Incentives to Improve Government Agricultural Extension Agent Performance: A Randomized Controlled Trial in Bangladesh

https://www.grips.ac.jp/list/jp/facultyinfo/kijima-yoko/This study provides empirical evidence on how financial and non-financial incentives improve service delivery of government agricultural extension agents. A randomized control trial was conducted in 40 sub-district agriculture offices in Bangladesh, with 807 agricultural extension officers, randomly allocated into five groups (one control and four treatment). The financial incentive was a one-time monetary reward, while the non-financial incentive was recognition by the district director. In the nonfinancial incentive treatment, we added another treatment in which the two worst performers, instead of best, are selected for inspection. In the financial incentive, we created another treatment where the best performer is selected based on performance level. We find that financial and non-financial incentives have positive effects of equal magnitude, about one standard deviation of total performance index on average. Giving a disadvantage to better performers does not decrease effort by better performers.The data collection and the randomized controlled trial was financed by the GRIPS Emerging State Project of the Japan Society for the Promotion of Science (JSPS KAKENHI Grant Number 25101002).technical repor