Introduction

This paper will focus on issues and problems that is related with credit risk models and VAR models in banking sectors. The problems of both the models has been discussed in this paper. The last part of this paper has provided certain recommendations for the banks to avoid such problem related to the risk management models. The main objective of this report is to eliminate the issues associated with risk assessment models in banks.  

Discussion

Part A

Risks in Banking

Banking sectors are exposed to variety of risks that needs to be controlled by following the government regulations. This is a very important matter because it can lead to failure in banks and may impact the privacy of millions of people. It also affects the investor’s decisions and will be not interested to invest in the banks and hence, can cause lower profits for the banks (Alodayni, 2016).

Credit Risk Model

Credit Risk Model is very important in solving the various problems related to credit risks. Credit risk is the risk of borrower not paying back the loan, credit card and some other type of loan (Doss, 2017). Credit Risk model calculates the chances of defaults of borrowers on loans. If he/she fails to repay the loans, it calculates how much loss the lender will bear from the outstanding amount. Banks faces credit risks also from financial instruments like interbank transactions, foreign exchange transactions, trade financing, options and many more (Yamanaka & Kinoshita, 2018).

Problems in Credit Risks Model

• Credit Concentration- Credit Risk Model is only concerned with only specific borrower of any specific sectors. This can cause credit concentrations of the connection of the borrowers (Yang & Lv, 2019. It doesnot identifies the risks that is related to risk of large exposure. For example, in case of any credit risks in particular sector or group of entities, the whole sector will be affected, which will automatically create loss to the banks (Khokhlova, Kretova & Burov, 2019). This model doesnot calculates the loss in this scenario. It doesnot distributes its lending practices among all the business sectors.
• This model evaluates all the information that the bank needs. This information includes; the borrower’s credit history, capacity of borrower to repay the loan, total capital of the borrower, loan conditions and the collaterals of the borrower. But, there are various companies and borrowers who donot furnish sufficient information. The information may be incorrect and can cause higher credit risk for the company (Garcia, Sanchez & Marques, 2019). In this case, if the banks lends money to the company, then it will ultimately poses a higher credit risk for the banks. Credit Risk Model doesnot evaluate this, whether the information is complete or correct. It only calculates the credit risks on the basis of the credit history that is provided by the borrower or any company.
• Credit risk model only checks the collaterals that the borrowers have to secure their loans. They mitigates the risks exposure to their properties or assets. But, it doesnot analyse the value of assets. The value of assets can deteriorate over a particular period of time (Khashei & Torbat 2019). It doesnot monitor the performance of the value of assets. This will ultimately create credit problems for the banks. There can be also frauds related to these collaterals. Credit risk model doesnot verify these circumstances. Hence, this is one of the most important problem of this model.
• Credit risk model doesnot completely evaluates the cyclic performance of an industry. In some cases, certain industries can go through depression period but may cause boom period in the future. It only looks on the current trends of the business, but doesnot calculates the future slumps of the industry performances.

Therefore, Credit risk model doesnot measures the above problems and this causes various credit problems for the banks (Chen & He, 2017). This only implement a sound risks management system in order to maximise their returns from the loans, but doesnot focus on minimising the above risks of the banks.

VAR Models

VAR models or Value at Risk Models are a type of statistical measures that helps in measuring & quantifying the financial risks that is associated with a portfolio or a firm over a specific period of time. This model is widely used by banks to determine the ratio of potential loss in the portfolio of the financial firms (Elhorst, Gross & Tereanu, 2018). This model helps to control these financial risks. It identifies the probability of loss that can occur in a specific time frame. For example, it can identify the probability of loss associated with a trading assets by assessing the cumulative risks of the assets positions from different trading department. Banks estimates this measures by using historical simulation methods and Monte Carlo methods.

