The formal credit supply gap in the MSME sector in India amounts to ~INR 2.9 lakh crore. Of this, Micro Enterprises form ~INR 2.25 lakh crore i.e. 80% of the gap. The absence of or poor quality of business records, low asset base, and inadequate capital negatively impacts the ability of micro and small enterprises to access formal financial services. The ability of Financial Institutions (FI) to effectively underwrite microenterprise loans depends on accurate assessment of the loan applicant’s ability and intent to repay. Traditionally, financial institutions use their past knowledge and experience of financing microenterprises to carry out such assessments. This approach works fine till the time the loan sizes remain small and the institutions remain limited in scale. However, as they begin to target clients with higher ticket loans and achieve scale, they are prone to face some challenges like:
Lack of standardisation of decision-making process due to the difference in experience and biases of the decision makers. This leads to adverse client selection and might negatively impact portfolio quality
Lack of understanding about the tangible and intangible characteristics of applicants that could have an impact on his/her repayment behavior. This will also negatively impact the portfolio quality
One possible solution to the above challenges is the adoption of a Statistical Credit Scoring Model (SCM). An SCM predicts the probability of an applicant to default based on predictive data modeling. In this video, we discuss the rationale for developing a Statistical Credit Scoring Model, the key issues to remember while developing an SCM and the benefits the tool can bring to financial service providers who intend to serve thin file customers.