Competiton and bank stability in asean countries an empirical analysis

The effect of competition measured by H-statistic, Lerner index and HHI on Z-score and equity ratio as measure of bank stability in ASEAN from 1995-2015.. 55 Table-9b: The effect of com

UNIVERSITY OF ECONOMICS ERASMUS UNVERSITY ROTTERDAM

HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES

VIETNAM – THE NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

COMPETITION AND BANK STABILITY IN

ASEAN COUNTRIES: AN EMPIRICAL

BY

Ms VU THI QUYNH

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, Nov 2017

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

VIETNAM THE NETHERLANDS

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

COMPETITON AND BANK STABILITY IN ASEAN COUNTRIES: AN EMPIRICAL

A thesis submitted in partial fulfilment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

By

Ms VU THI QUYNH

Academic Supervisor:

Dr NGUYEN THI THUY LINH

HO CHI MINH CITY, Nov 2017

Declaration

―I certify the content of this dissertation has not already been submitted for any degree and is not being currently submitted to any other degrees

I certify that, to the best of my knowledge, any assistance received in preparing this

dissertation and all source used, have been recorded in this dissertation.‖

Signature

Vu Thi Quynh Date: Nov 1st, 2017

I also thank all my friends who always stand by my side with encouragement.

Lastly, I would like to thank my family for supporting not only in this thesis but also in my life

Abstract

Investigating the relationship between competition and bank stability has been at center of academic and policy issues, intensifying after occurrence of system banking crises over the last three decades The global trend towards consolidation in the financial industry, the banking sector in the ASEAN region experienced a dramatic shift from deregulation to regulation based on the experience of 1997 Asian Financial crisis Hence, this study examines the impact on bank stability of competition in ASEAN market by taking into account crisis periods Using unbalanced dataset from more 200 commercial banks across 7 ASEAN countries over the period from 1995 to 2015, this study provides empirical evidence supporting the competition point of view that a non-linear or U-shaped nexus between competition and bank stability Banks were found to be highly volatile and lost capitalization during the 1997 Asian crisis, whereas the 2007-2008 global financial crisis stage did not directly affect Asian banks Besides, smaller banks in this region were more competitive than large banks over same period and they may contribute to improve financial soundness On the other hands, the findings also provide some recommendations for policymakers in ASEAN economics

Key words: Competition, Stability, Market power, Concentration, Fragility

vi

CONTENTS

CHAPTER 1: Introduction 1

1.1 Problem statement 1

1.2 Scope of the study 2

1.3 Research objectives and research questions 4

1.4 Research methodology 5

1.5 Contributions and implications 5

1.6 Limitations and structure of the thesis 6

CHAPTER 2: Literature review 7

2.1 Theoretical literature 7

2.1.1 Bank stability and competition 7

2.1.1.1 Charter value 8

2.1.1.2 The efficiency–stability debate 9

2.1.1.3 The concentration–contestability debate 11

2.1.1.4 Moral hazard and adverse selection problem 12

2.1.2 Main arguments about relationship between competition and bank stability 13

2.1.2.1 The traditional competition-instability view 13

2.1.2.2 The modern competition-stability view 15

2.1.2.3 The ambiguous 15

2.2 Empirical literature 16

2.3 Hypothesis construction 18

2.3.1 Competition and bank stability 18

2.3.2 Effect of financial crises on the competition and bank stability relationship 19

CHAPTER 3: Methodology 21

3.1 Model specifications 21

3.1.1 General specification 21

3.1.2 Specific regression equations 22

3.2 Description of the variables 23

3.2.1 Dependent variables 25

3.2.1.1 Bank Z-score 25

3.2.1.2 Capitalization ratio 27

3.2.1.3 Non-performing loan ratio 27

3.2.2 Measures of bank competition 28

3.2.2.1 Lerner index 29

3.2.2.2 The Panzar-Rosse H-statistic: Non-Structural approach 30

3.2.2.3 The Herfindahl-Hirschman Index (HHI): Structural approach 32

3.2.3 Other control variables 33

3.2.3.1 The bank specific indicators 33

3.2.3.2 Regulatory indicators 34

3.2.3.3 Macro-economic indicators 35

3.2.3.4 Instrumental variables 37

3.3 Estimation method: Two-stage Least Square (2SLS) 38

3.4 Data sources 41

CHAPTER 4: Empirical analysis 43

4.1 Descriptive statistics and correlation structure 43

4.1.1 Description statistics 43

4.1.1.1 Competition and stability in ASEAN banks 43

4.1.1.2 Characteristics of ASEAN banks 44

4.1.1.3 Cross-country differences in bank regulatory policies 44

4.1.2 Competition and bank stability 46

4.1.3 The variable correlation 49

4.1.4 The first-stage regression 51

4.2 Results and discussion 53

4.2.1 The effect competition on stability 53

4.2.2 The impact of Asian crisis and global financial crisis on competition and bank stability relationship 60

4.2.3 The effect of Market Entry Conditions and Bank regulatory factor 63

4.2.4 Bank-level and country-level factors on competition-stability nexus 64

4.3 Robustness checks 65

CHAPTER 5: Conclusion 67

5.1 Conclusion 67

5.2 Policy implications 68

5.3 Limitations and suggestions for further research 69

REFERENCES 71

APPENDICES 75

viii

LIST OF TABLES AND FIGURES

Table-1: Summary some previous studies 20

Table-2: Summary of the variables used in the analysis 24

Table-3: ASEAN and country wise descriptive statistic of the variables 45

Table-4: Yearly average of H-statistic, Lerner index, HHI based on loan and CR3 based on loan for ASEAN-5 during 1990-2014 46

Table-5: Pearson pair wise correlation matrix of independent variables used in the analysis 50

Table-6: Variance Inflation Factor (VIF) 50

Table-7: The first-stage regression between instrumental variables and Z-score as a measure bank competition 51

Table-8: The comparison between FEM, REM and Pooled model The P-value shows the significant level at 1% and 5% 52

Table-9a The effect of competition measured by H-statistic, Lerner index and HHI on Z-score and equity ratio as measure of bank stability in ASEAN from 1995-2015 55

Table-9b: The effect of competition measured by H-statistic, Lerner index and HHI on NPL ratio as a measure of bank stability in ASEAN from 1995-2015 58

Table-10: The combined effect of bank competition and crises 61

Table-11: Summary findings 65

Fig 1: Log of Z-score, H statistic and Lerner index of ASEAN during 1995-2015 48

Fig 2: The correlation between LnZ-score, NPL ratio and Capital ratio of ASEAN banks from 1995-2015 49

Fig 3: ASEAN and other regions – Z-score and Concentration index 76

CHAPTER 1: Introduction

1.1 Problem statement

The impact of competition on bank stability has been an issue of active disputation

in both academic and policy circles for over three decades and especially since 1997-1998 Asian and the 2007-2008 global financial crises

This debate was intensified by the deregulation of branches and activity restrictions

in the United States during the late 1970s and early 1980s, lead to the international process of banking liberalization in both matured and emerging economies This makes more competition in the banking sector which large banks from advanced countries operating at low profit margin penetrate in developing countries with relatively high profit margin Meanwhile, some economists stated that these deregulation also positively impact on financial depth (Rice & Strahan, 2010), growth (Cetorelli & Gambera, 2001), income distribution (Beth at al., 2010b) and efficiency (Bertrand et al., 2007), and then led to the belief that the bank's fierce competition will promote a more efficient banking system Accordingly, competition is also seen as a pre-condition of efficient, innovative and developed financial system (Demirguc-Kunt & Peria, 2010; Weill, 2013; Apergis, Fafaliou, & Polemis, 2016)

Unfortunately, there is a fact that, coupled with more competition through financial liberalization is the emergence of systemic banking crises in the last two decades of the 20th century and leading regulatory failures to bring the banking system in discipline have raised concerns among policy makers and academics regarding the subsequent effect

of competition on bank stability in the banking system Consequently, the concern about the extent to which competition is responsible for these crises is still no consensus as to whether high or low competition leads to bank stability in the banking system

