Base III and bank lending channel evidence in BRICS nad OECD countries

In contrast, in the 5 countries of OECD, there are no evidence to prove time dummy variable and the liquidity regulations effect on bank lending channel, nevertheless, the capital requir

UNIVERSITY OF ECONOMICS ERASMUS UNVERSITY ROTTERDAM

HO CHI MINH CITY INSTITUTE OF SOCIAL STUDIES

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

BASEL III AND BANK LENDING CHANNEL:

EVIDENCE IN BRICS AND OECD COUNTRIES

BY

NGUYEN THI HONG VAN

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

HO CHI MINH CITY, December 2017

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

HO CHI MINH CITY THE HAGUE

VIETNAM THE NETHERLANDS

VIETNAM - NETHERLANDS PROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

BASEL III AND BANK LENDING CHANNEL: EVIDENCE IN BRICS AND OECD COUNTRIES

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

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

DECLARATION

“I certify the content of this dissertation has not already been submitted for any degree and is not being currently submitted for any other degrees I certify that, to the best of my knowledge, any help received in preparing this dissertation and all source used, have been acknowledged in this dissertation.”

Signature

Nguyen Thi Hong Van Date: December 15th 2017

ACKNOWLEDGEMENT

Foremost, I would like to express my sincere gratitude to my supervisor Dr Le Ho An Chau, for her patience, motivation, enthusiasm, sympathy, immense knowledge, and for giving

me valuable advice Her guidance helped me at all the time of research and writing of this thesis

In addition my advisor, I would like to thank Prof Nguyen Trong Hoai and Dr Pham Khanh Nam who have their expertise view with me, the valuable experience in research, and Dr Truong Dang Thuy who has provided the practical econometric technique, a valuable knowledge

ABSTRACT

Basel committee on banking supervision (BCBS) promulgated Basel III regulations with more tightened operating conditions in the banking sector This improvement of BCBS has brought about a lot of arguments Some researchers suppose that Basel III regulations raise the marginal cost and reduce bank lending, but others believe Basel III regulations can improve the banking system and increase the financial shocks absorbability To investigate the impact of Basel III on bank lending channel, this study utilizes data collected from 391 commercial banks

in 10 countries (BRICS and 5 countries of OECD) from 2011-2016

The empirical results from three stages least square under generalized structural equation model (GSEM) show that there is the disparity impact of Basel III regulations on bank lending channel between BRICS and 5 countries of OECD Specifically, there is the significant negative impact of Basel III liquidity regulations and time dummy variable but no significant capital regulation effects on bank lending channel in BRICS countries The possible reason is that capital ratios (Tier 1 ratio, Common equity tier 1 ratio, and leverage ratio) of commercial banks

in BRICS were higher than Basel III regulations in the period 2011-2016 They have not had too much pressure on adapting capital requirements of Basel III regulations In contrast, in the 5 countries of OECD, there are no evidence to prove time dummy variable and the liquidity regulations effect on bank lending channel, nevertheless, the capital requirements of Basel III have a significant negative impact on bank lending channel

Keywords: Basel III, Bank lending channel, Capital requirements, Liquidity regulations,

BRICS, OECD

JEL Classification: G32, G38, E52, E42

TABLE OF CONTENTS

DECLARATION i

ACKNOWLEDGEMENT ii

ABSTRACT iii

TABLE OF CONTENTS iv

LIST OF TABLE vi

LIST OF FIGURES vii

LIST OF ACRONYMS viii

CHAPTER 1 INTRODUCTION 1

1.1 Problem statement 1

1.2 Research objectives: 2

1.3 Research methodologies and data 3

1.4 Research contribution 3

1.5 The thesis structure 3

CHAPTER 2 LITERATURE REVIEW ON BANK LENDING CHANNEL AND BASEL III REGULATIONS 5

2.1 Theoretical review on bank lending channel and Basel III regulations 5

2.1.1 Bank lending channel 5

2.1.2 Basel III on bank regulations 9

2.1.3 Basel III effects on bank lending channel 10

2.2 Empirical review on the effect of Basel III regulations on bank lending channel 13

2.3 Hypothesis construction and the conceptual framework 18

2.3.1 The conceptual framework 18

2.3.2 The hypothesis construction 18

CHAPTER 3 RESEARCH METHODOLOGY 20

3.1 Data sources 20

3.2 Research methodology 20

3.2.1 Model specification 23

3.2.2 Measurement variables 26

CHAPTER 4 THE EMPIRICAL RESULTS 29

4.1 Data descriptions 29

4.2 The impact of Basel III regulations on bank lending channel 36

4.2.1 The impact of Basel III regulations on bank lending channel in 10 countries 36

4.2.1 The impact of Basel III regulations on bank lending channel in BRICS 38

4.2.1 The impact of Basel III regulations on bank lending channel in 5 countries of OECD 41

CHAPTER 5 CONCLUSIONS AND POLICY IMPLICATIONS 44

5.1 Concluding remarks 44

5.2 Policy implications 45

5.3 The limitation and further researches 45

REFERENCES 46

APPENDIX 54

LIST OF TABLE

Table 3.1 The descriptions’ definition and measurement unit of variables 24 Table 4 1 Descriptive statistics of variables 29 Table 4 2 The correlation table 35 Table 4 3 Bank lending channel with each effect of Basel III regulations in total 10 countries 37 Table 4 4 Bank lending channel with each effect of Basel III regulations in BRICS 38 Table 4 5 Bank lending channel with each effect of Basel III regulations in 5 countries of OECD 41

LIST OF FIGURES

Figure 2 1 The simplified measurement for NSFR 12

Figure 2 2 Scenario to approach the NSFR 13

Figure 2 1 The conceptual framework 18

Figure 3 1The recursive relationship 21

Figure 4 1 The loan growth distribution of 10 countries from 2011 to 2016 30

Figure 4 2 The loan growth distribution of 5 countries in OECD from 2011 to 2016 31

Figure 4 3 The loan growth distribution of BRICS from 2011 to 2016 31

Figure 4 4 The interest income ratio distribution of 10 countries from 2011 to 2016 32

Figure 4 5 The interest income ratio distribution of 5 countries in OECD from 2011 to 2016 33

Figure 4 6 The interest income ratio distribution of BRICS from 2011 to 2016 33

LIST OF ACRONYMS

BCBS Basel committee on banking supervision

BIS The Bank for International Settlement

BRICS Brazil, Russia, India, China, South Africa

DSGE Dynamic Stochastic General Equilibrium

GSEM Generalized Structural Equation Model

OECD The Organisation for Economic Co-operation and Development

SVAR Structural Vector Autoregressive

VECM Vector Error Correction Model

CHAPTER 1 INTRODUCTION 1.1 Problem statement

The financial crisis of 2007-2008 began on Wall Street, New York, and then spread over the global economy After the financial crisis in 2008, many researchers and policymakers recognize that the global financial crisis was caused by the weakness of banking systems Therefore, they propose that banks should reform to strengthen the financial system and encourage more the prudent lending behavior in upturns The Bank for International Settlements (BIS)1 rapidly developed Basel2 III regulatory framework in 2009 which was upgraded from Basel II with more focus on banking sector monitoring, change in the definition of bank capital, capital requirements, risk coverage, leverage ratio and liquidity management The regulations of Basel III aim to raise the quality and quantity of banks’ regulatory capital base and to improve the risk coverage of the banking sector (Bezoen, 2015) In addition, the enaction of Basel III also conducts the new provisions of liquidity management However, there has been a lot of the controversy about the negative effect of raising capital requirements and liquidity regulations as

it may lead to a slow credit growth and choke off asset-price pressures before a crisis occurs On the other side, many authors argue that Basel III regulations have the positive impact on the economic performance and fluctuations in long term Hence, the BIS extend the point of time to complete the applied process of Basel III from 2015 to 2019 However, Angelini et al (2015) argue that in long-term via Basel III regulations commercial banks can absorb the financial shocks

