5. Contribution of the research
Scientifically:
The research results supplement the theoretical basis of factors affecting the provisioning of credit risk provisions at Vietnamese commercial banks.
Maybe you are interested!
-
Credit risk management at Vietnam International Commercial Joint Stock Bank - Hanoi Branch - 12 -
Credit risk management for personal loans at Nam A Commercial Joint Stock Bank - Quang Ninh Branch - 13 -
The Role of Risk Management in Credit Activities at Commercial Banks -
Credit risk management at Military Commercial Joint Stock Bank, Hue Branch - 12 -
Current Status of Credit Risk Management at Vietnam Bank for Agriculture and Rural Development
In practice:
The research results are the scientific basis for accountants at commercial banks to record and present financial statements on credit risk provisions in the most reasonable, complete and timely manner to create the premise for making management decisions of bank managers. Besides, it is also the basis for internal control of banks, auditors and investors can check and evaluate the accuracy and reasonableness of credit risk provisions presented in the bank's financial statements.
For researchers:
The completion of the study will help the researcher improve his/her scientific research ability along with improving his/her knowledge and experience in the field of credit risk accounting at commercial banks.
6. Topic structure
The content of the topic is divided into 5 chapters: Chapter I: Research overview Chapter II: Theoretical basis
Chapter III: Research design Chapter IV: Research results Chapter V: Conclusion and recommendations
CHAPTER I
RESEARCH OVERVIEW
1.1 Access to world research
In this section, the researcher presents empirical studies and research results that have been conducted around the world.
1.1.1 Research by Larry D. Wall and Ifterkhar Hasan (2003)
The paper analyzes the determinants of credit risk provisioning of banks, with the research sample selected from US banks and non-US banks including Canada, Japan and a group of 21 countries. The author uses a fixed effects model to identify and measure the impact of factors on the level of credit risk provisioning. The calculation equation used by the author is as follows:
LLA it = 1 + 2 NPL it + 3 NCO it + 4 LOAN it + 1 ER i,t-1 + 2 RETN it + 1 Y it + u it In which:
NPL it : ratio of bad debt to total assets at time t
NCO it : ratio of net loss value for the whole year t to total assets at time t LOAN it : ratio of total debt to total assets at time t
ER i,t-1 : equity to asset ratio for bank i at the end of the previous year
RETN it : ratio of pre-tax and reserve income to total assets of bank i at time t
Y it : year fixed effect variable set to 1 if observations start in year t and 0 otherwise.
This model is applied by the author to research US banks and non-US banks. The results show that all factors are determined to have an impact on the level of credit risk provisioning.
However, there are differences in statistical significance between the two regions. Specifically, US banks have lower statistical significance than other banks in terms of LLA, NPL and NCO. Similarly, the author continues to compare the US with Japan and Canada to show the different levels of impact of factors on provisioning levels in different countries. Through this, the study also shows that some variables that are thought to have a strong impact on banks in this country are not important factors in banks in other countries.
1.1.2 Research by Grace T. Chen and colleagues (2005)
The study uses data from 200 banks over a five-year period from 1995 to 1999. The study identifies and examines the relationship between the factors and credit risk provisioning. The results demonstrate the usefulness for users of bank financial statements in assessing bank performance more fully.
