The effect of credit growth on credit quality, evidences of commercial banks in dong nai province

The loans to these borrowers become lowquality credit.Under the assumptions that banks reduce the lending rate as well as lower the credit standards of customers, the occurrence of suppl

UNIVERSITY OF ECONOMICS ERAMUS UNIVERSITY ROTTERDAM

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

THE EFFECT OF CREDIT GROWTH ON CREDIT QUALITY:

EVIDENCE FROM THE COMMERCIAL BANKS IN DONG NAI

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

By

TRINH HOANG VIET

Academic Supervisor

Dr VO HONG DUC

I hereby declare that the thesis “T HE E FFECT OF C REDIT G ROWTH ON

C REDIT Q UALITY : E VIDENCE FROM T HE C OMMERCIAL B ANKS IN D ONG N AI ”, which

is submitted to Vietnam – Netherlands Programme, is my original research work.All of the contents which are not from my own work are cited carefully and clearly

A CKNOWLEDGEMENT section.

Signature

Trinh Hoang Viet

Ho Chi Minh City, 1st November

I wish to acknowledge the contributions of all VNP students in Class 20,especially Mr Vo Van Hung and Mr Nguyen Son Kien, for sharing learningexperiences and valuable academic materials.

I would also like to extend my thanks to all VNP Staffs for their enthusiasm ofassisting my study over the last two years

Finally, I would not forget to send my deepest thank to my parents whoalways encourage me to keep up with my study objectives

This research is conducted to examine and quantify the effect of credit growth

on credit quality for the commercial banks in Dong Nai In relation to this possibleeffect, theoretical framework of the so–called “three shifts” in the credit market islikely to explain that credit growth might have positive or negative effect on credit

quality These three shifts are generally known as: (i) a supply shift (an expansion in

bank loan supply by lowering credit standards), (ii) a demand shift (an increase inloan demand to optimize business activities) and (iii) productivity shift (a positivechange in macroeconomic conditions) In addition, empirical evidence confirms thatrapid credit growth of commercial banks could lead to a deterioration orimprovement of credit quality The macroeconomic context for banking industryindicates that the decline in credit quality after a period of growth might be areflection of (1) negative changes in the macroeconomic determinants which have abad influence on the business activities of borrowers; and (2) informationexternality which makes banks hardly gain efficiency in evaluating their customers

This study utilizes the data of 29 commercial banks operating in Dong Naiprovince for the period from 2009Q3 to 2014Q4 The econometric technique ofDifference GMM for dynamic panel data model is adopted in order to examine theeffect of credit growth on credit quality in the context of the commercial banks inDong Nai This study finds empirical evidence to confirm that credit growth causesthe decrease in credit quality after three quarters to one year In addition, this effect

of credit growth in the long run is found in this study

These findings obtained from this study reflects that: (i) commercial banks inDong Nai might have lowered their credit standards to increase their lending tobusiness customers and individuals; (ii) the conditions of local economy in DongNai might have not been really favorable for business activities during the researchperiod; (iii) and the information externality in the loan market might have distorted

