Resilience of the group lending model to a COVID-19 induced shock: evidence from an Indian microfinance fund

Purpose - The authors study the effect of an exogenous shock in the form of Coronavirus lockdowns on individual default and on default contagion within the microfinance (MF) sector in India. The authors rely on proprietary data obtained from an MF institution for the period from Nov 2019 to Dec 2020. The authors show that default increased to 95.29% in the month of April 2020, when Covid lockdowns were fully in place. However, borrowers bounced back thereafter, either making full or partial payments, so that defaults had fallen to 5.92% by December 2020. Static features of the group lending model like peer monitoring and joint liability help explain 90% of the monthly deficit during Covid lockdowns among uneducated borrowers. Dynamic features such as contingent renewal help explain why defaults were cured quickly through timely repayments. Finally, there is an absence of default contagion at the district level. Indeed, lagged own default explains 96.6% of variation in individual default, rather than contagion through group, village or district-level defaults. The authors conclude that the MF sector is resilient to exogenous shocks like the pandemic. Design/methodology/approach - The authors use time series panel regressions, as well as cross-sectional regressions. Findings - The authors find that borrower defaults increased significantly to 95.29% during the month of April 2020, when Covid lockdowns were fully in place. However, borrowers bounced back almost immediately, either making full or partial payments, such that defaults had fallen to 5.92% by December 2020. The group lending model does remarkably well in explaining defaults even during Covid lockdowns. Among the majority (92%) of borrowers who are residents of rural districts, the group lending model appears to blunt the impact of the exogenous shock on rates of default. Indeed, panel regressions demonstrate that the group lending model helps explain 90% of the monthly deficit among uneducated borrowers. Logistic regressions indicate that the group lending model is less persuasive among relatively affluent borrowers residing in semi-urban or urban areas who have some formal schooling. Contingent renewal is shown to be an effective disciplining mechanism when a group does default due to the Covid lockdowns. The authors find that groups who defaulted in April 2020 but repaid the outstanding balance within the next two months were more likely to receive subsequent loans from the lender. On the other hand, groups who defaulted in April 2020 and did not repay the outstanding balance until December 2020 did not receive follow-on financing. Finally, the authors find that lagged individual default is the primary source of individual default, rather than contagion through group, village or district-level defaults. Research limitations/implications - The limitation of the study is that it is confined to a single MF institution in India. Social implications - The authors conclude that the social capital that is the foundation of the group lending model succeeds in limiting both the risk and contagion of default from an exogenous shock, such as the Covid pandemic. Originality/value - To the best of the authors’ knowledge, the authors are the first to examine defaults in the Indian MF sector during the Covid lockdowns in April 2020.