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 MASTER OF ARTS IN DEVELOPMENT ECONOMICS

UNIVERSITY OF ECONOMICS HO CHI MINH CITY VIETNAM ERAMUS UNIVERSITY ROTTERDAM INSTITUTE OF SOCIAL STUDIES THE NETHERLANDS 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 MASTER OF ARTS IN DEVELOPMENT ECONOMICS By TRINH HOANG VIET Academic Supervisor Dr VO HONG DUC HO CHI MINH CITY, December 2015 DECLARATION I hereby declare that the thesis “THE EFFECT CREDIT QUALITY: EVIDENCE FROM OF CREDIT GROWTH THE COMMERCIAL BANKS IN ON DONG NAI”, 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 in this thesis I certified that the contents of this thesis have not been and are not being submitted for any other degrees This thesis was done under the supervision and guidance of Dr Vo Hong Duc, Economic Regulation Authority, Western Australia and the Edith Cowan University, Australia Any other contributions to this thesis are presented in the ACKNOWLEDGEMENT section Signature Trinh Hoang Viet Ho Chi Minh City, 1st November 2015 In my capacity as the supervisor of this thesis, I certified that the statements above are true to the best of my knowledge Signature Dr Vo Hong Duc Date: i ACKNOWLEDGEMENT I would like to express my great appreciation to VNP Lecturer Team for their dedication in teaching, which brought to me useful knowledge and experiences in my academic career I would like to express my deep gratitude to Dr Vo Hong Duc, my research supervisors, for his patient guidance and useful critiques of this research work I would like to offer my special thanks to Mr Pham Quoc Bao and Mr Pham Thanh Huu for facilitating me to collect necessary data during my internship in The State Bank of Viet Nam, Dong Nai Branch 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 learning experiences and valuable academic materials I would also like to extend my thanks to all VNP Staffs for their enthusiasm of assisting my study over the last two years Finally, I would not forget to send my deepest thank to my parents who always encourage me to keep up with my study objectives ii ABSTRACT 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 possible effect, theoretical framework of the so–called “three shifts” in the credit market is likely 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 in loan demand to optimize business activities) and (iii) productivity shift (a positive change in macroeconomic conditions) In addition, empirical evidence confirms that rapid credit growth of commercial banks could lead to a deterioration or improvement of credit quality The macroeconomic context for banking industry indicates that the decline in credit quality after a period of growth might be a reflection of (1) negative changes in the macroeconomic determinants which have a bad influence on the business activities of borrowers; and (2) information externality which makes banks hardly gain efficiency in evaluating their customers This study utilizes the data of 29 commercial banks operating in Dong Nai province for the period from 2009Q3 to 2014Q4 The econometric technique of Difference GMM for dynamic panel data model is adopted in order to examine the effect of credit growth on credit quality in the context of the commercial banks in Dong Nai This study finds empirical evidence to confirm that credit growth causes the 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 in Dong Nai might have lowered their credit standards to increase their lending to business customers and individuals; (ii) the conditions of local economy in Dong Nai might have not been really favorable for business activities during the research period; (iii) and the information externality in the loan market might have distorted the accuracy of customers evaluation in relation to their financial capacity iii Keywords: credit growth, credit quality, commercial bank, difference