Banking crises that have occurred around the world over the past 100 years have led to significant changes in legislation and regulation, as well as changes in the coefficients of a number of indicators of the stability of the banking system.

On March 9, 2023, clients of Silicon Valley Bank (USA) remotely withdrew $42 billion from their accounts using digital identification, which amounted to 85% of the bank's deposits. Due to this situation, by the end of the working day the bank had liquidity problems amounting to minus 1 billion dollars, and on the same day the bank was declared bankrupt. Modern technology allows customers to quickly withdraw funds from banks, but the influence of regulatory bodies on banking operations lags behind progress [1].

The result of banks' activities in attracting funds for short periods for investments in long-term assets makes them weak and tightly tied to depositors. Douglas Diamond and Philip Dybvig, winners of the 2022 Nobel Prize, noted in their research that the imperfect security of deposits gives rise to several types of a Nash equilibrium in the financial system. This situation is especially justified in the era of online and internet banking development. In addition, in their opinion, full guarantee of deposits will help prevent bank transfers [2].

Decree of the President of the Republic of Uzbekistan DP-60, dated January 28, 2022, “On the Development Strategy of the New Uzbekistan for 2022-2026” outlines the following key directions, goals, and objectives: “To develop stress-testing models, with the support of the International Monetary Fund, aimed at identifying potential risks in the economy and assessing their impact on the stability of the banking system. This includes taking measures to create macroprudential buffers to ensure the resilience of banks against potential financial losses” [3].

In the context of the modern banking system and economic transformations, commercial banks face several challenges in forming a stable resource base. The current situation is complex, primarily due to the limited amount of available funds held by individuals and legal entities. Consequently, banks need to explore new opportunities for forming and effectively utilizing their resource base [4].

Currently, several methods and models are used to assess the overall financial stability of commercial banks. These methods and models primarily differ in terms of the scope of information and data, evaluation principles, complexity of calculations, implementation processes, timelines, and the accuracy of the results. Depending on the types of bodies conducting the assessments, national, rating agency, and international methodologies are used to evaluate the financial stability of commercial banks.

The methods and models used to assess the overall financial stability of commercial banks have been studied in detail in scientific research by a number of renowned scientists, and the methods and models they have developed have a number of distinctive features.

The main examples of indicative models that are important for assessing the financial stability of commercial banks are the works of such scholars as Edward Altman, James Olson and Robert Merton, Kromonova V.S. as well as such models as the CAMELS rating. Their developments serve as an important theoretical and practical basis for assessing and strengthening the financial stability of commercial banks.

The Altman Z-model (Altman Z-Score) is a financial model (formula) developed by the American economist Edward Altman, designed to predict the probability of bankruptcy of an enterprise [5].

The formula is based on a combination of 4-5 key financial ratios that characterize the financial position and performance of an enterprise. The formula was originally proposed by Altman in the 1960s. Later, the author proposed variations of this formula taking into account the industry-specific features of organizations.

The Altman 4-factor Z-model is used for non-manufacturing companies. The formula for the four-factor model is as follows:

Z-score = 6.56T1 + 3.26T2 + 6.72T3 + 1.05T4 (1)

For manufacturing companies, the Altman 5-factor Z-model is used. The formula for the Altman 5-factor model is:

Z-score = 0.717T1 + 0.847T2 + 3.107T3 + 0.42T4 + 0.998T5 (2)

Where:

T1 = Working Capital / Assets

T2 = Retained Earnings / Assets

T3 = EBIT / Assets

T4 = Equity / Liabilities

T5 = Revenue / Assets

Altman's know-how consists of selecting indicators and, most importantly, weights (coefficients) by which these indicators and estimates of the resulting value are multiplied in the formula. Altman made his conclusions based on an analysis of American enterprises over a number of years [6].