Whether the new framework of the Basel or the upgrade of the Basel II to the Basel III induces the slow loan growth and harms the revival of the economy? To adopt the capital requirements and liquidity regulations of Basel III, the commercial bank can adjust their balance sheet by increasing the capital and decreasing the risk-weighted assets including lending volume The decreasing of lending volume leads to the increasing the lending rate Angelini, Neri, and

1 The Bank for International Settlement (BIS) is the international financial organization to serve central banks in their pursuit of monetary and financial stability, to foster international cooperation in those areas and to act as a bank for central banks

2 The Basel is the set of international banking regulations issued by BIS

Panetta (2011), Modigliani and Miller (1958), suppose that higher requirement capital leads to the higher marginal cost of loans, so the loan's interest will increase Kashyap, Stein, and Hanson (2010) find that the loan interest upper 6 basis point after applied Basel III for the United State’ banking system Hence, Basel III regulations directly increase the loans interest income ratio and decrease loans volume Besides that, the loans interest income ratio increase leads to the decreasing of bank volume So the Basel III regulations induce decreasing the loans volume on direct and indirect sign Cosimano and Hakura (2011) suppose that the Basel III regulations prevent the recovery of the economy after the financial crisis However, Angelini et al., (2015) argue that in long-term via Basel III regulations commercial banks can absorb the financial shocks As the instruments of the monetary policy, Basel III is examined in several dimensions, and angles The foundations of the international regulatory framework of the Basel III for banks are capital requirements and liquidity regulations On one hand, researchers investigate the impact of Basel III on the capital requirement (Kashyap et al., 2010, Cosimano and Hakura 2011, Howarth and Quaglia 2013) They prove the capital requirements of Basel III raise the bank lending rate and reduce the bank loan volume On the other hand, Basel III is studied in term of the impact of the liquidity regulations on the economy (Giordana and Schumacher 2011, Bonner, Van Lelyveld and Zymek 2015) They suppose that the liquidity regulations also have the same impact on the economy as that of capital requirements Empirical studies consider one sector of Basel III the capital requirements or the liquidity regulations This study considers Basel III by combining these two sectors to measure the impact of both capital requirement and liquidity regulations on the economic growth via bank lending channel

In theory, monetary policy can be transmitted into the economy through several channels Basel III indirectly affects bank lending channel by increasing the marginal cost of the bank loan (Giordana and Schumacher 2011) This thesis focuses on the impact of the Basel III on bank lending channel for commercial banks in 10 member countries of the Basel committee during the period from 2011 to 2016 Differentiate from previous studies, this paper employs both capital requirements and liquidity regulators in Basel III and utilizes loans rate and loan growth as two dependent variables to examine the volatility of the bank lending

1.2 Research objectives:

The objective of this study is to investigate the bank lending channel effect of Basel III

on the economies via of 10 member countries belonging to the Basel Committee during the

period 2011-2016 Specifically, this study aims to test the direct effect of two elements of Basel III capital requirements and liquidity regulations on loan growth and indirect effect on loan growth via lending rate Thenceforth, we can figure out the accurate impact of Basel III on each factor of bank lending channel

In order to obtain these objectives, this study aims to seek the answers to the following research questions:

Do new regulations on capital and liquidity in Basel III have any effect on the economy via bank lending channel? If so, are the impacts negative or positive?

Which factors of Basel III, i.e the capital requirement or liquidity regulations have higher marginal effects on the economies?

The answers from this study will provide important policy implication regarding the argument on whether the expanding schedule of Basel III is the right decision?

1.3 Research methodologies and data

To investigate the direct and indirect impact of Basel III regulations on bank lending channel, and solve the causality relationship between of loan growth and bank lending rate, this study utilizes Three-Stage Least Square (3SLS) estimation based on Generalized Structural Equation Modeling (GSEM) Besides that, this study employs time dummy variable to study if the bank lending channel is different before and after applying Basel III regulations

Data of this research is collected for 391 commercial banks in 10 countries (BRICS and 5 countries of OECD) from 2011-2016 from Orbis Bank Focus Moreover, data for macroeconomics variables are collected from World Bank and OECD database

1.4 Research contribution

This study provides empirical evidence of the Basel III effects on member economies via bank lending channel The main contribution of this thesis is to analyze both direct and indirect effects on the loan growth and lending rate, and address the endogeneity problem of these variables Additionally, this thesis examines the Basel III regulations more comprehensively in both capital requirements and liquidity regulations dimensions and compare the difference between two groups of countries: BRICS and OECD The findings from this study will suggest important policy implication for central banks

1.5 The thesis structure

This thesis includes five chapters, which can briefly be presented as follow:

Chapter 1 presents the problem statement and the overview of this study

Chapter 2 presents the related theories and the empirical evidence with the focus on the bank lending channel and Basel III regulations

Chapter 3 discusses the research methodology, in which the formulated model and explanations of data measurements are examined In addition, the econometric technique for research objectives will be clarified

Chapter 4 presents the empirical results

Chapter 5 presents the general findings and the discussions The policy implications, the limitations and the potential improvements for future studies are also discussed

CHAPTER 2 LITERATURE REVIEW ON BANK LENDING

CHANNEL AND BASEL III REGULATIONS 2.1 Theoretical review on bank lending channel and Basel III regulations

2.1.1 Bank lending channel

2.1.1.1 The introduction of bank lending channel

Bank lending channel (BLC) is one of the supplement transmission channels of the credit channel to strengthen the impact of the monetary policy on economic variables through bank credit supply In the money market, bank lending is the main capital market instrument to mobilize capital BLC approaches all of the elements in the economy for example households, entrepreneurs, financial institutes, governments, etc Moreover, commercial banks which provide BLC into the economy occupies mainly the market share in the financial market So, monetary policymaker can promulgate policies to obtain their target via BLC Bernanke and Blinder (1988) indicate that if policy maker contracts the monetary policy, the credit supply also decreases and vice versa

Monetary policy ↓→ funds of commercial banks ↓→ credit supply ↓→ investment, consumption ↓→ output ↓

The bank lending channel operates through the credit supply of the commercial bank, so BLC complies some existed conditions in the economy, commercial bank, and customer characteristic:

Firstly, enterprises depend on funds from the commercial bank and commercial banks can’t replace entire funds from deposits (Bernanke & Blinder 1988) Bank loans hold mainly position in the capital structure of the enterprises because firms can easily and fast approach money from the bank with low cost than other funds Besides that, enterprises can salvage the financial leverage by using bank loans to get gain from tax shield If firms won’t employ bank loans, BLC won’t exit Gomez-Gonzalez and Grosz (2006) indicate the small enterprises depend

to the capital from bank loans more than large firms, because of the lack access to securities markets and banks have a comparative advantage in obtaining customers information than other investors Commercial banks are the financial intermediations, so deposits always hold major capital of commercial banks If commercial banks employ other funds instead of deposits, the monetary policies can’t affect the credit supply (Bernanke & Gertler 1995)