GMM, dynamic model, non–performing loan iv TABLE OF CONTENTS DECLARATION i ACKNOWLEDGEMENT ii ABSTRACT iii TABLE OF CONTENTS v LIST OF TABLES vii LIST OF FIGURES viii CHAPTER INTRODUCTION 1.1 PROBLEM STATEMENT 1.2 RESEARCH OBJECTIVE AND QUESTION 1.3 RESEARCH SCOPE AND METHODOLOGY 1.4 THESIS STRUCTURE CHAPTER LITERATURE REVIEW 2.1 THE MACROECONOMIC CONTEXT FOR BANKING 2.1.1 Main Characteristics of Banks 2.1.2 Shock and Vulnerability of Banking System 2.1.3 The Effect of Macroeconomic Determinants .6 2.1.4 Credit Growth and Vulnerability of Banking System 2.2 CREDIT GROWTH AND CREDIT QUALITY THROUGH DIFFERENT SHIFTS 2.3 CONTROL FACTORS FOR CREDIT QUALITY 17 2.4 PREVIOUS EMPIRICAL STUDIES 20 2.5 THE CONCEPTUAL FRAMEWORK 26 CHAPTER METHODOLOGY AND DATA 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 3.3.2 Econometric Problems 31 v 3.3.3 Estimating The Long–run Coefficients 33 3.3.4 Econometric Specification 35 3.3.5 Hypothesis testing .35 CHAPTER RESULT AND DISCUSSION 38 CHAPTER CONCLUSIONS 46 REFERENCES 49 vi LIST OF TABLES TABLE 2.1 Change in credit standard, credit growth and credit quality 16 TABLE 2.2 Summarization of the literature 25 TABLE 3.1 Necessary items and their account type 29 TABLE 3.2 The expected signs of variables used in the research 30 TABLE 3.3 The calculation of variables 30 TABLE 3.4 Hypotheses need testing 37 TABLE 4.1 Descriptive statistics 39 TABLE 4.2 Estimation result 41 vii LIST OF FIGURES FIGURE 2.1 The macroeconomic context for banking FIGURE 2.2 Supply shift 10 FIGURE 2.3 Demand shift 13 FIGURE 2.4 Productivity shift 15 FIGURE 2.5 Different shifts in the macroeconomic context for banking 16 FIGURE 2.6 The effect of credit growth on credit quality 26 FIGURE 4.1 Deposit growth rate and deposit interest rate 39 viii CHAPTER INTRODUCTION 1.1 PROBLEM STATEMENT Commercial bank is one of the most important financial intermediaries in the economy Their main functions are mobilizing and lending money to allocate financial resources for households, firms and other economic entities The typical problem of banks is that when borrowers could not use well the money they had borrowed, a credit risk arose One of the main causes is that banks lower their credit standards to attract more borrowers Although it might be a good opportunity to boost credit growth in the present time, banks will face a higher probability to deal with non–performing loans (NPLs) in the future However, if loans expansion is due 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 some circumstances 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 that commercial banks attempt to boost credit growth for profit objective whereas households and firms who borrow money from banks have to face with difficulties in business activities This leads to a consequence that credit growth may reduce credit quality The problem is whether the profit target of commercial banks by boosting 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 getting more and more important for not only commercial banks but also central bank and policy makers For commercial banks, it may help them to consider the appropriate time of loosening or tightening credit standards and the decision to expand or limit lending activities For central bank, it will help to control the loans growth of In addition, the “too–big –to–fail” hypothesis contributes to marginal effect of credit growth on credit quality The coefficient of credit growth 𝐂𝐑𝐄𝐃𝐈𝐓 with its interactive term 𝐂𝐑𝐄𝐃𝐈𝐓 × 𝐋𝐀𝐑𝐆𝐄𝐒𝐓, is interpreted as follows: 𝟒 𝟒 ∑ 𝛄𝐣 ∆𝐂𝐑𝐄𝐃𝐈𝐓𝐢𝐭−𝐣 + ∑ 𝛅𝐣 ∆(𝐂𝐑𝐄𝐃𝐈𝐓𝐢𝐭−𝐣 × 𝐋𝐀𝐑𝐆𝐄𝐒𝐓𝐢𝐭 ) 𝐣=𝟏 𝐣=𝟏 𝟒 = ∑(𝛄𝐣 + 𝛅𝐣 𝐋𝐀𝐑𝐆𝐄𝐒𝐓𝐢𝐭 )∆𝐂𝐑𝐄𝐃𝐈𝐓𝐢𝐭−𝐣 , ሺ10ሻ 𝐣=𝟏 If 𝐋𝐀𝐑𝐆𝐄𝐒𝐓 = 1, the coefficient of 𝐂𝐑𝐄𝐃𝐈𝐓 would be (𝛄𝐣 + 𝛅𝐣 ) and the variance is calculated as 𝐕𝐚𝐫(𝛄𝐣 + 𝛅𝐣 ) = 𝐕𝐚𝐫(𝛄𝐣 ) + 𝐕𝐚𝐫(𝛅𝐣 ) + 𝟐𝐂𝐨𝐯ሺ𝛄𝐣 , 𝛅𝐣 ሻ The long–run coefficient of 𝐂𝐑𝐄𝐃𝐈𝐓 corresponding to 𝛄𝐣 and 𝛅𝐣 is calculated as formula (7), the