Content uploaded by Philippe Le Billon
Author content
All content in this area was uploaded by Philippe Le Billon on Jul 24, 2015
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
Disponible en: http://redalyc.uaemex.mx/src/inicio/ArtPdfRed.jsp?iCve=81215371002
Redalyc
Sistema de Información Científica
Red de Revistas Científicas de América Latina, el Caribe, España y Portugal
Le Billon, Philippe
Natural Resource Types and Conflict Termination Initiatives
Colombia Internacional, núm. 70, julio-diciembre, 2009, pp. 9-34
Universidad de los Andes
Colombia
¿Cómo citar? Número completo Más información del artículo Página de la revista
Colombia Internacional
ISSN (Versión impresa): 0121-5612
infcisoc@uniandes.edu.co
Universidad de los Andes
Colombia
www.redalyc.org
Proyecto académico sin fines de lucro, desarrollado bajo la iniciativa de acceso abierto
Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
Natural Resource Types and
Conict Termination Initiatives
Philippe Le Billon
University of British Columbia
Abstract
There is mounting evidence that natural resources can inuence the likelihood, course and
outcome of armed conicts. Much of these relationships depend on the institutional setting in
which the conict and resource exploitation occurs, and the specic characteristics of resources
involved. This paper examines the relevance of two broad resource characteristics—lootability
and legality—for conict termination initiatives. Observing revenue sharing, economic sanction
and military interventions in a total of 26 conicts between 1989 and 2006, the paper suggests
that resource characteristics can affect the effectiveness of resource-related conict termination
instruments.
Keywords
war • natural resources • sanctions • revenue sharing • military interventions
Tipos de recursos naturales e iniciativas para la nalización de conictos
Resumen
Existe creciente evidencia de que los recursos naturales pueden inuenciar las probabilidades,
la trayectoria y el resultado de los conictos armados. Muchas de estas relaciones dependen del
marco institucional en el cual el conicto y la explotación de recursos ocurren, y de las caracterís-
ticas especícas de los recursos implicados. Este documento examina la relevancia de dos grandes
características de un recurso —saqueabilidad (lootability) y legalidad (legality)— para las iniciati-
vas de nalización de conictos. Observando la repartición de ingresos, las sanciones económicas
y las intervenciones militares en 26 conictos entre el año 1989 y el año 2006, este documento
sugiere que las características de un recurso pueden afectar la efectividad de los instrumentos de
nalización de un conicto relacionado con los recursos naturales.
Palabras clave
guerra • recursos naturales • sanciones • repartición de ingresos • intervenciones militares
Recibido el 5 de mayo de 2009 y aceptado el 13 de octubre de 2009.
Philippe Le Billon es profesor asociado de Geografía en
el Instituto Liu de Asuntos Globales de la Universidad de
Columbia Británica, Vancouver, Canadá.
philippe.lebillon@geog.ubc.ca
Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
Natural Resource Types and
Conict Termination Initiatives
Philippe Le Billon
University of British Columbia
INTRODUCTION
Much attention has been devoted to the relationships between natural resource
wealth and armed conicts since the mid-s (Bannon and Collier ; Bal-
lentine and Nitzschke ; Nitzschke and Studdard ). e United Nations
Security Council () has taken an unprecedented number of measures to curtail
access to revenues by targeted groups and help foster a durable transition to peace
(Cortright and López ). Natural resources do not have the monopoly of war
nancing, but this priority reected an upward trend of resource-funded hostili-
ties since the s until the mid-s, and a consensus that insurgent access to
resource revenues tends to prolong conicts (Ross ). is paper reviews the
potential importance of the type of resources involved in hostilities for conict
termination. If specic instruments are more eective for some resources than
others, matching them may improve the chance of ending a conict. e success
or failure of resource-focused instruments may also inform arguments about the
role of dierent types of resources in the prolongation of conicts.
e paper rst provides a brief overview of large- and medium- studies on
natural resources and conict duration, making no general claims about small-
studies. It then discusses potential linkages between resource types and conict
termination, before reviewing the record of the three main types of resource-
focused instruments used for conicts between and . Following a
discussion of their relative eectiveness, the paper concludes with a discussion
of ndings, policy implications and avenues for future studies.
HOW DO RESOURCES INFLUENCE (OR NOT) WARS?
Studies linking natural resources and armed conicts until the late s have
generally sought to examine resource scarcity or environmental degradation
eects, mostly through using comparative case studies (Dalby ). Using re-
source dependence as proxy, Collier and Hoeer (; ) initially argued that
greed was a widespread motivation amongst belligerents, before suggesting that
resources constituted a favorable opportunity, setting the context for violence to
12
LE BILLON
escalate into armed rebellion. ese arguments received widespread attention,
including in the public media and international policy circles. e ndings and
their interpretation, however, have been challenged—notably in terms of the
interpretation of relationship (e.g. resource dependence being also a possible
proxy for grievances), of the characteristic of the variable (e.g. being too broad,
not accounting for some resources such as diamonds, or reecting dependence and
its institutional causes rather than resource wealth), and of the unit of analyses
(e.g. not disaggregating the location of resources and conict), see (Buhaug and
Gates ; de Soysa, ). Several studies have re-examined this relationship
by checking for robustness (Fearon ), using alternative resource variables
(Humphreys ; Ross ), disaggregating types of resources (Lujala et al.