In terms of banking sector in emerging countries, competition has been increasing due to the trend of penetration into the banking market of large foreign banks from advanced countries This prompted domestic banks to accelerate the consolidation process to protect their market power, leading to the appearance of "too big to fail" banks and the moral hazard incentives are more likely to exploit government aid (Berger & Mester, 2003)

Furthermore, businesses depend significantly on banks for external funding to secure their operations (Adam, 2008) As a consequence, bank stability is a major concern for policy marker in the formulation of important policies In recent decades the relationship between competition and bank stability has been researched empirically focusing on both developed and developing nations

Nevertheless, the findings of those studies ended up with conflicting experimental results keeping this link still a puzzle In addition, in order to resolve the consolidation logically, regulators need to consider how it could affect their overall goal of the financial system, which is to maximize social welfare

This study assesses again this relationship, or further effect of competition on financial crisis on the competition-stability nexus that may be impaired by crisis Crisis may lead the banking sector to adopt divergent reform strategies, such as capital regulation, activity restriction, and consolidation which may change the market power or competition and risk taking behavior of the banks

1.2 Scope of the study

Understand how competition can form the incentives of the credit institutions for risk taking is essential for a stable and efficient banking system that can finance timely profitable investment opportunities and remain sustainable economic growth By the above importance roles, few researches have focused on the impact of competition on the stability of banking system in the ASEAN context This study also contributes to the full implementation of this gap by using samples of 226 commercial banks from 7 countries

of ASEAN1 area These countries include Cambodia, Indonesia, Lao PDR, Malaysia, The Philippines, Thailand and Vietnam

In terms of this study‘s territorial scope, the ASEAN region has seen an attractive sample providing a fertile laboratory to analyze this relationship, because its banking industry has been experienced liberalization via foreign bank penetration in early 1990s, followed by deregulation, regional economic integration, and tremendous consolidation in late 1990s as port 1997-98 Asian financial crisis strategies Specifically, the banking sector in this area has been sequentially and significantly affected from the Asian crisis in

1 The Association of Southeast Asian Nations (ASEAN) is a regional organization comprising ten Southeast Asian nations that promotes intergovernmental cooperation and facilitates economic integration amongst its members Since its formation on August 8, 1967 by Indonesia, Malaysia, the Philippines, Singapore, and Thailand, membership has expanded to include Brunei Darussalam, Cambodia, Lao PDR, Myanmar, and Vietnam.

1997, severer than any other emerging areas Indonesia, Thailand has been countries being affected the most (1998-1999, NPLs ratio from 33% to nearly 50% (World Bank)) Malaysia, Laos and the Philippines have experienced a sudden price collapse, while Singapore and Vietnam are almost unaffected

In global financial crisis, this region was also influenced due to most of the economy here where this export-oriented and highly dependent on international trade and foreign investment Southeast Asia region exhibited that the degree of vulnerability depends on openness to international capital flows, on the business model for banks funding and access to the real estate sector In the short run, the export sector, the stock market and the money markets, banking and financial services are strongly affected by the global financial crisis The freezing of the financial system in developed countries has led to the reduction or narrowing of a series of activities in manufacturing, business or personal consumption as well as the belief of the people in decline

Besides, in the process of deregulation to deregulation, many policy makers are also imposing market entry restrictions and bank regulation and supervision Hence, the banking sector has been strong entry barriers to the activities and the limited import allowances In some countries, it is dominated by the intermediate state, which would benefit from assistance in case of distress Barriers to the emergence of banks in origin countries to increase after the government stopped licensing new bank established from September 8, 2008 (Vietnam) In addition to the provisions of its charter capital, as the number of years required to continuously be profitable, the new bank established under the close monitoring of the State Bank

In addition, the barriers to entry banking sectors are also reflected through market segment, the target market that banks are targeting, brand value and customer base, customer loyalty that bank customers have built for such barriers, the bank market is mainly national However, for the countries signed a number of international trade agreements such as WTO and the ASEAN Economic Community (AEC) have posed enormous challenges for the banking sector in the adjustment process and reforms to move towards a banking system of sustainable development and stability

In brief, ASEAN banking market is distinctive by the following reasons Firstly, ASEAN‘s central banks pushed commercial banks towards consolidation in order to attain bank stability Secondly, the governors of ASEAN central banks endorsed the ASEAN Banking Integration Framework (ABIF) in 2011 which will be implemented

initially among ASEAN-5 including Indonesia, Malaysia, the Philippines, Singapore and Thailand by 2020 ABIF may extend competition and make the banking market enhance efficiency and attain economy of scale (Yamanaka, 2014) However, the resulting increased competition in the regional banking market might push small banks towards further consolidation in order to strength their domestic existence and better compete with regional banks (Asia Development Bank, 2013) This initiative towards high concentration may allow regional banks to enjoy high monopoly power which is issue of concern for policy makers and bank regulators, because high monopoly power leads to increase loan interest rates which undermine easy access to credit and financial inclusion, and put bank stability of the region at risk due to the tightknit regional banking system In addition, despite a good number of studies that examined the nexus between competition and bank stability, only few cover emerging Asia

The crises have prompted a broad debate to find out the structural and regulatory policies that can improve the resilience of the banking market, both in matured economies and emerging countries In fact, the banking sector of the ASEAN countries has undergone a dramatic transformation in the three decades from deregulation to re-regulation based on the experience of the Asian financial crisis In these debates, there are two typical arguments as well as its solution, "competition-fragility‖ and ―competition-stability‖ What of market power or market competition will help stabilize the financial and banking system run better? Thus, the structure of the banking market and the level of competition among financial intermediaries is an important aspect for establishing policies and regulations, from government, central banks to supervisory authorities This paper uses a sample of 226 commercial banks in ASEAN countries during 1995-2015 which captures 1997-98 Asian financial and 2007-08 global financial crises

to investigate the relationship between competition and bank stability though applying various risk measures Indeed, this study also extend to examine whether financial crises and too-big-to-fail subsidies influence, or some regulatory framework such as deposit insurance, activity restrictions influence this nexus in ASEAN countries

1.3 Research objectives and research questions

This study aims to analyze the relationship between competition and bank stability

in ASEAN countries consist of Cambodia, Indonesia, Lao PHD, Malaysia, Myanmar, Thailand and Vietnam in the context that bank consolidation raised massively to face the consequences of the 1997 Asia crisis and 2008 crisis In particular,

- Does competition affect bank stability in ASEAN countries?

- Does the 1997 Asian crisis and 2008 global financial crises impact on recent crises

on the relationship of competition and stability?

1.4 Research methodology

This study employs the Two-Stage Least Squares (2SLS) with fixed-effect regression estimator in static IV model Competition measures include the non-structural approaches (H-statistic, Lerner index) and traditional approach proxies for bank concentration (HHI index and CR3 (concentration ratio of large three banks in loan market) For proxy of bank stability, the Z-score is used as the main measure Besides, Non-performing loans and Capitalization ratio are applied also as two alternative measures of financial distress/stability and risk-taking measurement (Fu et al, 2014; Berger, 2009)

1.5 Contributions and implications

The banking system is increasingly contributing significantly to the economic development of a region, as enterprises now view the bank as their external source of funds to sustain their operations The international process of banking liberalization also entailed the arrival of large foreign banks into the emerging markets, which affected stability and increased competition in the banking sector In addition, according to fact observation, this internationalization trend has also gone hand in hand with the increased emergences of systemic banking crises in the last two decades of the 20th century

Thus, in an attempt to say this study contributes to literature in several ways Firstly,

it contributes to the debate on the competition / stability relationship by providing new evidence from emerging ASEAN countries Secondly, the study uses a long panel data covering 21 years from 1995 to 2015, which captures the 1997-98 Asian and the 2007-08 global financial crises This allows the development of an expanded database to examine the connection between competition, crisis and bank stability To the best of my knowledge, all relevant documents that examine the relationship between competition and bank stability that provide further analysis of the effects of financial crises or based on pre-global financial crisis, exceptions of Fu et al (2014), only limited to the focus of the 2007-08 global financial crisis In this study, effect of 1997-1998 Asian crises also provided

