Access to Credit and Agricultural Productivity: Evidence from Kenyan Maize Farmers
Description
The primary data for this study came from a survey designed to determine the impact of credit on maize productivity in Kenya. According to Yadav & Rao (2024), 400 smallholder maize farmers in five sub-counties of Trans Nzoia, Kenya, were surveyed between 2020 and 2023. The study focused on 200 maize smallholders who received agricultural credit (treatment group) and 200 who did not receive credit (control group). By Wanzala et al., (2023), the 200 quasi-control group was chosen using a propensity matching score (PMS) that runs a logistic regression of the observable characteristics listed in Table 1. PSM was used to prevent selection bias because the treatment variable (access to agricultural credit) was not randomized. The authors also use sensitivity analysis with "Rosenbaum bounds" to assess the impact of hidden bias. All the farmers in this study produced maize for household consumption, demonstrating that they were credit constrained. Farmers who produced improved maize seed but were not credit-constrained were excluded from the study. This is because these smallholder farmers do not face the same technical production constraints and risks. The treatment group was drawn from the Chereng'any, Kwanza, and Saboti sub-counties. The control group was drawn from the Endebess and Kiminini sub-counties. To reduce potential agricultural input spillovers to the control group in their sublocations, this study separated the treatment and control groups with a 6 km buffer zone (Wanzala et al., 2023).
Files
Steps to reproduce
primary data collection