Poverty reduction programs inspired by BRAC's graduation approach aim to develop both tangible productive assets and intangible psychosocial assets such as self-confidence and the aspiration for upward mobility. In the broader global context, recent estimates underscore the urgency of poverty alleviation efforts, especially in a post-COVID-19 world. The surge in extreme poverty, coupled with disproportionate setbacks faced by the vulnerable, emphasizes the necessity of targeted interventions like the graduation approach, including fiscal support.

The first chapter of this thesis seeks to enhance our understanding of how psychosocial factors operate and shape the impact of graduation programs. After formulating a set of hypotheses about the effects of psychosocial constraints based on a dynamic optimization model that considers the choice between a low-income casual wage-labor occupation and a higher-earning entrepreneurial activity, this paper utilizes a randomized controlled trial of a graduation program implemented in the pastoralist regions of Northern Kenya. Key empirical findings reveal that the estimated highly favorable average treatment effects conceal significant heterogeneity, as beneficiaries who initially experienced severe depressive symptoms gained little from the program. The saturation design of the RCT also enables the identification of substantial spillover effects on the asset accumulation of women who were not enrolled in the graduation program. Spillovers are also shown to positively impact non-beneficiary women's preference for upward economic mobility, providing a plausible explanation for their capital accumulation despite not receiving direct support from the graduation program. The paper extracts implications from these findings for the cost-effective design and implementation of graduation programs.

The second chapter delves into the spillover effects of the REAP program through the female kinship network within local communities. In contrast to the previous chapter, which employed a saturation measure on community-level interaction with treated women to assess the impact of treatment intensity on REAP-eligible women, this study investigates spillover effects on control group women whose relatives benefited from the program. The findings reveal increased business assets, savings, and living standards among non-treated women whose relatives were recipients of the REAP program. Importantly, spillover effects are more pronounced for non-head control female respondents compared to female household heads. The study also draws a comparison between spillover effects at the kinship network and community levels. Similar effects are observed in business assets, with unique spillover effects on savings through the kinship network. Overall, this chapter underscores the significance of network structures in poverty alleviation program spillover effects and calls for inclusive strategies tailored to vulnerable groups and their specific needs. The study enhances our comprehension of local network dynamics and provides insights into the success of interventions within developing economies and the intricacies of spillover mechanisms.

The third chapter delves into the long-term impacts of the REAP program in pastoralist regions of Northern Kenya, taking into account spillover effects. The research discovers substantial positive effects of the program on women's productive assets and financial savings even four years into the program. However, severe depressive symptoms among beneficiaries lead to varying treatment effect trajectories. With the inclusion of endline data, the study demonstrates continuous treatment effects, revealing an asset accumulation and savings trajectory that peaked at around 30 months into the treatment and then declined by about 20% by 48 months. A positive trajectory for income over 48 months of the program is also observed. Importantly, the spillover impact found at the midline is not sustained at the endline. The distinctive characteristics of the program, including direct initial cash transfers and collaborative business management, introduce complexities in impact estimation, exacerbated by external shocks such as floods, locusts, and COVID-19. Challenges in measuring jointly owned business assets arise from group-based enterprises. These findings provide insights into the intricate dynamics of poverty alleviation interventions in challenging contexts.