Keywords: credit growth, credit quality, commercial bank, difference GMM,

dynamic model, non–performing loan

TABLE OF CONTENTS

D ECLARATION i

A CKNOWLEDGEMENT ii

A BSTRACT iii

T ABLE OF C ONTENTS v

L IST OF T ABLES vii

L IST OF F IGURES viii

C HAPTER 1 I NTRODUCTION 1

1.1 PROBLEM STATEMENT 1

1.2 RESEARCH OBJECTIVE AND QUESTION 2

1.3 RESEARCH SCOPE AND METHODOLOGY 2

1.4 THESIS STRUCTURE 3

C HAPTER 2 L ITERATURE R EVIEW 4

2.1 THE MACROECONOMIC CONTEXT FOR BANKING 4

2.1.1 Main Characteristics of Banks 4

2.1.2 Shock and Vulnerability of Banking System 6

2.1.3 The Effect of Macroeconomic Determinants 6

2.1.4 Credit Growth and Vulnerability of Banking System 7

2.2 CREDIT GROWTH AND CREDIT QUALITY THROUGH DIFFERENT SHIFTS 9

2.3 CONTROL FACTORS FOR CREDIT QUALITY 17

2.4 PREVIOUS EMPIRICAL STUDIES 20

2.5 THE CONCEPTUAL FRAMEWORK 26

C HAPTER 3 M ETHODOLOGY AND D ATA 27

3.1 MEASURING CREDIT QUALITY 27

3.2 DATA COLLECTION METHOD 29

3.3 ECONOMETRIC METHODOLOGY 30

3.3.1 Dynamic Panel Data Estimator 31

3.3.3 Estimating The Long–run Coefficients 33

3.3.4 Econometric Specification 35

3.3.5 Hypothesis testing 35

C HAPTER 4 RESULT AND D ISCUSSION 38

C HAPTER 5 CONCLUSIONS 46

R EFERENCES 49

LIST OF TABLES

T ABLE 2.1 Change in credit standard, credit growth and credit quality 16

T ABLE 2.2 Summarization of the literature 25

T ABLE 3.1 Necessary items and their account type 29

T ABLE 3.2 The expected signs of variables used in the research 30

T ABLE 3.3 The calculation of variables 30

T ABLE 3.4 Hypotheses need testing 37

T ABLE 4.1 Descriptive statistics 39

T ABLE 4.2 Estimation result 41

LIST OF FIGURES

F IGURE 2.1 The macroeconomic context for banking 8

F IGURE 2.2 Supply shift 10

F IGURE 2.3 Demand shift 13

F IGURE 2.4 Productivity shift 15

F IGURE 2.5 Different shifts in the macroeconomic context for banking 16

F IGURE 2.6 The effect of credit growth on credit quality 26

F IGURE 4.1 Deposit growth rate and deposit interest rate 39

to an increase in the demand, this growth will not essentially lead to bad loans.Therefore, credit growth may be a reflection of credit quality under somecircumstances.

When the economy is in the stage of recession, it is certain to affect negatively

on the financial market especially the banking system The general picture is thatcommercial banks attempt to boost credit growth for profit objective whereashouseholds and firms who borrow money from banks have to face with difficulties

in business activities This leads to a consequence that credit growth may reducecredit quality The problem is whether the profit target of commercial banks byboosting credit growth would be efficient or it just increases the NPLs which bring

no profits or even losses

The determination of the effect of credit growth on credit quality is gettingmore and more important for not only commercial banks but also central bank andpolicy makers For commercial banks, it may help them to consider the appropriatetime of loosening or tightening credit standards and the decision to expand or limitlending activities For central bank, it will help to control the loans growth of

commercial banks This is to prevent potential banking crisis when the credit quality

is too low In some cases, understanding the true influence of credit growth oncredit quality could help to recognize the real state of the economy in order to applymacroeconomic policies more efficiently

In a report of The State Bank of Vietnam, Dong Nai branch (2015), the currentdebit balance of total loans is about 100 billion VND This source of capital ismainly concentrated on prior fields and industries of Dong Nai province Theoverdue loans are kept at a safe ratio of 2.32 percent However, the businessactivities of firms still meet many difficulties when commercial banks apply newstandards of classifying debts Besides, commercial banks are detected of infringingcredit regulations such as appraising credit documents carelessly; misevaluatingcustomer financial capacity and collaterals; and not supervising capital usageclosely Under these conditions above, commercial banks in Dong Nai might have

to face a high chance of potential credit risk while lending activities is more andmore expanded

1.2 R ESEARCH O BJECTIVE AND Q UESTION

This research is to investigate the influence of banks’ credit growth on theircredit quality under the control of some characteristics of banks To achieve theresearch objective, this study attempts to answer the following question:

Does a positive change in the commercial banks’ credit growth lead to a negative change in banks’ credit quality in the case of Dong Nai banking system?

