Food Insecurity and Conflict Events

Since time immemorial food shortages and food price increases have sparked social unrest. The degree of any resultant conflict, depend on a host of factors including the degree of inter-personal and inter-group inequality, the income level of the country, its relative urban and youth population share, its polity, quality of governance, the degree of social protection or government expenditure and state capacity. Our identification strategy is dictated by our hypothesis that the transmission mechanisms between food prices and conflict, are mediated by these variables. Our results suggest that food prices, and especially food price volatility, the two main explanatory variables for our research, do contribute statistically significantly to conflict risk in the variety of forms with which we have chosen to measure conflict: armed conflict or civil war measured by the number of battle deaths, conflict events measured by political events in Africa (ACLED), the urban social disorder dataset and our own project data collection efforts on events related to food price related riots or protests. In the case of food price related protest events, the variable that best explains it is the level of food prices rather than its volatility. The level of development of the country and the capacity of the state also matter, as greater affluence and government spending cushion food price shocks. Our results indicate that greater democracy can engender more conflict, especially when it comes to protests, riots or other disturbances, but after a certain threshold of higher democratic development greater democracy is conflict abating. The broader V-DEM participatory index can also encourage more protest, but it does not enhance civil war related fatalities. Our governance variables are more significant when it comes to conflict events and protests short of civil war. The importance of governance in this regard is highlighted by the fact that when we interact food price volatility or level with governance, the food price volatility or food price level coefficient sometimes becomes statistically insignificant. This emphasizes the salience of the mediating role of institutions and state capacity when it comes to conflict induced by food price volatility. An innovation of our study is the inclusion of inequality. We deploy two metrics of inter-personal or inter-household vertical inequality: the GINI coefficient of income inequality and the V-DEM egalitarian index which is a broader measure of how egalitarian a society is. They emerge with the correct expected signs suggesting that more inequality breeds more conflict. We also utilize measures of inter-group or horizontal inequality in its political dimensions. These result in expected signs; more horizontal inequality results in greater conflict and is quite often statistically significant. We also employ a different definition of food insecurity only looking at positive food price spikes, utilize count data methods for monthly data on conflict events, as well as test for endogeneity between food prices and conflict events.