The first person to use the logistic regression method to create a model for diagnosing the risk of bankruptcy of enterprises was J. A. Olson in 1980 [7]. According to this model, the Z index is calculated using the following formula:

Z= -1.3-0.4 X1+0.6X2-1.4X3+0.1X4-2.4X5-1.8X6+0.3X7-1.7X8-0.5X9 (3)

Where:

X1 - the natural logarithm of the ratio of total assets to the gross national product deflator index;

X2 - the ratio of total liabilities to total assets;

X3 - the ratio of working capital to total assets;

X4 - the ratio of current liabilities to current assets;

X5 - 1 if total liabilities exceed total assets, otherwise it is 0;

X6 - ratio of net profit to total assets;

X7 - ratio of revenue from core activities to total liabilities;

X8 - 1, if net profit has been negative for the last two years, if vice versa, the indicator is 0;

X9 - the ratio of the difference between net profit in the last reporting period and net profit in the previous reporting period to the sum of net profit in the last reporting period, taken modulo, and net profit in the previous reporting period, taken modulo.

The Z index is used to find the probability of bankruptcy risk using the logistic regression formula:

Despite the variety of evaluation methods and techniques available today, there is no ideal model or single set of indicators that allows for the most accurate determination of a bank's financial stability or the strength of its resource base. Therefore, in this research, we decided to develop an indicative model for ensuring the stability of commercial banks' resource base [12].

The implementation of this model consists of four sequential stages: the first stage identifies the components of indicators, the second stage identifies the indicators themselves, the third stage identifies the integral indicators, and the fourth stage assesses the level of stability (Table 1) [13].

Three groups of levels of financial stability of commercial banks are shown based on the optimal value of integral indicators; based on expert recommendations, the recommended values of components are calculated:

1. financially stable commercial banks (index value - 35% and higher);

2. balanced financially stable commercial banks (index value - 25-35%);

3. financially unstable commercial banks (index value - less than 25%).

Based on the research results and the conclusions we have drawn, in the fourth stage, based on the obtained values of the integral indicators, we have divided the stability of the commercial banks' resource base into the following stability groups:

1. High Stability: Irb > 80%

2. Stable: 60% < Irb ≤ 80%

3. Moderate Stability: 40% < Irb ≤ 60%

4. Low Stability: 20% < Irb ≤ 40%

5. Weak: Irb ≤ 20%

The purpose of dividing the integral indicators into 5 stability groups is that the strength of the resource base of commercial banks is a separate and relatively narrow approach to expressing the financial stability of banks.

In general, other criteria may be used by commercial banks themselves or by regulators and rating agencies.

According to our proposed classification of resistance groups, the value of this index reflects the moderate stability of resistance [15].

It is also recommended to select the following strategies for banks depending on the group of sustainability indicators characterizing the strength of the resource base of commercial banks (Table 4).

In our opinion, we have concluded that the advantages of using an indicative model in assessing the stability of commercial banks' resource base are as follows:

Firstly, geometric mean values were used in calculating the integral indicators of the stability of commercial banks' resource base. Unlike the arithmetic mean, the advantage of using the geometric mean value is that it takes into account the asymmetry of the distribution. This is especially important when calculating indicators with significant dispersion of deviations due to the use of quantities of different dimensions. The use of a weighted arithmetic mean instead of the geometric mean in financial stability indicators (as well as the inclusion of weights in the calculations) may lead to a loss of universality of the indicators. Also, changes in the indicators of ensuring the stability of commercial banks' resource base depend on the specific activities of each bank, which necessitates adapting the selected indicators to the characteristics of these banks. In addition, the use of geometric mean indicators in the calculation allows adding other indicators without making additional adjustments to the indicators. When arithmetic mean indicators are used, the weight of each indicator needs to be revised by adding a new one or removing an old one from the indicator, which is a very labor-intensive process when using a large number of indicators.

Secondly, it is related to the high speed, simplicity, ease of understanding of the analysis, and the convenience of collecting the necessary data by a wide range of interested users associated with a small number of indicator systems.

Thirdly, it is the existence of the possibility for banks to assess the state of their resource base based on the general system of indicators that characterize the stability of their resource base.

Fourthly, there is also the possibility of using the results of the analysis to compare the state indicators of the resource base of different commercial banks.

At the same time, this model also has a number of shortcomings, and the result of the indicative model only reliably reflects the current state of the commercial bank's resource base, while the characteristics of other indicators (term, quantity, non-financial indicators, etc.) are not taken into account.