1.3 R ESEARCH S COPE AND M ETHODOLOGY

The research is carried out in the scope of credit growth and credit quality of

29 commercial banks in Dong Nai province, Vietnam The data is collected in theperiod from 2009Q3 to 2014Q1

The main methodology of this study is quantitative analysis Due to theavailability of panel data, the limitations of common estimation methods and theobjective of capturing the changes of banks’ credit growth and credit quality, thisresearch applies the method of Difference GMM for the dynamic panel data model.Besides, the dynamic model could be used to generate long–run coefficients whichreflect the equilibrium of the effects of credit growth.

1.4 T HESIS S TRUCTURE

This thesis includes five chapters Chapter 1 introduces the background andmotivation of the research on the effect of credit growth on credit quality Chapter 2reviews related theories, previous empirical studies and builds a conceptualframework for the research Chapter 3 presents the data collection method andquantitative techniques for producing necessary results Chapter 4 shows the resultsand discussions Chapter 5 summarizes main research findings, brings out policyimplications, raises some limitations and suggests further studies

C HAPTER 2

This chapter introduces the two main theories related to the effect of creditgrowth on and credit quality First, the macroeconomic context for banking is toexplain how this effect could occur Second, the so–called three shifts are to explainthe detail channels in which credit growth could affect credit quality in the creditmarket This chapter also presents some theories about the effect of some bankcharacteristics on credit quality as control factors Besides, some related previousempirical studies and the conceptual framework for this research are also presented

2.1 T HE M ACROECONOMIC C ONTEXT FOR B ANKING

As the role of a financial intermediary, banks have a large contribution on theentire economy in the aspect of finance or anything relating to money Bankingitself is an industry in the economy Therefore, banks have their own characteristicswhich are unique and different from other industries In order to analyze bankingsystem under the impact of the macroeconomic context, this research is firstly tointroduce the three main bank characteristics which have close relationship to creditgrowth and credit quality

Banks have extremely high leverage (1) Banks mostly use other people’s

money for their portfolio Similarly, banks primarily mobilize capital for lendingactivities According to Gavin and Hausmann (1996), bank leverage has twoimplications First, bank operations are very sensitive to the volatility of themacroeconomic determinants due to very thin capital They may become insolventafter small negative changes of the economy Second, high leverage may bring to aproblem relating to the benefit of bank shareholders and debt–holders Bank

managers often generate risky portfolio to bring the highest benefit for shareholderswhile debt–holders is limited in their capital recovery in case of insolvency.

Banks are easy to become illiquid (2) First, the term of deposit liabilities tend

to be higher than the term of loan assets Borrowers – for example, firms andhouseholds need a long time to finance their business activities while the depositorshave right to withdraw money at any time In case of time deposits, the depositorsare still allowed to withdraw money as long as they accept low or no interest rate.Second, if banks attempt to manage the term of their loans, the borrowers still havelonger time in paying the debts There is a temporary solution in which borrowerscould roll over their loans by borrowing the new to service the old ones in the samebank This action is not always allowed or even illegal, so it affects strongly andnegatively on profitability of the borrowers, which causes a decrease in creditquality This characteristic raises a problem that banks have to plan for theiradditional reserves These reserves are used to reduce the illiquidity which mayoccur under the influence of adverse macroeconomic shocks

Banks cannot evaluate exactly their borrowers in the expansionary phase of the economy (3) Gavin and Hausmann (1996) indicated that “good times are bad

times for learning” about the truth of financial capacity of the borrowers Theadvantages of the economy may be one of the reasons of lending booms Theborrowers could easily borrow money from a bank to service the debts in anotherbank Therefore, most of the borrowers appear in good state with banks althoughtheir financial capacity may be different In this case, banks have difficulty indetermining which loans may potentially become NPLs

These characteristics above are to imply that the decision of boosting creditgrowth should be considered carefully especially in the disadvantageous conditions

of macroeconomic environment Banks have very high probability in the decline ofcredit quality, which is the source of illiquidity as well as banking crisis

