Cost-effectiveness of community-based gendered advisory services to farmers: Analysis in Mozambique and Tanzania.

Rigorous impact evaluations on agricultural interventions in the developing world have proliferated in research of recent years. Whereas increased care in causal identification in such analyses is beneficial and has improved the quality of research in this field, much of the literature still fails to investigate the costs needed to achieve any benefits identified. Such understanding, however, would be crucial for drawing policy and programmatic conclusions from the research and for informing the allocation of public investments. Cost-effectiveness analysis (CEA) subjects both the cost side and the effects side of agricultural and rural interventions to technical scrutiny and unifies both sides in order to compare the relative cost-effectiveness of different modalities of a programme, of efforts to reach different target groups, or of efforts to achieve different outcomes. CEAs, while present in the health and education sectors, remain rare in agricultural and rural development research. This study contributes to filling the knowledge gap by conducting CEAs in a particular type of programmatic work in the agricultural sector-namely, interventions conducted as field experiments that bring a gender lens to community-based advisory services in African rural areas. Specifically, we consider two such programmes-one in Mozambique in which such advisory services aim to improve sustainable land management (SLM) practices in agricultural production, and the other in Tanzania to advise farmers on their land rights. Using CEA methods combined with econometric analysis based on randomised controlled trials, we find that the gendered modality is consistently more cost-effective than the basic modality when considering varied outcomes and target groups. However, for any given modality, it is more cost-effective to improve outcomes for men than for women. The structure of costs in the agricultural extension programme further allowed for a simulation of how cost-effectiveness would change if the programme were scaled up geographically. The results show that expansion of the basic modality of the SLM programme leads to improvements in cost-effectiveness, while the gendered modality displays nonlinear changes in cost-effectiveness along the expansion path, first worsening with initial scale-up and subsequently improving with further expansion.