Financial Asset Accumulation by Poor Adolescents Participating in Child Savings Accounts in Low Resource Communities in Uganda

This dissertation examined savings attitudes and financial asset accumulation of poor and vulnerable school-going AIDS-orphaned adolescents involved in a subsidized matched child savings program in Uganda and being cared for by a living parent (adolescents who have lost one parent) or by an adult guardian within an extended family (for adolescents who have lost both parents). More specifically, the study tested (1) whether participation in a subsidized matched savings program had an independent effect above and beyond the effect of individual and family characteristics on adolescents' saving attitudes and self-reported financial asset accumulation; (2) whether family characteristics (i.e. family relations, family financial socialization, and household demographics) moderated the effect of participation in a subsidized matched child savings program on adolescents' saving attitudes and self-reported financial asset accumulation; and (3) whether the adolescents' future orientation and family financial socialization served as mechanisms to transmit the effect of the participation in a subsidized matched child savings program on adolescents' self-reported financial asset accumulation. Grounded in an integrated theoretical framework of classical and behavioral economics, family financial socialization theory, and the institutional theory of saving, this study used longitudinal analyses of data on 346 dyads (adolescents and their guardians) collected in a experimental cluster-randomized controlled trial. The study found that adolescents' saving attitudes (both reported willingness to save and reported confidence in saving), although not affected by participation in a subsidized matched child savings program, were significantly associated with family relations, family financial socialization, caregiver's gender, and adolescent's gender and educational aspirations. Adolescents' self-reported saving was significantly affected by participation in a subsidized matched child savings program; this effect was direct, and neither moderated nor mediated by any of the family characteristics, nor by adolescent's future orientation. The adolescents' self-reported amount saved is significantly affected by participation in a subsidized matched child savings program. This effect is weakened by the number of children in the household: the more children in the household, the weaker the effect. In addition, the effect is potentially mediated by the guardian saving for the adolescent. The findings contribute to the institutional perspective on saving arguing that saving is not only a function of individual characteristics, but also institutional opportunities. Findings may help inform programs and policies facilitating asset-building initiatives for youth in developing countries, particularly in sub-Saharan Africa.