variance is in formula (8) In the case of long–run coefficient of (𝛄𝐣 + 𝛅𝐣 ), the formula would be calculated as: 𝛄𝐋𝐑 + 𝛅𝐋𝐑 𝟒 𝟒 𝐣=𝟏 𝐣=𝟏 𝟏 𝟏 =( × ∑ 𝛄𝐣 ) + ( × ∑ 𝛅𝐣 ) , ሺ11ሻ 𝟏−𝛂 𝟏−𝛂 The corresponding variance of this long–run marginal effect (11) would be: 𝐋𝐑 𝐕𝐚𝐫ሺ𝛄 +𝛅 𝐋𝐑 ሻ 𝐂𝐨𝐯(𝐂, ሺ𝟏 − 𝛂ሻ) 𝐕𝐚𝐫ሺ𝛂ሻ 𝐃𝟐 𝐂 [ ] , ሺ12ሻ = × − 𝟐 + ሺ 𝟏 − 𝛂ሻ 𝟐 ሺ 𝟏 − 𝛂ሻ 𝟐 𝐃𝟐 𝐂 ሺ 𝟏 − 𝛂ሻ 36 𝟒 𝟒 𝐃 = 𝐕𝐚𝐫 (∑ 𝛄𝐣 + ∑ 𝛅𝐣 ) 𝐣=𝟏 𝐣=𝟏 𝟒 𝟒 𝟒 𝟒 = ∑ 𝐕𝐚𝐫 (𝛄𝐣 ) + ∑ 𝐕𝐚𝐫 (𝛅𝐣 ) + 𝟐 ∑ ∑ 𝐂𝐨𝐯(𝛄𝐣 , 𝛄𝐥 ) 𝐣=𝟏 𝐣=𝟏 𝟒 𝐣=𝟏 𝐥=𝟏,𝐥≠𝐣 𝟒 𝟒 𝟒 + 𝟐 ∑ ∑ 𝐂𝐨𝐯(𝛄𝐣 , 𝛅𝐥 ) + 𝟐 ∑ ∑ 𝐂𝐨𝐯(𝛅𝐣 , 𝛅𝐥 ) 𝐣=𝟏 𝐥=𝟏 𝟒 𝐣=𝟏 𝐥=𝟏,𝐥≠𝐣 𝟒 𝐂 = ∑ 𝛄𝐣 + ∑ 𝛅𝐣 𝐣=𝟏 𝐣=𝟏 Table 3.4 presents four null hypotheses which need testing as follow: TABLE 3.4 Hypotheses need testing Hypothesis Meaning H01 : γj = Credit growth does not have any effect on credit quality at lag j H02 : γj + δj = Credit growth does not have any effect on credit quality at lag j for the three largest banks H03 : γLR = Credit growth does not have any long–run effect on credit quality H04 : γLR + δLR = Credit growth does not have any long–run effect on credit quality for the three largest banks 37 CHAPTER RESULT AND DISCUSSION This chapter presents the estimation results and discusses the main findings According to the estimation result, credit growth has negative effects on credit quality at the lag level of and and the long–run coefficient of credit growth is also statistically significant This proves for the existence of supply shift and the information externality in the credit market, and the negative impact from the macroeconomic conditions Figure 4.1 is to compare the growth rate of deposit liabilities of Dong Nai banking system (the growth rate data only capture commercial banks’ branches which were in Dong Nai from 2010Q1 to 2014 Q4) with the quarterly deposit interest rate in Viet Nam As can be seen from the figure, the condition stated in the literature is hold when the deposit growth rate is higher than deposit interest rate It implies the inexistence of potential shock or vulnerability of Dong Nai banking system Therefore, credit growth has not been a great concern for commercial banks during this period However, this figure only brings some evidences to confirm that net resource might be transferred in the direction from depositors to banking system If credit growth does not perform well, it would become a large pressure for commercial banks to boost their mobilizing activities in the future, which could disrupt the condition of deposit growth rate and interest rate and reverse the direction of net resource 38 FIGURE 4.1 Deposit growth rate and deposit interest rate 11 -1 Deposit growth rate Deposit interest rate Source: The State Bank of Vietnam, Dong Nai branch International Monetary Fund eLibrary Data Table 4.1 below presents a summary of descriptive statistics for the sample used in this empirical study TABLE 4.1 Variable Descriptive statistics Observation Mean Standard Deviation Minimum Maximum NPL_SL 551 0.0348 0.0394 0.0000 0.1977 CREDIT 551 0.0414 0.1570 –0.6192 0.5467 COST_EFF 551 0.9050 0.1627 0.4036 1.4639 ROA 551 0.0089 0.0133 –0.0312 0.0534 SIZE 551 0.0345 0.0443 0.0007 0.2641 LEVERAGE 551 0.8120 0.1655 0.3893 0.9938 LARGEST 551 0.1034 0.3048 0.0000 1.0000 ∆NPL_SL 522 0.0010 0.0225 –0.1577 0.1530 ∆CREDIT 522 –0.0004 0.2148 –0.9063 0.9854 Number of Banks 29 Number of Quarters 19 39 Table 4.2 presents the estimation result of one–step Difference GMM method with robust standard errors First, test for second order autocorrelation AR(2) shows that 𝛆𝐢𝐭−𝟏 does not correlate with 𝛆𝐢𝐭−𝟐 due to an existence of the high p–value obtained from the analysis Test for first order correlation AR(1) is as expected with small p–value (lower than 0.05) Therefore, the tests for instrument variables are valid Second, both Sargan Test and Hansen Test provide very high p–value, which proves that the null hypothesis of valid instrument variables might not be