Ukraine2%20Mcguirk.png
VoxEU Column Development Politics and economics

War in Ukraine, world food prices, and conflict in Africa

The humanitarian catastrophe unfolding in Ukraine has rightly commanded the attention of policymakers worldwide. However, Russia’s invasion of Ukraine will likely have consequences that echo far beyond the borders of either country. This column draws on recent research to discuss how the war’s impact on food commodity prices may shape the distribution of violent conflict in Africa. The authors predict an overall increase in inter-group conflict, yet this encompasses large spatial variation across countries, with the top agricultural producers exhibiting a decrease in conflict due to higher wages.     

Editors' note: This column is part of the Vox debate on the economic consequences of war.

Together, Russia and Ukraine supply almost 30% of global wheat and barley exports. Ukraine also accounts for 14% of globally traded corn and 75% of globally traded sunflower oil, a key cooking fuel. The supply of these commodities has been severely impeded by the ongoing conflict. In Ukraine’s case, this is due in part to the disruption of Black Sea shipping routes. In Russia’s case, it is due to the effect of sanctions along the agricultural supply chain. Together with extreme weather patterns worldwide (e.g. heat in parts of India, the US, and France; historic droughts in East Africa; and flooding in China) and rising protectionism, this has led to sharp increases in staple food prices that confer significant welfare losses on poor households in developing countries (Arezki 2022, Artuc et al. 2022, Porto and Rijkers 2022). 

To put these recent changes into historical context, in Figure 1 we plot the FAO Food Price Index in real terms from 1961 to 2022. The current figure of 145.5 is a record high, easily surpassing previous spikes in the 1970s and early 2010s. The recent growth is driven mostly by cereals and vegetable oils. 

Figure 1 FAO Real Food Price Index, 1961-2022

 

From conflict in Europe to conflict in Africa 

Understanding the downstream effects of these food price shocks on conflict in Africa is imperative. State fragility and recurrent civil wars are costly impediments to economic development in many African countries. Identifying how and why economic fluctuations affect civil conflict can help to inform policies that promote peace. 

One mechanism predicts that higher commodity prices will reduce conflict in areas that produce the given commodity. This is because rising productivity in the affected sector ought to increase wages and thus increase the opportunity cost for the marginal worker of participating in illicit or risky economic activities, such as joining armed groups (Dal Bo and Dal Bo 2011). This prediction is therefore particularly relevant to labour-intensive sectors, such as coffee production in Colombia and crop agriculture more generally in Africa (Dube and Vargas 2013, Berman and Couttenier 2015, McGuirk and Burke 2020). 

However, as we document in McGuirk and Burke (2020), there are also countervailing mechanisms through which higher staple food prices can increase conflict. Since food occupies a large share of household expenditure in Africa (around 40% on average), the net effect of food price shocks for a given individual will depend critically on whether one is a net producer or a net consumer of the relevant commodities. Sufficiently high food prices could conceivably force a net consumer to turn to risky economic coping strategies in order to maintain a necessary caloric intake, especially in the absence of conventional financial smoothing mechanisms. Thus, just as rising prices may induce marginal workers to avoid participating in armed groups in areas where crops are produced, they may also induce marginal workers to join armed groups in areas where crops are consumed. 

We find evidence of these countervailing effects in our paper. We examine the impact of rising food prices on the incidence of inter-group conflict battles in Africa at the level of a 0.5-degree grid cell (an area of 55km x 55km at the equator). We create two shift-share instruments to distinguish between the channels: a ‘producer price index’ (PPI) that combines temporal variation in world food prices with cross-sectional variation in crop production across cells; and a ‘consumer price index’ (CPI) that instead uses cross-sectional variation in crop consumption across countries. We estimate that a one standard deviation rise in the PPI reduces conflict in a cell by 17.2% of the mean, while a one standard deviation rise in the CPI increases conflict in a cell by 8.6%. Our estimates indicate that the countries most at risk of conflict through the CPI effect are Rwanda, The Gambia, Sierra Leone, Somalia, Swaziland/Eswatini, Central African Republic, Djibouti, Mozambique, South Africa, Zimbabwe, Ghana, Niger, and Mali.

We can use these estimates to predict the specific effect of increases in wheat and maize prices from January to April 2022, which we assume to be due primarily to the war in Ukraine. Through the PPI effect, conflict falls by 1.7%. Through the CPI effect, conflict increases by 6.17%. Since the PPI effect is only relevant in areas where crops are produced, we estimate the weighted average effect of the Russian invasion to be an increase in inter-group conflict in Africa of 5.3%. 