Secondly, the policies of the central bank can adjust bank lending supply of commercial banks (Kashyap & Stein, 1994) It depends on many elements such as the level independence of the central bank, creditability, transparency

Besides that, monetary policy will be transmitted through bank lending channel, when the reduction of long-term real interest income ratio is ineffective (Bernanke & Gertler, 1995) Farinha and Robalo Marques (2001) suppose that BLC exists when banks cannot perfectly shield transaction balances (deposits) from changes in reserves; and there are no close substitutes for money in the conduct of transactions in the economy

2.1.1.2 Determinants of bank lending channel

BLC is affected by both macroeconomic and microeconomic factors (Altunbas, Gambacorta and Marques-Ibanez, 2010, Angeloni and Ehrmann, 2003) Policy rates affect bank lending interest income ratio (Cottarelli and Kourelis, 1994, Mojon and Peersman, 2001, Sander and Kleimeier-Ros, 2004, De Bondt et al., 2005, Berg et al., 2006) However, Berger and Udell, (1992) indicate that the loans rates and credit supply of each commercial banks have a different reaction to the volatility of policy rates because of different bank’s characteristics

Macroeconomic conditions

The level wealth of bank lending channel transmission is different among countries because of the differences in macroeconomic conditions Mishra and Montiel (2013) suppose that BLC exits in most of the developed and developing countries However, BLC in the developing countries is stronger than developed countries because of the weakness of the international and domestic financial markets (Mishra, Montiel and Spilimbergo, 2014) In developing countries, financial markets are inefficient so investors capture capital mainly via bank loans, so central banks intervene deeply in bank loans (Mengesha and Holmes, 2013) Consequently, BLC in developing countries is stronger than other channels Conversely, BLC is weaker in high – developed financial markets

BLC depend on the economic openness of each country If a country is less open, the international financial markets will be weaker Consequently, BLC is stronger than other countries and vice versa Escrivá and Haldane (1994) suppose that in the openness countries, BLC may be weaker because of the impact of the external shock into the domestic economy In addition, BLC depend on the economic structure In addition, Cecchetti (1999) suppose that BLC

is affected by the scale of the economy, the development of infrastructure Juselius and Toro

(2005) find the evidence for the impact of volatility of the economic structure on BLC in Spain Moreover, BLC depend on public debt Woodford (1990) suppose that the effective management

of public debt will help balance the inefficiency in the intermediate financial system Besides that, BLC depend on the degree of independence of central bank Dornbusch, Favero and Giavazzi (1998) indicate that BLC will be stronger when the central bank is more independence because of the efficient transparency and accountability of central bank

Banking regulations

The central banks manage and monitor capital mechanism via BLC Commercial banks directly react to the central bank regulations (Honda, 2004) Commercial banks can break safety hurdles and join new riskier activities, when the central bank doesn’t consider risky actions of commercial banks in monetary policy conducting (Zhu, 2007), this makes BLC stronger and vice versa In the late 1980s, the Basel banking supervision committee issued Basel I to create minimum capital adequacy standard, then Basel II (2006) improve the safety regulations on minimum capital adequacy After that, Basel III (2010) focus on clarifying the capital definition, raising the quality, quantitative of capital requirements and enact the new liquidity regulations (BCBS, 2010)

BLC is affected by regulations on the interest income ratio Central banks employ regulations on interest income ratio mechanism via promulgating the ceiling on the interest income ratio which prevents the excessive competition among commercial banks, limit the risky investment activities of commercial banks So the regulations on ceiling interest income ratio would make BLC weaker Mertens (2008), prove that BLC weaker when the United States apply the ceiling interest income ratios

The competition in banking sector

BLC will be weaker if the banking sector is less competitive (LeRoy, 2014) Fungáčová, Solanko and Weill (2014) indicate that BLC will be stronger if banking sector is the monopoly competitive market And they also suppose that BLC will be weaker, if banking sector is the oligopolistic market

Microeconomic determinants

Besides macroeconomic determinants, microeconomic determinants also effect on BLC (Bernanke and Blinder, 1988) BLC in different commercial banks is different because of the difference of commercial bank characteristics: (i) the commercial banks’ size, (ii) the

commercial banks’ capital, (iii) the commercial banks’ liquidity, (iv) the commercial banks’ risk, and (v) the commercial banks’ customer characteristics, etc

The commercial banks’ size

BLC is weaker in large commercial banks Larger commercial banks have a more competitive advantage and powerful in the financial market So they can tolerate any kind of shock and the chance of monetary policy more easily, and their development is stable In contrast, small commercial banks have less competitive advantage and powerful in the financial market, so they can be impacted by the volatility of the monetary policy (Kashyap and Stein,

2000, Altunbaş, Fazylov and Molyneux, 2002, Kakes and Sturm, 2002, Altunbas, Gambacorta and Marques-Ibanez, 2012)

The commercial banks’ capital

The capital of commercial bank has the negative relationship with the bank lending channel The banking system would be healthier with the higher capital ratio (Mingo, 1977, Mingo and Wolkowitz, 1977) However, raising capital ratio is equivalent to restricting the fundraising Consequently, the behavior of commercial bank behave is cutting loans supply (Kishan and Opiela, 2000, Kishan and Opiela, 2006, Altunbaş, Fazylov and Molyneux, 2002)

The commercial banks’ liquidity

The liquidity of commercial bank has a negative impact on the bank lending channel Higher liquidity commercial bank has higher ability to fight against the economic shocks, so to raise the commercial banks’ liquidity they reduce the liquidity risk and loans supply (Kashyap and Stein, 1994) So, the asset of commercial bank more liquidity the bank lending channel weaker (Peek and Rosengren, 1995)

The commercial banks’ risk

The risk of the commercial bank has a negative impact on the bank lending channel Hoshi and Kashyap (2000) indicate that banks’ risks restrict the loan supply because of their flexibility However, the commercial bank can pretermit the Moral Hazard issue to capture more loans with higher risk (Krugman, 2008)

The commercial banks’ customer characteristics

The bank lending channel can be affected by their customer characteristics: (i) financial barriers, (ii) borrowing capacity, (iii) and size of enterprises, etc (Dedola and Lippi, 2005, Peersman and Smets, 2002, Tena and Tremayne, 2009, Koop et al., 1996) Commercial banks

have more power to grant loans for small and medium enterprises (SMEs) because of their less capacity to approach other capital sources and higher risk than large enterprises (Gertler and Gilchrist, 1991, Gertler and Gilchrist, 1993, Mojon et al., 2002) When the borrowing capacity of enterprises is lower, there are not many chances for them to capture the capital So commercial bank has more advantage to grant bank loans If the customers have more financial barrier that means their borrowing capacity is restricted In that case, the bank lending channel will be stronger (Dale and Haldane, 1995, Stoneman and Canepa, 2002)