rejected These tests indicate that the problem of endogeneity would be solved and the coefficients in the model could be used for further analysis According to the estimation results, the coefficients of lagged independent variable (∆𝐍𝐏𝐋_𝐒𝐋𝐢𝐭−𝟏 ) is positive and significant, so credit quality in the present might associate with itself in the past It means higher (lower) credit quality might lead to higher (lower) credit quality after one quarter of a particular year The lagged variables of credit growth are also positive and significant at lag level of and It means the decision of expanding lending activities of commercial banks in Dong Nai has negative influence on credit quality after three quarters to one year As stated in the Circular No 02/2013/TT–NHNN, customer loans will be recorded as NPLs if they are overdue from 91 days and above It is about two–quarter shorter than the two significant lag levels of credit quality Therefore, new loans of commercial banks in Dong Nai might have very high chance of becoming NPLs after two quarters from the overdue threshold In the shorter period of one and two quarters, the effect of credit growth on credit quality is not statistically significant 40 TABLE 4.2 Estimation result Dependent variable Baseline model Model (1) Model (2) Model (3) Model (4) Full model ∆NPL_SLit−1 0.4825** 0.5697*** 0.4030 0.4969** 0.5606*** 0.5119*** (0.2117) (0.1967) (0.2766) (0.2131) (0.1453) (0.1462) 0.0092 0.0071 0.0051 0.0096 0.0055 0.0113 (0.0098) (0.0120) (0.0112) (0.0097) (0.0107) (0.0085) 0.0142 0.0061 0.0087 0.0141 0.0158* 0.0174* (0.0096) (0.0142) (0.0107) (0.0096) (0.0087) (0.0096) 0.0247*** 0.0149 0.0184* 0.0242*** 0.0261** 0.0248*** (0.0086) (0.0101) (0.0091) (0.0085) (0.0097) (0.0091) 0.0416*** 0.0268 0.0354*** 0.0401*** 0.0444*** 0.0368*** (0.0071) (0.0198) (0.0098) (0.0074) (0.0064) (0.0134) 0.0172 0.0363 0.0269 0.0235 0.0126 0.0652 (0.0279) (0.0404) (0.0385) (0.0306) (0.0362) (0.0551) 0.0941 0.1057 0.1011 0.1023 0.0953 0.1953* (0.0691) (0.0741) (0.0649) (0.0704) (0.0689) (0.1079) 0.0566 0.0587 0.0607 0.0571 0.0555 0.0629 (0.0535) (0.0514) (0.0675) (0.0656) (0.0459) (0.1965) 0.0193 0.0394 0.0251 0.0208 0.0154 –0.0254 (0.0483) (0.0692) (0.0737) (0.0602) (0.0404) (0.2058) Credit growth ∆CREDITit−1 ∆CREDITit−2 ∆CREDITit−3 ∆CREDITit−4 ∆ሺCREDITit−1 × LARGESTit ሻ ∆ሺCREDITit−2 × LARGESTit ሻ ∆ሺCREDITit−3 × LARGESTit ሻ ∆ሺCREDITit−4 × LARGESTit ሻ 41 Cost efficiency ∆COST_EFFit−1 ∆COST_EFFit−2 ∆COST_EFFit−3 ∆COST_EFFit−4 –0.0086 0.0823 (0.0205) (0.0736) –0.0465* 0.0709 (0.0269) (0.0779) –0.0189 0.1105 (0.0426) (0.0879) –0.0538 0.1672 (0.1161) (0.1319) Profitability ∆ROAit−1 ∆ROAit−2 ∆ROAit−3 ∆ROAit−4 0.2330 1.0349 (0.2272) (0.7915) 0.3698 1.0998 (0.2211) (0.8471) 0.2590 1.1002 (0.3152) (0.9268) 0.3494 1.2169 (0.5254) (1.0555) Bank size ∆SIZE 42 0.2782 2.4969 (0.8120) (3.1992) Leverage ∆LEVERAGEit−1 ∆LEVERAGEit−2 ∆LEVERAGEit−3 ∆LEVERAGEit−4 –0.0375* –0.0554 (0.0190) (0.0370) 0.0404 0.0364 (0.0260) (0.0260) 0.0248 0.0279 (0.0250) (0.0257) 0.0257 –0.0038 (0.0484) (0.0659) Long–run coeficient ∆CREDIT ∆ሺCREDIT × LARGESTሻ 0.1733* 0.1274 0.1133 0.1750* 0.2091** 0.1851** (0.0906) (0.1623) (0.0916) (0.0908) (0.0817) (0.0720) 0.3617 0.5580 0.3582 0.4049 0.4067 0.6104 (0.3971) (0.5411) (0.3960) (0.4687) (0.4402) 1.0702 Testing for autocorrelation and validity of instruments AR(1) 0.041 0.017 0.155 0.039 0.005 0.027 AR(2) 0.279 0.381 0.338 0.286 0.280 0.577 Sargan Test 0.574 0.499 0.780 0.580 0.659 0.641 Hansen Test 0.537 0.338 0.763 0.502 0.798 0.998 22 26 26 23 26 35 Number of instruments Standard errors are presented in parentheses Symbol *, ** and *** represent for significant level at 10%, 5% and 1% 43 As presented on Table 4.2, all the coefficients of credit quality variables interacting with dummy ∆(𝐂𝐑𝐄𝐃𝐈𝐓𝐢𝐭−𝐣 × 𝐋𝐀𝐑𝐆𝐄𝐒𝐓𝐢𝐭 ) are not statistically significant It means the effect of credit growth on credit