Fall - 2012 War1

Published on December 20th, 2012


Natural Resources and Civil War

By Tugba Gunes

In this article I address the proposed link between abundance of mineral and fuel resources, and the onset of civil war. Does mineral and fuel abundance, as claimed by dominant figures in the field, in fact lead to civil war? And if so, what are the relevant theories that shed light on initiating forces? Lastly, is there any convincing evidence?

The end of the Cold War, despite initial optimism of a New World Order, did not abolish organized violence; if anything it changed the rules of the game and not the game itself. Violence took on a different shape. Perhaps not for the better, may I add.

The works of the two researchers Paul Collier and Anke Hoeffler were ground-breaking sets of empirical analysis on the causes of civil war and large-scale violence. Hence, the international policy community welcomed them as new lenses on the relatively new forms of violence experience in the post-Cold War Third World. To better understand the claims put forward by Collier and associates, I see it as useful to map out some of the arguments on the issue of large-scale political violence and its link to natural resources.


Greed and Grievance

The 1998 article by Collier and Hoeffler argues that war occurs if the incentives for rebellion are sufficiently large relative to costs.  Collier and Hoeffler use the share of primary commodity exports to GDP as a proxy to measure the availability of lootable resources, and further point to the key role of share of primary commodity exports to GDP in explaining the risk of civil war.

A country that is very dependent on primary commodity exports, where a quarter of its national income derives form it, is statistically four times more at risk of conflict than a country with opposite characteristics. Especially mineral and oil rich states are vulnerable to violent rebellion as they often rely heavily on one primary commodity, which make up great parts of the national income.

Greed-motivated forms of violence can, in its simplest form, be explained through the existence of a valuable prize which motivates individuals to use time and effort to claim this prize. Mineral resource rents are one such prize that motivates violence. The idea is that grievance related forms of violent rebellion suffers from collective action problems, as “justice” is a public good and hence victim to freeriding tendencies.

Collier therefore argues that greed motivated forms of violence are more viable as the benefits go straight into the pocket of those who take part in the acts of violence. This because, greed appeal to people’s demands for instant gratification for loot, and nicely overcomes difficulties related to collective action and free riding. The line of reasoning made by Collier and Hoeffler is that natural resources, which make up significant parts of primary commodity exports to national income, offer rebel groups with incentives to capture the state apparatus, and in doing so also the opportunity to collect rents from location specific lootable resources, and in effect do “well out of war”.

Although the results presented by Paul Collier and Anke Hoeffler detect correlation between both grievance and greed motivated forms of conflict, the proxy for greed-available natural resources (for loot), shows a significantly greater correlation to conflict then that of grievance. On the note of grievance versus greed we could bluntly sum up the results of Collier and Hoeffler by stating that there are fewer martyrs then opportunists in the eye of valuable prize.

In short the basic line of argument on the role of natural resources is that it presents rebel groups with the opportunity to collect rents from location specific resources, and therefore incentives to capture the state apparatus. The incentives for rebels are dependent on two factors: 1) probability of victory; 2) expected gains upon victory; and 3) the expected negative outcomes: of 1) duration of war; and 2) cost of rebel coordination.  Natural resources can therefore be viewed as a “honey pot” for greedy rebels were state capture is ultimately only a means towards the end aim of loot.


Mineral and oil resources and civil war: is there any convincing evidence?

With regards to above mentioned theory, how do we go about evaluating its usefulness in the context of the real world? Is it enough that these theories make sense on a theoretical level? Or do we need to look further into the empirics? As the theories of Collier and Hoeffler are very much empirically anchored, the “validity” of their theory making will ultimately unravel in an investigation into their definition, coding, and correlation considerations. Three methodological concerns are therefore raised in order to conceptualize on the evidence presented by Collier and Hoeffler:

Firstly, a principal concern is tied to the definition of civil war.  What constitutes violence, let alone civil war? And how does one go about defining civil war in such a way as to be able to measure its existence? The definition used by Collier and Hoeffler depicts civil war as:

An internal conflict with at least 1,000 combat-related deaths, with both an identifiable rebel organization and government forces suffering at least five percent of these casualties.”