; Rustad et al. ) and conict levels (Buhaug and Rød ).
e picture emerging from these recent studies is that the relationship is
generally less robust than was initially found, except in the case of oil. Both oil
dependence (oil exports as a percentage of ) and oil abundance (rent per
capita) are positively correlated with the risk of war (Fearon ; Ross ).
Humphreys () nds this relationship aected by the level of oil dependence
and abundance (medium dependence and abundance present higher risk), the
phase of the oil cycle (production is riskier than discovery), and the strength of
institutions (oil increasing risk for weak states but reducing it for strong states).
Basedau and Lacher () nd that the increased risk of high oil dependence
is counterbalanced by high abundance. e location of oil and type of conict
also matter, with overlapping conict and oil areas being associated with longer
governmental conicts (over central government), but not with territorial (i.e.
secessionist) ones (Lujala et al. ). e presence of oil in conict areas would
also increase the number of deaths resulting directly from hostilities, whereas the
presence of oil within the country but outside the conict area tends to decrease
it (Lujala ). Besides oil, Lujala et al. () nd that secondary diamonds
(surface or alluvial diamonds) have some eect on ethnic conicts. Snyder
() stresses in this regard that the type of resource exploitation institutions
plays a major role for building political order challenged by the particularities
of lootable resources—with joint private/public exploitation oering the best
outcomes. Le Billon () suggests that high diamonds abundance and industrial
exploitation seem to reduce armed conict occurrence. No statistically signi-
cant relationship was found for timber (Rustad et al. ), while Humphreys
() suggests that agricultural commodities are under-examined and would
be positively correlated with a higher risk of conict; Dube and Vargas. ()
qualifying this relationship as they nd that higher coee prices reduce the likeli-
hood of conict in Colombia—with the opportunity cost eect (leaving coee
production employment for rebellion) trumping the rapacity eect (increasing
13Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
prices exacerbating incentives to illegally appropriate coee rents through rebel-
lion). Narcotics tend to strengthen rebel groups in terms of capacity compared
to states, thereby prolonging conicts (Cornell ).
RESOURCES AND CONFLICT TERMINATION
Of most relevance to this study are analyses of conict duration, risk of renewed
conict, and conict termination processes. Several empirical studies have sug-
gested that the availability of valuable resources in a conict area tends to prolong
hostilities and undermine conict termination eorts (Doyle and Sambanis
; Stedman ; Fearon ; Ross ; Lujala et al. ). Humphreys
() nds, in contrast, that oil or diamond productions are associated with
shorter conicts (for a critique, see Collier and Hoeer ), and highlights
the diversity of mechanisms possibly at play:
Feasibility mechanism
•
, with resources providing a ow of revenues enabling bel-
ligerents to continue ghting;
Balance of power
•
mechanism, with resources prolonging wars when easily acces-
sible to the weaker party, or shortening them when more exclusively accessible
to the stronger party;
Conict premium
•
mechanism, with belligerents opportunistically pursing eco-
nomic agendas in a rewarding resource context;
Fragmentation •mechanism, with economic interests and distrust over resource
revenue sharing within armed groups fostering a breakdown of discipline, al-
legiance switching, or the crowding out of ideologically driven belligerents by
opportunistic ones (see Weinstein ). Hostilities may be prolonged through
more diuse economically-driven violence and undermining peace negotiations,
but also shortened by militarily weakening fragmented groups if they face a more
capable military force;
Peace-buying
•
mechanism, whereby resource revenues can provide an incentive
to participate in and abide by a peace process. Such redistribution may, however,
provide an incentive for further fragmentation among armed groups and diu-
sion of armed violence;
International stakes
•
(or international conict premium) mechanism, whereby
resources can inuence the interests and capacity of regional or international
actors. Interest groups in neighbouring states in particular can economically
benet from a conict and seek to prolong it, while major commercial interests
can also seek to end a conict to protect or access resources; and
Resource enclave
•
(or sparse economic networks) mechanism argues that, since
economic sectors with dense economic linkages across divided communities would
promote conict termination, the enclave nature or sparse economic linkages of
many resource sectors would tend to prolong conicts.
14
LE BILLON
ese mechanisms may have divergent eects on conict duration. Ross ()
nds only weak support for the conict premium mechanism and its negative
eect on the success of negotiated settlement. Humphreys () nds support
for the balance of power, fragmentation (through military weakening) and the
international stakes mechanisms, but no support for the conict premium and
peace-buying mechanisms. Finding a positive statistical association between oil
discovery in conict areas and increased conict duration, Lujala () argues
in support of the conict premium mechanism. Assessing the likely impact of
dierent initiatives on the various mechanisms may assist to conrm this argu-
ment: some initiatives may be more suited to the types of mechanism at play in
a conict, and help realize their potential contribution to conict termination.
Examining the eect of resource dependence on peacebuilding success, Doyle and
Sambanis (, ) nd that these are signicantly and negatively associated,
and suggest that “easily looted resources provide incentives for new wars, which
would reduce the probability of [peace-building] success.” Collier, Hoeer et al.