quality of three largest commercial banks is not different from the others Moreover, almost bank characteristics which are used as control variables in this research are not statistically significant in explaining the change of credit quality ∆𝐋𝐄𝐕𝐄𝐑𝐀𝐆𝐄𝐢𝐭−𝟏 is significant at 10% but not as expected Besides, despite the insignificant coefficients (except for ∆𝐂𝐎𝐒𝐓_𝐄𝐅𝐅𝐢𝐭−𝟐), the signs of cost efficiency and profitability are all positive at all lag levels It partially reflects the potential existence of “Skimping” and “Pro–cyclical credit policy” hypothesis in Dong Nai banking system In the long–run, the coefficient of ∆𝐂𝐑𝐄𝐃𝐈𝐓 is positive and significant It indicates that credit growth of commercial banks in Dong Nai may gradually reduce credit quality The total effect of credit growth is much larger than individual effect from lag levels Similar to the individual lags of credit growth interacting with dummy, the long–run coefficient of ∆ሺ𝐂𝐑𝐄𝐃𝐈𝐓 × 𝐋𝐀𝐑𝐆𝐄𝐒𝐓ሻ is not significant, which prove that the influence of credit growth on credit quality is the same among largest banks and the others The estimation results have presented the negative relationship between credit growth and credit quality of commercial banks in Dong Nai, which is represented in both short–run and long–run timeframe This relationship reflects two main implications First, based on the literature, there may be a supply shift in the loan market It appears that commercial banks in Dong Nai are willing to lower their credit standards in order to boost lending activities This action from the commercial banks results in the high chance of lending money to borrowers who have bad financial capacity Second, there would be the existence of adverse effects from the macroeconomic conditions which result in disadvantages for business activities of the economic entities, especially households and small and medium 44 enterprises They would hardly gain positive cash flows to finance their liabilities Thus, overdue loans increase Third, there would be the existence of information externality in the loan market Commercial banks in Dong Nai may mis–evaluate their new customers Firms, households and other borrowers may be able to approach many sources of loans, so they can use borrowed money from somewhere to pay the debts in somewhere else The effect of credit growth on credit quality is not adjusted by the characteristic of large bank size This proves for the inexistence of “too–big–to– fail” hypothesis in the case of commercial banks in Dong Nai Although some banks have much larger size than others, they seem to have partially similar credit growth policy In the aspect of this hypothesis, it means that large banks not have over–confidence psychology (which makes bank believes in the protection of The State Bank of Vietnam) in lending activities In more details, banks whether large or small size have the same change in credit quality corresponding with their change in credit growth This is contrary to the expectation that larger banks often suffer from greater change in credit quality than the smaller banks Vietnam banks in general and Dong Nai banks in particular play a very important role in the financial market If one of them collapses, it would influence other banks due to the problems of cross–deposit among banks and negative psychology of people, and the whole Vietnam banking system would be in trouble Therefore, it may not reasonable to conclude that commercial banks not believe in the intervention from The State Bank to prevent them from failure In the scope, methodology and available data of this research, there is temporarily no evidence to confirm this hypothesis 45 CHAPTER CONCLUSIONS The relationship between credit growth and credit quality could be explained through three shifts in the loan market First, the supply shift represents for the fact that banks