Trade, Storage and Climate Extremes: Evidence from Sub-Saharan Africa

Global warming is associated with changes in the frequency, intensity, spatial extent, duration and timing of temperature and precipitation extremes. Agriculture is particularly vulnerable to such climate extremes, resulting at times in large production losses and a strong reliance on market mechanisms to ensure local food availability. However, the impact of trade and storage market mechanisms on local food security has not yet been clearly established, theoretically nor empirically. This study aims to address this gap and focuses on Sub-Saharan Africa, a hotspot for food insecurity and climate extremes and characterized by high trade costs and diverse market policies. A theoretical model combining household level consumption smoothing with market level trade and storage is presented. Variable stock levels and trade balances are shown to have a direct relation with intertemporal household asset transfers. Based on a numerical simulation exercise, three qualitative predictions are derived that can be taken to the data. First, when households are isolated from markets, climate extremes have a stronger upward pressure on local food prices. Second, as a consequence of the negative local price-harvest relationship, storage and trade have a smaller buffering effect on local food security in isolated markets. Third, trade and storage can act both as substitutes and complements, depending on the temporal and spatial correlation of climate extremes. The predictions are empirically tested by estimating local food security impacts of dry and wet climatic conditions in 1,400 secondary administrative units across 12 countries in SSA over the time period 2010 – 2016. Results of local and country-wide climate extremes, interacted with local travel times to market and country-level trade and stock dynamics are broadly consistent with the theoretical predictions. Future work on adaptation to future variability under climate change should explicitly consider the interaction between micro- and macro-level dynamics, as well as the spatial and temporal correlation of climate shocks.