The effect of using indigenous and scientific forecasts on arable farmers’ crop yields: Evidence from Rwenzori region, Western Uganda

Abstract

Climate change is already a reality, but adaptation is still low in sub-Saharan Africa as reported in several quantitative studies. One of the reasons for low adoption has been cited as lack of information including weather information. This paper investigates the role of weather information comparing arable farmers using a hybrid of Scientific forecasts and indigenous forecast with farmers solely using indigenous forecast in bolstering adaptation. In this study, we assess whether farmers who use hybrid forecasts (HF) are better off in terms of crop productivity than their comparable counterparts who use only IF based on two quasi-experimental estimations using the Propensity Score matching estimation approach. Farmers who use both scientific forecasts (SF) and indigenous forecasts (IF) (hybrid forecasts) are better off in terms of crop productivity than their comparable counterparts who use only IF. Based on a cross sectional household survey of 580 smallholders in the Rwenzori region of Western Uganda, we find that farmers who used HF had significantly higher crop yields compared to farmers that only depended on IF forecasts. The crop yields for maize and beans were significantly higher for farmers who used hybrid forecasts than those who used IF only. The average effect on the treated was over 300 kg for maize and beans. Farmers who used HF had also better access to improved crop varieties and also showed better response to effects of rainfall onset changes. Rural institutions such as access to agricultural extension, credit, improved crop varieties are impact pathways of climate information. Two policy implications can be drawn from this study's findings. First is that there is a need to increase farmers' access to credible climate information in developing countries. Climate information is associated with higher crop productivity hence contributing to farmers' resilience and welfare improvement. The second policy implication is that it’s imperative to integrate IF and promote the use of hybrid forecasts in national meteorological systems. Increasing the meteorological station and rain gauge density in farming communities, capacity building of farmers in management, analysis and use of locally generated rainfall data from rain gauges and validating IF with locally generated rainfall data will enhance the adaptive capacity to climate change in arable farming.