2.1.2 Shock and Vulnerability of Banking System

The action of using mobilized money for lending requires a necessarycondition in which the growth rate of deposit liabilities is higher than the depositinterest rate However, if this condition does not hold, in principle, banks still have asolution for this problem by calling their borrowers for paying their mature debts.Actually, banks have limitations to do so because it depends too much on the ability

of the borrowers Consequently, there would be a net resources transfer frombanking system to depositors in the form of withdrawal and interest The largeamount of this net resources transfer will create a shock in the banking system And

if this amount is large enough, the banking system would collapse, which create thevulnerability (Gavin and Hausmann, 1996)

2.1.3 The Effect of Macroeconomic Determinants

The shock from large change in the net resources transfer may originate fromthe changes of the economy When there is a negative economic surprise from one

or some macroeconomic determinants, there would be two cases: (i) the borrowerscannot service their debts due to the reduced efficiency in their business activities;(ii) banks are limited to investing activities especially lending and become illiquiddue to the decline in deposit demand or increase in withdrawal Both cases lead tobanking crisis in the form of insolvency In the first case, credit quality decreases andbanks are not able to recover enough principal and interest to finance their depositliabilities In the second case, banks hardly meet the demand of withdrawals Theseconsequences may lead to potential financial vulnerability in the future However, ifbanks can boost their mobilizing activities, they would have a source of liquidity forwithdrawal demand of depositors Besides, banks would have more time to deal withtheir NPLs

2.1.4 Credit Growth and Vulnerability of Banking System

As discussed above, the macroeconomic determinants do not have direct andcomplete effect on the banking system Their affect is mainly on the businessenvironment and the depositors’ behaviors The core question is the reason whybanking system becomes too fragile to suffer from the negative changes of theeconomy It is easy to understand that borrowers’ business activities are stronglyinfluenced by these changes If banks have very close relationship with theirborrowers, absolutely, they are also influenced Rapid credit growth would be atypical proxy for this close relationship The more banks expand their loans, themore they rely on their borrowers

Boosting credit growth is closely related to the third characteristic of banks.Once they recognized the good appearance in the ability of their borrowers, they arewilling to lend more This creates a link between credit growth and the vulnerability

of the banking system However, credit growth should be considered as a signal ofeconomic development than a cause of vulnerability The next question is in whatcircumstances credit growth performs its negative aspects The answer would beconcerned about information problems

Gavin and Hausmann (1996) believed that “…it is very difficult for bankers toobtain information about the creditworthiness of borrowers” (p 14) First, due to theeconomic expansion, the borrowers can perform well on their capital and gainpositive cash flow This would be an advantage opportunity to offer loans not onlyfor the existing customers but also new borrowers Banks would have very limitedinformation about their new borrowers Thus, the probability of misevaluating themmay be quite high, which causes potential decline in credit quality in the future.Second, the plenty of loans supply during the economic expansion helps theborrowers to approach more lenders As stated above, banks offer the loans and theborrowers use these loans as a source of paying debts in other banks These loansaccidentally and adversely impact on other banks’ information, which create an

information externality in the credit market This type of externality also leads to

credit misevaluation and potentially low credit quality

Figure 2.1 summarizes the macroeconomic context for banking The scope of

this research concentrates on the “lending” and “paying debts” direction in this

figure The negative relationship of credit growth and credit quality might reflect

two situations Firstly, the adverse shocks from macroeconomic determinants would

make business activities become inefficient, which obstructs the ability of paying

debts Secondly, the good signals in the expansionary phase of the economy might

create information externalities for banks to evaluate their customers

F IGURE 2.1 The macroeconomic context for banking

Depositors

Mobilizing

Paying debts Evaluating Information externality

2.2 C REDIT G ROWTH AND C REDIT Q UALITY T HROUGH D IFFERENT S HIFTS

Theoretically, the nature of credit growth may not relate to its quality It meanscredit growth might not directly have any influence on the change in credit quality.However, the amount of loans which banks decide to lend would depend greatly onthe performance of themselves and their customers For instance, banksunderestimate the risk of their borrowers and are willing to lend more Therefore,the relationship of credit growth and credit quality might exist

One of the earliest studies on the theoretical link between credit growth andcredit quality is present by Clair (1992) This link comes from banks lowering theircredit standards to attract more borrowers This action may lead to low creditquality in the future Besides, when banks boost credit growth but they do not haveany appropriate strategies to administrate their borrowers’ loans usage Creditquality would decline