() do not nd that primary commodity export dependence has signicant
eect on the duration of civil wars; yet they nd that lower commodity prices
reduce conict duration—sanctions able to lower prices through closing ‘open’
markets should thus help shorten conicts. Beyond such contradictory ndings,
several studies examine the sensitivity of conict duration to the characteristics
of the resources, their location with regard to the conict, as well as the type of
conict involved and initiative deployed (see Table ). Using a large- analysis
that spatially disaggregates resources and conict areas, Buhaug et al. ()
nd that the presence of resources within a conict zone increases the duration
of conict—ending hostilities around resource areas should thus be a priority.
Examining the specic case of ‘lootable’ or ‘contraband’ goods, Fearon ()
nds that, compared to a total sample of civil wars between and ,
those in which rebels had access and relied heavily on contraband goods such as
narcotics or gems lasted about ve times longer—cutting access to such resources
should thus also contribute to hastening peace. Based on thirteen case studies
of civil wars involving natural resources in the s, Ross () nds that ac-
cess to resource wealth by rebel groups lengthened eight of them, but shortened
two conicts as a result of military interventions by regional powers and had a
mixed eect on two others as rebel groups defected to the government, in part to
maintain access to resource revenues in the face of mounting military pressure.
Ross’ ndings seem to conrm that access to resource revenues by the weaker
party prolongs conicts; they oer weak support, however, for the argument that
resources oer a nancial incentive to oppose a peace settlement through current
gains on present or future control of resources. Given these studies, one would
expect that resource-related peacebuilding initiatives—such as wealth sharing,
15Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
military interventions and economic sanctions—would contribute to conict
termination and post-conict stability. As discussed below, these initiatives are
not likely to have the same eect within all conict environments—notably with
respect to resource type.
Many studies have examined relationships between conict termination initia-
tives in general and conict settlement. Licklider () nds for a sample of
civil wars between and that the risk of renewed conict is higher for
negotiated settlement than a military victory, thereby suggesting that interven-
tions should aim at facilitating military victory rather than negotiations. Using
a sample of civil wars between and , Collier, Hoeer et al. ()
points out that only military interventions on the rebel side shorten civil war.
Several studies have found that foreign interventions tend to increase the dura-
tion of civil wars (Regan ); although towards negotiated but not military
settlement (Balch-Lindsay et al. ). Regan, Frank et al. () qualify these
ndings by arguing that well-timed diplomacy conducted by some international
actors, alone or combined with other initiatives, can help reduce conict dura-
tion. Besides these general studies, at least four studies have examined conict
Table 1. Major N-studies on resources and conict duration
Authors Coverage Resource measure Finding
Collier et al. 2004 55 civil wars
1960–1999
Pre-conict percent
of primary exports
in GDP.
No signicant effect.
Collier et al. 2004 55 civil wars
1960–1999
Decline in primary
exports prices.
Reduces duration.
Fearon 2004 128 civil wars
1945–1999
Major reliance of rebel
forces on contraband
goods.
Prolongs duration.
Ross 2004 13 civil wars
1990–2000
Resource wealth. Mostly prolongs
duration.
Regan and Aydin 2005 153 civil wars
1945–1999
Availability of
extractable resources
(diamonds and other
gemstones, opiates).
Opiates production
prolongs duration,
gemstone do not.
Humphreys 2005 Post 1960
Fearon and Laitin
dataset
Oil or diamond
production.
Shortens duration.
Binningsbo and Rustad
2008
PRIO-Uppsala Armed
Conicts
1946–2003 dataset
Involvement of
resources in conict.
Prolongs duration,
especially land and
petroleum.
Buhaug et al. 2009 PRIO-Uppsala Armed
Conicts
1946–2001 dataset
Availability of
lootable resources in
conict area.
Prolongs duration.
Lujala 2010 PRIO-Uppsala Armed
Conicts
1946–2003 dataset
Presence of oil and
diamonds in conict
area.
Prolongs duration.
16
LE BILLON
termination initiatives either targeting resources or deployed in a specic re-
source context. Using a selection of cases, Stedman (, ) nds that no
peace agreement in which international actors were assigned a prominent role
on the ground has been successfully implemented where there are “valuable,
easily marketable commodities such as gems or timber,” and suggests that such
resources “not only provide armies with the means for continued ghting, they
also become the reward against which they weight the benets of peace.” In his
study of conict duration and mode of termination, Fearon () notes that
conicts are about four times longer in cases where ethnic minorities took up
arms to protect their access to land, fuel or mineral resources against migrants
or the state. Fearon () suggests that negotiated settlements would be more
problematic in such contexts by increasing the stakes of one party and suspi-
cions about reneging by the other.
Territorial conicts may not be the only ones
prolonged by the availability of resources.
Arguably, if resource revenues are only available to the stronger party, then
resources should not prolong but shorten conicts by providing the stronger party
more economic leverage to achieve a military victory or to ‘buy peace’ (Le Billon
). Humphreys () nds that both oil and diamond export dependence
is associated with shorter wars by making military victory easier, but not by
obstructing negotiated settlements. Oil revenues are generally much more ac-
cessible to governments than to rebels, even if rebel forces can steal oil or extort
protection rents or kidnapping ransoms from oil companies. Diamond revenues
may be as much accessible to rebels than governments, but Le Billon ()
nd that military victory over diamond-funded rebels is also easier, suggesting
that deriving revenues from diamonds could have a deleterious eect on rebel
organizational structure (see also Weinstein ; ; Le Billon ).
