natural logarithm of total assets (SIZEit); Natural logarithm of total assets squared (SIZEit^2 ); liquid assets/ total asset(LIAit); liquid assets / total Loans (LLRit); liquid assets / short term liabilities (LADSit); the ratio of equity to total assets (ETAit); the ratio of loan loss provision to loans, (LLPit) Macroeconomic variables include change

in GDP (GDPit); change in inflation (INFit); Dummy variables the impact of crisis on banking

performance efficiency (D_CRIS) The dependent variables include P it(NIM, ROA, ROE); with α (the constant), i (bank), t (year), u (the error)

Table 3.2: Relationship between dependent and independent variable in the model of the the impact of liquidity risk on Performance Efficiency in case of South-East Asia

14

SUMMARY VARIABLES TO BE USED IN THE MODEL 2

(The impact of liquidity risk on Performance Efficiency in case of South-East Asia)

Performance Efficiency

ROA

Return on Assets Bassey & Moses (2015), Anbar & Alper (2011) Ferrouhi

(2014), Arif & Nauman Anees (2012), Growe et al., (2014)

BankScope

(2014); Arif & Nauman Anees (2012); Growe et al., (2014);

Anbar & Alper (2011)

(2006), Bunda & Desquilbet (2008); Shen et al., (2009) (+) BankScope

NLST Loans/deposits + Short term liabilities Munteanu (2012), Ferrouhi (2014), Growe et al., (2014);

Anbar & Alper (2011); Ayaydin & Karakaya (2014) (+) BankScope

lag (P t-1 ) Ayaydin & Karakaya ( 2014); Lee & Hsieh (2013); Perera et

al., (2013); Growe et al., (2014)

(+) BankScope

market power , improving technology

efficiency at low cost

Log (total Assets) Munteanu (2012), Lee & Hsieh (2013); Anbar & Alper

(2011); Ferrouhi (2014); Growe et al., (2014)

(+) BankScope

will reduce profits

Log (total Assets)^2 Shen et al., (2009); Growe et al., (2014); Ayaydin & Karakaya

(2014); Lee & Kim (2013)

(-/+) BankScope

vary across countries, but generally include cash, government securities, interbank deposits, and short-term

liquid assets/ total asset Kosmidou et al., (2005), Poposka & Trpkoski (2013), Shen et

al., (2009), Ferrouhi (2014); Growe et al., (2014); Anbar &

Alper (2011); Ayaydin & Karakaya (2014)

(+) BankScope

Notes: (-) negative correlation, (+) positive correlation, (- / +) nonlinearity Source: summary from research results of the author

marketable securities Lower liquidity means higher risk

liabilities

Shen et al., (2009); Ferrouhi (2014); Growe et al., (2014);

Anbar & Alper (2011); Ayaydin &Karakaya (2014)

(-) BankScope

(2014), Anbar & Alper (2011)

(-) BankScope

ETA

the higher the capital, the lower the risk the ratio of equity to total assets Shen et al., (2009), Ferrouhi (2014); Growe et al., (2014);

Anbar & Alper (2011); Ayaydin & Karakaya (2014)

(+) BankScope

the bank's profitability

Loan loss reserves/Total loans Ayaydin & Karakaya (2014); Shen et al., (2009);

Trujillo-Ponce (2013); Growe et al., (2014)

(+)

ADB

of each year

Consumer Price Index Ayaydin & Karakaya ( 2014); Shen et al., (2009); Sufian

& Chong (2008); Ferrouhi (2014); Growe et al., (2014);

Anbar & Alper (2011)

16

CHAPTER 4: EMPIRICAL RESULTS

4.1 Factors affecting Liquidity Risk

4.1.1 Descriptive statistics

Table 1.1: Descriptive statistics for variables, case study of South East Asia countries

Table 4.2: Descriptive statistics for variables, case study in Viet Nam

4.1.2 Analysis of correlation coefficient

Table 4.3: Correlation between independent variables in the model of factors affecting Liquidity Risk, case study of South East Asia countries

Table 4.4: Correlation between independent variables in the model of factors affecting Liquidity Risk, case study in Vietnam

4.1.3 Analysis and discussion of results, case studies of South East Asia countries

To evaluate factors affecting Liquidity Risk in banks, the study used 12 different regressions (Table 4.5) The study used the F, LM, Hausman test to select the appropriate model for the analysis VIF ratio is less than 20, so the model does not exist multi-collinear phenomenon The P-values of F, LM test are less than 5% (<0.05), there is evidence to reject the hypothesis Hausman's test for the p-value (Prob> F) of the model is less than 0.05 (Table 4.5), this suggests that the FEM model is more appropriate than REM

Wooldridge test and Wald test with P-value (<0.05) indicates the presence of variance and self-correlation in FEM, which results in inefficient regression coefficients The author continues to use the SGMM method to estimate the model, which eliminates the problem of variance, autocorrelation or endogenity so the estimation result will be effective and durable The results of the final analysis are based on the SGMM method Sargan Test

to test the over-identifying of tool variables The results show that the p-value coefficient is greater than 0.05, indicating that the tool variable used in the GMM model is suitable The results are robust and fully analy (Table 4.5)

Table 4.5: Factors affecting bank liquidity risk, case study of Southeast Asian countries (Appendix)

Models (1): LIQUIDITYRISK t = f(α, LIQUIDITYRISK t-1, SIZE it , SIZE it ^2, LIA it , LLR it, LADS it ETA it , LLP it , NIM it GDP it , INF it , M2 it ,D_CRIS t , u)

Dependent variable: Liquidity risk (FGAP, NLTA, NLST)

Independent variable: (SIZE- natural logarithm of total assets; SIZE^2- natural logarithm of total assets squared; LIA- the ratio liquid assets to total assets; LLR- the ratio liquid assets / total Loans, LADS- the ratio liquid assets / short term liabilities; ETA- the ratio of equity to total assets.; LLP- the ratio of loan loss provision to loans, GDP- Annual percent change of GDP, INF- Annual percent change of inflation, M2 – Annual percent change of money supply; D_cris - Dummy variable )

Database from 2004 to 2016 Estimation technique: OSL, FEM, REM và SGMM.