2.1.2 Basel III on bank regulations

2.1.2.1 Basel III introduction

Basel III is the comprehensive set of reform measures for the international banking system, upgraded from Basel II, when Basel Committee on Banking Supervision (BCBS) perceives the deficiency of the previous version Basel III was promulgated in 2009 after the financial crisis in 2008 Perceiving the weak capital quality and liquidity in global banking system mainly contributes to the financial crisis in 2008 BCBS upgraded Basel II to Basel III to improve the health of banking system, increase the capability absorbing the financial shocks Basel III include the general operating criterions of the international banking system to increase the quality and level of bank capital, strengthening the regulations in the banking sector and strengthening overall risk management Based on Basel III, commercial banks raise capital requirements, leading to more resilient financial buffer (Admati and Pfleiderer, 2010) Besides that, commercial banks must hold the more liquid asset, so the bank funding will be more stable Consequently, the risk of bank default is decreased However, Basel III also damp on the economic growth through the effect of bank lending channel (Angelini et al., 2011) Meanwhile, the increased capital requirement, the marginal cost of using the capital of banking system increase, so the bank lending reduction

2.1.2.2 The chance of Basel III

Because of the unclear capital definition, lack liquidity management, and pro-cyclical effect of Basel II, the Basel III regulations achieve two main objectives: (i) strengthen the liquidity of banking system, (ii) raise the capability absorbing financial shocks Basel III regulations focus on clarifying the capital definition, tightening the capital requirement and augmenting liquidity to prevent system risk Specifically, Basel III regulations neglect the Tier 3

capital, raise the Tier 1 ratio, Common Tier 1 equity ratio, enact the Leverage ratio and Liquidity Coverage ratio, and Net stable funding ratio

New capital definition and requirements

Basel III regulations assert commercial banks raise their common equity including common share and retained earnings Basel III regulations abolish tier 3 capital and increase the minimum of Tier 1 ratio from 4 percent to 6 percent Simultaneously, the minimum of Common equity tier 1 ratio also raises from 2 percent to 4.5 percent Besides that, Basel III regulations enact the new leverage requirement with the minimum of the leverage ratio is 3 percent The tightening capital requirements reduce the risk exposure, counterparty credit risk and raise the capital buffer However, to adapt the capital requirements commercial banks carry the higher financial cost and lower shareholder’s profit

New liquidity ratios

Basel III regulations enact new liquidity requirements mainly via liquidity coverage ratio (LCR), and net stable funding ratio (NSFR) LCR is the criteria for holding liquid assets of the commercial bank The liquidity coverage ratio equals stock of high-quality liquidity assets to the total net cash outflows over the next 30 calendar days It expresses the magnitude of high liquidity assets for each next 30 calendar days total net cash outflows unit Basel committee on banking supervision (BCBS) suggest bank should maintain LCR more than 100 percent, that means bank should maintain the high-quality liquidity assets more than the total net cash outflows over the next 30 calendar days NSFR is the criteria for holding long-term assets of the commercial bank Under the liquidity requirements of Basel III regulations, commercial banks can counteract the economics shocks and limit the liquidity risk Angelini et al (2015) suppose

that: “the net stable funding ratio addresses the maturity mismatches between bank’s assets and

liabilities” The net stable funding ratio equals the available amount of stable funding over the

required amount of stable funding Basel III regulations require commercial bank maintain NSFR at least 100 percent, that means commercial bank have to hold the available amount of stable funding more than the required amount of stable funding

2.1.3 Basel III effects on bank lending channel

There are many researchers argue that the Basel III regulations have a negative impact on recover economy Basel III regulations tighten the commercial banks’ operating conditions on capital and liquidity These regulations raise the marginal cost of funding So commercial banks

react by raising the bank lending rates and reducing the credit growth When the credit growth decreases, the investment and consumptions decrease Consequently, the output decreases

2.1.3.1 The effects of Basel III capital requirements on bank lending

Some recent studies have mentioned the cost channel when they study the bank lending channel (Gaiotti and Secchi, 2006, Adolfson et al., 2005), it transmits monetary policy through the effect of credit supply on lending interest income ratio and the effects on input capital cost (Tillmann, 2008) Based on the Risk-adjusted rate of return on capital (RAROC), the marginal cost of loans is measured by the equation (Cosimano and Hakura, 2011):

𝑀𝐶 =𝐷

𝐴(𝑟𝐷+ 𝐶𝐷) + 𝐶𝐿+𝐴 − 𝐷

𝐴 𝑟𝐾 (1)

𝐾′= 𝐴 − 𝐷 (2) Where: MC is the total marginal cost, M is the marginal revenue, D is the deposit, A is the total assets, rL is the loan rates, rD is the interest income ratio on deposits, rK is the return on equity, CL is the cost of loans, CD is the cost of deposit, K is the capital, K’ is the future capital,

L is the bank loans

The equation (1) expresses the component of marginal cost in commercial bank operation In particular, the equation (1) figure out the future capital fraction is a positive impact

on total marginal cost That means, in case the return on equity constant, the capital fraction increase, the total marginal cost also increase

Basel III regulations raise the capital standards for commercial banks Based on the equation (1), the total marginal cost of commercial bank increases belongs to the higher capital requirements

Cosimano and Hakura (2011) also conduct the equation for loan rates:

𝑟𝐿 = 𝑏0+ 𝑏1𝑟𝐷+ 𝑏2(𝐶𝐿+ 𝐶𝐷) + 𝑏3𝐾′

𝐴 + 𝑏4log(𝐴) + 𝑏5𝑀 + 𝜀1 (3) The demand for loans:

𝐿 = 𝑐0− 𝑐1𝑟𝐿+ 𝑐2𝑀 + ε2 (4)

As the equation (3), an increase in capital fraction would lead to increase the loan rates And the equation (4) shows that the increase in loan rates would lead to decrease the bank loans Barth III and Ramey (2002) argued that monetary policy not only effects on demand side but also on the supply side of the economy through the cost channel The change in monetary policy will alter the credit supply and lending rates of commercial banks This process alters the

marginal product cost and then affects output prices of enterprises (Chowdhury et al., 2006, Ravenna and Walsh, 2006) However, the transmission of the cost channel is different between industries Barth III and Ramey (2002) found that the cost channel is the most important channel

in some industries in the United Kingdom due to these sector characteristics That is consistent with the study of Dedola and Lippi (2005), they found that sector characteristics affect monetary policy transmission, and GDP is affected by impacts of monetary policy on both sides of supply and demand The existence of the cost channel greatly affects the optimal monetary policy (Ravenna and Walsh, 2006) Because any change in monetary policy that transmits through IRC and BLC may well be destroyed if a cost channel exists So central bank is argued to keep a stable policy in monetary policy conducting (Chowdhury et al., 2006)

2.1.3.2 The effects of Basel III liquidity regulations on bank lending

The liquidity requirements for commercial banks are the regulations which require commercial banks have the ability to repay their liabilities immediately Basel III regulations introduce two new factors (LCR and NSFR) to indicate the liquidity regulations for commercial banks LCR expresses the magnitude of high liquidity assets for each next 30 calendar days total net cash outflows unit NSFR is the criteria for holding long-term assets of the commercial bank

King (2010) conduct the simplified measurement for NSFR as the equation (5):