There are several concerns related to the use of battle-related deaths as an indicator of civil war. For one, it can be a misleading indicator of civil war because many current conflicts are not heavily combat oriented, instead more than 90 percent of casualties are civilian. Moreover, the operational definition of 1000 battle-related deaths also obscures the difference in population size and in effect favors countries with larger populations. A country with a larger population will in absolute terms suffer greater then a smaller country with the same percentage of causalities, and herein lays a significant inaccuracy in measurement of impact.

Secondly, the relationship between the ethos of large-scale violence and war is rather ambiguously and arbitrary operationally defined in the literature. Collier and Hoeffler’s threshold is of 1000 battle-related deaths. While others propose 25 battle-related deaths a year to better capture the range in war (for instance eco-violence). To further extend the strain of thought one could even argue that a battle-related death (hot war) is not the sole criteria and measure of civil war.

Using the wisdom of Thomas Hobbes, state weakness could just as well be an initiating source of violence, as war exists “not only in battle or in the act of fighting, but in a tract of time wherein the will to contend by battle is sufficiently know”. Henceforth the wider definition of civil war is open to a wide array of (philosophical) interpretations.

Third, correlation does not necessary equal causality. It might well be that, for instance a high threshold for civil war, like the one posited by Collier and Hoeffler, only detect a correlation with mineral abundance because in the time it takes for a conflict to grow into a civil war most other industries have “fled the country”. And as Collier and Hoeffler only recognize the initiation of civil war when at least 1000 battle-related deaths have occurred, they in effect ignore the role of preceding violence before the 1000 threshold is reached. Moreover also neglecting the possibility of “reverse causality”, were large-scale violence and civil war might in fact be the initiator of mineral and fuel dependence and not the other way around.

Fourth, different datasets pave the way for different interpretations. Recent studies draw mainly on four datasets for civil war: Collier and Hoeffler, Fearon and Laitin, Elbadawi and Sambanis, and Gleditsch and his associates, all of which branch out from the Uppsala dataset.  The aforementioned datasets differ in their approach to: 1) defining the occurrence of civil war; 2) its end; and 3) in handling missing data. With regards to coding civil war onset Collier and Hoeffler detect 78, while Fearon and Laitin detect 97 civil wars. Furthermore other scholars such as Gleditsch et al. observe in total 111 civil wars in their dataset. These variations point to the ambiguous nature of defining civil war.



The initial dive into the world of “greed and grievance” has given some theoretical weight to conceptualizations on the assumed link between mineral and oil abundance, and civil war. What it has not done however is to give clarity. The civil war literature is indeed a tangled pot of theory and evidence. What we can extract from the literature is the distinction made between greed and grievance in motivating civil war. What is rather harder to do is to draw a clear-cut conclusion on the role of mineral and oil in initiating civil war and large-scale violence. Despite presenting theory positive to the “civil war link”, the end conclusion is not of confirmation, but of ambiguousness. This is mainly due to methodological challenges. Still the ambiguity of the literature seems, rather rightly, to reflect the complexity of the phenomena at hand.


Further reading:

Collier, Paul & Hoeffler, Anke (1998) “On the economic consequences of war”, Oxford Economic Papers. 51: 168–183.

Collier, Paul (2000) “Doing Well out of War: an Economic Perspective”, in Greed and Grievance: Economic Agendas in Civil Wars, (eds) M. Bergdal & D. Malone (London: Lynne Rienner).

John, Di Jonathan (2005) “Oil Abundance and Violent Political Conflict: A Critical Assessment”, Journal of Development Studies. 43: 961–986.

Ross, Micheal (2004) “What Do We Know About Natural Resources and Civil War”, Journal of Peace Research. 41: 337–356.


Cover photo by Brian Harrington, diamond photo by Steve Jurvetson, soldier photo by USMC Archives, lorry photo by Brian Harrington, gemstones photo by Mauro Cateb.




Tags: , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Back to Top ↑