Overall, recent studies on resources and conict termination underline the
relevance of commodity-related instruments for ending conicts. ey also
point to the importance of distinguishing between dierent type of resources
and conicts involved, their relative location, as well as the parties accessing
resources and the mechanisms prolonging or shortening the conict. Finally,
they tend to suggest that conicts involving resources are more likely to be suc-
cessfully settled through military victory than negotiated settlements resource,
1 This nding should be qualied, however. Stedman (2001) assesses that two peace agree-
ments were partial successes (Cambodia and Liberia) and that there were as nearly many
peace agreements that were partial successes or failures in the case of available and valu-
able ‘lootable’ resources than in their absence (Cambodia, Lebanon, Liberia, Sierra Leone,
Angola I and II, versus Bosnia, Sri Lanka, Rwanda, and Somalia).
2 Fearon (2004: 284) notes, however, that ‘the business synergies between rebel groups and
drug trafckers are so strong that any rebel group that can avoid destruction long enough
will eventually move into this area.’
17Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
and that policy initiatives should in general seek to deny resource access to the
weaker party rather than accommodate its demands.
RESOURCE TYPE AND CONFLICT TERMINATION INITIATIVES
e literature on conicts and natural resources identies several resource type
dimensions potentially inuencing their relationship, including those aecting
conict duration and preferred mode of conict termination (Le Billon ; Lujala
; Ross ). e most classic dimension in resource and war studies is that
of ‘scarcity,’ measuring the imbalance between ‘supply’ and ‘demand,’ and thus
the motivational element in a ‘struggle’ over ‘strategic’ or ‘vital’ resources. is
imbalance is not set only by the supply and demand for the resource itself, but by
the broader economic and livelihood context (e.g. conditions of poverty). Scarcity
is often interpreted through the categories of renewable and non-renewable,
with much of the literature emphasizing more robust links between large-scale
violence and non-renewable resources than for renewable ones (de Soysa ;
Binningsbo et al. ; eisen ). e argument that ‘abundant’ resource
would motivate conicts, although apparently contrary, builds on some of the
same assumptions, but shifts scale from local to global scarcity (hence a locally
abundant but still highly valuable resource), and often adds the dimension of
dependence.
e most well known dimension in the current literature is the idea of ‘loot-
ability,’ measuring the ease with which a rebel group could access revenue from
this resource. is aspect has been declined through terms such as accessibility
and appropriability, as well as variations on this theme such as obstructability
(i.e. ability to racket through threats of obstruction—see Ross ). All of these
relate to the ‘opportunity’ eect according to which a rebel group operating in a
more ‘opportune’ environment would be more viable. ere are many components
to this lootability: the materiality of the resources, its mode of exploration and
production, its spatial spread and accessibility to its revenues, and (il)legal and
(il)licit character along its value/commodity chain.
Materiality
•
of the resource inuences for example its ease of extraction and
transportation, and its price/weight ratio. e easier to extract and transport,
and the more valuable per weight, the more ‘lootable’ the resource.
Mode of exploration, production, and consumption
•
inuences its accessibility,
notably through labour, technical, and capital input, but also the social rela-
tions and nancial ows that take place around resources along the (global)
commodity chain.
Location, spatial spread and accessibility to its revenues
•
are dened by the com-
ponents presented above, as well as the geographical spread of resources (either
in the form of placer/reserves or suitable socio-ecosystems in the case of cash
18
LE BILLON
crops for example). Depending on the relative capacity of the belligerents to
capture revenues, point source or diuse resources can provide advantages to
some belligerents over others—with diuse resources generally favoring rebel-
lions over governments.
Livelihood impact
•
of the resource reects its importance for the survival of in-
dividuals or groups. is can relate to ‘vital’ resources such as water, not only in
terms of access but also of quality (e.g. mining related pollution).
Legal and licit character
•
is dened, respectively, in judicial and moral terms. What
is illegal by law can be considered licit by the population, such as narcotics at
some production sites, and thereby further advantage an ‘illegal’ actor such as
a rebellion vis à vis a government.
Identity and divisibility • characters dened by how much a resource is identied
by a social group as its own and by the divisible character of resource ownership
and benets, both of which relate in large part to its territoriality, mode of pro-
duction, and revenue ows. ese dimensions are less often examined, notably
by quantitative studies due to its ‘qualitative’ character (although not always
impossible to measure). Beyond simple dimensions of ‘property rights,’ this can
extend for example to cultural rights. e particularity of this dimension is the
indivisible character it can give to resources, especially with respect to land. is
notion of identity is, of course, socially constructed and historically contingent.
It can help explain, as discussed above, the longer duration of many secessionist
or ‘sons of soil’ conicts.
