As for microeconomic financial indicators of bank performance, the rise of the both RONA and the share of highly liquid assets increases the bank financial stability and makes the default less probable to occur. As we proposed in the hypothesis H1 («The increase in the amount of bank net assets accounts negatively to the probability of default for both «money laundering» and «economic reasons» default probabilities»), the problem called «too big to fail» is clearly seen in our model. Thus, the first hypothesis is confirmed.

As for the indicators referred to the allocation of Bank's Assets, we are troubled to interpret the insignificance of the share of overdue debt to loans to individuals and the same indicator for legal entities at least at the level of 10 percent. As for the loans to individuals, we are also troubled to interpret the negative influence of increasing the volume of retail lending to the default probability, because usually retail lending is the most risky one. As for the loans to legal entities, the indicator is statistically significant (at least at 10 percent level) for the «money laundering» bank default. The crediting of enterprises and organizations usually is much less risky, than retail crediting. This fact explains the negative influence of increasing the volume of legal entities lending to the default probability. As we proposed in the hypothesis H2 («The increase in the amount of the interbank loans given accounts negatively to the probability of default for both «money laundering» and «economic reasons» default probabilities »), our model confirm a fact, the banks involve in crediting other banks are more stable and less predisposed to the bankruptcy. Thus, the second hypothesis is confirmed.

As for the indicators referred to the Investments, we are troubled to interpret the negative influence of investments to principal notes on the default probability. The rise of the investments in bonds makes the default less probable to occur. Usually bonds are associated with smaller risk and more safe way to absorb interests. This fact makes cash flows more predictable and decreases banks default probability. Contrary to what was proposed in the hypothesis H3 («Investments in shares indicator will be statistically significant both «money laundering» and «economic reasons» default probabilities at least at 10 percent of statistical significance»), the investments in shares indicator has statistical significance at least at 10 percent only in «money laundering» model. Probably, investments in shares increases the default probability for banks involved in «money laundering», because they execute operations with illiquid securities and this fact makes them an object of the supervisory authority close attention and thus increases the license revocation probability. Thus, the third hypothesis is denied.

As for the indicators, referred to the Banks Liabilities, the interbank loans gained indicator appears to be statistically insignificant. The rise of the deposits of individuals makes bank defaults more probable. We suppose that the explanation of this phenomenon is the fact, that very often depositors provoke chain reaction of deposits outflow from the bank. The banking history knows a lot of such incidents, when this process ends with the bank default. Funds of enterprises and organizations indicator is statistically significant only for the «money laundering bank default». The rise of funds of enterprises and organizations in the bank increases the default probability of banks, involved in money laundering. Probably those banks mainly operates with enterprises that are parts in capital outflow chains. An increase of those funds signifies more aggressive capital outflow and thereafter more close attention of supervisory authority. This fact increases the default probability. Bonds and principal notes indicator is statistically significant only for the «economic reasons bank default». The rise of bonds and principal notes among banks liabilities increases the default probability of banks, because bank borrowed capital is increased with bank own capital remained at the same level. This fact increases the default probability.

Here we will make a resume on the hypotheses confirmation. Confirmed hypotheses:

H1: The increase in the amount of bank net assets accounts negatively to the probability of default for both «money laundering» and «economic reasons» default probabilities.

H2: The increase in the amount of the interbank loans given accounts negatively to the probability of default for both «money laundering» and «economic reasons» default probabilities.

Denied hypotheses:

H3: Investments in shares indicator will be statistically significant both «money laundering» and «economic reasons» default probabilities at least at 10 percent of statistical significance.

Conclusion

This work is a demonstration of econometric theory implementation into the bank default probability estimation. Our work showed that the division of bank defaults on the basis of different license revocation reasons seems to be appropriate taking into account the possibilities of econometric modelling and statistical packages.

The first conclusion consists in the fact, that we can estimate the bank default probability estimation using publicly available accountings. Quite high predictive quality of the models suggest the viability of this approach.

The second conclusion consists in the fact that the change in the macroenvironment influences differently on the probability of default for different reasons of the default.

The third conclusion consists in the fact that varying time lags helps to forecast bank default. In addition it should be mentioned that different default reasons mean different time lags due to various contribution of factors in the probability of bank default.

Of course, there is a wide field for further researchers to deepen the study of this issue.