Figure 2 3 The simplified measurement for NSFR

To approach the NSFR, the commercial bank can increase Tier 1 capital, lengthen the maturity of debt, raise holdings of high-quality investments, increase investments as a share of total assets, increase cash as a share of total assets Moreover, the commercial bank can reduce maturity of loans to corporates and retail to less than one year, reduce liabilities, reduce all other assets, issue debt and purchase government bond (King, 2010) These options direct decrease the bank loans and indirect decrease the bank loans via raising the marginal cost of funding

Figure 2.2 expresses the solution for commercial banks to approach the NSFR requirement with the least strategies of cost-benefit analysis (King, 2010)

Source: King (2010)

Figure 2 4 Scenario to approach the NSFR

2.2 Empirical review on the effect of Basel III regulations on bank lending channel

During the financial crisis in 2008, commercial banks reduce new loans because of the stresses on bank liquidity (Ivashina and Scharfstein, 2010) and weak capital health (Demirguc -Kunt, Detragiache, and Merrouche, 2013) After that, BCBS promulgated the upgrade version for Basel namely Basel II with more tightening the quality and quantitative capital requirements and introduce new liquidity regulations More tightening in capital structure and liquid asset would increase the economic shocks absorbing capacity Miles, Yang, and Marcheggiano, (2013) suppose that under Basel III framework capital requirements assert commercial banks raise their equity, then commercial banks have more cash for lending However, the lending spread would increase, and negative impact on bank lending volume (King, 2010) There are many empirical researches study the impact of capital requirement on bank lending channel Cosimano and Hakura (2011) examine the effect of Basel III new capital requirements on bank lending channel (bank lending rates and loan growth) By using GMM estimations for 100 commercial banks and bank holding companies from 2001 to 2009, their results show that the new capital requirements will raise the cost to commercial banks, bank holding companies and borrowers lead to the increase of the loan interest income ratio and reducing the loan growth Francis and Osborne (2012) indicate the capital requirements for banks in the United Kingdom harm on loan growth

Sutorova and Teplý, (2013) also indicate that commercial banks in the European Union have a negative impact on bank volume after applying Basel III regulations Blundell-Wignall and Atkinson, (2010) indicate that Basel III regulations don’t handle the core issues of commercial banks at that time, it just moves the risk buckets around with the derivatives to capture the capital requirements

There are many researchers studying the impact of Basel III liquidity regulations on bank lending channel Giordana and Schumacher, (2011) study the impact of Basel III liquidity regulations on bank lending channel By employing the Generalized Method of Moment (GMM) for the commercial bank in Luxembourg, they indicate that both Net Stable Funding Ratio and Liquidity Coverage Ratio make reducing on bank lending channel Therein, the Net Stable Funding ratio reduces bank lending channel more than the Liquidity Coverage Ratio Gunji et al (2010) consider the impact of the monetary policy on the bank lending for the different scale of Chinese bank from 1985 to 2007 They mention that under the monetary policy, the banks’ size and banks’ liquidity significant impact on bank lending But the capital of bank does not impact

on the bank lending with the varying monetary policy Van den End & Kruidhof (2013) investigate the systemic implications of the Liquidity Coverage Ratio and the bank reactions under the Basel III liquidity regulations As the results, they aware that lower liquid assets in the buffer, more benefit for the macroprudential instrument Their results prove that the Liquidity regulations of the Basel II might be not effective, and have the adverse side effect during times

of stress Bonner, Van Lelyveld, & Zymek (2015) study the determinants of the banks’ liquidity holdings with a combination of bank-specific and country-specific variables for 7000 banks from

25 OECD countries Their results show that the bank-specific and the country-specific determined the liquidity of the banks without the liquidity regulation of the international regulations In addition, they figure out that higher liquidity buffers lower the lending volumes and higher interest income ratios Duijm & Wierts (2016) examine the relationship between the new liquidity regulations under the Basel III and the bank balance sheet for banks in Dutch from

2003 to 2013 They figure out that to adopt the new liquidity regulations, the banks prefer adjusting the liabilities side than the assets side of the balance sheet In addition, their finding point to the significant role of secured financing for explaining the leverage and liquidity cycle

Most of the empirical studies on the Basel III regulations and bank lending channel converge in the developed countries At country level, For instance, Furfine (2000) examines the

bank's loan portfolios adjustment to adopt the regulatory minimum capital requirements, the relationship between the bank capital levels and lending with the sample from the commercial banks in the United States from 1989 to 1997 The author indicates the capital requirement have

a negative impact on the bank loans demand and loan growth Francis & Osborne (2009, 2012) study the relationship of banks’ capital regulation on credit supply in the United Kingdom They also indicate that the higher capital requirements constrain the growth of credit Gunji et al (2010) consider the impact of the monetary policy on the bank lending for the different scale of Chinese bank from 1985 to 2007 They mention that under the monetary policy, the banks’ size and banks’ liquidity significant impact on bank lending But the capital of bank does not impact

on the bank lending with the varying monetary policy Giordana & Schumacher (2011) study the impact of the Basel III liquidity regulations on the Bank lending channel for banks in Luxembourg from 2003 to 2010.They employ the Liquidity Coverage Ratio and the Net Stable Funding ratio as two proxies of the Basel Liquidity regulations Besides that, they add banks’ characteristics: the total loans of the bank, shares of the total assets, the monetary indicators: the first difference of the nominal short-term interest income ratio, output gap, the dummy variables for the last liquidity period as the controls variable to avoid the heteroskedasticity Their findings indicate the new liquidity regulations of the Basel III reduce the Banking lending channel However, the liquidity regulations of the Basel III also enhance the reaction of banks to the monetary policy shock or the economic shock Cosimano & Hakura (2011) investigates the impact of the new capital requirement of Basel III on the bank lending channel via bank lending rate and loan growth for bank holding companies from 2001-2009 in Japan, Denmark, and the United States Under GMM estimation results, they suppose that the marginal cost of funding will increase when capital requirements for banks are higher The banks’ reaction on bank lending channel varies in the different countries Francis & Osborne (2012) study the relationship between the bank’s capital, bank’s lending and the bank’s balance sheet management behavior for bank firms in the United Kingdom from 1996-2007 Their results point out that the capital requirements change the bank’s capital and the bank’s lending lean on the disparity of the actual and the target ratio Šútorová & Teplý (2014) study the Basel III regulations in European from 2005-2011 They suppose that the new regulations of Basel III constrain the credit growth and negatively impact on the European bank value Bezoen (2015) investigate the impact of Basel III regulations on bank lending channel in the European banks from 2009-2013 He examines the

impact of Basel III via separately three stage regressions: (i) for capital, (ii) for interest income, and (iii) for bank loans As his results, the higher capital requirements can raise the marginal cost

of funding, and consequence is the interest income ratio increase and the bank lending decrease Another study examining the impact of Basel III on bank lending rates in European indicates that Basel III capital requirements can’t defend the financial market Tightening the capital requirements, will raise the lending rates and reduce the bank loans Angelini et al (2015) examine the long-term impact of Basel III regulations on the economic performance and fluctuations in the United States, Canada, France, Germany, Italy, Netherlands, and Spain They indicate that via Basel III regulations commercial banks can absorb the financial shocks Nevertheless, Basel III regulations also encumber the bank lending channel