Most of the contemporary resource and conict literature has focused on the
nancial opportunities aorded by resources to belligerents in conict situa-
tions: the legality of a resource, and its accessibility or ‘lootability.’ e legality
of a resource refers to its legal status in domestic and international markets.
is legal character shapes specic opportunities for belligerents. In the case of
an illegal resource, a rebel group is advantaged compared to a government that
risks losing its international legitimacy and associated sources of support if it
engages in tracking. In the case of a legal resource, a government is advantaged
since the market should oer higher prices to a recognized authority rather than
an illegal one.
e accessibility of a resource is dened by the ease with which an armed group
can generate revenues from it, through exploitation, theft, as well as taxation or
extortion (see Table ). Several factors inuence this accessibility. Some relate to
the production and commercial characteristics of a resource: a resource is more
accessible when its exploitation requires less nancial, technological or labour
3 On the potential importance of the specic characteristics and modes of production of
different types of resources on the likelihood, type, or duration of armed conicts, see (Le
Billon 2001; Ross 2003; Snyder 2006).
19Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
inputs, and when the high price per volume ratio facilitates transportation. Other
factors relate to the geographical context and mode of exploitation of a resource:
a resource is more accessible when it is spread over a vast territory, in a terrain
propitious to insurgency, and along an international border, as well as when it
is exploited by high number of businesses vulnerable to protection rackets and
protected by ineective or corrupt security forces (Le Billon ).
Together, the legality and accessibility criteria may be used to dene four
categories of resources: illegal lootables (e.g. narcotics), legal lootables (e.g. al-
luvial diamonds, on-shore oil), legal non-lootables (e.g. o-shore oil), and illegal
non-lootables. Illegal non-lootables could include uranium, which exploitation is
mostly conducted by tightly controlled industrial mines and which trade comes
under regulation through the Nuclear Non-Proliferation Treaty and the national
legislation of Nuclear Suppliers Group members (akin to the voluntary agreement
of participants to the Kimberley Process Certication Scheme). Uranium from the
Democratic Republic of Congo, however, was identied by a group of experts
4 The low traceability of resources can also facilitate trading and the collaboration of business
intermediaries with rebel groups. Difculties over the identication of the origin of diamonds,
for example, delayed the effective application of sanctions and required signicant reforms in
rough diamond trading, notably through the Kimberley Certication Process Scheme.
Table 2. Resource categories and risk of accessibility by rebel forces
Resources
Risk of accessibility by rebel forces
Price range (USD/Kg)Production Theft Extortion
Illegal lootables
Narcotics High High High 5,000–6,000
Legal lootables
Alluvial gems and minerals High High High 20,000–500,000
Timber Medium Medium High 0.1
Agricultural goods Low Medium Medium 1.5 (coffee)
On-shore oil Low Medium High 0.12
Legal non-lootables
Kimberlite diamonds Low Medium Medium 20,000–500,000
Deep-shaft minerals Low Low Medium 2 (copper)
Water dams Low Low Medium n/a
Off-shore oil Low Low Low 0.12
Note: Approximate prices in producing countries during the 1990s, adapted from industry sources,
pers. com. from Gavin Hayman at Global Witness, and Auty 2004. For a discussion of the case of a
water dam, see Johnson Likoti 2007.
20
LE BILLON
as both ‘lootable’ through artisanal exploitation, and somewhat ‘legal’ given the
total absence of control on the site and the porous borders of the country.
Some
commodity-focused instruments may better address specic resources given
these distinctions; I review each type of instrument in turn.
Conict termination is often presented in three stages that are frequently over-
lapping each other: peacemaking consists of initiatives seeking to settle a conict
through negotiations (and possibly military intervention), peacekeeping consists in
preventing further hostilities through a military interposition between contending
parties, and peacebuilding consists in normalizing relations and reconciling contend-
ing parties. Diagram matches these three stages with the three main arguments
relating natural resources and armed conicts (see Le Billon ): the resource
curse argument suggesting that resource dependence negatively aects economic
performances and the quality of institutions—thereby supposedly increasing the
vulnerability of countries to armed conicts; the resource conict (resource dispute)
hypothesis positing that the resource itself, its discovery and its exploitation can
increase the likelihood of conicts and various forms of violence, notably through
disputes over rent allocation as well as the social and environmental impacts of
exploitation; and the conict resource (looting or resource nancing) argument for
which resources shape the opportunities and behavior of belligerents by nancing
their activities, dissociating them from a popular base and securing the support
of external actors. e resulting framework can help situate initiatives targeting
potential resource and conict linkages.
MILITARY INTERVENTIONS, SANCTIONS, AND REVENUE SHARING
ree main types of conict termination instruments targeting resources have
been used: military interventions taking over the control of resource production
areas from belligerents, economic sanctions against targeted belligerents, and
revenue sharing agreements between belligerents.
ese three types of instruments are not the only ones deployed to address
the role of resources in prolonging conicts, as indicated in Diagram . Some of
5 Letter dated 18 July 2006 from the Chairman of the Security Council Committee established
pursuant to resolution 1533 (2004) concerning the Democratic Republic of the Congo ad-
dressed to the President of the Security Council, S/2006/525, accessed at http://daccess-
dds.un.org/doc/UNDOC/GEN/N06/391/16/PDF/N0639116.pdf?OpenElement.
6 The resource curse argument remains debated, with Mehlum et al. (2006) nding that the
resource curse result disappears in a cross-section of countries once the quality of institu-
tions is accounted for.