Clair (1992) also indicated that credit growth may positively correlate to creditquality during the recovery or expansion phases of the economy or the structuralchanges in the financial markets – for example, reducing barriers between banksand borrowers to expand credit growth and reduce credit risk throughdiversification

Keeton (1999) had developed a theory about different shifts to investigate theeffect of credit growth on credit quality This study explained both negative andpositive relationship between credit growth and credit quality

This relationship is firstly explained by a supply shift in the loan market In

this research, supply shift means banks have decision on the willingness to lendmore and there are two ways for them to carry out The first is to reduce the lendingrate of new loans and the second is to lower the credit standards of these loans Tomake lending become easier, banks would overestimate the value of collaterals ofthe loans, accept to lend customers with low financial capacity or go through

projects of which cash–flow statement is not appraised carefully These actionslower the credit standards and put banks into a high chance of lending to theborrowers with low credit– worthiness The loans to these borrowers become lowquality credit.

Under the assumptions that banks reduce the lending rate as well as lower the

credit standards of customers, the occurrence of supply shift which boost credit

growth will tend to result in the low credit quality

F IGURE 2.2 Supply shift

r e Expected rate of return from loans

S Supply of loans from banks

D Demand of loans from the borrowers

S2

D

Figure 2.2 presents how the supply shift has effect on total amount of lending

and the level of credit standards In the left–hand side, the expected rate of return ofbanks is a function of credit standards This figure assumes that the credit–standardcould be measured in number z on the horizontal axis The high value of z showsthat the borrowers are in good state of servicing debts, for example, they have highvalue on their collaterals or their investment project is safe Banks’ lending decision

is made basing on the expected rate of return from loans which is measured on thevertical axis This expected rate of return depends on both lending rate and debtservicing capacity Good borrowers would bring high expected rate of return forbanks and it might be the same as the lending rate If there are any signs of not goodborrowers, the banks’ expected rate of return would be less than the lending rate.

From the side of credit standard, with each value of z, banks could derive amaximum expected rate of return This is reflected by the curve re z For a point ofany value of z, when the lending rate increases, the banks’ expected rate of returnwill increase However, the increase in the lending rate could not always raise theexpected rate of return, there would be a limit For example, when the lending rateincreases, good borrowers still have enough financial capacity in their project torepay for debts If the lending rate increases more and more, the borrowers tend toinvest in riskier project with the expectation of higher return for repayment Theseborrowers might become inferior and banks’ expected rate of return could notincrease any more Thus, the curve re z in the left–hand diagram shows themaximum expected rate of return This curve is upward sloping because banksexpect to earn more return from the borrowers with better credit standards byoffering them high lending rate

The curve re z could also be analyzed from the side of expected rate return For any given value of re, there would

be a minimum credit standard level of the borrowers For example, at the equilibrium point in the loan market, banks will expect for r 1e It is certain that banks could not give any credit to any borrowers with lower than z 1 because the expected rate of return will be less than r 1e no matter how high the lending rate is All borrowers from z 1 and higher could receive

standards would be a threshold for banks to decide whether they lend or not The higher the expected rate of return banks desire, the higher threshold of credit standards they set to their customers.

11

The right–hand side of Figure 2.2 describes the loan market which determines

the banks’ expected rate of return Banks are willing to lend more if their expectedrate of return from loans increases Thus, the supply curve is upward sloping Forthe demand curve, it is downward sloping due to two reasons Firstly, banks couldcharge lending rate basing on their expected rate of returns Higher lending rate maybring higher return However, the borrowers have to suffer high cost of capital andthey will borrow less Secondly, the negative slope of the demand curve could beexplained through the left–hand diagram When the expected rate of return of banksincrease, the threshold level of credit standard would be higher, then the number ofborrower who meets the credit standards would reduce

The loan market is in the equilibrium when the bank loan supply equals to the loan demand Before the supply shift, the supply curve is S 1 S 1 At the equilibrium, banks’ expected rate of return is r 1e and the total amount of loans is L 1

