Abstract:
The study examines how cost of credit, digital finance and bank capital shape MSME lending and performance in Kenya. Using FinAccess 2024 microdata and bank-level panel data (2011-2024), we estimate a control-function Probit, Mundlak Random Effects Logit and System-GMM model. Results show that perceived high borrowing costs are endogenous to loan access, but become insignificant once corrected. Firm age, size, formalization and female ownership significantly improve loan access. Digital finance usage for business transactions increases approval probabilities, while urban location widens access gaps. On the supply side, capital adequacy significantly anchors sustainable loan growth. Overall, MSME performance is driven more by structural firm characteristics and credit supply conditions than by cost perceptions alone.