Thirdly, the study examines the impact of recent banking crises on stability nexus to significant policy implications regarding consolidation, capitalization, and activity restrictions

competition-Fourthly, the research adds the regulatory variables such as activity restrictions, entry into banking, capital requirements, and deposit insurance in the economic specification since they can significantly affect the link between competition and stability (Agoraki, Delis & Pasiouras, 2011) Finally, this research controls endogenous ability between competition and bank stability by using 2SLS estimators that introduce stock market capitalization, financial freedom and property right as additional instrumental variables

1.6 Limitations and structure of the thesis

This study also could not avoid certain restrictions, typically database, measures and methodology The sample only captures the commercial banks, ignoring other banking activities such as cooperative banks, real estate and mortgage banks Investment banks are excluded in the analysis sample, to ensure some homogeneity in the type of and to allow for data comparability Besides, some regulatory variable cannot be collected adequately due to lack of reporting by countries which then causes to error measurement in the calculation of the variable Limiting measure of bank concentration: HHI deposits (A country-level indicator of bank concentration, represented by the Herfindahl– Hirschman deposits index, with higher values demonstrating higher market concentration (Berger, 2009) Other limitations of this writing that it does not analyze influence of foreign banks

at the host countries It not, that may provide a wedge between the measurement of market power for domestic versus foreign banks

This study is organized into five sections Section 1 provides a brief introduction about this topic Section 2 gives a detailed review of relevant literatures regarding the relationship between competition and bank stability / fragility Section 3 deals with the different approaches to measure competition and bank stability and the models used to investigate this relationship This part also shows the specific data sources The research findings and their explanations are reported in Section 4 Final parts such as conclusions, policy recommendations, limitations, and suggestion for further researches are contained

in section 5

CHAPTER 2: Literature review

This section contributes to the related theories and debates to the stability nexus by providing empirical evidence which support many conflicting views as well as introducing various approach method to investigate this relationship

competition-2.1 Theoretical literature

Banks are a service industry They contribute to economic growth not by producing physical goods but by providing financial means to facilitate the production of other industries An efficient banking system will make the biggest contribution to economic growth Hence, the impact of a competitive banking system or market power needs to be discussed in terms of allocation efficiency and effectiveness In other words, the banking sector is often seen as more vulnerable than other sectors If it does not work well, there will be potential for financial fragility (Northcott, 2004)

Instability in the banking sector may damage the entire economy by sinking credit facility, and distorting the interbank loan market and the payment system Therefore, a primary concern of banking regulators and policy makers is to formulate policies that foster bank stability and reduce the risk of bank failure

In analyzing the cause of fragility in banking industry, Keeley (1990) initiated show this academic debate theoretically and empirically that the banking failures in the U.S since the 1970s and 1980s were partly due to a rise in competition come from deregulation in the banking market here Keeley has explained that an increase in competition eroded monopoly rents and hence charter value

2.1.1 Bank stability and competition

The traditional literature on financial structure shows that market power is necessary to ensure stability in banking The explanation is that the best performing and profitable banks can withstand the shock of their balance sheet According to Northcott (2004), the probability of a bank receiving a shock will depend partly on the risk-taking behavior

More views show that highly leveraged companies are motivated to engage in risky behaviors or moral hazard behaviors If it does not expected, lenders will bear the cost (Jensen & Meckling, 1976) This agency issue is particularly strong and sensitive for

bank sector, as they tend to be highly utilized Many creditors can be depositors who are mall, widely dispersed and tend not to be fully informed about bank operations and potential risks The emergence of deposit insurance continues to diminish the incentives

of depositors to track bank boycott

Not only that, banks that are large enough to "too-big-to-fail" will be more likely to tolerate risky behavior by resorting to government subsidies if bad something happens

As such, large banks with market power are generally considered to have incentives to minimize risky behavior (the charter-value terms) and improve the quality of their assets (the screening theories) However, as seen in effective documentation, the problem is not that straightforward

2.1.1.1 Charter value

In a study, Keeley (1990) shows that the increase of bank failures in the United States in the 1980s was partly due to increased competition in the banking sector The main point of his argument is the relationship between charter value and risk behavior Charter value is the benefit that a bank‘s owners derive from its future activities and

it represents the opportunity cost for going bankruptcy To determine risky behavior, the bank must balance the gains of increasing risk-taking with the loss of ―charter value‖ if it fails Keeley observes that banks with power have market higher rents and hence have higher charter values This generates higher opportunity costs for bankruptcy, which reduces the activity of risky behavior

The rise in competition then leads to a reduction in the charter value, along with an increase in risk-taking Different authors have expanded the material, modeling different elements that influence charter values According to Besanko and Thakor (1993), the value comes from the exclusive information obtained from relationship lending in a unified market Besides, Perotti and Suarez (2002) suggest that the charter value is increased when policy makers have a policy of promoting the takeover of failed banks by solvent investments, leads to larger rents for incumbents

In other hands, any factor that raises the opportunity cost of bankruptcy will be consistent with charter value theory Hellman et al (2000) examine charter value in a regulated capital environment They argue that capital requirements put conflicting pressures on risk-taking incentives The higher capital requirement will reduce motivation

to take on risk by raising losses to shareholder if the bank goes bankrupt ("capital-at-risk"

effect) However, higher capital requirements also reduce charter value, putting pressure

on risk taking incentives (or called as "charter value" effect) Using the dynamic model where banks compete for deposits, Hellman et al suggest that, as long as deposit rates are determined freely, banks in competitive markets have motivation to raise their deposit rates to expand their deposit base ("market-stealing" effect) This reduces profits, declines charter values, and promotes risky behavior In addition, Repullo (2003) expands this model by modeling competition clearly in the deposit market Banks can invest in

"gambling" or "prudent" assets Repullo suggests that if there is no capital requirement, only a gambling balance exists in very competitive and very monopolistic markets The competitive structure can make a difference In particular, for a median market, either a gambling or a prudent risk-taking equilibrium exists However, Repullo later proposed that, in an environment with capital requirements, a "prudent" equilibrium would prevail Agusman et al (2006), Liu et al (2012) and W Soedarmono et al (2013) have the same findings that greater charter value in Asian listed banks fails to decrease banks‘ income volatility Higher bank competition results in greater stability in Asian banks Their findings are also in line with Schaeck et al (2009) who investigates the competition-stability connection applying the Panzar-Rosse approach Based on information both developed and developing countries, they find that a more competitive banking market is less prone to a systemic banking crisis and exhibit increased time to crisis

2.1.1.2 The efficiency–stability debate

The traditional view of the trade-off between efficiency and stability is that competition improves efficiency (and then simulate growth), but that market power is also necessary for stability Nevertheless, as stated above the documents show that problem is not that straightforward There is no consensus in the literature on whether perfect competition or market power would promote allocation efficiency, financial stability, growth and well-being, these makes the relationship between competition and bank stability still a puzzle:

In a traditional approach, perfect competition maximizes the amount of credit at the lowest price, and market power (measured by the likelihood of price above marginal cost) leads to a reduction in the supply level and higher prices

Where there is asymmetric information, market power could raise the bank's incentive to lend the relationship, benefiting people who are ambiguous as young companies that have no credit history or collateral

By directing credit to higher-quality projects first, screening can improve allocative efficiency The incentive to screen falls as the number of banks rises

By conducting credits to higher quality projects, screening can make better allocation efficiency However, the incentive for screening activities could be mitigated declines as the number of banks increases

The empirical literature largely focuses on the relationship between concentration (as a measure for market power) and profitability In some researches, concentration is indeed related to higher profits Nevertheless, there is evidence that the adverse impacts could be reduced (or eliminated) by a well-developed financial system or by policies that raises competition, such as low barriers to entry and few restrictions on bank activities Very few empirical studies have been done on the loan and screening theories While there is some evidence that lumpish borrowers benefit from a banking sector that represents market power degree, it is not conspicuous that such a sector enhances allocation efficiency across all firms, and thereby growth in the overall economy As a whole, it seems that a competitive environment is useful in motivating allocation efficiency, although the market concentration of banks may not be a good indicator of a competitive environment

A banking sector with greater market power may improve credit availability to some certain enterprises, and it may facilitate more efficient allocation by providing incentives to loan monitoring Productive efficiency, following the traditional framework,

is maximized by perfect competition

Studies discussed with regard to allocation efficiency expose that it is possible for a market can show both competitive behavior and concentration If so, it is able to benefit from both competition and economies of scale Is market power or competition necessary for stability?

