This thesis explores significant dimensions of ultra-poverty graduation through cash transfer
program. Drawing upon both qualitative and quantitative methodologies, the thesis provides a
comprehensive assessment of graduation models, beneficiary trust levels, the efficacy of behavioral
nudges and group investment dynamics, crucial elements for sustainable graduation outcomes for
Tanzania’s Productive Social Safety Net (PSSN) program. The thesis is organized around three
main empirical papers.
The first paper sets the foundation by reviewing global graduation models, particularly the
acclaimed BRAC ultra-poor graduation framework. Using comparative analysis, it identifies critical
success factors such as comprehensive graduation package and intensive coaching, as well as pitfalls
including inadequate contextual adaptation and insufficient market linkages. Stakeholder interviews
and surveys involving implementers, policymakers and beneficiaries reveal varied perceptions
regarding the concept of graduation, sustainability of program impacts and the efficacy of existing
interventions. Results indicate a strong preference for context-specific graduation strategies that
emphasize continuous capacity building and financial education. Therefore, program designs must
be carefully tailored to local socio-economic contexts and incorporate robust market analysis to
ensure sustainable outcomes and genuine beneficiary self-reliance.
The second paper quantitatively measures trust levels among beneficiaries participating in
Community Savings and Investment Promotion (COMSIP) groups. Trust emerges as a
foundational component influencing successful financial collaboration and effectiveness of savings
groups, critical for sustainable poverty alleviation outcomes. Using trust experiments and
structured surveys, the research identifies significant variations in trust levels influenced by factors
such as exposure to location, expectations and behavioral nudges exposure. While behavioral
nudges produced mixed effects, expectations-based framings demonstrated relatively stronger
associations with trust behaviour. Higher trust levels correlate with improved group financial
outcomes, including enhanced savings rates and greater investment in productive assets. These
findings suggest that livelihoods enhancement and financial inclusion programs can substantially
improve their outcomes by strengthening trust and cooperation among beneficiaries.
Finally, the third paper examines group-level trust, reciprocity and investment choices using group
investment games, surveys and qualitative case studies. The findings reveal that collective
investment decisions are shaped not only by individual preferences but also by group deliberation,
peer influence, leadership dynamics and shared lived experiences. Group settings generally
promoted stronger reciprocity and willingness to contribute to collective investments compared to
individual decision-making settings. The paper also identifies key characteristics of successful
groups, including balanced group composition, effective governance structures, trust, external
support and shared economic motivation.