| The purpose of this dissertation is to demonstrate that group lending with joint liability improves information flows regarding borrower creditworthiness and thus represents a significant innovation in delivering formal credit to collateral constrained, small scaled producers, a clientele poorly served by traditional, individual liability loan programs.Using an agency model, group credit is shown to be a valid means of signaling or transmitting information on borrowers' latent creditworthiness to a distant formal lender interested in reducing default exposure. The lender, through contract termination threats, induces the credit groups to screen and select members with a high average measure of creditworthiness. This relationship is mutually beneficial because the lender realizes better repayment rates and the borrowers are guaranteed access to credit and economize on transaction costs.Using survey data from an existing Costa Rican group credit program, repayment performance, information transfer, cost-effectiveness, and program viability are studied. Evidence is found supporting the hypothesis that groups engaging in screening outperform groups that do not with regards to loan recovery. Furthermore, groups seem to exploit to their advantage private information. In regression analysis controlling for observable creditworthiness measures, individuals belonging to screened groups had better repayment rates. This suggests that "reputation" can compensate for limited stocks of wealth. The group scheme thus performs a valuable service by correctly identifying creditworthy clients where other traditional credit evaluation techniques fail.The chief weakness of the case study program relates to cost recovery. Only 30% of the groups had positive internal rates of return under a strict standard of full cost recovery. When a relaxed standard, accounting for non-economic benefits and ignoring overhead cost, was used, 52% of the groups recorded positive rates of return. This weakness stems from the high fixed cost of group formation borne solely by the lender and the small loan portfolio size. The program may be justifiable only in social terms if private rates of return exceed the theoretical interest rates needed to make a larger number of groups viable. Sensitivity analysis shows that by varying several parameters, upwards of 77% of the groups can be made viable. |