When studies discuss distribution efficiency, it is possible for a market to exhibit both competitive behavior and concentration If so, it can benefit from both competition and economies of scale Market power required for stability?

The charter value assumes that a greater charter value declines a bank‘s incentive to take risks by raising the opportunity cost of bankrupt Market power has traditionally been associated with higher charter values, but any factor that increases the opportunity cost of bankruptcy would be consistent with the theory

Capital requirements could be useful in easing risk-taking behavior, regardless of the competitive structure of the market The literature on screening shows a similar conclusion A less competitive banking sector may have more incentive to screen loans, which enhances the quality of loan portfolios Nevertheless, regulations and policies that encourage transparent of bank portfolios and/or risk-based deposit insurance may raise the incentive to screen even in a more competitive environment

Hence, as market power may increase incentives to behave prudently, regulatory policies such as capital requirements, disclosure rules, and risk-based deposit insurance may provide incentives for banks to behave prudently even in a competitive market Indeed, this is consistent with a recent study that finds that banking crises are less likely

in more competitive and more concentrated banking systems

Neither extreme competitive (perfect competition nor monopoly) can be ideal or even possible Perfect competition may be even less possible in banking than in other sectors The standard IO assumptions certainly do not seem to use: small economies of scale relative to the market size, homogeneous products, perfect information, and free entry and exit (characterized as zero sunk costs) Thus, it may not be possible to completely eliminate market power in banking industry Therefore, the aim may not be possible to remove market power, but to facilitate an environment that encourages competitive behavior In this way, the potential costs of market power will be slightly reduced as realizing some benefits from residual market power (Northcott, 2004)

2.1.1.3 The concentration–contestability debate

Under the traditional SCP hypothesis2, the assessment of competitive behavior is relatively straightforward: more concentration is linked with higher market level Nevertheless, the experimental finding examining this hypothesis gives the conclusion not so simple

The contestability literature focuses on the competitive behavior of banks, rather than on concentration or the number of banks The growing consensus in this area is that

2

Market Structure Conduct Performance (SCP) hypothesis revisited using Stochastic Frontier Efficiency Analysis Its framework was derived from the neo-classical analysis of markets

contestability improves with less-severe entry restrictions, the presence of foreign banks,

a few of restrictions on the activities that banks can perform, and well-developed financial systems, the last two of which may indicate that competition from the non-bank sector is important

The literature on competitiveness focuses on the competitive behavior of banks rather than the concentration or number of banks The growing consensus in this area is improved competitiveness with less severe entry restrictions, the emergence of foreign banks, some restrictions on the activities that banks could perform, and the financial system develops well, the last two things that can show that competition from the non-banking industry is very important

Contestability is not necessarily related to concentration or the number of banks While it may seem counterintuitive that concentration and competition can exist together, certain characteristics of the banking sector may make this coexistence more understandable, such as the presence of asymmetric information and branches and the effect and use of new technologies On the other hands, the belief that some degree of market power was necessary to maintain stability in the banking sector led many countries to pursue policies that, implicitly or explicitly, restricted competition, following

to Northcott (2004) The literature is developing a better understanding of these issues (2004)

Competitiveness is not necessarily related to the concentration or number of banks While it may seem inconsistent that concentration and competition may exist together, certain banking characteristics may make this more understandable, such as presence The information and branches are disproportionate and influence and use new technologies

On the other hand, the belief that some degree of market power is needed to maintain stability in the banking sector has led many countries to pursue policies, implicitly or explicitly, restrict competition, in Northcott (2004) Literature is developing better understanding of these issues (2004)

2.1.1.4 Moral hazard and adverse selection problem

For credit markets, where exists asymmetric information, if the bank cope with excessive demand for loans by increasing interest rates (is seen as the price of a loan) to reduce demand can be affected by adverse selection problem and moral hazard

The adverse selection occurs because the bank could not understand the customer more clearly themselves This leads to asymmetric information, thus if the bank increases interest rates to limit demand for loans, good customers will not borrow, but bad customers will try to get loans because they know that if they can get another loan (for example, black market loans that the interest rate is very high, not even borrow) Therefore, as increasing the interest rates to limit the loan demand from customers, the bank has the ability to accumulate bad customers and expel the good customers

Moral hazard also comes from asymmetric information Since the banks are not able

to fully understand the business activities of borrowers than themselves, so after the loan action is completed, if the interest rate is higher than one they want, customers can change the purpose of using the loan to increase interest to make up for that higher interest rate The purpose of using such higher-yielding loans is often higher risk, so the ability to repay of customers is also affected badly

Therefore, in the credit market price could not always be used to limit the demand for loans from customers

2.1.2 Main arguments about relationship between competition and bank stability

The nexus between competition and bank stability is not clear in both theoretical and empirical literature Keeley represents the traditional competition-fragility view that excessive competition in the banking market reduces the market power and profit margin

of banks and puts them at high risk of bank failure and fragility in the banking market In contrast, Boy and Nicolo (2005) represents the modern competition-stability shows that excessive competition in the banking sector causes banks to reduce loan interest rates which will reduce moral hazard and the adverse selection problem of banks, lower their default risk, and strengthen bank stability Besides, the neutral or mutual view of Martinez-Miera and Repullo (2010) argues that both views above can coexist, and the nexus between competition and bank stability is supposed as non-linear or U-shaped link

In sum, there is no academic consensus on whether bank competition leads more or less stability in the banking system

2.1.2.1 The traditional competition-instability view

This view is grounded by Marcus (1984) and Keeley (1990), also known as the charter value hypothesis, suggests that more competition among banks leads to more fragility This hypothesis explains that the high competition in financial market weakens

market power, reduces the profit margin and capital buffer, and reduces the franchise value that encourages banks to adopt risk taking strategies to raise returns

Proponents of this view argue that in less competitive markets that large banks can gain better from economies of scale and scope, and can diversify better their portfolios compared to small banks (Diamond & Dybvig, 1983) More specifically, competition in the deposits market leads banks to implement risk-taking schemes due to the contraction

in bank‘s franchise value "Charter value" hypothesis supposes that an enhance in the degree of competition may cause both to a decrease in lending rates and a rise in deposit rates, with a consequent reduction in profit margins This would also lead to a decline in the franchise value of banks and encourage more risk-taking activities, which can have negative influences on the stability of individual institutions and of the whole system Similarly, Hellmann et al (2000) suggest that the removal of interest ceilings on deposits reduces the franchise value and motivates moral hazard behaviors by banks (e.g shift risks to depositors) Broecker (1990) supported this hypothesis by pointing out a negative correlation between the average credit quality and the number of banks in the system

In addition, many argue that a small amount of large banks is also easy to monitor and supervisor in a less competitive market Allen and Gale (2001) continue to argue that banks earn less information rent from the relationship with borrowers in competitive markets, which discourages to monitor them being cautious and may lead to increase moral hazard and adverse selection problem

Besanko and Thakor (1993) also propose that the higher the level of competition, the lower the information rents obtained from relationship lending that will increase the bank risk taking As a result, more bank competition diminishes the market power, simultaneously decrease profits margins, and lead to erode franchise values Such an environment with more pressure on profits, banks are more motivated to take on more excessive risks to enhance returns, contributing to the increased fragility of the banking industry (Boot & Thakor, 1993; Allen & Gale, 2004)

