Determinants of Customer’s Credit Risks in Commercial Banks (Case Study: Tejarat Bank of Neka, Mazandaran Province)

Document Type : پژوهشی

Authors

• Zahra Karimi Moughari
• hossein asadi gorji
• Mohammad Taghi Gilak Hakimabadi

Mazandaran University

https://doi.org/10.22067/pm.v22i10.22570

Abstract

Lending is a principal business activity for most commercial banks. The loan portfolio is typically the largest asset and the predominant source of revenue. As a result, it is one of the greatest sources of risk to a bank’s safety and soundness.
Credit risk is the risk of loss due to a debtor's non-payment of a loan or other line of credit (either the principal or interest (coupon) or both). Defaulting occurs when a debtor has not fulfilled his or her legal obligations according to the debt contract, or has violated a loan covenant (condition) of the debt contract, which might occur with all debt obligations including bonds, mortgages, loans, and promissory notes. Since financial innovation and derivatives grow rapidly in competitive financial industry, credit risk measurement and management become essentially important.
Credit risk is the primary financial risk in the banking system and exists in virtually all income-producing activities. How a bank selects and manages its credit risk is critically important to its performance over time; indeed, capital depletion through loan losses has been the proximate cause of most institution failures. Identifying and rating credit risk is the essential first step in managing it effectively.
Well-managed credit risk rating systems promote bank safety and soundness by facilitating informed decision making. Rating systems measure credit risk and differentiate individual credits and groups of credits by the risk they pose. This allows bank management and examiners to monitor changes and trends in risk levels. The process also allows bank management to manage risk to optimize returns.
The consistent use of analytic credit risk has many advantages. It improves an institution’s risk assessment time, speed, accuracy, consistency, bad debt reduction and prioritization of collections. Using analytic credit scoring, an institution can review its entire receivable portfolios in the same time as it would take to review just one account by traditional methods. Analytic credit risk assures accuracy since the review process is mostly free of human error. It offers consistency, by using the same set of rules and weighted variables over the entire portfolio. Scoring permits regular reviews of the entire account base, thereby, quickly and efficiently identifying accounts that require immediate attention, and isolating customers who warrant human intervention. The net effect is a substantial reduction in risk assessment time and a more systematic approach to collection.This paper examines the factors affecting the credit risk of real customers of Tejarat bank of Neka.The data that is used in this paper was extracted from 2545 loan files of real customers of Tejarat Bank of Neka who had got credit facilities during the years 1381 to 1390, and logistic regression was used to evaluate the data.
In this study, first we introduced the factors affecting credit risk of real customers of Tejarat bank and then defined the risk and presented the methods that can measure the risk. Then we extracted the data and used the Eviews software to estimate our model and finally analyzed the results.

Methodology
The statistical techniques used in this research are a logit method to estimate the probability functions. Logit model is one of the easiest statistical modelsthat is based on the analysis logit model (logistic), the financial ratios and other quantitative and qualitative variables for predicting the risk of non-repayment of loans are used. In this model, the probability of failure is showed in normal distribution:
This function is ideal for all levels of Z values between zero and one.
Variables include the dependent variable (including 2 cases: will the real customers pay or not pay their loan) and independent variables (including 6 variables that have an impact on the repayment).One of these independent variables is the period of repayment of loans or facilities and another one is the amount of loans that customers have got and also the rate of interest of loans and also collateral types that is got for thoes loans and two final variabels are mandatory or non-mandatory loans variable and loan types variable.We can see all these results in table below:

Results and Discussion
The result of this study shows that loans repayment period, interest rates of loans, and collateral types have significant effect on the probability of default; but mandatory or non-mandatory loans and loan amount do not have significant effect on the probability of default. Despite the significance of the coefficients, according to the sign of the coefficients, as expected, we found that the probability of default is reduced by increasing the loan repayment period and with a decrease in interest rates getting banking deposits as collateral for loans have the greatest negative impact on loan default regarding the loan types, Gharz-ol- Hasaneh and Mosharekat loans have the highest and lowest impacts on default possibility.

Keywords

• Credit Risk
• Validation
• Logit regression
• default

References

[1] Ayati Gazar, H. (2005). Comparison between regression and classification trees logit model and in the process of credit scoring for individual customers of a bank, Master thesis, University of Sharif,.(in Persian)

[2] Jamshidi, S. (2004). Methods of credit scoring of customers, Monetary and Banking Research Institute, Central Bank of the Islamic Republic of Iran, 25-37.(in Persian)

[3] Zakavat, S, M. (2002). Customer credit risk models Saderat Bank of Iran, Master Thesis, Institute of Banking..(in Persian)

[4] Fallah Shamsi, (2005). The model design and the credit risk in the banking system, Proceedings of the Sixteenth Conference on Islamic Banking, Islamic Banking Institute, 421-445,.(in Persian)

[5] Mansouri. (2003). Define and design a mathematical model of bank credits, the approach of classical models and neural networks, PhD thesis, University of Tarbiat Modadess.(in Persian)

[6] Kimiagari, M, A., Amini, M, J., Tabatabaei Arati, H., Hosseini, L. (2012). Thecredit risk of customers' repayment of credit facilities (the scope of the Branch of Bank Melli Iran). the Ninth International Conference on Industrial Engineering.(in Persian)

[7] Mousavi.Ebtehaj. (2007). Factors affecting the credit ratings of customers,Master's Thesis, Institute of Banking..(in Persian)

[8] Molla Ebrahimlo, M, H. (2005). Developing a credit scoring model for legal customers bank, Master thesis, University of Sharif,.(in Persian

[9] Jamshidi. (2006). Credit scoring of customers methods. Monetary and Banking Research Institute,.(in Persian)

[10] Hedayati.Safari.Kalhor.Bahmani. (2005). Domestic banking operations (Customize sources). Banking Institute, Central Bank of the Islamic Republic of Iran, 100-250.(in Persian)

[11] Akhbari, M. (2008). Banks Legal customer credit ratings with artificial intelligence approach, Master thesis, Isfahan University of Technology,.(in Persian)

[12] Rostamkolaei, S A. (2007).Fuzzydecisionmodelto evaluatea bank loanin the management ofelectronic banking, Master thesis, University of Shahid Beheshti,.(in Persian)

[13] Sabzevari, H. (2005). Credit scoring models to estimate and compare the logit and AHP (Case study: Legal customers

[14] Parsian Bank, Master thesis, University of Sharif, (in Persian)

[15] Jacobson T., Roszbach, K. (1998). planned one statistic model for measurement of loan portofilio risk and showed how this model can help to assessment of facilities payment politics.

[16] Akmal, M., Muhammad S. (2008). Technical Efficiency of the Banking Sector in Pakistan, SBP Research Bulletin, Vol.4, No. 1, Nov.

[17] Castro, V. (2013) Macroeconomic determinants of the credit risk in the banking system: The case of the GIPSI, Economic Modeling, 31, 672-683.

[18] Da, S,Marcos, S., Divino, Jose, A. (2013). The role of banking regulation in