The research of the group of authors has identified factors affecting the provisioning of credit risk. The group of authors has proposed a hypothesis including factors affecting the change in the level of provisioning for risk as follows: Bank size, bad debts, lending interest rates, the ratio of non-real estate loans to real estate, the rate of recovery of written-off debts, and net debt losses during the period. The relationship between the independent variables and the dependent variable is tested by the author in two ways: using charts for presentation and using regression analysis. The model used by the author for the research is as follows:
ALL it = A 0t + A 1t NPL it + A 2t NRE/RE it + A 3t INT it + A 4t CHAOFF it + A 5t SIZE it
+A 6t RECOVE it + e t In which:
ALL: is the dependent variable, measured by the ratio of credit risk provisions to total debt NP: is the ratio of bad debt to total debt
INT: is the ratio of interest income to average total outstanding loans
NRE/RE: is the ratio of non-real estate loans to real estate RECOV: is the total recovery from written-off debt over the previous year's total debt
CHAOFF: three-year average capital loss ratio over three-year average total debt SIZE: log of total assets
The research results show that all research factors have an impact on the credit risk provisioning ratio, except for the bank size factor, which is not statistically significant. Banks with a high bad debt ratio tend to increase the risk provisioning level. The riskier the loans, the higher the lending interest rate, leading to a positive correlation between the lending interest rate and the risk provisioning level. Real estate loans are safer than non-real estate loans, so the ratio of non-real estate loans to real estate increases, the provisioning level will increase. The debt recovery rate has a positive impact on the provisioning level, when this ratio increases, banks tend to increase the provisioning level. Experience with loan losses in banks also has a positive correlation with risk provisions, the more experience a bank has with past losses, the higher the provisioning level will be than banks with less experience.
1.1.3 Research by Asokan Anvàarajan et al. (2005)
By collecting full information, annual financial reports of Australian commercial banks in the period from 1991 to 2001, with a number of 50 commercial banks, including 10 listed banks and 40 unlisted banks. The purpose of the study is to examine whether Australian banks use credit risk provisioning costs in managing earnings, capital and signaling or not, and if so, what is the level of provisioning costs used for this purpose. The author uses OLS regression to analyze the following model: LLPR = a 0 + a 1 ΔLLA + a 2 ΔGDP + a 3 MCAP + a 4 EBT + a 5 LISTED + a 6 POST + a 7 TA + a 8 CFEER + a 9 LISTED * MCAP + a 10 LISTED * EBT + a 11 MCAP * POST
+ a 12 EBT*POST + a 13 LISTED * MCAP * POST + a 14 LISTED * EBT * POST
In there:
LLPR: credit risk provision cost on average outstanding debt ΔLLA = Loan loss spread/ total assets
ΔGDP = GDP gap, representing the change in economic growth MCAP = equity capital ratio before credit loss provisions minimum required capital EBT = earnings before taxes and provisions/average total assets
LISTED = Dummy variable (1 if listed commercial bank; 0 if unlisted commercial bank)
POST = Dummy variable (1 for post-Basel years 1996-2001, and 0 for pre-Basel years 1991-1995)
TA = log (total assets)
CFEER = Fee and commission income on total assets
LISTED *MCAP = interaction between commercial bank type and capital adequacy ratio LISTED *EBT = interaction between commercial bank type and income
MCAP * POST = interaction between capital adequacy ratio and dummy variable POST EBT * POST = interaction between income and dummy variable POST
LISTED*MCAP*POST = Interaction between commercial bank type and capital adequacy ratio and dummy variable POST
LISTED*EBT * POST = Interaction between commercial bank type and income and dummy variable POST
The results of the study show that Australian banks use risk provisions to manage earnings, with listed commercial banks doing more than unlisted commercial banks. In particular, earnings management is more evident after the implementation of the Basel Accord. However, the study does not find a relationship between risk provisions and capital management. In addition, the study finds that credit risk provisioning expenses are not considered as important in the post-Basel period compared to the pre-Basel period. This suggests that the income statements may not reflect the true economic reality underlying the figures.