Problems in VAR models

• Value at Risk model uses analytical and Monte Carlo method. These method is calculated using assumptions that the assets are normally distributed. It only estimates the risks. It doesnot calculates the exact risks associated with the assets. It is only assumed (Billio, Casarin & Rossini, 2019).
• This model donot have any standard protocol to evaluate the risks. The risks are calculated using probabilities and the potential loss determines the lowest amount of risk associated with the outcomes. This underestimated the other risk magnitudes that may be associated with the portfolios.
• This model provide descriptive statistics on the trading revenues from the trading accounts for the banks. It forecasts the estimations from the trading activities. For this, it uses daily profit & loss from the trading activities for analysing the trading risks. It is difficult to forecast the daily risks factors of the trading portfolios like interest rates risks, exchange rate risks, risks in commodity prices and equity risks (Carlini & Santucci de Magistris, 2019). It is difficult to measure all the market risk factors on a daily basis. It can only identifies the regulatory constraints of the market. Banks may make approximations to roughly estimate the figure. It did not properly the changes in the profit and loss volatility. Hence, it reduce the forecast advantage and harm the accuracy.
• The VAR models doesnot capture the variances in the market risks. It cannot evaluate the risks related to hedging effects or diversification effects of the portfolios (Atwood, 2019). Therefore, banks can only have limited exposures to stress on such portfolios.
• Since, VAR models identifies the regulatory standards, there are certain regulatory guidelines that may contribute VAR being conservatives. This can provide a limited changes in the volatility and may use simple procedure to calculate these risks. Hence this will affect the banks performance (Skripnikov & Michailidis, 2018). They will also be conservative and actual risks related to the trading accounts will be affected.
• This model is unable to capture the credit and liquidity risk of the banks.

Part B

Recommendations

Credit Risk Model

• Banks must distributes its lending practices at a broader perspectives to all the borrowers and business sectors. They must describe the level of risks that is associated in the bank’s portfolio with respect to concentration of the overall sector. They should calculate the concentration risk by observing the concentration ratio, which is the proportion of the portfolio the loan represents (Alam et al., 2019). They can also use herfindahl index, which measures the size of firm with related to the industries in order to calculate the degree of concentration of the particular sector.
• Banks should thoroughly check the borrower’s personal information in order to ensure that the information is sufficient and complete. They should build a strong relationship with the customers before extending any credit. This will help to determine the attitude of credit & clearly understand what the borrower expects from them. Banks can use credit agency reports as a tool to check the financial history of a company, bank reports of the company to evaluate the details of the bank’s relation with the company & borrowing capacity of the company, Review the financial statements to evaluate the profitability and liquidity position of the company.
• Banks must focus on calculating the value of the assets by calculating the collateral value of the assets. This can be done by evaluating the recent sale price of the asset. They can take advice from a qualified expert before lending a loan to the borrower. This will help the banks to secure their loans (Hargarter & Van Vuuren, 2019).
• Banks can also evaluate and monitor the performance of and cyclic trends of the loan portfolio by closely visualising the information in the charts & pictures to clearly understand the industry trend, industry concentration and evaluate to a large exposures.

Therefore, banks should focus on minimising the impact of credit risks by properly evaluating the factors and controlling the actions. They should focus on minimising the problems of the above three factors that includes; credit concentration, credit issuing problems and cyclic performances.

Value at Risk Model

• Banks should use this model to forecast and compare the different financial instruments. This model will help to compare between the trading accounts and identifies the instruments which is more risky than the other one. They should not use this model to forecast the risks associated with the trading accounts.
• Banks can use financial software’s of VAR models to properly estimate the probabilities instead of simply writing down the assumptions according to the historical data. Financial software will be more concise on the results as compared to manual prediction of the figures.
• Banks can successfully implement the VAR model by analysing the portfolio return distribution. They must appropriately identify the value of each portfolio. They must properly handle the disadvantages of VAR models before examining the value of portfolio.
• Banks should use this method only if all the approaches of Value at Risk Model gives similar result. If the results is different for the same portfolio with different methods. This will protect from market risks of the portfolio. They can do for sensitive analysis techniques to determine how the target variables are affected with the changes in input variables. This method will help to predict the results under certain market conditions. It will consider the market risks and determines the input variables (Hugonnier & Morellec, 2017). Banks can study the change in interest rates on bonds and accordingly predict the outcomes. Therefore, sensitivity analysis can be selected by back as an alternative of Value at Risk Model.
• Banks can use the coherent risk measure that is more sensitive to wrong estimation of the distribution. It can combine this risk measure with Value at Risk Model to get a more realistic and reliable measure of risks. The coherent risk measures will capture the current market events of the portfolio and measures the risks with the financial instrument.