7 Conict resources are dened by Global Witness, a UK-based NGO, as “natural resources
whose systematic exploitation and trade in a context of conict contribute to, benet from
or result in the commission of serious violations of human rights, violations of international
humanitarian law or violations amounting to crimes under international law,” see
http://www.globalwitness.org/pages/en/denition_of_conict_resources.html.
21Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
these instruments have been used to implement or complement the three main
types of instruments examined in this study, such as the Kimberley Process
Certication Scheme seeking to enable more eective sanctions against ‘conict
diamonds’ (Le Billon ). Some have also been deployed according to conict
prevention or human rights abuse accountability objectives, rather than conict
termination per se. Although rare, litigation against resource companies or in-
dividuals suspected of trading in conict commodities has occurred, in order to
implement and give future credibility to sanctions. Dutch timber merchant
Guus Van Kouwenhoven, for example, received an eight-year prison sentence
for violating a arms embargo imposed on the Liberian government.
Several
transparency instruments, most notably the Extractive Industries Transpar-
ency Initiative, have also been deployed to curb the likelihood of corruption and
revenue embezzlement.
Curtailing wartime access to resource revenues can take two main forms:
military interventions to capture resource production areas; or economic sanc-
tions preventing investments, technical inputs or the trading of resources. e
three main categories of military interventions relevant to this discussion are:
those conducted by domestic forces as part of the general conduct of war; those
conducted by external mercenary forces (or private military companies, s)
working under contract from one of the belligerents; and those conducted by
external military forces under a mandate from the , a regional organization,
or in the form of a ‘coalition of the willing’. ese divisions are not always clear,
especially when external governments intervene outside of a mandate using
foreign mercenaries.
Economic sanctions, or ‘commodity sanctions,’ seek to prohibit the import of
resources under the control of the sanctioned party, an alternative being restrict-
8 The sentence was subsequently overturned by a Dutch court of appeal. ‘Prole: Libe-
ria’s ‘Mister Gus’’, BBC News, 7 June 2006, accessed at http://news.bbc.co.uk/2/hi/
africa/5055442.stm.
Diagram 1. Peace initiatives and resource and conict linkages
Peacemaking Peacekeeping Peacebuilding
Resource curse
Transitional
‘reconciliation’
government
Trusteeship Institutional and
economic reforms
Resource conict /
dispute Resource sharing
Peacekeepers
deployment in
resource areas
Resource entitlement
and management
reforms
Conict resource Demobilization (DDR)
Certication
Sanctions
Military interventions Fair trade initiatives
22
LE BILLON
ing investment for, or export of production technology to, the sanctioned party.
Sanctioned parties have included governments (often through a country-wide
sanction regime) or non-state groups (mostly rebel movements). Sanctions
aiming at particular commodities (commodity sanctions) can be applied to a
country but in eect only target the actor most beneting from that commodity.
Commodity sanction regimes have also been applied to a country, but provide
for the exemption of commodities certied by the government, once a credible
system is in place. Sanctions targeting resources have been imposed by the Secu-
rity Council, regional associations of states such as the Economic Community of
West African States (), or individual governments such as the , as well
as business associations and non-governmental organizations (s)—through
market access restrictions through sectoral reforms or consumer boycotts, as in
the case of the Kimberley Process Certication Scheme for ‘conict diamonds,’
or industry guidelines for coltan in relation to the conict in the Democratic
Republic of Congo. Sanctions have been the most studied among the three types
of initiatives examined in this essay and, as reviewed by Cortright and López
(), ndings generally suggest that sanctions most often fail to change
the behaviour of their targets. Mack and Khan () argue that, nevertheless,
many sanctions had positive impacts in terms of stigmatizing and contain-
ing targets, notably in terms of funding opportunities. Furthermore, the use of
sanctions has much evolved since the end of the Cold War, with a greater use of
sanctions that are better targeted and implemented (Cortright and López
; Le Billon ).
Although sanctions have been imposed to alter the behaviour of their targets,
they have increasingly been used with the objective of curtailing the nancial
means available to rebel groups. e growing use of so-called ‘smart sanctions’
has allowed for selective targeting to maximize the impact on the selected
group, while lowering it on the general population. Measures such as the public
‘naming and shaming’ of sanction busters by expert panels and the threat
or implementation of ‘secondary sanctions’ on sanction busting states have
strengthened their implementation. e National Union for the Total Indepen-
dence of Angola (), for example, lost logistical and diplomatic support in
and , following exposure by expert panel reports. e imposition of
sanctions against the government of Liberia progressively eroded its support for
the Revolutionary United Front in Sierra Leone. Resource smuggling, however,
remained widespread under most sanction regimes, in part because of a lack of
enforcement on the ground and eective judicial action against sanction bust-
ers. e characteristics of resources have also played a role, in terms of ease of
transport, concealment, or low traceability. Yet the smuggling in the s of
vast quantities of petroleum from Iraq or logs from Cambodia illustrate that,
23Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
with the collusion of local authorities on both sides of an international border,
the bulkiness of a resource is less of a major factor.
