Analyzing Insolvency Drivers and Developing Credit Rating System for Small and Medium-sized Enterprises in Russia

Small and medium-sized enterprises (SMEs) play a key role in the Russian economy. However, banks and investors are reluctant to provide debt financing to these firms. This is underpinned by SMEs’ speculative credit quality and information asymmetry between borrowers and lenders. In this study, we aim to identify the insolvency drivers of Russian SMEs and compare them with those in other markets. The relevance of the study is underpinned by the scarcity of research in this field and the high demand for an accurate rating system for domestic SMEs. Logistic regression was selected as the modeling method. The sample contained 177 non-financial domestic SMEs over the period 2015–2019. The set of explanatory variables consisted of firm-specific financial, categorical, and macroeconomic factors. An accuracy ratio of >80% was achieved. We found that, unlike those in Asian emerging markets, financial factors explained around 70% of domestic SMEs’ credit health. Significant financial factors included profitability, debt leverage, and coverage ratios and the term structure of debt. Non-financial drivers included ownership of the firm by large businesses (or group of companies), firm size, and territory of operation within Russia. Among macroeconomic drivers, the unemployment level was the most significant driver of SMEs’ credit quality. In addition, we developed a rating system for domestic SMEs and determined the relative benchmarks from Expert RA and Moody’s agencies. We found that the existing scales of rating agencies did not provide the granular assessment of SMEs’ creditworthiness. This confirmed our hypothesis that distinct rating frameworks and methodologies for domestic SMEs in the Russian market are imperative. As shown in the literature, the greater the rating granularity and transparency, the more enhanced the debt market’s appropriate risk-return tradeoff analysis.

Credit rating agencies; Insolvency drivers; Probability of default; Risk management; Scoring model; Small and medium-sized enterprises

    Small and medium-sized enterprises (SMEs) play an important role in the Russian economy as they contribute to accelerating economic growth and increasing employment. However, their development is limited by restricted access to long-term funding. The recently organized security offering platform in Moscow Stock Exchange (MOEX)–Growth Sector (GS)–is aimed to provide SMEs access to the domestic debt capital market. This platform is aimed at breaking the monopoly of banks, channel government support, and enable access to unsecured and long-term funds. Nonetheless, the volume of bond issuance in GS remains low due to the high information asymmetry between borrowers and lenders.

        The assignment of credit ratings (CRs) to SMEs can alleviate this problem. However, existing rating agencies lack the methodologies that address the specifics of SMEs. A literature review indicated that there are several analogs of GS in some emerging and developed markets (Anwar et al., 2018). In Italy, there is a “mini-bond” market where SMEs can issue public debt (Altman et al., 2020). In China, there are three types of bonds designed for SMEs. Analysis of study showed that the effective way to reduce information asymmetry between lenders and borrowers is to assign CR to firms’ obligations (Anwar et al., 2018). As a result, the number of research in modeling defaults has been growing (Demeshev and Tikhonova, 2014).

        The number of research on this topic, particularly devoted to SMEs, however, remains low. Nonetheless, few studies have stressed the importance of considering non-financial factors related to sales, operations, or governance, as well as macroeconomic drivers of default (Lyukevich et al., 2020; Koroleva et al., 2020; Hol and Van der Wijst, 2008). Anwar et al. (2018) studied the frameworks of rating agencies in assessing the creditworthinss of SMEs. For example, these institutions in Singapore and Malaysia are predominantly focused on financial data. This could be attributed to the mature governance and reporting in these countries, as well as the availability of a reliable database of SME data. Conversely, in Thailand, the Philippines, or Indonesia, where reporting and governance standards of SMEs are still emerging, the institutions focused on non-financial drivers of SME’s insolvency (60%–70% of total assessment). From a practical standpoint, starting from 2021, Russia’s MOEX requires that all issuers or issues in GS be rated by the domestic rating agencies. However, the existing methodologies of rating agencies are tailored for large businesses and do not consider specific risks of SMEs.

        Among the modeling methods (MMs) of SME defaults, the most widespread are logistic and probit regressions (Demeshev and Tikhonova, 2014). They demonstrate good accuracy, including non-financial and macroeconomic variables, assume any form of explanatory factor distributions, and result in the interpreted scorings. Their disadvantages include susceptibility to multicollinearity and overfitting. However, in the most recent studies, artificial intelligence (AI) and hybrid methods have become widespread. Although these MMs have shown to have higher accuracy than that of regressions, they were often “black boxes” that reduced their application in practice. Fantazzini and Figini (2009) predicted SMEs’ default probability in China using the random survival forests (RSF) method. RSF performed better than the logistic regression for the “in-sample”; however, for the “out-of-sample,” performance evidence was the opposite. Demeshev and Tikhonova (2014) revisited differences in the predictive power of insolvency models for Russian SMEs and demonstrated that random forest outperformed logistic regression both “in-sample” and “out-of-sample.” The addition of non-financial information to the model led to improved forecasts. In turn, hybrid models gave a better and stable performance (Zhu et al., 2017; Li et al., 2016).