Let assume that banks desire to expand the total amount of loans which causes thesupply shift To do this, banks have to firstly reduce their credit standards for attractingmore borrowers On the right–hand diagram, the supple curve will shift to the rightfrom S1S1 to S2S2 The total lending would increase from L1 to L2. Then the expectedrate of return would decrease from r1e to r2e It means banks do not require such highrate of return Therefore, they not only charge lower lending rate for good borrowersbut also reduce the credit standards threshold to approach more borrowers This decline

in the credit standards is presented by the movement down along the curve re z Oncethe credit standard declines, banks have to suffer more borrowers with low capacity ofservicing debts, which causes NPLs or low credit quality The whole progress could bedescribes in brief as follow:

H YPOTHESIS A: Credit growth might negatively associate with credit quality

12

Keeton (1999) also supposed that the expansion in lending which occurs not

by a supply shift may have positive effect on credit quality There would be two

reasons First, the source of loan expansion is caused by positive shift of the

borrowers’ demand This positive demand shift may come from – for example, the

decision to change the capital structure of firms or projects in order to improve the

cash–flow In this case, the borrowers’ repayment capacity might become better,

which improved the credit quality Second, the source of loan expansion is still from

the positive demand shift but this shift come from the productivity of borrowers It

could be called productivity shift The case of productivity shift reflects favorable

conditions in the borrowers’ business activities, which boost credit growth first and

credit quality afterward

For demand shift, under the consumption that increase in the borrowers’ loans

demand does not related to their goodness in financial capacity – for example,

requiring loans from banks to avoid high interest rate in the capital market or

restructuring capital to reach optimal leverage ratio, the demand shift which boost

credit growth will lead to high credit quality

F IGURE 2.3 Demand shift

choice for the borrowers to achieve their objectives of – for example, restructuringcapital Therefore, the demand shift also raises the credit standards.

As can be seen in the right–hand diagram of Figure 2.3, the increase in the loans demand

demands increases, bank will expect more return by raising both lending rate and the minimum credit standards This action helps banks to avoid some bad borrowers and the credit quality would be improved.

In a demand shift, banks do not often realize demand shift, they still keep thelending rate and the threshold level of credit standards unchanged There would bemore and more borrowers who meet the credit standards desire to borrow money.Then, they realize the growth in demand and start to raise lending rate and tightencredit standards Lastly, the credit quality increases Therefore, the process would

be as follow:

For productivity shift under the consumption that the increase in loans demand

comes from the favorable conditions in business activities of the borrowers.Although the credit standards may decline in this case, boosting credit growth couldresult in high credit quality

When the productivity shift occurs – for example, firms have someimprovements in their technology, the input costs reduce or the economy is in goodcondition, the borrowers will need more credit to operate their business or invest innew projects In this case, banks might believe that most borrowers would havegood opportunities in business and obtain more cash inflows to service debts Thus,

14

their attitude to credit standards may change negatively and riskily It means banks would expect

higher rate of return than at the same level of credit standards or they are willing to loosen the

credit standards in order to earn the same expected rate of return Therefore, the curve re z shifts to

the left, which can be seen on the left– hand diagram of Figure 2.4 On the one hand, this shift of

the curve re z could attract more borrowers due to the decline in credit standards and on the other

hand, the productivity shift also increase the loans demand to meet the requirements of business

activities This would bring a significant positive shift in loans demand In the right–hand side of

Figure 2.4, total amount of loans will grow dramatically from L1 from L 2 and banks’ expected rate

of return increase from r 1e to r 2e Although banks loosen their credit standards to accept more bad

borrowers, these borrowers are not certain to be really bad because they still experience benefits

from productivity shift As a result, the credit quality is still improved However, there would be a

possibility that banks still safely keep the curve re z unchanged like the demand shift case.

