Community and private sector approaches to development in Kenya

Economic development policy increasingly emphasizes the role of communities and the private sector. Communities may enjoy informational advantages in providing services, while the private sector can provide considerable employment and growth opportunities. This dissertation presents three essays that investigate the impact of community and private sector approaches to development in rural Kenya. The first two chapters explore the design of community-based institutions that provide and maintain rural water infrastructure. The third chapter examines the productivity of teams in commercial agriculture. Chapter 1 studies the extent to which women's participation in local public goods management is enhanced by advocacy efforts and how increased women's participation affects water infrastructure maintenance. Gender advocacy can be a useful means of boosting women's participation, with little distortion in the effectiveness of these committees in delivering public goods. However, there is no evidence that enhanced participation by women results in greater project sustainability. Chapter 2 examines the health impacts and valuation of improved water sources, and the potential effects of alternative property rights institutions. Improvements in water infrastructure lead to large improvements in source water quality, increased use of improved water sources, moderate gains in home water quality and to a one quarter fall in reported child diarrhea incidence. Households appear willing to spend an additional 100 hours per year to walk to improved water sources. Simulations suggest that a social planner would only improve water sources with many nearby households. Allowing landowners to charge households for protected water only if they continue to provide access to unprotected water is Pareto improving relative to the status quo. Chapter 3 studies the productivity of teams of casual workers in Kenya's commercial agriculture sector. Workers choose to sort into more ethnically homogenous teams than would be expected with random matching, but ethnically diverse teams are more productive even after controlling for individual fixed effects. Workers may prefer to trade off higher earnings for opportunities to socialize or for mutual insurance, both of which may be easier to provide within tribal groups. Productivity may thus be improved by facilitating inter-ethnic team formation or by offering formal insurance schemes.