Gender Differences in Access to Extension Services and Agricultural Productivity

Purpose:This article contributes new empirical evidence and nuanced analysis on the gender difference in access to extension services and how this translates to observed differences in technology adoption and agricultural productivity.Approach:It looks at the case of Ethiopia, where substantial investments in the extension system have been made, but the coverage and effect of these investments on female and male producers are not well-understood. This article employs a cross-sectional instrumental-variable regression method using a regionally representative data set of more than 7500 households in four major regions in Ethiopia during the 2010 main season.Findings:Female heads of households and plot-managers are less likely to get extension services through various channels and less likely to access quality services than their male counterparts after controlling for other factors. Receiving advice from extension agents is positively related to adoption of improved seed and fertiliser for both female and male, as hypothesised. However, beyond their influence through fertiliser and improved seed use, visits by or advice from agents are not significant or negatively significant in all productivity models estimated for females and males, which is in contrast to past studies. In some crop-productivity models estimated, it is the perceived quality of agents’ visits and access to radio that appear to be positively significant factors in explaining productivity levels for both females and males.Practical implications:Results highlight the need for stratified productivity models by gender and crop in future research. In terms of policy implication, results highlight the need to focus on quality of service and alternative channels of information, such as radio, to improve productivity.Originality/value:This article utilises a large-sample data set; uses the instrumental-variable regression method to address selection bias and endogeneity issues in productivity models; and stratifies the analyses to account for differentiated production functions by gender and crop.