Besides that, there are also several researches for banks belong to the group of countries For instance, Matousek and Sarantis (2009) investigate the bank lending channel and monetary transmission in Central and Eastern European countries from 1994 to 2003 They find that the banks’ size and banks’ liquidity have a significant influence on bank lending channel Berrospide

& Edge (2010) suppose that the impact of the bank capital on the bank lending lean on the relationship between the financial conditions and the real activity from 1992 to 2009 for 165 international bank holding companies By approaching VAR model, they prove that the bank capital modest effect on the bank lending Allen et al (2012) mention and review the impact of the reform Basel III regulation They figure out that Basel III reform will decrease the credit and reduce economic activity, but in the long term, Basel III keep banker more careful in their management Van den End & Kruidhof (2013) investigate the systemic implications of the Liquidity Coverage Ratio and the bank reactions under the Basel III liquidity regulations As the results, they aware that lower liquid assets in the buffer, more benefit for the macroprudential instrument Their results prove that the Liquidity regulations of the Basel II might be not effective, and have the adverse side effect during times of stress Bonner, Van Lelyveld, & Zymek (2015) study the determinants of the banks’ liquidity holdings with a combination of bank-specific and country-specific variables for 7000 banks from 25 OECD countries Their results show that the bank-specific and the country-specific determined the liquidity of the banks without the liquidity regulation of the international regulations In addition, they figure out that higher liquidity buffers lower the lending volumes and higher interest income ratios Duijm & Wierts (2016) examine the relationship between the new liquidity regulations under the Basel III

and the bank balance sheet for banks in Dutch from 2003 to 2013 They figure out that to adopt the new liquidity regulations, the banks prefer adjusting the liabilities side than the assets side of the balance sheet In addition, their finding point to the significant role of secured financing for explaining the leverage and liquidity cycle

A number of works have been done to test the effect of Basel on bank lending channel in different models such as: (i) Vector Autoregressive (VAR), (ii) Structural Vector Autoregressive (SVAR), (iii) Error Correction Model (ECM), (iv) Vector Error Correction Model (VECM), (v) Dynamic Stochastic General Equilibrium (DSGE), and (vi) Semi – structure models Each model coincides with each research approaches

The bank lending channel was studied via VAR (Bernanke and Blinder, 1992, Cogley and Sargent, 2005, Den Haan et al., 2007, Tillmann, 2008, Krusec, 2010, Auel and de Mendonça, 2011, Montes, 2013, Mengesha and Holmes, 2013) As the upgrade of VAR, SVAR also is employed for studying the bank lending channel (Bernanke and Blinder, 1992, Christiano

et al., 1996)

Fountas and Papagapitos (2001) using ECM to investigate the credit channel in the European Union They found that in the different countries in the European Union the lending indicator has different important level VECM is used to study the relationship between demand and supply of bank lending de Mello and Pisu (2010) study the bank lending channel transmission in Brazil by using VECM They employ the stock market, bank balance sheet, macroeconomic factor into their model, and they found that the interbank rates have a negative impact on bank lending Hülsewig, Winker, and Worms (2004) investigate the credit channel transmission in both demand and supply sign in Germany They figure out the existence of loan supply depends on loan interest income ratio Kakes, (2000) examine the bank lending in the Netherlands, and they found that bank lending channel in Netherland is not an important in the monetary transmission

Thompson and Cowton, (2004) study the bank lending in the United Kingdom by employing Semi – structural model Matthews, (2013) study the risk management for commercial banks in China by using semi-structured

Roger and Vlček (2011) study the costs of increased capital and liquidity regulations by employing DSGE model Gerali, et al (2010) study the credit supply by using DSGE Beau, Clerc, and Mojon (2012) investigate the relationship between monetary and macroprudential

policies by using DSGE model Angelini, et al., (2015) study the Basel III regulations effect on the macroeconomic in long –term

2.3 Hypothesis construction and the conceptual framework

2.3.1 The conceptual framework

This study adopts the analytical framework as the figure 2.3

Figure 2 2 The conceptual framework

2.3.2 The hypothesis construction

The Basel III capital requirements effect on bank lending channel

Basel III regulations require commercial banks raise their capital quality and quantitative The Basel III capital requirements impact on bank lending channel through the cost channel in both supply and demand side (Barth III and Ramey, 2002) Modigliani and Miller (1958) suppose that higher capital requirements lead to the higher marginal cost of loans Consequently, the loans interest income ratio increase and the loan growth decrease Besides that, in case increasing the capital requirements, commercial banks can adjust their banks’ portfolio (Rochet, 1992) To capture the capital requirements, commercial banks can cut directly their loan in case the cost of issuing shares too high (Van den Heuvel, 2002)

H1: The capital requirements have significant negative effects on loan growth

The Basel III liquidity regulations effect on bank lending channel

Basel III regulations require commercial bank maintain their higher liquidity assets Peek and Rosengren (1995) indicate that commercial banks with more liquid assets will make the bank lending lower Raising the liquidity of commercial banks assets, bank loan can be substituted by

Basel III regulations

Capital requirements

Bank interest income

Time dummy variable

s

the other assets with higher liquidity (Kashyap and Stein, 2000) So the lending volume would be reduced by the liquidity regulations

H2: The liquidity regulations of Basel III have a significantly negative impact on the loan

growth

The time effect of before and after applying Basel III on bank lending channel

In 2013, 10 countries (BRICS and 5 countries of OECD) already applied Basel III regulations After applying Basel III regulations, commercial banks have to raise their capital and liquid assets Based on the theory of bank capital and bank liquid asset, bank lending would reduce after applying Basel III regulation This study employs time dummy variables to investigate the effect of after applying Basel III regulations with the value equal 1 if the period time from 2014 to 2016 and equal 0 if the period time from 2011 to 2013

H3: After applying the Basel III regulations the loan growth significant reduce

comparing with before applying Basel III regulations

Different with the previous researches, this thesis nay contribute to filling the literature gap in analyzing both direct and indirect effects on the loan growth and lending rate, and addressing the endogeneity problem of these variables Additionally, this thesis examines the Basel III regulations more comprehensively in both capital requirements and liquidity regulations dimensions and compare the difference between two groups of countries: BRICS and OECD

CHAPTER 3 RESEARCH METHODOLOGY 3.1 Data sources

In order to study the impact of Basel III on the bank lending channel, this study employs unbalanced panel data of 391 commercial banks in 10 countries include: BRICS (Brazil, Russia, India, China, South Africa) and 5 countries in OECD (Australia, Canada, Japan, Singapore, Switzerland) which already applied Basel III as the progress report3 on implementation of the Basel regulatory framework in 2013 Data included microeconomic elements of commercial banks, collected from Orbis Bank Focus Besides that, data also included macroeconomic elements as control variables, collected from World Bank and OECD database

3.2 Research methodology

To investigate the direct and indirect impact of Basel III regulations on bank lending channel, this study employs Generalized Structural Equation Model (GSEM) GSEM clarified the causal relationship between the dependent and independent variables via the system of equation Different from other models of previous studies, this thesis utilizes two dimensions of the structural equation model as the following:

The system of equation expresses the mechanism impact of X1 and X2 exogenous variables on the Y1 and Y2 dependent variables Besides that, Y1 is the endogenous variable in the equation (1) and the dependent variable in the equation (2) Similarly, with the vector and matrix design, this study employs Loan growth as Y2, lending rate as Y1, and Basel III regulations as the X2 which have both impacts on Loan growth and lending rate The consequence, the simultaneous equation regression via the system equation of GSEM can integrate the direct impact of Basel III regulations on bank lending channel and the indirect of Basel III regulations on bank lending channel through lending rate So this study utilizes GSEM method in the study which clarifies both cases of the transmitting mechanism And in this case, it

is called the “binary-outcome models” for GSEM technique Based on the Paxton et al (2011), the system of equation of GSEM also demonstrated the following diagram

3 See the appendix 4

Sources: Paxton et al (2011)

Figure 3 2The recursive relationship

In previous studies, bank lending channel was studied in different models such as Vector Autoregressive (VAR), Structural Vector Autoregressive (SVAR), Error Correction Model (ECM), Vector Error Correction Model (VECM), Dynamic Stochastic General Equilibrium (DSGE), and Semi – structure models Each model coincides with each research approaches

VAR is steady-state equilibrium remains method unaffected by prudential policies (Angelini et al., 2015) For bank lending channel, researchers employ VAR with macroeconomic variables and time – series data By using VAR method, the research model can be omitted variables, because the variables of VAR model have to had a strong economic theory support (Van Aarle, Garretsen and Gobbin, 2003) SVAR is an upgrade of VAR, SVAR improvement compared to VAR are: (i) SVAR can distinguish the structural shock effect and unstructured shock effect, (ii) SVAR provides variance decomposition and impulse response function tool for analyzing monetary transmission

Similarly with VAR and SVAR, ECM also can be employed for bank lending channel with macroeconomic variables and time – series data But ECM can solve the cointegrated and non-stationary time series data (Granger 1981) ECM can examine the speed of returning long-term equilibrium after changing an independent variable VECM is the upgrade of ECM, estimates long-term and short-term relationships a small set of macro variables VECM disentangle loan demand and loan supply factors in the steady state However, VECM does not allow constructing the counterfactual experiments, and VECM also uses the basis of relationships observed in historical data to predict the effect of a change in economic policy, so agent’s expectations are not fully modeled The limitation of all theoretical models that is the possibility of bank’s default can’t reckon This means that the expected output associated with

the reduction in the frequency and severity of banking crises, could not measure (Kiyotaki and Moore 1997), indicate that macro models can feature endogenous bank’s default, but can’t feature bank’s capital or liquidity

For forecasting and policy analysis, central banks and other economic agencies also approach bank lending channel by using Semi – Structural model (Angelini et al., 2015) Semi – Structure is the model’s new steady state can be straightforward to compute Although Semi – Structural used by policymaker, they do not model the interaction between the financial sector and the economy Semi – Structural model only uses the basis of relationships observed in historical data to predict the effect of a change in economic policy, so agent’s expectations are not fully modeled

Moreover, most of researches approach bank lending channel by using the simulations belong to the last generation of the DSGE family, where the bank’s balance sheets and credit markets are modeled explicitly Researchers utilize DSGE to build up the framework for investigating the impact of capital and liquidity regulations on bank lending channel in both demand and supply side (Smets and Wouters 2007) DSGE can integrate the counterfactual experiments and policy scenarios into a conceptual framework DSGE can measure the steady-state value and long-term volatility of macroeconomic variables However, DSGE is the complicated estimation with a lot of fully and partially calibrated models, and DSGE for bank lending channel miss several aspects: (i) bank default, (ii) treating endogenous risk (Angelini et

al 2015)

Three crucial elements of the new regulatory framework in Basel III are higher capital ratios, higher quality of capital and tighter liquidity requirements To investigate the effect of new regulation of Basel III on bank lending channel, this study integrates three elements into models and employ dummy variables for time to analyze the effect of before and after apply Basel III regulations Most studies in bank lending channel focus on large and comprehensive national data This study observes bank lending channel for 391 banks in ten countries include: BRICS (Brazil, Russia, India, China, South Africa) and 5 countries in OECD (Australia, Canada, Japan, Singapore, Switzerland) which already applied Basel III

There are other estimation methods in panel data such as pooled least square (PLS), fixed effects model (FEM), or random effects model (REM) to estimate parameters PLS can’t handle heteroskedasticity, autoregression, and endogeneity problems in case these issues occur in the

model In addition, FEM and REM can solve heteroskedasticity, but FEM and REM can’t solve endogeneity problem Blundell-Bond system GMM (Blundell and Bond, 1998) can handle endogeneity, heteroskedasticity, serial correction and identification (Hall, 2003) This study employs three states least square – sub-method of system GMM, because of the causal relationship between bank lending volume and bank lending rate

To investigate the direct and indirect impact of Basel III regulations on bank lending channel based on the microeconomic principles from commercial banks, this study utilizes Three State Least Square estimation to estimate this study’s econometric model belong GSEM and panel data This study builds up GSEM with two functions for bank loans and loans interest income ratio and there is the endogeneity issue among bank loans and lending rate Three State Least Square is the combination of Seemingly Unrelated Regressions and Two State Least Squares So Three state least square can handle the heteroskedasticity, autoregression and endogenous problem in panel data (Arellano and Bond, 1991) and estimate for a system of equation Hessou (2017), utilize three state least square to study the effect of Basel III capital buffer requirements and credit union prudential regulation on bank lending in Canada

3.2.1 Model specification

This study conducts the applied model to test the impact of Basel III on the bank lending channel by employing GSEM Inheritance of the modifying: (i) the empirical models for loan’s rate of Cosimano & Hakura (2011), Sutorova and Teplý (2013), and (ii) the empirical models for the loan growth of Kashyap and Stein (2000), the simultaneous equation regression of this study

as the econometric models below:

IIi,t = αi + β1CRi,t+ β2LRi,t + β3LGi,t-1+ β4IERi,t + β5NIERi,t + β6LLPi,t + β7INBRi,t+

β8IRSi,t+ β9IFLi,t-1+ β10TAi,t-1 + β11YEARt + β12NPLRi,t+ µi,t

LGi,t = αi + β1LGi,t-1 + β2CRi,t +β3LRi,t+ β4IIi,t + β5LLPi,t + β6TAi,t-1 + β7GDPRi,t-1 +

β8MSGi,t-1 + β9YEARt + β10NPLRi,t + β11IBBi,t+ µi,t

Where:

LG is the loan growth

II is the interest income rate

Basel III regulations proxies:

YEAR is the time dummy variable

CR is the capital requirement

LR is the liquidity requirement

Control variables:

GDPR is the GDP growth

MSG is the money supply growth

INBR is the interbank rate

IRS is the interest income ratio spread

IFL is the inflation

INBR is the interbank rate

TA is the total asset

LLP is the loan loss provisions

NPLR is the non-performing loans to assets ratio

IBB is the interbank borrowing

IER is the interest expensive rate

NIER is the noninterest expensive rate

Table 3.1.The descriptions’ definition and measurement unit of variables

Dependent variables

Loan growth (LG) LG is the annualized change in total loans

Loan growth is defined for bank lending

Percentage

Independent variables

YEAR YEAR is the dummy variables for time

YEAR equal 1 in the period after 2013 and equal 0 in the period before 2013

Percentage

Liquidity regulations LR is the liquidity regulations of Basel III Percentage

Variables Definition Unit of measurement

(LR) regulations LR is defined by Liquidity asset

Inflation (IFL) IFL is the percentage change in CPI Percentage

Interest income ratio

spread (IRS)