F IGURE 2.4 Productivity shift

As discussed above, productivity shift make banks loosen credit standards and

make borrowers demand large amount of loans The question of whether the change

of credit standards or credit growth happens first would depend on who could

15

realize the productivity shift first Anyway, credit quality is improved lastly The process would be as follow:

H YPOTHESIS B: Credit growth might positively associate with credit quality

T ABLE 2.1 Change in credit standard, credit growth and credit quality

Type of shift Order of change

H.A: Credit growth might negatively associate with credit quality

Supply shift Credit standards ↓ Credit growth ↑ Credit quality ↓

H.B: Credit growth might positively associate with credit quality

Demand shift Credit growth ↑ Standards ↑ Credit quality ↑

Productivity shift Credit growth ↑ Credit standards ↓ or ↑ Credit quality ↑

Table 2.1 summarizes the relationship between credit growth and credit

quality with the appearance of level of credit standards Besides, the order of change

of different shifts is to confirm that a change in credit growth in the past may lead to

a change in credit quality in the future This effect is not contemporaneous

Figure 2.5 shows the position of different shifts in the macroeconomic context

for banking and they do not exist simultaneously First, banks lower their credit

standard in evaluating customers to make a supply shift Second, the demand of

capital from borrowers for improving their business activities would create a

demand shift Finally, the advantages from macroeconomic determinants would

generate a productivity shift for business activities.

F IGURE 2.5 Different shifts in the macroeconomic context for banking

Depositors

Principals

Evaluatingcustomers

Source: Author’s summarization

2.3 C ONTROL F ACTORS FOR C REDIT Q UALITY

“Bad management I” hypothesis

This hypothesis was developed by Berger and DeYoung (1997) to indicate thatmeasured cost efficiency could affect credit quality Low measured cost efficiencymay be a signal of bad management which could be considered as, first, banks’credit appraisal process is weak and they may offer loans for the bad borrowers orinvest in projects of which cash flow is inflated Second, banks’ valuation skill isnot accurate, which may cause the underestimation for collateral values and bankscould not recover their low quality loans This problem may be related to thephenomenon that the compromise between the banks’ asset pricing department andthe borrowers may exist Third, the banks’ customer supervision is

loose that borrowers may use the loans with improper purposes – for example,borrowers use loans to invest in high risk and high return projects instead of projectswhich are appraised by banks Due to these three reasons, the credit quality woulddecrease.

“Skimping” hypothesis

Berger and DeYoung (1997) also showed the adverse relationship betweenmeasured cost efficiency and credit quality Banks may have more short–termoperating costs to monitor their current loans to avoid NPLs in the future In otherwords, skimping on some short–term costs for monitoring loans may lead to lowcredit quality Banks would be less cost efficient if they are willing to spend somecosts which prevent them from suffering higher losses from NPLs in the future.However, this trade–off reflects a decrease in measured cost efficiency and anincrease in credit quality afterward

Both “Bad management I” and “Skimping” hypotheses relate to the measured

cost efficiency There may be a variable which could capture it However, it isnecessary to distinguish these two hypotheses The difference is in the cost, one iswaste because of consequences of bad management and one is useful to preventlosses in the future

“Bad management II” hypothesis

Similarly to “Bad management I” hypothesis, banks’ bad management

including limitations of credit appraisal process, valuation skill and customersupervision may be reflected by low performance which could be recognized by lowprofitability on equity or assets In contrast, banks with better performance wouldhave high quality skills of administrating lending activities, which could reduceproblem loans or increase credit quality in the future (Louzis, Vouldis and Metaxas,2011)

“Pro–cyclical credit policy” hypothesis

Banks credit policy depends on the goal of not only maximizing profitabilitybut also keeping their reputation at good state According to Rajan (1994), bankscannot show the market their efficient loan portfolios as well as how highperformance their customers are The only thing that market could observe is banks’earnings As a result, banks are willing to inflate their earnings to maintain the goodperception of the market Inflating earnings would be easy to do if banks use aliberal credit policy This kind of credit policy is also called “pro–cyclical” because

it is correlated with the demand conditions – for example, banks keep offering newloans to the bad borrowers to continue their business although they had problemloans before This action not only conceals the existence of NPLs but also convincethe market that banks still gain profits from lending However, it is not safe forbanks in the future Therefore, pro–cyclical credit policy might lead the increase inperformance but then, the credit quality would decline

