Abstract
This paper examines the drivers of credit uptake, with a focus on how non-conventional data can reduce information asymmetry, among credit-invisible-smallholder farmers in Vizianagaram and Srikakulam districts of Andhra Pradesh, India. It leverages both traditional and alternative data, mapped to the 5Cs of credit, to develop a weighted alternative credit scoring model for financial inclusion. Using primary data from 96 smallholder farmers, each survey response was mapped to Character, Capital, Capacity, Collateral, and Conditions. Weighted scores were computed (Character25%, Capital-15%, Capacity-20%, Collateral-20%, Conditions-20%), and credit appraisal was conducted based on these scores. Statistical and comparative analyses were performed. A broad questionnaire capturing demographic, agronomic, infrastructure, and digital-behaviour related data was used for credit risk assessment. The study finds that alternative indicators such as digital literacy, social capital, and social welfare scheme participation significantly improve the prediction of creditworthiness. Only 18% of farmers were “Approved”, 70% were “Conditionally Approved”, and 12% were “Declined” under the new model. These results indicated a significant ‘partially eligible’ profiles where specific documentation and monitoring can convert conditional approvals into institutional credit uptake. The model enables lenders to better assess underserved farmers, suggesting that integrating alternative data can enhance rural credit access. This is among the first empirical studies in India to operationalize the 5C’s using both traditional and alternative data for smallholder credit scoring. The scoring system acts as a transparent, explainable, traceable, and auditable Decision Support System (DSS) pipeline for lenders serving marginalized communities.
Keywords: Alternative Credit Scoring, 5C’s of Credit, Financial Inclusion, Rural Finance, Smallholder Farmers.