It may be deepen in sense of using advance econometric approaches, such as multinomial regression analysis or neuronets. Probably correct specification may lead us to the results that will outperform binary choice logit regression analysis.

Also further researches may involve new predictors in the analysis. Actually the most promising trends are found in the use of factor analysis, which allow to combine the set of factors into the smaller set or even one factor.

As another form of widening this research, it is possible to make binary choice modelling on the basis of premodelling clustering through underlying economic reasons or the use of data science approaches, for example cluster analysis.

At the end, we want to state, that our work is a necessary basis for the following works in this field, because it demonstrates the opportunity to make econometric regression analysis using publicly available information at present days. The debates on this issue are found promising and the relevance of bank default probability estimation is not in doubt.

References

1) Иванов В. В., Фёдорова Ю. И. (2015). Результаты моделирования вероятности наступления дефолта банка на примере российской банковской системы. Экономика и современный менеджмент: теория и практика, 6 (50), 6-20.

2) Карминский А. М., Костров А. В. (2014). Совершенствование моделей вероятности дефолта российских банков: использование рейтингов и панельных данных. XIV Апрельская международная научная конференция по проблемам развития экономики и общества: в 4-х книгах. Книга 1. М.: 538-546.

3) Пересецкий А. А. (2007). Методы оценки вероятности дефолта банков. Экономика и математические методы, 43 (3), 37-62.

4) Пересецкий А. А. (2013). Модели причин отзыва лицензий российских банков. Влияние неучтенных факторов. Прикладная эконометрика, 30 (2), 49-64.

5) Altman E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. Journal of Finance, 23 (4), 589-609.

6) Altman E. I., Haldeman R., Narayanan P. (1977). Zeta analysis: A new model to identify bankruptcy risk of corporations. Journal of Banking and Finance, 1 (1), 29-54.

7) Bondarenko M. V., Maria Semenova M. V. (2018). Do high deposit interest rates signal bank default? Evidence from the Russian retail deposit market. Financial Economics, WP BRP 65/FE/2018, 1-28.

8) Bovenzi J. F., Marino J. A., McFadden F. E. (1983). Commercial bank failure prediction models. Economic Review, 68, 14-26.

9) Cole R. A., Gunther J. W. (1995). Separating the likelihood and timing of bank failure. Journal of Banking and Finance, 19 (6), 1073-1089.

10) Cole R. A., Gunther J. W. (1998). Predicting bank failures: A comparison of on- and off-site monitoring systems. Journal of Financial Services Research, 13 (2), 103-117.

11) Estrella A., Park S., Peristiani S. (2000). Capital ratios as predictors of bank failure. FRBNY Economic Policy Review, 6 (2), 33-52.

12) Godlewski C. J. (2007). Are ratings consistent with default probabilities? Empirical evidence on banks in emerging market economies. Emerging Markets Finance and Trade, 43 (4), 5-23.

13) Izan H. Y. (1984). Corporate distress in Australia. Journal of Banking and Finance, 8 (2), 303-320.

14) Jagtiani J., Kolari J., Lemieux C., Shin H. (2003). Early warning models for bank supervision: Simpler could be better. Economic Perspectives, 27 (3), 49-60.

15) Kavussanos M. G., Tsouknidis D. A. (2016). Default risk drivers in shipping bank loans. Transportation Research Part E: Logistics and Transportation Review, 94 (C), 71-94.

16) Kolari J., Glennon D., Shin H., Caputo M. (2002). Predicting large US commercial bank failures. Journal of Economics and Business, 54 (4), 361-387.

17) Lennox C. (1999). Identifying failing companies: a reevaluation of the logit, probit and DA approaches. Journal of Economics and Business, 51 (4), 347-364.

18) Martin D. (1977). Early warning of bank failure: A logit regression approach. Journal of Banking and Finance, 1 (3), 249-276.

19) Ohlson J. A. (1980). Financial ratios and the probabilistic prediction of bankruptcy. Journal of Accounting Research, 18 (1), 109-131.

20) Peresetsky A. A., Karminsky A. M., Golovan S. V. (2004). Probability of default models of Russian banks. Competitiveness and modernization of economy, 1, 407-417.

21) Peresetsky A. A., Karminsky A. M., Golovan S. V. (2011). Probability of default models of Russian banks. Economic Change and Restructuring, 44 (4), 297-334.

22) Raffaella Calabrese & Paolo Giudici (2015). Estimating bank default with generalised extreme value regression models. Journal of the Operational Research Society, 66 (11), 1783-1792.