expand their lending activities by lowering their credit standards In these circumstances, banks will face a high probability of lending money to bad borrowers Second, the demand shift originates from the demand of the capital restructuring as well as taking advantage of the lower cost capital from banks than somewhere else in capital market Banks would tighten their credit standards for this shift, which reduces credit risks in the future Third, the productivity shift in favorable conditions of the economy reflects the efficient business activities of households, firms and other economic entities who had borrowed money from banks The ability of financing the debts would be improved In addition, the negative nexus between credit growth and credit quality is also explained by adverse macroeconomic shocks on borrowers and the information problems, in which banks misevaluate new customers and as a consequence, the banks not take any care of negative effects of their loans on other banks’ information In accordance with the empirical results obtained from this study on the sample of 29 commercial banks in Dong Nai from 2009Q3 to 2014Q1, credit growth has negative effect on credit quality in both short run and long run This partially proves for the existence of the supply shift in the loan market in the context of Dong Nai Commercial banks in Dong Nai appear to lower their credit quality in this researched period to boost lending In addition, the local economy of Dong Nai is not a good environment for business activities Moreover, the credit externality may exist in the loan market These findings may derive some policy implications which can be presented as follows: 46 For the Commercial banks It is not the right time for boosting credit growth rapidly in Dong Nai The commercial banks have to be more cautious in appraising investment projects, evaluating collaterals and supervising their customers’ capital usage especially new customers in order to avoid bad borrowers Moreover, commercial banks should focus greatly on projects on advantageous fields in Dong Nai such as agriculture, production of exports, labor intensive and high technology industries In the relationship with customers, they should deploy and maintain solutions for their customers to overcome difficulties in business activities; and classify borrowers depending on their state of credit relationship to apply appropriate incentive policies In addition, commercial banks need to control new NPLs, continue to handle old NPLs and prepare sufficient provision for credit risk Finally, commercial banks should keep track of the state of the local economy to implement credit policies timely and efficiently For The State Bank of Vietnam, Dong Nai branch (SBV–DN) The State Bank of Vietnam, Dong Nai branch (SBV–DN) needs to supervise more closely on credit growth of commercial banks, especially commercial banks with high NPL ratio Most importantly, the SBV–DN should classify commercial banks into groups based on their lending activities to target the appropriate rate of credit growth, avoiding high credit risk of weak commercial banks In addition, the SBV–DN has to catch up with the conditions of local economy as well as the real state of commercial banks’ activities in order to enhance the efficiency of administration and operation Finally, they should co–ordinate with the local governments at various levels to improve local economy, create good business environment for commercial banks, households and enterprises 47 For local government at various levels Local governments should ease the legal environment for commercial banks, households and enterprises First, they need to facilitate households and enterprise in the proceedings of business registration Second, they could co–ordinate with relevant departments to support for enterprises’ outputs and stabilizing the markets