Other ideas that the contagion impact is more prominent than competitive markets

as all banks are price takers and banks may not be encouraged to provide liquidity to troubled banks (Sáez & Shi, 2004) In addition, too-big-to-fail and safety net subsidies are also considered for the banking market It means that exists the large banks with high concentration/ market share and contributes in explain a negative relationship between bank competition and stability A number of recent empirical researches consistent with

this view find that more competition (measured by the Lerner index) is linked with a higher risk loan portfolio (measured by nonperforming loans) in Spain (e.g Jimenez et al., 2007)

2.1.2.2 The modern competition-stability view

Modern view or "competition-stability" viewpoint assumes that more competition leads to greater stability in financial institutions In other words, view of bank market power raises the risk taking Boyd and Nicolo (2005) states that high market power banks enjoy lower competition in the loan market It means that encourage them to set high interest rates for borrowers, thereby increasing their (borrowers) risk exposure and insolvency risk or risk taking tendency They continue to find evidence that the bank will face the problem of moral hazard and adverse selection and lose solvency because the risk

is ultimately shifted from the borrower to the banks, as the bank risk and the borrower's risk are perfectly correlated

However, if the opposite inference, the banking market with an increase in competition, by lowering the interest rates charged banks, or mentioning better credit conditions for borrowers This would make them easier to repay bank loan, so banks face lower credit risk for their loan portfolio and improve the stability of individual institutions (Boyd et al., 2006)

In addition, Acharya, Gromb and Yorulmazer (2012) demonstrate that large banks

in concentrated markets receive subsidies from policy makers through "too big to fail" or

"too important to fail" programs It means change their risk taking motivation, encourage

to more risk, thereby more vulnerable The recent credit crunch is empirical evidence that large banks are difficult to monitor because their complexity and high political affiliation (Johnson & Kwak, 2011) Moreover, large banks in a focused market affect others through the contagion impact Hence, the failure of large banks in a concentrated market

to the whole system should be fragile

2.1.2.3 The ambiguous

However, some authors argue that competition-stability is linked in a non-linear or U-shaped manner by allowing imperfect correlation in loan defaults (Martinez-Miera & Repullo, 2010; Hakenes & Schnabel, 2011) They state that this non-linear and U-shaped link is as high market power in less competitive loan markets, which makes banks to set high interest rates for borrowers In result, it raises the insolvency risk due to the risk

shifting effect; on the other hand, it also enhances the bank‘s profitability due to margin effect

Besides, Berger, Klapper and Turk-Ariss (2009) also claim that competition and concentration can coexist and can concurrently induce stability or fragility Hence, the impact of increased competition could go either ways, depending on other factors and the existing level of competition In sum, competitive – fragility view and competition – stability view are supposed that both are not opposite perditions, and rather both can apply simultaneously if the risk-taking can be hedged by a high capital buffer

Nevertheless, there are also some studies find that actually no specific relationship between banking market concentration and bank risk (Yeyati & Micco, 2007)

On average, transnational studies provide mixed results related to the stability of competition Nevertheless, there is evidence that market concentration and market competition can co-exist and this affects bank stability through several various channels

2.2 Empirical literature

There exist a large number of empirical literatures investigating the correlation among banking competition, bank stability, bank market power or concentration across multiple economies by applying various measures and techniques For example, Schaeck, Cihak and Wolfe (2009) sampling of 31 countries in the financial crisis during 1980-2005 shows that the opportunities and time to crisis are declined under competition after directing for banking concentration which demonstrates a positive link with bank stability Another cross country analysis by Yeyati and Micco (2007) which takes a sample of commercial banks from 8 Latin American countries during 1993-2002 and finds that competition (proxied by the Panzar and Rosse 1987, H-statistic) is positively related to bank risk (represented by the Z-score) but no relationship between banking market concentration and bank risks, hence favor to competition-fragility hypothesis

Berger et al (2009) which samples of more than 8000 banks from 23 industries countries over the period 1999-2005 and state that banks with greater market power have less overall risk This research applies the Lerner and HHI index as a measure of market power and bank market concentration and Z-score as a proxy of overall bank risk Although their findings support to the competition-fragility paradigm, they also suggest that banks with market power results in riskier loan portfolios (as captured by non-performing loan ratios)

Another result of Berger et al (2009) is that banks hold more equity to protect their charter value from risks arising out of loans Anginer et al (2014) uses Lerner index and contingent claim pricing model of Merton (1974) to investigate the relationship between banking market competition and banks‘ insolvency risk Their study takes a sample of

1850 banks from 63 countries around the world The results suggest banking competition

to be positively related to bank stability (the findings remain unchanged when still using market concentration as a representative for market competition)

Liu et al (2012) studies the similar connections for banks operating in 4 South East Asian countries (Indonesia, Malaysia, The Philippines and Vietnam) in period of 1998-

2008 Their analysis applies a variety of bank level risk indicators such as loan loss provisions, loan loss reserves, volatility of earnings, and natural log of Z-index and Pangar-Ross H index as a competition proxy The findings show evidence that competition does not lead to financial fragility Besides, the concentration is negatively connected with bank risk, while regulatory restrictions positively impact bank fragility

A recent article by Fu et al (2014) find the connection between bank competition (measured by Lerner and HHI) and bank stability (measured contingent claim pricing model of Merton (1974)) This analysis is based on bank-specific data from 14 Asia Pacific countries in the period of 2003-2010 and uses both accounting based and market based measures of bank stability Their study claims that Lerner index is inversely associated to banks risks while exists a positive link between concentration and bank fragility

On the other hand, study of Martinez-Miera and Repullo (2010) has explored the possibility of some nonlinear relationship Freixas and Ma (2014) have been investigated on the effect of competition on various aspects of bank stability Boyd et al (2006) and De Nicolo and Loukoianova (2006) both explore that the Z-score index, an inverse measure of bank risk, lower with banking market concentration (measured by HHI index), suggesting that the risk of bank insolvency increase in more concentrated markets

Moreover, business cycle theory assumes that during banks apply conservative approaches in credit management, shrink low extension, and focus on building a capital buffer (Jokipii & Milne, 2008) Such activity can help banks to decrease loan exposure and moral hazard behavior and improve stability Nevertheless, Cook (2008) observes that a few banks suffered from a moral hazard issue during Asian Financial Crisis

Therefore, the impact of competition on bank stability becomes doubtful in period of financial crises, when crises change the risk taking initiative of banks In this situation, banks may apply a risk taking policy benefit from safety- net subsidies or the risky averse policies to decline moral hazard

On the average, cross country research papers offer mixed findings regarding competition-stability relationship by using different samples, risk measures and competition proxies Nevertheless, there is evidence that both market concentration and market competition can coexist and that these impact bank stability through various channels Summary table of some previous theories as well as popular measures for main

variables are shown in Appendix 1

2.3 Hypothesis construction

Based on theoretical and experimental evidence, two hypotheses have been constructed as following:

2.3.1 Competition and bank stability

As the review above, it was found that there is no consensus in the literature on relationship between competitive structures and bank stability Competition is important for efficiency, but market power can also bring some benefits The risk of bank insolvency will be archived when taking into account the simultaneous effects of charging lower rates from competition of banks

Author group supporting the negative effect of competition on bank stability such as Allen and Gale (2001), Berger et al (2009), and so on argue that competition reduces the charter value and raise bank‘s incentive for moral hazard (e.g shift risks to depositors), and thereby cause more risk-taking activities However, if considering this argument in one direction, competition also simultaneously decreases the interest rate and banks with more competitors will provide better credit products for borrowers This help banks limit the credit risk for their portfolio and thereby contribute to promote bank stability (Boyd et al., 2006; Noman, Gee and Isa, 2017)

As a whole, the margin effect (related to bank‘s profitability) and the risk shifting effect (related to bank risks) could simultaneously influence on this nexus and it is not straightforward to predict its magnitude in the ASEAN banking environment in part due

to adverse selection and moral hazard problems of banks here

H1: There is a non-linear relationship between competition and bank stability

2.3.2 Effect of financial crises on the competition and bank stability relationship

There has many recent studies investigating the effect of crises such as Asian 1997 and global financial crisis on bank stability Williams and Nguyen (2005) find that the time period following the 1997 Asian crisis was featured by significant changes in the Asian banking sector W Soedarmono (2013) explores that during the crisis time, if a bank exhibits some higher degree of market power, this will contribute to decrease risk of moral hazard behavior On the other hand, Noman, Gee and Isa (2017) studying ASEAN-

