| Microfinance has successfully spread across most of the developing world and has provided access to credit to over 100 million asset-poor clients. Despite this accomplishment, there is little evidence that it has resulted in sustained income growth and poverty reduction. In light of this problem, researchers and practitioners are seeking to develop microfinance contracts that achieve both high repayment rates and promote high return investments. Two tools hoped to achieve these ends in an agricultural context are joint liability and index insurance. This dissertation seeks to investigate the separate and combined impacts of these two financial technologies on credit market participation and investment behavior.;In Chapter 1, I provide a brief introduction to the two financial technologies and set the stage for the three main essays. In Chapter 2, I investigate the willingness to pay for index based rainfall insurance-backed agricultural loans in northern Ghana using a contingent valuation methodology. The results demonstrate a preference for insurance payouts made directly to the farmer over those made to the bank, a high willingness to pay to avoid basis risk, and that insurance results in an overall reduction in demand for agricultural loans. In Chapter 3, I investigate the role that social capital plays in risk taking under joint liability loans. I develop a theoretical model of joint liability group credit and demonstrate that shame penalties and altruism between symmetric pairs of borrowers reduce risk taking, while risk taking between asymmetric pairs of borrowers can rise due to free riding. Results from a framed field experiment and a social network survey administered in Tanzania lend empirical support for this theoretical model. In Chapter 4, I investigate the impact of index insurance-backed contingent credit on risk taking under individual and joint liability loan contracts. I develop a theoretical model that predicts increases in risk taking under contingent credit when default penalties and risk aversion are sufficiently high, and these results hold under both individual and joint liability. I use a framed field experiment conducted in rural Tanzania to support the model empirically. I show that index insurance-backed contingent credit has a robust positive impact on risk taking under both individual and joint liability loan contracts that increases with default penalties and risk aversion. |