23) Scott J. (1981). The probability of bankruptcy: A comparison of empirical predictions and theoretical models. Journal of Banking and Finance, 5 (3), 317-344.

24) Westgaards S., Wijst van der N. (2001). Default probabilities in a corporate bank portfolio: a logistic model approach. European Journal of Operational Research, 135, 338-349.

25) Wiginton J. C. (1980). A note on the comparison of logit and discriminant models of consumer credit behavior. Journal of Financial and Quantitative Analysis, 15 (3), 757-770.

Abstract

The object of this paper is to investigate the influence of both different microeconomic financial and macroeconomic factors on the bank default probability. Moreover, we differentiated the different reasons for recalling the banking license by Russian Central Bank through two binary choice models. The data stands for five years of Russian banking sector development: from the first quarter of 2014 to the fourth quarter of 2018. The contribution of our work to the bank default probability estimation can be described in two points. Firstly, we use the combination of factors, which differs from combinations of factors in other works. Secondly, we use two simultaneous binary choice models differentiating various reasons of default. This issue is rarely seen in such researches, because other papers most commonly ignore differing the reasons of recalling the license.

Целью данной работы является исследование влияния как различных микроэкономических финансовых, так и макроэкономических факторов на вероятность дефолта банка. Более того, мы дифференцировали различные причины отзыва банковской лицензии Центральным Банком России с помощью двух моделей бинарного выбора. Данные представлены за пять лет развития банковского сектора России: с первого квартала 2014 года по четвертый квартал 2018 года. Вклад нашей работы в оценку вероятности дефолта банка можно описать в двух пунктах. Во-первых, мы используем комбинацию факторов, которая отличается от комбинации факторов в других работах. Во-вторых, мы используем одновременно две модели бинарного выбора, дифференцирующие различные причины дефолта. Данный аспект редко встречается в подобных исследованиях, поскольку другие статьи чаще всего игнорируют разделение причин отзыва лицензии.

Appendix

Dataset for license revocation with wording «economic reasons». Histograms

Table 1 RONA

Table 2 Natural logarithm of Net Assets

Table 3 Highly Liquid Assets to Net Assets

Table 4 Loans to Individuals to Net Assets

Table 5 Overdue Debt to Loans to Individuals to Net Assets

Table 6 Interbank Loans Given to Net Assets

Table 7 Loans to Enterprises and Organisations to Net Assets

Table 8 Overdue Debt to Loans to Enterprises and Organisations to Net Assets

Table 9 Investments in Shares to Net Assets

Table 10 Investments in Bonds to Net Assets

Table 11 Investments in Principal Notes to Net Assets

Table 12 Deposits of Individuals to Net Assets

Table 13 Funds of Enterprises and Organisations to Net Assets

Table 14 Interbank Loans Gained to Net Assets

Table 15 Bonds and Principal Notes to Net Assets

Dataset for license revocation with wording «money laundering». Histograms

Table 1 RONA

Table 2 Natural logarithm of Net Assets

Table 3 Highly Liquid Assets to Net Assets

Table 4 Loans to Individuals to Net Assets

Table 5 Overdue Debt to Loans to Individuals to Net Assets

Table 6 Interbank Loans Given to Net Assets

Table 7 Loans to Enterprises and Organisations to Net Assets

Table 8 Overdue Debt to Loans to Enterprises and Organisations to Net Assets

Table 9 Investments in Shares to Net Assets

Table 10 Investments in Bonds to Net Assets

Table 11 Investments in Principal Notes to Net Assets

Table 12 Deposits of Individuals to Net Assets

Table 13 Funds of Enterprises and Organisations to Net Assets

Table 14 Interbank Loans Gained to Net Assets

Table 15 Bonds and Principal Notes to Net Assets

Dataset for license revocation with wording «economic reasons». Descriptive statistics

Table 1 Macroeconomic variables

Table 2 Microeconomic financial variables

Table 3 «Assets» variables group

Table 4 «Investments» variables group

Table 5 «Liabilities» variables group

Dataset for license revocation with wording «money laundering». Descriptive statistics

Table 1 Macroeconomic variables

Table 2 Microeconomic financial variables

Table 3 «Assets» variables group

Table 4 «Investments» variables group

Table 5 «Liabilities» variables group

Размещено на Allbest.ru