Last but not least, they could direct Executive Office to speed up process of recovering NPLs This would help commercial banks in injecting capital into the market continuously, which stimulates the local economy In addition to the main empirical findings, this research still has some limitations First, the research scope is only in Dong Nai province Hence, the result will not reflect the convincing outcomes for the whole banking system in Vietnam Second, the period of available data is not sufficient to capture the lags of very long time There are many former researches which indicated that the effect of credit growth on credit quality may be even three years Further studies may use data of yearly financial statements and the scope for commercial banks in Vietnam and other countries in the world 48 REFERENCES Arellano, M., & Bond, S (1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations The review of economic studies, 58(2), 277-297 Arellano, M., & Bover, O (1995) Another look at the instrumental variable estimation of error-components models Journal of econometrics, 68(1), 29-51 Berger, A N., & DeYoung, R (1997) Problem loans and cost efficiency in commercial banks Journal of Banking & Finance, 21(6), 849-870 Blundell, R., & Bond, S (2000) GMM estimation with persistent panel data: an application to production functions Econometric reviews, 19(3), 321-340 Caporale, G M., Di Colli, S., & Lopez, J S (2014) Bank lending procyclicality and credit quality during financial crises Economic Modelling, 43, 142-157 Valverde, S C., Ibanez, D M., & Fernandez, F R (2011) Securitization, bank lending and credit quality: the case of Spain Clair, R T (1992) Loan growth and loan quality: some preliminary evidence from Texas banks Economic Review, Federal Reserve Bank of Dallas, Third Quarter, 922 Kunt, A D., & Detragiache, E (1998) The Determinants of Banking Crises in Developing and Developed Countries Staff Papers - International Monetary Fund, 45(1), 81-109 Foos, D., Norden, L., & Weber, M (2010) Loan growth and riskiness of banks.Journal of Banking & Finance, 34(12), 2929-2940 Gavin, M., & Hausmann, R (1996) The roots of banking crises: The macroeconomic context Washington, D.C.: Inter-American Development Bank, Office of the Chief Economist Jiménez, G., Lopez, J A., & Saurina Salas, J (2007, September) How Does Competition Impact Bank Risk Taking? In EFA 2007 Ljubljana Meetings Paper Judson, R A., & Owen, A L (1999) Estimating dynamic panel data models: a guide for macroeconomists Economics letters, 65(1), 9-15 49 Keeton, W R (1999) Does faster loan growth lead to higher loan losses?.Economic Review-Federal Reserve Bank Of Kansas City, 84, 57-76 Louzis, D P., Vouldis, A T., & Metaxas, V L (2011) Macroeconomic and bankspecific determinants of non-performing loans in Greece: A comparative study of mortgage, business and consumer loan portfolios Journal of Banking & Finance, 36(4), 1012-1027 Nguyen, L (2015) Government ownership and bank stability: The impact of governance, political influence and control of corruption in developed and developing countries (Doctoral dissertation) Monash University, Australia Rajan, R G (1994) Why bank credit policies fluctuate: A theory and some evidence The Quarterly Journal of Economics, 399-441 Roodman, D (2006) How to xtabond2: An introduction to difference and system GMM in Stata Center for Global Development working paper, (103) Salas, V., & Saurina, J (2002) Credit risk in two institutional regimes: Spanish commercial and savings banks Journal of Financial Services Research, 22(3), 203-224 State Bank of Vietnam, Dong Nai branch (2015) Bao cao hoat dong ngan hang thang 5/2015 [Banking report 5/2015] Stern, G H., & Feldman, R J (2004) Too big to fail: The hazards of bank bailouts Brookings Institution Press Stuart, A., Ord, K (1998) Kendall’s Advanced Theory of Statistics Arnold, London 50