“Diversification” hypothesis

Louzis et al (2011) believed that banks could enhance their credit quality ifthey have diversification opportunities For example, banks have good ability inseeking good business projects or investing in stocks of companies with highpotential development The proportion of capital for lending would decline, whichlimit the probability of lending the bad borrowers However, diversifyinginvestment fields other than lending may bring high return but they need a very longtime to recover both principals and interests Liquidity risk will happen if banks’customers want to withdraw money because the capital structure of banks is mainlyliabilities from mobilizing Therefore, large banks will have more diversificationopportunities which hardly affect their liquidity In contrast, bank with small sizehave less chance to diversify due to liquidity risk

“Too–big–to–fail” hypothesis

According to Stern and Feldman (2004), large banks believe in government’sintervention when they are in failure because they have enormous influence onfinancial market Therefore, larger banks are willing to increase more leverage andattempt to lend more This act of credit expansion increases the chance ofapproaching low quality borrowers (Louzis et al., 2011)

Under this hypothesis, the effect of leverage on credit quality would benegative and adjusted by banks size When banks increase an amount of leverage,credit quality of larger banks would decline more than smaller banks It is similar tolending aspect which banks boost credit growth

Nguyen (2015) believed that this hypothesis could be appropriate only forsome largest banks It means only some largest banks could be able to receivesupport from government while they are in distress Due to this reason, “somelargest banks” is a characteristic which could be used as an alternative for banksize–adjustment under “too–big–to–fail” hypothesis

H YPOTHESIS C: The effect of credit growth on credit quality from large

commercial banks might be larger than the smaller banks.

2.4 P REVIOUS E MPIRICAL S TUDIES

Clair (1992) is one of the earliest authors who investigated the relationshipbetween credit growth and credit quality of banks in Texas using annual data from

1980 to 1990 The author used the loan loss ratio and NPL ratio to measure creditquality The independent variables were divided in to three groups including: (i)credit growth, (ii) financial characteristics (bank assets, bank equity, business loansand real estate loans) and (iii) business conditions (non–agricultural employmentgrowth) There were three types of credit growth used simultaneously in the model:internal growth, growth through bank merger and growth through bank acquisition

The effect of credit growth was considered in contemporaneous and 1, 2 and 3–yearlagged variables The method of ordinary least square (OLS) was applied in themodel The main results indicated that credit growth (internal growth and growththrough bank acquisition) had significantly improved the credit quality atcontemporaneous and one–lag year in both cases of using loan loss ratio and NPLratio However, the credit growth through bank merger had the inverse effect in case

of loan loss ratio but no significant impacts on NPL ratio In addition, most offinancial characteristics and business conditions are also good control variables

In a study of determining factors of banking crises conducted by Kunt andDetragiache (1998), one of the conditions which contributed to banking crises is thehigh ratio of non–performing assets (exceeding 10 percent) This study used annualdata of all market economies in the period 1980 – 1994 and applied a technique ofmultivariate logistic regression to analyze There are three groups of independentvariables including macroeconomic (GDP growth, external terms of trade, realinterest rate, inflation and government surplus), financial (ratio of money supply M2

to foreign exchange reserves, ratio of domestic credit to private sector to GDP, bankliquidity and credit growth), and institutional (GDP per capita, index of the quality

of law enforcement and a dummy for the presence of an explicit deposit insurancescheme) variables in the model One of financial variables – credit growth – is usedwith the lag of two year and it has significant and positive influence on theprobability of banking crises According to the result, 10 percent in the increase ofcredit growth may lead to 5.4 percent in the increase of probability of crisis

Another study on the effect of credit growth on credit quality was carried out

by Keeton (1999) This study uses quarterly time–series data in the two separateperiods: 1967 – 1983 and 1990 – 1998 (missing data in the 1984 – 1989) The data

is collected from Senior Loan Officer (SLO) conducted by the Federal Reservesince 1967 The method of vector auto–regression (VAR) is used for the two VARsystems The first one includes loan growth, credit standards and GDP gap and the