This paper presents empirical evidence of whether participation in
community water resource management improves household agricultural
income using a case study of irrigation co-management
under the Bwanje Valley Irrigation Scheme in Malawi. An endogenous
switching regression was applied on a dataset of 412 farmers
to correct the outcome (net agricultural income) for sample selection
bias. Propensity score matching was then used to measure the
impact of participation in irrigation co-management on net annual
agricultural income. Despite variations in the magnitude of benefits
among different groups, empirical evidence suggests that net
annual agricultural income of the poor, youths and female-headed
farmers participating in the scheme would have worsened had they
not participated in the scheme.