    To conclude, the literature review demonstrated that SME insolvency drivers vary significantly from country to country. The majority of the studies are focused on developed or emerging markets in Asia, and only a few, although outdated, studies covered Russia’s SME. Therefore, this study aims to close these gaps by identifying the insolvency drivers of Russian SMEs. The novelty of this study is to identify the insolvency drivers that are particularly inherent to SMEs in Russia. We tested the non-financial (business, corporate governance, and macroeconomic) drivers that were not previously considered for Russian SMEs in the literature. We tested the hypothesis that the maturity of reporting and governance in the country directly affects the share of non-financial drivers in the SME scoring models. We tested the assumption that the existing methodologies and scales of rating agencies did not consider specific risks of SMEs. We compared identified insolvency drivers of domestic SMEs with those in other advanced and emerging markets. The results can be used by investment practitioners to assist in developing rating scales and methodologies for SMEs. They can also be interesting for researchers who are studying the differences in SME default drivers across various markets. 

    This study aims to identify the insolvency drivers of SMEs in Russia. We compared our findings with those in other markets and found that, unlike those in Asian emerging markets, financial factors explained around 70% of domestic SMEs’ credit health. Meaningful financial factors included gross profit margin, return on equity, debt leverage, and coverage ratios and the term structure of debt. Non-financial drivers included ownership of the firm by large businesses (or group of companies), firm size, and territory of operation within Russia. Among macroeconomic drivers, the unemployment level was the most significant driver of SMEs’ credit quality. In addition, we developed a rating system for SMEs and determined the relative benchmarks from Expert RA and Moody’s rating agencies. The benchmark indicated that the existing rating scales did not provide the granular assessment of SMEs’ credit health and gave a very coarse evaluation of their default probabilities. This confirmed our hypothesis that distinct rating frameworks and methodologies for domestic SMEs in the Russian market are imperative. Future research concerning expanding the list of non-financial explanatory factors toward the business and corporate governance practices of entrepreneurs, a detailed analysis of the impact of government support on SMEs’ credit quality, and comparing the ability of other empirical methods, including AI and hybrid models, to predict defaults of SMEs in various emerging markets is necessary.

    The research is partially funded by the Ministry of Science and Higher Education of the Russian Federation under the strategic academic leadership program “Priority 2030” (Agreement 075-15-2021-1333 dated 30.09.2021)

Altman, E.I., Esentato, M., Sabato, G., 2020. Assessing the Credit Worthiness of Italian SMEs and Mini-Bond Issuers. Global Finance Journal, Volume 43(C), https://doi.org/10.1016/j.gfj.2018.09.003

Altman, E.I., Sabato, G., Wilson, N., 2010. The Value of Non-Financial Information in SME Risk Management. Journal of Credit Risk, Volume 6(2), pp. 95–127

Anwar, S.M.R., Mustafa, R., Tanzo, T.T., 2018. Revisiting the Existing Credit Rating Methods of Small and Medium Enterprises. Indonesian Journal of Business Finance and Accounting, Volume 1(1), pp. 31–44

Demeshev, B., Tikhonova, A., 2014. Dynamics of Predictive Power of Insolvency Models for Russian Small-Medium Enterprises: Wholesale and Retail Trade. Journal of Corporate Finance Research, Volume 8(3), pp. 4–22

Eko, W.H., Heti, M., Teguh, S.I., 2019. Improving Credit Scoring Model of Mortgage Financing with Smote Methods in Sharia Banking. Russian Journal of Agricultural and Socio-Economic Sciences, Volume 8(92), pp. 56–67

Fantazzini, D., Figini, S., 2009. Random Survival Forests Models for SME Credit Risk Measurement. Methodology and Computing in Applied Probability, Volume 11(1), pp. 29–45

Hol, S., Van der Wijst, N., 2008. The Financial Structure of Non-Listed Firms. Applied Financial Economics, Volume 18(7), pp. 559–568

Karminsky, A.M., Mistrulli, P.E., Stolbov, M.I., Shi, Y., (Eds.), 2021. Risk Assessment and Financial Regulation in Emerging Markets' Banking. Trends and Prospects. Springer International Publishing, p. 396

Koroleva, E., Baggieri, M., Nalwanga, S., 2020. Company Performance: Are Environmental, Social, and Governance Factors Important? International Journal of Technology, Volume 11(8), pp. 1468–1477

Li, K., Niskanen, J., Kolehmainen, M., Niskanen, M., 2016. Financial Innovation: Credit Default Hybrid Model for SME Lending. Expert Systems with Applications, Volume 61(C), pp. 343–355

Liu, Y., Li, S.Z., Wu, W., Huang, R., 2013. Dynamics of a Mean-Shift-Like Algorithm and its Applications on Clustering. Information Processing Letters, Volume 113(1–2), pp. 8–16

Lyukevich, I., Agranov, A., Lvova, N., Guzikova, L., 2020. Digital Experience: How to Find a Tool for Evaluating Business Economic Risk. International Journal of Technology, Volume 11(6), pp. 1244–1254

Mishchenko, A.V., Chizhova, A.S., 2008. Methodology of Credit Risk Management and Optimal Formation of a Loan Portfolio. Financial Management, Volume 1, pp. 91–105

Senaviratna, N.A.M.R., Cooray, T.M.J.A., 2019. Diagnosing Multicollinearity of Logistic Regression Model. Asian Journal of Probability and Statistics, Volume 5(2) pp. 1–9

Zhu, Y., Xie, C., Wang, G.J., Yan, X.G., 2017. Comparison of Individual, Ensemble and Integrated Ensemble Machine Learning Methods to Predict China's SMEs Credit Risk in Supply Chain Finance. Neural Computing and Applications, Volume 28(1), pp. 41–50