5 banks, find that the impact of the AFC was much more severe and intensive than the GFC In particular, they support the substantial effect of the AFC in 1997-98 on competition-bank stability nexus due to eroding bank capitalization In the AFC, ASEAN-5 banks have absorbed too much risk and loss capitalization to a greater extent, which made them more vulnerable or fragile from losing the charter value and face higher moral hazard in the form of gambling for resurrection and looting In contrast, during the GFC, ASEAN-5 banks were less resilient to risk-taking initiatives due to high capitalization and effectively manage their loan portfolio based on the experiences accumulated from the previous AFC Besides, Soedarmono (2013) also analyzes impact

of crises in emerging markets in Asia find that the 1997–1999 period shows that lower competition has a stabilizing effect which is opposite to what they find in the overall period of study and in the 2007–2009 period

Hence, from the foregoing identifies, the second hypothesis is proposed as follows: H2: The 1997 Asian crisis and 2008 global financial crisis have impact on competition- bank stability nexus

Table-1: Summary some previous studies

No Author Title Methods Data Main findings

Applying a squares (3SLS) simultaneous equation model,

three-stage-least-following Keeley (1990),

978 banks in 55 emerging and developing countries over an eight year period 2000–2007

Competition increases

stability as diversification across and within both

interest and non-interest income generating activities of banks increases

Fixed effect, IV (2SLS) regression setup

A sample of 17,055 banks from

79 countries consisting of a mix

of developed and developing countries over 1994– 2009

An increase in competition will have a larger impact on banks‘ fragility in countries with stricter activity restrictions, lower systemic fragility, better developed stock exchanges, more generous deposit insurance

(2008)

8,235 banks in 23 developed nations over the period 1999–

2005

Supporting competition-fragility‖ view—

banks with a higher degree of market power also have less overall risk exposure

(2014)

Bank competition and financial stability in Asia Pacific

GMM estimator, using the method developed by Bharath and Shumway (2008)

Focusing on commercial banks

in 14 Asia Pacific economies over 2003-2010, with 4069 observations The subsample for listed banks includes 1500 observations

Greater concentration fosters financial fragility and that lower pricing power also induces bank risk exposure after controlling

for a variety of macroeconomic, bank-specific, regulatory and institutional factors

Consider an economy with

two dates (t = 0, 1) and three

classes of risk-neutral agents: entrepreneurs, banks, and depositors

a U-shaped relationship between competition and the risk of bank failure generally obtains

Two step system GMM of Arellano

and Bover (1995) and Blundell and Bond (1998)

2527 observations at 180 commercial banks in ASEAN-5 nations

Higher competition measured fosters financial stability of the banking sector controlling a number of bank specific, regulatory and macro-economic factors Deposit insurance promotes financial stability, whereas, activity restrictions are associated with greater fragility

2SLS estimations, departing from Uhde and Heimeshoff (2009)

636 commercial banks established in

11 countries in Asia

A higher degree of market power in the banking market is associated with higher capital ratios, higher income volatility and higher insolvency risk of banks The 1997 Asian crisis that has directly affected Asian banks, market power in banking has a stabilizing impact

Fu et al (2014) and Berger et al (2009) Therefore, the study applies the following general regression form/ baseline specification

Bank stability= F (Competition, Competition2, Bank Controls, Regulatory

In the equation (i), Bank stability each bank over the period is captured by Z-score, Equity ratio and NPL ratio Competition denotes level of competition or market power which is measured with both non-structural measure of competition, Panzar-Rosse H-statistic and Lerner index, and structural measure of competition, HHI index

Bank control indicates bank characteristics consisting of bank size (SIZE), assets composition (LOAN), operational efficiency (COREV), Loan loss provisions (LLP) Bank Regulatory Control and Macro Control are country-specific variables for the quality of the regulatory and supervisory framework and macro-economic control In particular, bank regulatory control variables consist of deposit insurance, activity restriction, capital requirements and entry into banking requirement

The macro-economic control variables consists annual inflation rate and real GDP growth Moreover, both 1997-98 Asian financial crisis and 2008-2009 global financial crisis also examined in the interaction of competition on bank stability of ASEAN by incorporating crisis dummies Here, two crises dummies are included, where one for capturing 1997-98 Asian financial crisis which takes 1 if the year is 1997 and 1998, otherwise zero; and another one for capturing 2008-2009 global financial crisis which takes 1 if the year is 2008 and 2009, otherwise zero

In the above form (i), the coefficient value of Competition and Competition2 are examined, such as, if Bank competition is proxied by H-statistic, a positive and significant coefficient value of both Competition and Competition2 for Z-score and

Equity ratio as dependent variable provide evidence to support competition- stability hypothesis, where higher competition makes banks to take less insolvency risk and to be more bank stability On the contrary, negative and significant coefficient value of both Competition and Competition2 provide evidence to favor view of competition- fragility

In addition, a different sign of both coefficients imply that non-linear U-shaped link between competition and bank stability

3.1.2 Specific regression equations

Firstly, this study estimates the following panel regression that is consistent with the previous literature (Boyd et al., 2006; Soedarmono et al., 2011):

Stabilityijt= α0 + α1Competitionijt + α1Competition2ijt + βBankControlijt + ƟMarcoControljt + γCrisisDummyt + εijt (1)

In the equation (1), i=1…N, j=1…J, t=1…T, N refers the number of individual banks, J refers the number of countries, T defines the period of time (years), and α, β, Ɵ, and γ are estimated parameters

The three dependent variables of interest are considered to assess bank stability These include the ratio of bank capital, the ratio of nonperforming loans and an insolvency risk measure The insolvency risk is proxied by natural logarithm of the Z-score (Houston et al., 2010; Laeven and Levine, 2009 and many others) The ratio of bank capital is calculated by the ratio of total equity to total assets (ETA) The NPLs ratio is measured by ratio between non-performing loans and total loans

Secondly, to investigate the impact of both Asian crisis and Global financial crisis

in bank competition- stability nexus, specify the following equation:

Stabilityijt= α0 + α1Competitionijt + α2Competitionjt · AsiaCrisist +

α3Competitionjt · GlobalCrisist + βBankControlijt + ƟMarcoControljt + γCrisisDummyt

Where, AsiaCrisis and GlobalCrisis are crisis dummies which capture 1997-98 Asia financial and 2007-08 Global financial crises, respectively The interaction terms are considered as indicator of Bank Competition/ Market power, adjusted for effect of crises

in a given market

This equation add the interaction terms of the main explanatory variable for bank competition/market power, with crisis dummies to examine competition-stability nexus

adjusted for effect of crises in a given market These variables are used as the relevant subcomponent, their coefficients‘ meaning depend on the specification for the dependent variable

Thirdly, to analyze the effect of bank regulatory with the nexus bank market risk taking behavior, specifying the following equation:

power-Stabilityijt= α0 + α1Competitionijt + α2Competitionjt · RegulatoryControlijt + βBankControlijt + ƟMarcoControljt + γCrisisDummyt + εijt (3) Where, RegulatoryControl is a country-specific variable for market entry conditions, and quality of the regulatory and supervisory framework In particular, the study considers some indicators from the World Bank Survey on Bank Regulation and Supervision, concerning various aspects of impact of depositor protection schemes and the barriers to entry (entry into banking requirement, capital requirements, and activity restrictions) to financial intermediaries In this respect, the interaction terms are considered as an indicator of bank competition/market power, adjusted for the entry conditions, bank regulatory in a given market

In sum, the key explanatory variables include indicators of price competition at the bank –level (Lerner Index) and H-statistic, HHI (at the country-level) The Lener index measures the market power of a bank; whereas H-statistic measures for intensity of competition and HHI index measures the concentration of baking market Besides, these above equations also include, as control variables, bank-level balance sheet variables at the bank-level, indicators of macroeconomic and market entry conditions at country-level, and specifically, instrumental variables are taken to test and treat endogenous problem

