| Description |
This report summarizes the findings from the baseline survey for Food and Nutrition Security
Enhancement Project (FANSEP) in Nepal. The main findings are:
. Most households in the sample rely on cultivation of crops to meet their nutritional needs, especially cultivation of cereals. Most are smallholder farmers with less than 0.5 hectares of land cultivated on average.
. At the time of the baseline survey, a relatively small proportion of farmers reported membership in farmers groups. If membership does not increase among the target sample, it will be difficult to measure the impact of the program. If 90% of target enrollees join groups, we would be able to detect an impact on household incomes of 20% or more. However, the current reported rate of enrollment among the target population was only 30%. Ensuring that early starter groups have joined groups and are receiving interventions should be a priority.
. The Proxy Means Test scores correctly predicted households with low incomes, and using PMT scores to determine program eligibility was justified to improve beneficiary targeting.
. The timing of when the IE can measure impacts using the most rigorous possible design, depends on when the later starter groups would be phased in. As long as the 40 communities in late starter groups have not been phased into FFS interventions, impacts can be compared using a randomized control design. Given the ongoing recruitment of early starters remains a priority, keeping these 40 communities of late starters until last maximizes the potential for the IE. In the event that delays related to COVID-19 and other constraints prevent the project from reaching the originally planned number of recipients, ensuring that all other communities besides the 40 late starters are reached first would give the most rigorous possible evaluation with the longest possible impact horizon. Ideally, FANSEP should delay implementation in ”late treatment” villages until fall of 2022 to maximize the probability that the IE can detect program impacts in RCT sample before this final set of villages is enrolled in the interventions. |
