Modeling of Optimal Credit Limits in Microfinance Organizations

In an unstable economic situation, the population has a high demand for money, which leads to an increase in credit risks of microfinance organizations (MFIs). This requires the development of system management solutions to minimize them. The need for a systematic approach to risk management in MFIs is caused by the peculiarity of the development of the microcredit market in the Russian Federation: limited time resources when implementing decision-making systems (DSS) due to the short loan term, insufficient qualifications of risk mana­gement compared to the banking sector, lack of resources of technical specialists, an increase in the degree of market regulation by the Central Bank (CB). One of the ways to manage credit risk, especially in microfinance organizations, is to set limits on loans issued depending on the degree of risk of the borrower and the expected profitability. This article is devoted to the study of the issue of setting credit limits and their impact on credit risk in the entire portfolio of MFIs. The purpose of this article is to develop a systematic mathematical approach to credit risk management by establishing optimal credit limits in MFIs. The article presents a methodo­logy and a practical example of modeling limits on the example of an MFI, which is in the top 10 of the Russian online microfinance market. Based on real historical data, a mathematical model is built for the segments of primary and repeat borrowers, which allows adjusting the limit policy of the organization to reduce the level of risk, considering the profitability of each customer segment. The weighted least squares method is used as a mathematical tool to estimate the coefficients of polynomial regression, as well as a logistic regression model. The scientific novelty of this article consists in the application of a separate mathematical model for setting limits in the MFI DSS, in addition to scoring. The practical significance of this article is the possibility of using the obtained model as an adviser in the formation of credit and risk policy of a particular MFI.