3.2 Description of the variables

Table 2 below provides a summary of the variables consisting of dependent, explanatory and instrumental variables used in this analysis incorporating their definition, sources and expected sign

Table-2: Summary of the variables used in the analysis

Variable Definition and data source Source Expect- ed sign

Dependent variables

measure of the distance of a bank from insolvency Higher value indicates better solvency of a bank, and higher stability and less overall bank risk

BankScope, own

calculations

means higher credit risk and more risky loan portfolio Unit: %

BankScope

assets A higher value indicates lower loan portfolio risk and better capitalization Unite: %

BankScope

Variables measuring competition and concentration

Lerner index A bank-level indicator of bank competition which is a markup of price over

marginal costs Higher value indicates less competition and higher market power in banking sector

BankScope, own

calculations

+

revenue with respect to input prices Higher value of H-statistic indicates higher competition and lower market power

BankScope, own

calculations

Loans index which is the sum of square of market share of loan of commercial banks Higher value indicates greater market concentration

BankScope, own

calculations

+

largest banks in each country (Schaeck & Cihák, 2007) Higher value indicates greater market concentration (This index used for robust)

BankScope,ow

n calculations

Bank Level Variables

Bank Regulation and Supervision (WB)

banking

requirements

Whether various types of legal submissions are required to obtain a banking license, within a range between 0 and 8 Higher value means more entry requirements

Bank Regulation and Supervision

market capitalization Unit: %

Heritage Foundation

Property

Right

An index that takes a value from 0-100 indicating the level to which private property right is protected by the laws A higher score indicates more economic freedom and strong protection of property right of the individuals The Heritage Foundation property rights protection index A higher value signifies weaker protection

Heritage Foundation

3.2.1 Dependent variables

In order to study the nexus between bank stability and competition, it is necessary to select suitable measures of both bank stability and competition Bank stability is

frequently measured in the literatures in negative term such as systematic and/or

individual bank distress, because it threats financial safety net of a banking system (Bech,

Demirguc-Kunt, & Maksimovic, 2004) Moreover, several distress of systemic banking

distress begins with an individual bank distress (Beck, 2008) Similarly, failure of a large

bank due to liquidity shortage spillovers to the entire banking sector as observed in the

recent 2007-08 global financial crisis Therefore, this study uses the following different

risk exposure indicators as dependent variables to proxy for bank stability: the Z-score

index as an inverse measure of overall bank risk/insolvency bank risk; the volume of

nonperforming loans to total loans (NPLs) to account for loan portfolio risk5, and the

equity to total assets ratio (E/TA) for the bank‘s capitalization level

3.2.1.1 Bank Z-score – measure of bank solvency/default risk/bank bank stability

The background theory of the Z-score is based on the work of Roy (1952), which measures the distance of a bank from insolvency Subsequently, most researchers used the

Z-score index as the main proxy to the bank soundness (as in e.g., Berger, Klapper and

Turk-Ariss (2009); T Beck et al (2009); Laeven and Levine (2009), Demirguc-Kunt and

Huizinga (2010); Houston et al (2010); Soedarmono et al (2013); Beck, De Jonghe &

Schepens (2013) and Schaeck and Cihák (2014) and many other) Therefore, this study also uses Z-score index as the primary measure of bank stability

The risk of insolvency determines the risk that a bank unable to meet the obligations because it has a negative net value This can happen when a bank suffers some damage from its assets due to debts on securities, loans, or other banking activities, but then the institution's capital is enough to make up for those losses In that case, the bank could not fulfill its obligations defaults and loses its charter value

Hence, insolvency is a condition in which loss exceeds equity, such as: -π > E, where π stands for profit and E stands for equity (Roy, 1952) Then dividing this function

by A (total assets), get: E/A < -ROA, while E/A is the equity asset ratio and ROA is the return on assets Thus, the probability of insolvency can be represented as probability (E/A < -ROA) The inverse of the probability of insolvency is (ROA + E/A)/δ(ROA), where δ(ROA) is the standard deviation of ROA

Consequence, the Z-score is literally determined as the inverse of the probability of insolvency, and denoting an individual bank‘s soundness It describes these two factors because it is calculated as the sum of the equity-asset ratio (bank capital) and the return

on assets (bank profitability), divided by the standard deviation of the return on assets (profit volatility) In sum, the Z-score is defined as follows:

ijt

ijt ijt

ijt

ROA

ROAA A

E Z





 /

Where Z is a measure of stability of bank i, in country j, at time t In order

to decrease the influence of the changes in assets during the year, this study uses the Return on Average Assets (ROAA) which is available from Bankscope Additionally, to allow for the variability of indicator, this study computes the standard deviation

of the Return on Average Assets for all available sample period of bank The Z-score is inversely related to the probability of a bank‘s default A bank becomes insolvent when its asset value falls below its debt and the Z-score demonstrates the number of standard deviations that a bank‘s return must fall below its expected value to reduce equity and make the bank insolvent (Fu et al., 2014) A greater Z-score value indicates the low probability of a bank‘s financial distress, better solvency of a bank indicating less overall bank risk, and a higher stability or financial soundness The value of Z reduces with an

enhancing level of profitability and capitalization, and drop with an increase in the earnings volatility

3.2.1.2 Capitalization ratio

The study also uses two alternative proxies of financial distress and risk taking measurement, such as NPL ratio and equity ratio These alternative risk measures are used to investigate whether the change in financial soundness occurs due to a change in credit risk, or an increase in capitalization

Berger et al (2009) proposed capturing the equity ratio as an indicator of risk taking, arguing that high capitalization may offset the negative consequence of high insolvency risk on financial institutions‘ overall risk In addition, the view of competition-fragility argues that high market power induces banks to earn monopoly rent, which may be used

as a capital buffer to enhance capitalization Afterward, a number of studies also apply the equity ratio as a risk-taking indicator, such as the study of Laeven and Levine (2009), Soedarmono et al (2013) or Fang, Hasan, and Marton (2014), etc

The bank-level capitalization (ETA) is measured the r\atio of total equity to total assets (W Soedarmono et al., 2013) The higher ratio implies lower bank risk and may

enhance bank stability by offsetting banks‘ risk taking initiative (Berger et al, 2009) 3.2.1.3 Non-performing loan ratio

Similar to the capital ratio, the Non-performing loan (NPLs) ratio also is considered

as an alternative measure of risk taking or financial distress, following the work of Jiménez, Lopez, and Saurina Salas (2010); Agoraki et al (2011) and Amidu and Wolfe (2013) NPLs ratio represents for a bank‘s default risk or credit risk position in banking system This is because credit risk is the main banking risk, and the increase in bad debt

in the bank's loan portfolio In other words, credit risk is the risk that the borrower unable

to repay the bank In this sense, the debtor may be the receiver of a bank loan, the issuer

of a debt security, or even another bank borrowing in the interbank market

Given the primary focus of banking activity on credit provision, this study analyzes the impact of price competition on the quality of bank lending by understand the credit risk of the loans extended to customers Generally, an addition in price competition means a fall in the lending rates charged by banks to borrowers Nevertheless, this may impact the credit risk of the loan portfolio in two different ways In one case, corresponding to the argument in Boyd and De Nicolò (2005), the decrease in lending

rates may encourage the credit conditions for borrowers by making it easier for them to repay bank loans and then by declining the probability of default on bank credit

NPLs ratio is calculated by the ratio of non-performing loans to gross loans Where the non-performing loans are computed by the sum of substandard loans, doubtful loans and loss loans (provided available from Bankscope) A higher ratio indicates a bank‘s higher tendency to keep a riskier loan portfolio, which undermines the bank‘s financial soundness

