PDF Natural Resources, Conflict, and Conflict Resolution ...

[Pages:36]Natural Resources, Conflict, and Conflict Resolution: Uncovering the Mechanisms

Macartan Humphreys1 March 2005

1 macartan.humphreys@columbia.edu My thanks to Nicholas Sambanis, Iain Lustik, Robert Bates, Doron Hadass, Bryan Graham, David Hecht, Michael Ross, Jim Fearon, Habaye ag Mohamed, and Jeremy Weinstein for generous help with ideas and with data and to Brendan McSherry for terrific research assistance.

1

Abstract The interpretation of the resource-conflict link that has become most publicized--the rebel greed hypothesis--depends on just one of many plausible mechanisms that could underlie a relationship between resource dependence and violence. In this paper, I catalogue a large range of rival possible mechanisms, highlight a set of techniques that may be used to identify these mechanisms, and begin to employ these techniques to distinguish between rival accounts of the resource-conflict linkages. To do this I use finer natural resource data than has been used in the past, gathering and presenting new data on oil and diamonds production and on oil stocks. Using this data I find evidence that (1) Conflict onset is more responsive to the impacts of past natural resource production than to the potential for future production, supporting a weak states mechanism rather than a rebel greed mechanism; (2) the impact of natural resources on conflict can not easily be attributed entirely to the weak states mechanism and in particular, although there is some evidence that the effects of natural resources are exacerbated when states are weak, the impact of natural resources is independent of state strength; (3) the link between primary commodities and conflict is driven in part by agricultural dependence rather than by natural resources more narrowly defined, a finding consistent with a "sparse networks" mechanism; (4) Natural resources are associated with shorter wars, and natural resource wars are more likely to end with military victory for one side than other wars. This is consistent with evidence that external actors have incentives to work to bring wars to a close when natural resource supplies are threatened. It is also consistent with weak organizational structures mechanisms. I find no evidence that resources are associated with particular difficulties in negotiating ends to conflicts, contrary to arguments that loot-seeking rebels aim to prolong wars.

2

1. INTRODUCTION

In early April 1975 President Tombalbaye of Chad appealed on national radio for popular vigilance, warning that members of the army were plotting a coup against him. He explained that if anyone wanted to know why a coup was being plotted, that the answer lay with the oil in the Doba fields in the south. This appeal turned out to be Tombalbaye's last public address. On 13 April he was killed during Chad's first successful coup d'etat.2 Unfortunately for Tombalbaye, he had had increasingly strained relations with France, his chief military backer, ever since he allowed US corporations to prospect for oil in the ex-colony. The US corporation was successful where French prospectors had failed. And France took umbrage.3 Ever since, oil has had a striking prominence in the intrigues of Chadian politics. The government of Hiss?ne Habr? is reported to have received US support in exchange for his support of US oil corporations while politicians and NGOs argue that the present president, Idriss D?by, previously Habr?'s right hand man, was offered military support from France's Elf if he would overthrow the Habr? r?gime and give France a stake in the southern oil fields.4 The oil in the south has also been seen to be fuelling southern ambitions for a federalist state, if not for outright separation. The leader of the present Southern opposition, Ngarlejey Yorongar has been incarcerated for accusing the

2 This explanation is provided by Mohammet Sally, a leader of FROLINAT, the dominant Chadian rebel group at the time (Interviews, N'djamena, April 2003). The leader of the coup, Wadal Abdelkadar Kamougu?, denies any connection between the coup and Doba oil (Interviews, N'djamena, April 2003). 3 On French frustration see Nolutshungu (1996). Jacques Foccart (1999) describes how by 1969 the involvement of US corporations in Chad was seen as an insult to the prestige of the French army.

3

previous southern leader, Wadal Kamougu?, of allowing himself to be co-opted by the government in exchange for money from Elf.5 In short, in the eyes of Chad's political leaders, control of oil revenues has been central to Chadian politics for almost thirty years; it has made and broken political leaders, has incited violence, and has shaped political agendas.

But what is perhaps most striking about Chadian oil is that, up to 2003, not a single drop had been pumped.

The role of oil in Chad's politics illustrates some of the complexity of the linkages between natural resources and conflict. Contrary to popularized images of resource conflicts in Africa, oil did not lead to rival warlords establishing resource-funded local monopolies in Chad--the stories that link resources to conflict here center on gaining tight control of the state rather than on the creation of chaotic environments. And, while natural resources may produce conflict by leading to shadowy states, the weakness of Chad's state structure cannot be attributed to its dependence on oil revenues to buoy it; until recently none has been earned. Chad's experience comes closer to what Michael Ross (2002) terms a "booty futures" story--in which resources matter because revenues can be raised in advance to gain control of them. In such a context, policy prescriptions emanating from conflict research--that rebel financing needs to be cut off through the infiltration of quasi-criminal trade routes and robust opposition to rogue states--fall wide of the mark.

4 This claim, emanating apparently from the leader of the opposition, Yorongar, has been argued by Agir ici ?Survie (1999) and Verschave (2000). And, according to Verschave (2000) D?by also had at least tacit military support from the French troops stationed in N'djamena.

4

The problem is that unless we understand the mechanisms linking resources to conflict, the advice of conflict scholars will be of limited use to the policy community. And unless we test these mechanisms we will be unsure of the generality of processes we observe in individual cases, such as those seen in Chad. In this paper I explore the diversity of mechanisms that may link natural resources to conflict, and develop and employ strategies to identify which mechanisms are likely to be in operation when. The paper is organized as follows. Drawing largely on the experiences of Sahelian and West Africa states as well as work by scholars who have studied a wider set of cases and identified relevant mechanisms (notably, Le Billon 2001, Snyder 2002, Ross 2002, 2003, 2004a) I catalogue a series of mechanisms that may link natural resources to conflict onset and conflict duration. With the aim of engaging with the econometric literature, the task then is to find ways econometrically of differentiating between the effects of these rival mechanisms. I discuss four strategies for doing so, indicating how the different strategies may be used for the problems at hand. In the final section I turn to the data, building on the model developed by Fearon and Laitin (2003), I begin the task of identifying the work of rival mechanisms on conflict onset and duration and testing a core set of the relations discussed in previous sections.

2. NATURAL RESOURCES AND CIVIL WARS: AN EMBARRASSMENT OF MECHANISMS Highly influential research by Paul Collier and Anke Hoeffler at the World Bank6 suggests that countries whose wealth is largely dependent on the exportation of primary commodities ? a category that includes both agricultural produce and natural resources ? are highly prone to civil

5 Agir ici ?Survie (1999) and interviews with Yorongar (N'djamena, April 2003).

5

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download