Conclusion

Therefore, it can be concluded from the above report that, there are various problems associated with the risk measurement model in banking sectors. The most important problem in credit risk model is inefficient data management, lack of proper risk assessment tools and less risk visualization & reporting in the borrower’s information’s. The banks are only focusing only on maximising the revenues and not on minimising the risks. It is recommended that banks should focus on minimising the three problems that includes; credit concentration, credit issuing problems and cyclic performances. They can take advice from a qualified expert before lending a loan to the borrower. This will help the banks to secure their loans. They should visualise the information in the charts & pictures to clearly understand the industry trend, industry concentration and evaluate to a large exposures. It is also reviewed that Value at Risk Model (VAR) model estimates the normal distribution probabilities, which is not accurate with respect to market risk of the portfolios. It is recommended that, banks can combine coherent risk measure with Value at Risk Model to get a more realistic and reliable measure of risks. The coherent risk measures will capture the current market events of the portfolio and measures the risks with the financial instrument.

References

Alam, M. K., Ab Rahman, S., Mustafa, H., Shah, S. M., & Hossain, M. S. (2019). Shariah governance framework of Islamic banks in Bangladesh: Practices, problems and recommendations. Asian Economic and Financial Review, 9(1), 118.

Alodayni, S. (2016). Oil prices, credit risks in banking systems, and macro-financial linkages across GCC oil exporters. International Journal of Financial Studies, 4(4), 23.

Atwood, J. (2019). Applications of Relaxed Constraint (RC) Models in Portfolio Optimization Subject to VaR, cVaR and Related Risk Constraints (No. 2217-2019-1648).

Billio, M., Casarin, R., & Rossini, L. (2019). Bayesian nonparametric sparse VAR models. Journal of Econometrics, 212(1), 97-115.

Carlini, F., & Santucci de Magistris, P. (2019). On the identification of fractionally cointegrated VAR models with the F (d) condition. Journal of Business & Economic Statistics, 37(1), 134-146.

Chen, W., & He, X. (2017). Pricing credit default swaps under a multi-scale stochastic volatility model. Physica A: Statistical Mechanics and its Applications, 468, 425-433.

Doss, S. (2017). Evaluation of Deposit Insurance Fund Adequacy Using Credit Risk Model—An Indian Experience. Chinese Business Review, 16(5), 211-233.

Elhorst, J. P., Gross, M., & Tereanu, E. (2018). Spillovers in space and time: Where spatial econometrics and global VAR models meet.

García, V., Sánchez, J. S., & Marqués, A. I. (2019). Synergetic application of multi-criteria decision-making models to credit granting decision problems. Applied Sciences, 9(23), 5052.

Hargarter, A., & Van Vuuren, G. (2019). Measuring conduct risk in South African banks. Qualitative Research in Financial Markets.

Hugonnier, J., & Morellec, E. (2017). Bank capital, liquid reserves, and insolvency risk. Journal of Financial Economics, 125(2), 266-285.

Khashei, M., & Torbat, S. (2019). A Hybrid Intelligent Classification Model Based on Multilayer Perceptron Neural Networks and Fuzzy Regression for Credit Scoring Problems. Computational Methods in Engineering, 37(2), 97-111.

Khokhlova, G., Kretova, N., & Burov, V. (2019, November). The problems of investment activity of entrepreneurship and methodological aspects of credit risks assessment. In IOP Conference Series: Materials Science and Engineering (Vol. 667, No. 1, p. 012038). IOP Publishing.

Skripnikov, A., & Michailidis, G. (2018). Regularized Joint Estimation of Related VAR Models via Group Lasso. J Biostat Biometric App, 3(1), 104.

Yamanaka, S., & Kinoshita, M. (2018). A structural credit risk model based on purchase order information (No. 18-E-11). Bank of Japan.

Yang, Y., & Lv, X. (2019, September). Research on SME Credit Risk Control Platform Based on Supply Chain Finance. In The 4th International Conference on Economy, Judicature, Administration and Humanitarian Projects (JAHP 2019). Atlantis Press.