3.2.2 Measures of bank competition

The level of competition cannot be directly observed in the banks in particular or firms in general because of involvement of some unobservable factors such as specific cost of firm‘s products and firm‘s reaction in response to actions of its competitors (Tabak, Fazio, & Cajueiro, 2012) Therefore, several methods have been developed to measure the level of competition such as Hirschman Herfindahl Index (HHI), the degree

of market concentration (CRn), Panzar Rosse-H statistic and Lerner index

Based on the structural approach, HHI and CRn are classified into the traditional industry organization, and H-statistic and Lerner index are categorized to new empirical industrial organizational (NEIO) or non-structural approaches Structural approaches (such as HHI and CRn) measure competition considering indirect proxies such as market structure and market share based on the market concentration ratio Conversely, the non-structural approaches (Lerner index and H – statistic) estimate competition based on the direct proxies applying the bank-specific data and particular assumption about the competitive behavior of the bank

Although these measures importantly estimate competition, there exits some confusion in the documentation concerning what a particular measure can capture competition For example, among the different approaches HHI and CRn measure concentration, Lerner index measures market power and H-statistic measures competition (Leon, 2014) Nevertheless, Xu, Rixtel, and Leuvensteijn (2013) argue that market power negatively relates to competition, but positively relates to concentration On the contrary, Lapteacru (2014) shows evidence supporting that high market power is not a result of low competition and high concentration Consequence, there is a lack of concurrence among the regulators and economists with respect to the best measure of competition Because of that, this study applies both structural approach and non-structural approach to capture competition of ASEAN banks

3.2.2.1 Lerner index

This study primarily use Lerner index researched by Lerner (1934) which is seen as

a popular measure of competition, and extensively used in recent banking literature though it is an indicator of the level of market power, e.g Jiménez, Espana, and Lopez (2007), Berger et al (2009), Uhde and Heimeshoff (2009), Nguyen, Skully, and Perera (2012), Amidu and Wolfe (2013), Liu and Wilson (2013), T beck et al (2013), Soedarmono et al (2013) and Fu et al (2014), etc

The Lerner index represents the markup of price over the marginal cost The deviation of price from marginal cost is considered a market power Following T beck et

al (2013), the Lerner index is a proxy for current and future profits stemming from pricing power As such, it fits well with the theoretical concept of banks‘ franchise value Lerner Index estimation described as follow:

Where, Lerner index is denoted as the difference between output prices and marginal costs (relative to prices) This estimation help studies the pricing behavior of banks producing a single output and proxy the price of by the ratio of total revenues to total earning assets for bank i in country j at time t, in which total revenue is the sum of interest income, non-interest operating income and equity-accounted profit/loss operating income

Besides, MC is the marginal cost of the total assets and obtained from the estimation of the trans-log production function at the bank level for each ASEAN country, using the methodology of (Berger et al., 2009), including bank and fixed effects:

+ + + ∑ + ∑ ∑ +

Where, C defines the total operating costs calculated by total interest expenses and total non-interest expenses (personnel expenses, other operating expenses), expressed in millions of USD Q stands for the amount of total assets in millions of USD, representing output quality The three input prices are then used to capture the price

of fixed assets (w1), the price of labor (w2), and the price of funding (w3) W1 is computed as the ratio of other operating expenses to total assets, w2 is the ratio of personnel expenses over total assets and w3 is the ratio of total interest expenses to

the total amount of deposits, money market, and short-term funding The cost function is estimated separately for each country in the sample over the sample period to account for potential technological differences among the countries, following Berger et al (2009) Additionally, data collected have different units for each ASEAN country So, the

absolute value variables in the above equation such as LnC and LnQ, the currency units

will be converted from the local currency into the US dollar through the exchange rate quoted on December 31, 2010 Other variables are expressed in proportions, so no conversion is performed

The coefficient of Equation (*) is used to compute the marginal cost for bank i in country j at time t, using the following equation:

In sum, greater values of the Lerner index indicate less bank competition A greater Lerner index value implies banks‘ higher market power, to set the product price over marginal cost and low competition

3.2.2.2 The Panzar-Rosse H-statistic: Non-Structural approach

In order to capture bank competition, this also use H–statistic of Panzar and Rosse (1987) which is based on the reduced-form revenue equation, using easily available bank level variables to compute the bank‘s market power for product price setting This study applies one of versions published recently by Panzar and Rosse (2012) That is the P-R price equation controlling for firm size, means estimating a revenue or price equation with proxy for bank size as a control variable This specification has been popular by Molyneux et al (1994), Bikker and Groeneveld (2000), Bikker and Haaf (2002), Claessens and Laeven (2004), Yildirim and Philippatos (2007), and Schaeck et al (2009),

Moch (2013) The reduced-form revenue regression model in determining H-statistic for each calendar year separately for each ASEAN country as follow:

+ ∑

+ ∑

+ +

Where ijt defines bank i in country j at time t; subscript Ln stands for the natural

logarithm; is the error term; TR, TA denote the income or total revenue and total asset, respectively Besides, three input prices consisting of w1, w2 and w3 define similarly to the above trans-log production function W1 is the ratio of administrative and other operating expense to total assets as a measure of the price of fixed assets; w2 is the ratio

of personnel expenses to total assets as a measure of the price of labor; and w3 is the ratio

of total interest expenses to the total amount of deposits, money market, and short-term funding as a measure of the price of borrowed funds

In addition, LnX1, LnX2, and Bank size, were added as bank- specific control variables that are expected to influence the banks‘ revenue function Where X1, X2 are the ratio of customer loan to total assets, ratio of equity to total assets, respectively Bank size is computed by the natural logarithm of total assets in millions of USD

In short, the H-statistic is a measure of the degree of competition in the banking market It measures the elasticity of banks revenues relative to input prices In other words, it is calculated as a sum of the elasticity of bank‘s total revenue, with respect to the above input prices Hence, H-statistic for each calendar year separately for each ASEAN country captured as follow: ̂ + ̂ + ̂

The H-statistic may take a value from -∞ to +1 A larger H-value indicates the change in input prices‘ greater impact on total revenue and more market competition

In special case as perfect competition, the H-statistic = 1, implies that the proportion

of increase in input prices and total revenue is the same In other words, an increase in input prices raises both marginal costs and total revenues by the same amount under perfect competition, and thereby the H-statistic equals 1 This is because the firm exist the market if it does not cover input prices

On the other hands, H-statistic under a monopoly takes either a zero or negative value, which means that a rise in input prices results in an increase in marginal costs, a

fall in output, and a decline in bank‘s total revenue Finally, when H-statistic takes a value between 0 and 1, the system operates under monopolistic competition

3.2.2.3 The Herfindahl-Hirschman Index (HHI): Structural approach

As noted above, this work also specify as alternative measure of competition which

is the HHI index Despite structural approaches have been criticized by Berger, Demirguc-Kunt, Levine, and Haubrich (2004) and Shaffer (2004) as inappropriate and misleading measures of competition, some recent researches such as Jiménez, Lopez, and Saurina (2013), Jeon and Lim (2013), Xu et al (2013), Anginer et al (2014) use concentration measures as proxy of competition, because, concentration and competition could coexist and indicate a banking system‘s stability and fragility (Berger et al., 2009) Here, this study uses HHI in loan market as a supplementary competition measure This index is captured by a sum of the squares of the market share of each given bank in the loan market Hence the HHI indicates are calculated on a country level and higher

concentration ratio of large banks (CR3) in loan market for robustness testing This CR3 may be representative of the banks‘ position in terms of market concentration and is computed as a portion total loan held by the top three banks in each country over the total loan of all banks in that country

The reason for the structural distinction between market power and market concentration; for example, Claessens and Laeven (2004) point out that even if market concentration may be a good indicator of market structure, highly concentrated markets may be quite competitive, because loans and deposits of the bank as in a competitive setting or due to the open market for new entrants

This work recognizes that banking markets do not always correspond to national boundaries Some banking products, such as wholesale credit and off-balance sheet loans

to large corporations are competing on an international basis Other products, such as retail deposits and small business loans, are more often competitive on a local basis, and the level of national banking market capacity may not coincide with the market power exercised at the local level Nevertheless, HHI calculations at the national level are relatively standard in the literature More importantly, the Lerner indices in this analysis

do not depend on this assumption on market definition