We illustrate these relationships graphically using updated raw data in Figure 2. For simplicity, we use the FAO Food Price Index (again in real terms), which is publicly available and easy to track over time. We simply plot the relationship between the food price index on the x-axis and the natural log of inter-group conflict event fatalities in a cell-year on the y-axis.1 We label these events ‘factor conflict’, as they typically capture conflict between organised armed groups contesting the control of territory. 

To distinguish between the countervailing effects, we split the sample into two groups of cells. ‘Agricultural cells’ are defined as those in the top decile for harvested area, which implies that at least 22% of a cell’s land area is used for crop production (Monfreda et al. 2008). These cells contain around 42% of Africa’s population. ‘Other cells’ are the rest. The plots in Figure 2 imply that a 50-point price spike – a magnitude similar to the change between 2019 and 2022 – is associated with a 5.8% decrease in fatalities in agricultural cells and a 1.8% increase in fatalities in other cells.  

Figure 2 Relationship between FAO Real Food Price Index and fatalities from inter-group ‘factor conflict’ in agricultural versus other cells

 

A second countervailing effect relates to what are commonly termed ‘food riots’. Scholars have long documented the role of rapidly rising staple food prices in the outbreak of riots, demonstrations, looting, and even peasant rebellions throughout history (Bellemere 2015, Ubilava 2022). These actions differ from inter-group factor conflict in that they are typically more atomistic and uncoordinated decisions executed with a view to influencing policy (through demonstrations), obtaining output (through looting), or otherwise expressing grievances due to an acute shock to inequality that often accompanies food price spikes. These events can arise in both agricultural and non-agricultural cells due to the presence of net consumers in both. We label them as ‘output conflict’ events, measured as riots, demonstrations, or other violence against civilians in the ACLED dataset.2

We estimate that a one standard deviation increase in the PPI and in the CPI respectively raise the probability of output conflict by 18.9% and 14.4%. Unlike the case of factor conflict, here the price shock leads to more conflict in both agricultural and non-agricultural areas.

We again illustrate this relationship graphically using updated data on the FAO Food Price Index in Figure 3. We show that in both types of cells, higher prices lead to more output conflict deaths. The overall effect is thus unambiguous. 

Figure 3 Relationship between FAO Real Food Price Index and fatalities from ‘output conflict’ in agricultural versus other cells

 

Conclusion

In summary, Russia’s invasion in Ukraine has led to historically sharp increases in staple food items. These in turn are likely to affect the spatial distribution of conflict events in Africa over the coming year. We predict that inter-group ‘factor conflict’ events will be driven away from the most productive agricultural areas and towards areas with less crop production. Our estimates suggest that rising prices will also contribute to the higher likelihood of ‘output conflict’ – smaller-scale riots, demonstrations and/or civilian violence in both food-producing and food-consuming areas. Policies that improve the productivity of agriculture in Africa could potentially protect both producers and consumers from the harmful effects of international price volatility in future. 

References 

Arezki, R (2022), “War in Ukraine, impact in Africa. The effect of soaring energy and food prices”, VoxEU.org. 

Artuc, E, G Falcone, G Porto and B Rijker (2022), “War-induced food price inflation imperils the poor”, VoxEU.org, 1 April.

Bellemare, M F (2015), “Rising Food Prices, Food Price Volatility, and Social Unrest”, American Journal of Agricultural Economics 97(1): 1–21.

Berman, N and M Couttenier (2015), “External Shocks, Internal Shots: The Geography of Civil Conflicts”, The Review of Economics and Statistics 97(4): 758-776.

Dube, O and J F Vargas (2013), “Commodity Price Shocks and Civil Conflict: Evidence from Colombia”, The Review of Economic Studies 80(4): 1384–1421.

McGuirk, E and M Burke (2020), “The Economic Origins of Conflict in Africa”, Journal of Political Economy 128(10): 3940-3997.

Monfreda, C, N Ramankutty and J A Foley (2008), “Farming the planet: 2. Geographic distribution of crop areas, yields, physiological types, and net primary production in the year 2000”, Global Biogeochemical Cycles 22(1).

Porto, G and B Rijker (2022), “The food crisis has no respect for borders”, VoxEU.org, 20 May. 

Ubilava, D, J Hastings and K Atalay (2022), “Agricultural Windfalls and the Seasonality of Political Violence in Africa”, Preprint.

Endnotes

1 The conflict data is from the Uppsala Conflict Data Program (UCDP) (https://ucdp.uu.se/). 

2 See https://acleddata.com/#/dashboard