The impact of weather shocks on food security in Uganda: a panel regression analysis

Climate change endangers food security globally and in Uganda particularly. Our objective was the assessment of weather shocks’ impact on the food security of smallholder households in Uganda. We used the Uganda National Panel Survey (UNPS) data. A multistage cluster sampling procedure had been employed to recruit respondents. We used the periods 2013/2014, 2015/2016 and 2019/2020 to construct an unbalanced panel with 3,935 observations. We employed Poisson and logit fixed effects panel regressions. Food security was measured as the number of meals consumed by a farming household each day or alternatively the likelihood of experiencing a food shortage. Our main result is that weather shocks unequivocally reduce households’ food security. Furthermore, households that are large, female-headed, lack literacy or assets or reside in Northern Uganda or in rural areas tend to suffer from reduced food security. In contrast, non-agricultural (business) activities improve the food security of households. Households with low asset ownership or residing in the North are hit hardest by weather shocks in terms of the number of meals, while households with middle asset ownership are hit hardest regarding food shortages. For development policy, our results are in favour of region-specific support for climate adaptation and for non-agricultural (business) activities, the improvement of literacy via better education and support for building up assets or starting non-agricultural (business) activities.