Experimental Design
The empirical strategy of this pilot project consists of a combination of quasi-experimental and experimental methods. We focus on the following three elements: i) the impact of the FDP, ii) the spillover effect of the FDP, and iii) the impact of relationship-building interventions.
First, we evaluate the impact of the FDP. All farmers in the catchment areas are scored at baseline based on deliveries and other observable characteristics. Our sample is split into two groups: 60% of the farmers above the corresponding threshold within their zones (FDP treatment), and the remaining 40% of farmers scoring below the threshold (FDP control). To overcome potential self-selection bias, we exploit the fact that all farmers were scored against pre-determined criteria and were not able to ex-ante manipulate scores to participate in the FDP. By focusing on those farmers just above and below the threshold, we can use a regression discontinuity design (RDD) framework to causally identify the FDP impacts.
Second, we randomized 32 zones in our sample frame into two groups. The first group includes 16 zones with 80% invited farmers (high-intensity zones), and the second group covers 16 zones with 40% invited farmers to the FDP (low-intensity zones). Focusing on the bottom 20% of the farmers in terms of farmer score, this empirical design permits to measure the spillover effects by controlling for the intensity of invited farmers (40% or 80%) within each zone.
Third, we evaluate the impact of two organizational relationship-building interventions using experimental methods. We assess whether having a designated “fixed” relationship manager and whether communication by a familiar versus unfamiliar individual affects the main outcomes.