IRS is the spread between the loans rate and deposit rate

Percentage

Inter-bank rate (INBR) INBR is the interest income ratio for the

short-term loans among commercial banks

Percentage

Total Asset (TA) TA presents the commercial bank scale This

study employs the logarithm of total asset

2013, Bezoen, 2015) The data of interest income ratio and loan growth are collected from the Orbis Bank Focus

Independent variables:

The foundations of the international regulatory framework of the Basel III for banks are the capital requirements (CR) and liquidity regulations Basel III employs many regulations to dominate CR via Tier 1 ratio (TIER 1), Common equity tier 1 ratio (CET 1), and the leverage ratio (LEV) Tier 1 ratio measure the core capital of the commercial bank Tier 1 ratio equals Tier 1 capital (mainly include the common equity tier 1 and the additional tier 1) over the risk-weighted assets Common equity tier 1 ratio is a segment of Tier 1 ratio CET 1 is the proportion

of common equity tier 1 (mainly include common shares and retained earnings) and the weighted assets Leverage ratio expresses the investment strategy of commercial banks Leverage ratio equals the capital measure (mainly is the tier 1 capital of the risk-based capital framework) over the exposure measure (mainly is the on-balance and non –derivative exposure) Cosimano and Hakura (2011) also employ the leverage ratio to represent the capital requirements have a negative impact on the bank lending and indicate the leverage ratio has a negative impact

risk-on the bank lending Through TIER 1, CET 1 and LEV, Basel III set up the standard for capital

and assets of the bank to dominate the risk of the banking system To adapt Basel III, banks adjust the balance sheet to suitable with the regulations via increasing equity capital or restricting the bank lending Hence, banks’ behavior to capital requirements of Basel III changes the bank lending rate and loan growth via TIER1, CET1 and LEV indicator The data for capital requirements are collected from the Orbis Bank Focus

For the liquidity regulations, there are two main proxies: (i) Liquidity Coverage Ratio (LCR) and (ii) the Net Stable Funding Ratio (NSFR) for liquidity regulations used by previous studies (Giordana & Schumacher 2011, Bonfim & Kim 2014, Schmitz 2011, Van den End, & Kruidhof 2013, De Haan & van den End 2013) As the Basel III phase-in arrangements 4 (BCBS, 2013), in 2013, LCR and NSFR didn’t impulse to apply yet So this study employs the liquidity asset ratio (LAR) as the proxy for liquidity regulations LAR equals to the fraction of the liquid assets over the deposits and short-term funding Before applying the liquidity regulations, commercial banks have many adjustments to raise the liquidity assets ratio and contemporaneously reduce the bank loans with the high credit risk To raise the LAR, commercial banks behave by direct cutting their loans with higher risk to buy liquid assets Besides that, they can restrict their deposit and short-term funding, simultaneously increasing their equity to sponsor for bank assets In this case, the marginal cost funding of commercial banks would be increase lead to the increase of the lending rate and decrease of the loan growth

So liquidity regulations have a negative impact on the bank lending rate and loan growth The data for liquidity regulations are collected form Orbis Bank Focus

Besides that, this thesis employs dummy variables for time (YEAR) applying Basel III to indicate the impact of before and after applying Basel III on bank lending rate and loan growth The point of time to apply Basel III in this area is 2013 So YEAR equals 1 if year more than

2013, and YEAR equals 0 if YEAR less than 2014 After applying Basel III regulations, commercial banks face a lot of pressure about the quality of capital and the liquid of assets So loan growth reduces after applying Basel III regulations

Moreover, the bank lending can be affected by another factor such as the macroeconomic and the characteristics of bank (Gunji and Yuan, 2010, Kishan and Opiela, 2000) This thesis employs macroeconomic and banks’ characteristics indicators as control variables to solve the heteroscedasticity For the macroeconomic segments, this study employs: (i) GDP Growth

4 See the appendix 1

(GDPG), (ii) Money supply growth, (iii) Inflation, (iv) Interbank rate (Gomez-Gonzalez and Grosz, 2006), (v) Interest income ratio spread GDP growth incites the demand for bank loans via the increase of consumption and investment (Bernanke and Gertler, 1995, Kashyap and Stein, 1994) Based on the mechanism of money multiplier, money supply has a positive impact on bank lending channel This study employs the narrow money supply (M1) growth The inflation and the interbank rate have the positive impact on the loans interest income ratio (Lepetit et al., 2008) The macroeconomic variables are collected from World Bank and OECD database

And for the microeconomics segments, this study employs: (i) Size, (ii) Loan loss provision (LLP), (iii) Inter-bank borrowing (IBB), (iv) Non-performing loans to assets ratio (NPLR), (v) Interest expensive rate (IER), (vi) Non-interest expensive rate (NIER) (Bezoen, 2015) IER and NIER are the main ingredients of interest income ratio Kashyap and Stein (1995), suppose that the commercial banks’ size, and the loan loss provision ratio (LLP) impact

on the bank lending rate and loan growth The commercial banks’ size can measure by the logarithm of the total assets, and the LLP can measure by the proportions between the loans loss provision and the total Inter-bank borrowing is the short-term funds by other financial institutes NPLR is the non – performing loans to assets ratio NPLR express the proportion of the default loans and loans without procure the interest Inhering Bezoen (2015), NPLR is employed as a control variables to investigate bank lending The data for microeconomic variables are collected from the Orbis Bank Focus

Because of the theory of macro-latency in monetary policy (Ehrmann, 2005), the banks’ characteristics, macroeconomic control variables, and policy rate variable are used mainly with the one year lag to fit the endogenous error Simultaneously, because of the causality relationship between bank loan and loans interest income ratio, this study utilizes the one year lag for Loan growth to solve the endogeneity of the model

CHAPTER 4 THE EMPIRICAL RESULTS

4.1 Data descriptions

This thesis employs microeconomic variables for 391 commercial banks and macroeconomic variables in 10 countries over the period from 2011 to 2016 Due to the availability of bank-specific data, this thesis utilizes an unbalanced panel dataset The data descriptions variables in this analysis presented in Table 4.1:

Table 4 1 Descriptive statistics of variables

BANK LENDING CHANNEL

BASEL III REGULATIONS

COMMERCIAL BANKS’ CHARACTERISTICS

MACROECONOMIC FACTORS

From table 4.1, the average loan growth nearly equal 11 percent annually for 391 commercial banks with total 1683 observations from 2011 to 2016 In this sample, there are 364 observations have negative growth in the total of 1683 observations, major in the period 2014-

2016 with 280 observations have negative growth and just 84 observation in the period

2011-2013 have negative loan growth Simultaneously, there are 210 observations have negative loan growth in BRICS with 202 observations in the period 2014-2016, and 154 observations have negative loan growth in 5 countries in OECD with 78 observations in the period 2014-2016 So the data show that in 5 countries of OECD negative loan growth value occur throughout the

period 2011-2016, but in the BRICS the negative loan growth value just focus in the period after

2013 In the overall context, loan growth decrease after 2013 The trend of loan growth is expressed in the figure below:

Source: Author analysis

Figure 4 1 The loan growth distribution of 10 countries from 2011 to 2016