Send message to Author

Paper short abstract

Four-year randomised control study of an anti-poverty programme in Malawi examining whether gender-targeted assets plus couples' training produces sustained impacts on household welfare, women's empowerment, and intimate partner violence among the ultra-poor.

Paper long abstract

We present findings from a four-year follow-up of a four-arm randomized controlled trial in rural Malawi that tests whether internal household frictions can be as binding as external resource constraints for the ultra-poor. The Graduation program provided a productive asset package targeted either to women, to men, or to women with an added couples' cooperation and empowerment training, with a fourth control group.

Our 17-month results showed that gender targeting alone does not alter household income or consumption, though male targeting shifts benefits toward men. In contrast, adding couples' training to a female-targeted program produced a Pareto improvement: household income rose, driven by higher female business earnings and greater male participation in productive decisions, with no reduction in women's empowerment. Mechanism analysis revealed that gains arose from improved coordination and shifts in gendered labour allocation, not from changes in information or shared vision.

This follow-up, conducted four years post-intervention (five years post-cash transfer), examines whether these economic and empowerment effects persist in the medium term. This extended analysis will determine whether economic empowerment translates into sustained changes in household welfare, women's empowerment, and violence, and whether the mechanisms of improved coordination and labour reallocation continue to drive outcomes or evolve over time. By examining both the persistence of production frontier shifts and changes in intra-household dynamics, we offer new evidence on whether enhanced cooperation can generate lasting improvements in household welfare and the long-term foundations of poverty traps.