UCLA

This dissertation explores some of the causes of low-savings by the poor, which are linked to the persistency of poverty. The first two chapters study self-control problems, one of the channels that might explain the inability of the poor to save. Whereas the third chapter evaluates whether lack of access to safe savings mechanisms can be an explanation for insufficient savings.

The first chapter has the goal of testing the broadly adopted assumption in economics that people apply a unique discount rate to the utility from different sources of consumption. The essay uses unique data from two surveys collected in rural Uganda to elicit time preferences over different goods based on both hypothetical and real choices from approximately 2,400 individuals. The data reject the null of equality in discount rates under a number of different modeling assumptions.

These results have important theoretical and policy implications. In the first place, they provide support for the idea that time-inconsistent behaviors can be observed even if individuals do not exhibit hyperbolic preferences, the standard explanation in the literature for non-constant discounting. In addition, good-specific discounting, under certain conditions, can explain the persistence of poverty. The second chapter focuses on this last point and provides an application for the findings in the first chapter. It presents evidence that the poor spend a smaller share of their income than the rich on goods with higher discount rate, which implies a stronger disincentive to save, a necessary condition to predict the existence of a poverty trap in self-control models.

The third chapter evaluates the impact of expanding access to safe saving instruments. Using panel data from a national sample of Mexican households, it studies the effects of a program expanding financial access on savings for beneficiaries of social programs. The article uses the exogenous source of variability generated by a national electronic payments program that opened savings accounts to a group of social beneficiaries. The findings indicate that the access to saving accounts increased reported total savings significantly, with larger effects for households living in rural localities, and non-significant effects for those living in urban localities.