Causal mapping of food price-conflict connections

Much has been written about the connection between food prices and social unrest, particularly in the aftermath of the “Arab Spring” of 2011.  As a contributing factor to insecurity in food supply, and therefor to food price volatility, plant diseases are one of many partial causes of food insecurity.  In a world where there is enough food for everyone to have a sufficient diet, food insecurity can be thought of as primarily a political problem; the distribution of food, like that of wealth, is far from equitable and the correlation between poverty and food insecurity is no more than a clear marker of the underlying inequality in welfare across the global population.  The political mechanism for solving both problems is redistribution; an answer that it seems is only slightly more unpalatable to the “have’s” in the case of money than food.  The FAO estimates that 34 countries (27 in Africa) are in need of food assistance.

The relationship between food insecurity and social unrest – in its most extreme form civil war – is not linear.  In as much as food shortages and high food prices can act as a spark to ignite civil unrest, as some argue they did in the Arab Spring, conflict, once started, also has a measurable negative effect on agricultural productivity.  For example, in a report for the FAO, based on a synoptic analysis of conflicts in 38 countries which experienced conflicts between 1961 and 2000  Teodosijević  (2003) estimated that several indices of agricultural activity and food production were significantly higher in the five years prior to conflict than during conflict or in the five years immediately following conflict.

We showed in 2011 that it is easier to correlate the effects of pest complexes on crop yield than the effect of individual pests and diseases.  In any agricultural system farmers typically face characteristic pest complexes in each crop that is grown.  These pest complexes usually require multiple decisions and actions on the part of farmers, and the evolution of stable relationships between the pest complexes and farming in which volatility of yield loss is minimized demands social stability over extended periods of time.  Conflict obviously can disrupt this stability, and furthermore, the multi-component nature of pest complexes probably contributes to the difficulty of the task of rebuilding productive agricultural systems in the aftermath of conflict.

The effects of war on on food security are clearly more acute than the effects of pests and diseases in nearly every situation and co-occur with so many other negative effects on welfare, that a direct comparison between the two makes little sense.  However, as components in a larger network of causal factors it is possible to explore how they interact and their effects on food insecurity are mutually reinforcing.

Cognitive map of causal relationships among factors connecting conflicting, food, conflict, and crop pest complexes
Cognitive map of causal relationships among factors connecting conflicting, food, conflict, and crop pest complexes

Figure 1 shows a causal map for these interacting effects. The map was generated by noting causal relationships mentioned in the various articles, blog posts and official documents, plus our own experiences of dealing with pest complexes in various agricultural systems.  The map encodes broad generalizations about causal relationships.  The open arrows indicate negative relationships (for example, International aid reduces hunger) and the closed arrows indicate positive relationships (for example, conflict causes hunger).

Causal maps can be used as the starting point for debate and deliberation – different people might have different, justified, views of the direction, polarity or existence of different causal links in a map. Or, they may disagree about the list of factors included in a map (too few or too many? Which should be excluded or added?).  All of these issues, and more, can be examined in a structured way and, furthermore, the map can be used to generate dynamic analyses that indicate the way in which a system comprising causal connections of this structure will tend to behave.

Figure 2. Implied system dynamics when conflict and high food prices are absent from the initial conditions
Figure 2. Implied system dynamics when conflict and high food prices are absent from the initial conditions



When we look at the implied dynamics of the causal map in figure 1 we find that it implies an extreme sensitivity to “initial conditions”.  If we initiate the system with neither conflict nor high relative food prices, with social infrastructure and with no population displacement, it generates a stable outcome in which agricultural production, food supply, and social infrastructure increase, while hunger, relative food prices and pest complexes decrease (Figure 2).

The optimistic picture presented in figure 2 reverses if the system starts with either conflict or high relative food prices (or both) present.  In that case (see Figure 3) the system becomes locked into a pattern in which hunger, conflict, population displacement, and high food prices increase, international aid is needed and, agricultural production, food supply and social infrastructure decline.  In spite of several feedback loops in the system it does not appear to have any self-correcting tendency suggesting the need for external intervention to change the behaviour, or for the system’s structure to change eventually as the accumulating negative effects trigger larger scale events.

food conflict 3
Figure 3. Implied system dynamics when conflict and high food prices are present initially

The take-home message is one that it probably doesn’t require a formal analysis to make clear: starting conflicts and letting food get so expensive that people feel compelled to resort to civil unrest to address the issue, are really bad ideas.