Positive incentives can address the linkages between resources and wars. Rather
than seeking to curtail revenue access to belligerents, revenue can be made ac-
cessible to ‘former’ belligerents. Sharing resource revenues, in other words, can
‘buy peace.’ is type of positive incentive encompasses a broad range of options.
Resources constitute divisible goods, especially in terms of revenues and to a
lesser degree in terms of ownership (especially if considering state sovereignty),
and are thus amenable to self-enforcing sharing agreements. Divisibility can be
arranged according to territorial, organizational, or commercial criteria. A rst
option is to simply leave the armed group in—at least partial—control of the
territory and resources it is holding, for example as part of a local autonomy or
secession agreement or even as part of a sanction regime as in the case of the
oil-for-food program in Iraq. A second option is to oer the armed group new
resource concessions, the control of resource businesses, or lucrative government
positions overseeing resource sectors. A third option is to establish a broad shar-
ing agreement for resources through scal legislation. In this regard, any conict
settlement could be considered as involving a sharing of resource revenues as
long as opposing parties are allowed to have an input into governing. However,
in this analysis I only consider the cases in which natural resources constituted a
major nancial stake in the conict and in which agreements had an important
resource dimension (although not always incorporated into formal documents,
see below). ese agreements can take place at various levels, concerning an
entire rebel movement as part of a comprehensive peace agreement, or only
regional units as part of a local ceasere or defection process.
As with military interventions and economic sanctions, there are ethical di-
mensions to the use of sharing agreements, since those beneting from these
agreements (or at least negotiating them) include individuals or groups bearing
responsibility for war crimes and occupying positions of power through force
rather than consent and popular representation. Buying peace, in other words,
could be perceived as rewarding violence (Le Billon ). e trade-o is of
course curbing further abuses that could result from the absence of such agree-
ments. Although sanctions and military interventions should have the ethical
advantage of punishing rather than rewarding war criminals, in practice both
also often bring about suering for the general population.
As the cases of Sudan ( Khartoum Agreement), Liberia ( Abuja Ac-
cord), Sierra Leone ( Lomé Agreement), or Angola ( Lusaka Protocol)
illustrate, sharing revenue initiatives face in practice many risks of failure. e
parties to the sharing agreement may not encompass all actors with a capacity
to prolong the conict. e incapacity of a party to enforce the agreement within
24
LE BILLON
its own ranks can lead to a resumption of the conict by new factions rejecting
the agreement. Such agreement can also motivate rebellion in other aggrieved
regions (e.g. in Darfur). Finally, a party can be duplicitous and use such agreement
for tactical use to rearm, reorganize or relocate troops to achieve its objectives
by military means.
ere remains much debate about the eectiveness of these strategies, and more
generally about the use of force, sanctions or negotiations. In the following section
I present an assessment of initiatives conducted since , before discussing
possible factors inuencing the relative eectiveness of these initiatives.
MEDIUM-N STUDY (26 CONFLICTS BETWEEN 1989 AND 2006)
Resource revenues did nance belligerents before the end of the Cold War, but
in most cases their relative importance was much lower given the nancial and
military involvement of foreign powers. Furthermore, the number of inter-
national commodity-focused initiatives only sharply increased from the early
s onwards. I thus select commodity-focused international initiatives taken
between and , identifying armed conicts in which at least one
resource-focused initiative was used. In total, resource-focused initiatives
are surveyed (see also Le Billon and Nicholls ). A number of caveats and
limitations to this dataset need mentioning.
First, with regard to sharing initiatives, I only consider in this analysis the cases
in which a reference to the control of a key resource by the opposing party is
included in a publicly available settlement agreement, as in the options outlined
above. Yet revenues are generally fungible and other types of economic incen-
tives may be oered in addition to, or as a substitute for, resource revenues in a
sharing agreement. Furthermore, not all nancial deals appear publicly and any
condential agreement would not appear in this dataset. e selection criterion
thus reects the diculty of identifying other types of agreements, either be-
cause they are clandestine or because they occur at a smaller scale and fail to be
reported. e sharing initiatives identied were concluded between opposing
armed groups, with the exception of three cases that were set up unilaterally
by central governments, in part to address secessionist agendas (in Angola for
Frente de Libertação do Estado de Cabinda () and in Indonesia for both
the Free Aceh Movement () and the Free Papau Movement ()). Finally,
I assess the eectiveness of these initiatives through only three main criteria:
successful implementation, conict outcome after one year and peace stability
after ve years.
Second, I limit analysis of economic sanctions to those mandated by the
Security Council. is selection is motivated by the fact that sanctions are
currently the sole means of legally and internationally imposing a market access
25Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
denial, with the exception of the prohibition of specic commodities through
international agreements, as in the case of narcotics, or voluntary agreements
and peer monitoring as in the case of the Kimberley Process Certication Scheme.
sanctions can thus be considered more comprehensive than other types
of sanctions and related initiatives, even if they have often been less strictly
implemented than sanctions originating from individual states (for example
sanctions); and advocacy was often key in bringing about a more eective
implementation of sanctions, as in the case of ‘conict diamonds.’ Of the
seven sanction initiatives identied, only one involves a ban on the export of
production material to the targeted group—the Taliban in Afghanistan—since
narcotics were already illegal on the international market. ird, I only consider
external military interventions, including major mercenary interventions and
foreign government-mandated interventions that were publicly reported. Of
the selected military interventions, ve involved international mercenaries
groups (or private military companies).