1.1.4 Research by Ruey-Dang Chang et al. (2008)
With data source taken from banks listed on the Taiwan stock market in the period from 1996 to 2004 with 164 observations. The authors investigated the relationship between credit risk provisions and 6 indicators of bank performance in the period from 1996 to 2004 under the control of bank type, ownership status and asset size. In addition, the study also investigated whether bank managers intend to use credit risk provisions as a tool to manage earnings. The authors performed regressions on the variables of income before tax and provisions, next year's earnings, bad debt, bad debt ratio, bad debt provision ratio and tested the hypotheses through the following model:
DLLP t =γ 0 + γ 1 GOVERN t + γ 2 CATA t + γ 3 lnASSET +γ 4 BP_EARN t + γ 5 EARN t+1
+ γ 6 BIS t + γ 7 R_NPL t + γ 8 NPL t + γ 9 R_COVER t + ε t In which:
DLLPt : credit risk provision cost in year t
GOVERNt : is a dummy variable (equal to 1 when it is a state bank, otherwise it is 0) CATAt : is a dummy variable (equal to 1 when it is a commercial bank, otherwise it is 0) lnASSETt : log (total assets)
BP_EARNt : pre-provision income
EARNt+1 : next year's income
BISt : capital ratio (bank capital is at least 8% of credit risk)
NPLt : total bad debt
R_NPLt : bad debt ratio
R_COVERt : bad debt provision ratio (eg: provision expense over total bad debt)
t : year
The research results show that BP_EARNt, EARNt+1, lnASSETt, NPLt are variables that are correlated with DLLP at a significance level of 1%. In particular, BP_EARNt is positively correlated with DLLP, showing that bank managers tend to increase credit risk provisioning costs to clear bad debts as well as reduce bad debt ratios. Income
The following year also has a positive correlation, indicating that managers use credit risk provisioning costs to intentionally increase the following year's income. On the contrary, lnASSETt shows a negative correlation, meaning that the larger the bank size, the smaller the credit risk provisioning costs. The variable NPLt has a positive correlation with credit risk provisioning, when bad debts increase, managers will increase credit risk provisioning costs. However, the bad debt ratio is not statistically significant in the model, and the capital adequacy ratio is also rejected to affect credit risk provisioning. The results also found no evidence of the influence of GOVERNt CATAt on the level of credit risk provisioning in banks.
1.1.5 Research by Mahmuod O. Ashour et al. (2011)
To conduct this study, the authors collected secondary data from the financial statements of 7 Palestinian banks for 5 years from 2006 to 2010 with 35 observations. The purpose of the study is to examine whether Palestinian bank managers engage in lending and risk provisioning decisions to beautify the income statement or capital management. To test the research hypothesis on the factors affecting banks' credit risk provisioning, earnings smoothing ability, debt-to-equity ratio and legal reserves, the authors used the multi-stage regression model of Zoubi & Al-Khazali (2007) after being modified. The model has the following form:
LLP = CROA + LD + DE + RD + LOGTA + CAR + TYPE
In there:
LLP: provision level on total debt
CROA: earnings before tax and reserves on total assets LD: ratio of loans and investments on customer deposits DE: ratio of total outstanding loans to equity
RD: difference between current bank reserves minus required reserves / equity
LOGTA: log (total assets)
CAR: bank's capital adequacy ratio minus minimum capital adequacy ratio TYPE: Dummy variable (1 if Islamic bank and 0 otherwise)
The results show that there is no evidence that bank managers use loan loss provisions to smooth their income statements. The results also reject the hypothesis that the higher the total loan to equity ratio, the lower the level of loan loss provisions. However, the results show that managers reduce loan loss provisions when there is a requirement to increase legal reserves above the bank's current reserves. Similarly, the ratio of loans and investments to customer deposits is negatively correlated with loan loss provisions, because the higher the ratio, the more external capital the bank needs. Therefore, banks have an incentive to reduce loan loss provisions to reduce the perceived risk of customers in order to attract external capital. In addition, there is no evidence to demonstrate the difference between Islamic banks and conventional banks.
1.1.6 Research by Mohd Yaziz Bin Mohd Isa (2011)
The study uses data from 12 banks collected over a period of 14 years, starting from 1996 to 2009. The author found that there was no literature that addressed the reasons why Malaysian commercial banks failed to make adequate and reasonable provisions for their credit risks during the economic crisis that began in July 1997. Therefore, the author conducted a study to investigate the factors that influenced the failure to make adequate and reasonable provisions for credit risks.
The study uses the fixed effect model as follows:
LLP it = B 0 + B 1 NPL it + B 2 RC it + B 3 II it + B 4 NP it + B 5 LA it + B 6 GDP it + e it
In there:
LLP: Credit Loss Provision Expense