Impact of FinTech Lending on SME Credit Risk in Emerging Markets

Thesis Abstract

This study investigates how FinTech lending channels influence SME credit risk in emerging markets, addressing the rising concern that alternative finance platforms may alter risk profiles through borrower selection, underwriting practices, and dynamic repayment behaviors. The problem stems from limited longitudinal evidence on whether FinTech lenders improve SME access to credit while maintaining portfolio quality, or whether intensified competition and data-driven scoring strategies shift risk concentrations toward more opaque borrower segments. The aim is to quantify the impact of FinTech lending on SME credit risk and identify mechanisms through which digital lending platforms affect risk outcomes in comparison with traditional banks. Specific objectives are (1) to assess differences in default rates between SMEs financed by FinTech lenders and those financed by traditional banks; (2) to examine the role of borrower characteristics, credit scoring models, and loan terms in shaping default probabilities; (3) to evaluate the moderating effect of firm age, sector, and macroeconomic conditions on the FinTech–credit risk relationship; (4) to analyze post-disbursement repayment behavior and early warning signals captured by platform data; and (5) to derive policy and practice implications for risk management, regulation, and SME financial inclusion. The methodological approach adopts a mixed-methods design, combining quantitative panel data analysis with qualitative interviews to triangulate risk determinants and mechanism pathways. The population comprises SME borrowers across two to three emerging-market economies with active FinTech lending ecosystems over a five-year window (2019–2023). A purposive sample of 1,200 SME loan records from major FinTech lenders and 1,200 matched SME loans from traditional banks will be constructed, augmented by 60 in-depth interviews with risk managers, credit analysts, and platform data scientists. Quantitative analysis will employ survival analysis and logistic regression models to estimate default probabilities and hazard rates, controlling for firm-specific, loan-specific, and macroeconomic covariates. Advanced techniques such as propensity score matching, difference-in-differences, and fixed-effects panel regressions will be used to address endogeneity and selection bias. For robustness, machine-learning-based risk scoring models (e.g., gradient boosting, random forest) will be compared against conventional logistic regression, with explanatory variables including cash flow signals, alternative data indicators (digital footprints, social profiling), and macro variables. Qualitative data will be analyzed using thematic analysis to extract insights on underwriting practices, data quality, borrower experiences, and platform governance, with coding cross-validated by multiple researchers to ensure reliability. Key expected findings include (a) FinTech lending is associated with lower average default rates for SMEs with limited credit history, attributable to richer alternative data and more flexible underwriting; (b) elevated risk concentration persists among subsectors with volatile cash flows (e.g., hospitality, agriculture) and among microenterprises lacking verifiable financial statements; (c) repayment behavior on FinTech platforms exhibits more dynamic patterns, with higher early-stage repayment intensities but greater susceptibility to macroeconomic shocks; (d) platform-specific risk controls, such as real-time credit monitoring and automated early warning systems, significantly reduce hazard rates when key risk signals are detected promptly. The study also anticipates heterogeneity in effects across countries due to regulatory stringency, data access, and credit infrastructure maturity. The study contributes to knowledge by (i) providing causal and descriptive evidence on how FinTech lending reshapes SME credit risk in emerging markets, (ii) clarifying the mechanisms through which alternative data and underwriting practices influence risk, (iii) offering comparative insights between FinTech and bank lending risk profiles, and (iv) informing regulators and practitioners on effective risk management, disclosure, and consumer protection in digital credit markets. The main conclusion is expected to be that FinTech lending can reduce SME credit risk under supportive regulatory environments and robust risk analytics, but requires stringent governance, continuous data quality improvements, and targeted oversight for high-risk sectors. Policy recommendations include harmonizing data-access frameworks, mandating standardized reporting of platform risk metrics, and promoting transparency in algorithmic decision-making, while managerial guidance emphasizes integrated risk analytics that combine traditional financial indicators with alternative data streams.