Working Paper: CEPR ID: DP10146
Authors: Abhijit Banerjee; Esther Duflo; Richard Hornbeck
Abstract: We investigate the puzzle of microfinance: that loans generate large measured returns for businesses, yet loan take-up is low and the businesses often close. We analyze a randomized trial that bundled microfinance loans with a cheap health insurance policy. Requiring clients to purchase insurance substantially lowered loan renewal. The insurance was useless, due to administrative failures, but reduced loan renewal negatively impacted clients? businesses. Clients' decision to incur substantial business losses, rather than pay modest insurance premiums, implies the substantial financial gains from microfinance loans are dissipated by unmeasured costs and provide little net value to microfinance clients.
Keywords: Development; Health Insurance; Microcredit; Microenterprises; Microfinance; Revealed Preference; Welfare
JEL Codes: O12; O16; O19
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Decrease in loan renewal rates (G21) | Decreased sales (D49) |
| Decrease in loan renewal rates (G21) | Decreased profits (D33) |
| Decrease in loan renewal rates (G21) | Decreased investments in assets and workers (E22) |
| Requirement to purchase health insurance (G52) | Decreased client welfare (I38) |
| Dropout of clients from microfinance program (I21) | Decreased sales (D49) |
| Dropout of clients from microfinance program (I21) | Decreased profits (D33) |
| Dropout of clients from microfinance program (I21) | Decreased investments in assets and workers (E22) |
| Requirement to purchase health insurance (G52) | Decrease in loan renewal rates (G21) |