Among the initiatives examined, external military interventions were the
most frequent, followed by revenue sharing, and sanctions. Four conicts
were addressed through all three types of instruments, and eight through two
types of initiatives. To assess their potential eectiveness in terms of conict
settlement in general, I use three criteria: eective implementation and status
of the conict after one and ve years. Implementation success represents the
achievement of operational objectives, specically: the institutionalization of the
agreement in the case of sharing; curtailment of trade in the case of sanctions;
and control of resource production area in the case of military interventions.
Eectiveness has been assessed through a review of situation reports and
expert panel investigations, as well as think tank, civil society and press reports.
As such, these assessments remain tentative and at times subjective. e one
and ve year lags assess the immediacy and sustainability of a potential eect on
conict termination. I do not argue that peace is the result of the implementation
of instruments, but simply assess the occurrence of both events.
I nd that among the dierent types of instruments, those most successfully
implemented were military interventions and revenue sharing mechanisms,
while sanctions were lagging.
is result is not surprising since sharing involves
willing—if sometimes duplicitous—parties; military intervention is generally
9 As noted above, these results are tentative since they derive from somewhat subjective and
non-standardized measures. Furthermore, in part because of the small number of cases, I do
not control for the inuence of one initiative on the implementation effectiveness of the
other. Arguably military interventions can be more effective following sanctions that have
weakened a party. Such military intervention, in turn, can affect the likelihood of a successful
sharing agreement. In 14 cases only one instrument was used; in 8 cases two were used; in
four cases three instruments were used (Angola-UNITA, Cambodia, Liberia, and Sierra Leone).
26
LE BILLON
used when there are reasonable chances of success, especially in the case of
military interventions by western powers; and sanctions represent a limited
instrument of coercion which has furthermore been criticized for being poorly
enforced and used as a default policy option. When examining the potential eect
of instruments on a resolution of a conict, however, peace was achieved within
a year for about half of the successfully implemented revenue sharing agree-
ments, sanctions, and military interventions. is proportion increases for all
instruments after ve years, but sanctions were associated more frequently with
durable peace than revenue sharing, and military interventions. is suggests
that, whereas military interventions are the most frequently used and success-
fully implemented, their potential contribution to peace seems lower than that
of the two other instruments when these are also successfully implemented.
Military interventions were more successfully implemented against states than
non-state groups, but these successes were more frequently followed by war
than for non-state armed groups. Sanctions seem to have been more success-
fully implemented and followed by peace in the cases where they targeted whole
countries or governments, rather than non-state groups. Sharing agreements all
involved state and non-state groups (or the separatist government).
Turning to the criteria of resource type, Table provides an assessment for the
major dierent resources. Regrouping resources further into three major types, I note
that revenue sharing agreements have been used for all three types. e implementa-
tion success rate of revenue sharing agreements is highest for illegal lootables and
non-lootables, but their association with a stable peace is strong only in the case of
illegal lootables. Sanctions have also been used for all three types of resources, but
mostly for lootable goods. Sanctions failed to be implemented in the only case of
illegal lootable resource, have a low implementation success rate for other lootables
and a medium one for non-lootables. Association with peace stability is also nil for
illegal lootables, but medium for the other two categories. Military interventions
were used for all three types of resource categories. Military interventions were suc-
cessfully implemented in most cases, but most strongly for non-lootable resource, a
result that also appears to be associated with peace stability.
Overall, this survey of instruments indicates that illegal lootable resources have
been most successfully dealt with through revenue sharing mechanisms. ere are
several reasons for such a nding. First, a high level of implication of state ocials
10 Although sanctions move a commodity from a legal to an illegal category, for the purpose of
analysis I maintain the pre-sanction legal status of the commodity. This allows to measure
the effect of the sanctions on commodities with pre-existing systems of trade controls (i.e.
against illegal commodities), from systems specically established against legal commodi-
ties under sanctions. Since the Kimberley Process Certication Scheme was established in
2003, conict diamonds are in effect illegal commodities. I have nevertheless maintained
conict diamonds, even after 2003, in the ‘legal’ category for this analysis.
27Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
in the narcotics sector and the inuence of drug cartels in the aairs of the state
are relatively common within producing countries—as suggested by the extensive
literature on ‘narcostates’—some of these relations providing (in the eyes of some
protagonists) a degree of ‘political stability’ (McCormack ; Le Pichon ). Second,
clandestine operations by state agents have been nanced through illegal but lucrative
sectors, some of which have involved alliances with insurgent groups—as alleged in
the case of the —occasionally leading in the entrenchment of narcotics interests
through continued support for (former) allies (Dale-Scott and Marshall ; McCoy
). ird, narcotics income can play a signicant economic role, especially during
economic downturns, thus leading to acquiescence or even complicity on the part of
authorities to sustain the economy (for the case of Mexico in the late s, see Block
); an argument that also applies during ‘reconstruction.’ Empirically, this nding
is mostly driven by cases of cease-re agreements taking place between the govern-
ment of Burma/Myanmar and several insurgencies involved in drug production and
tracking (Sherman ; see also Snyder, this volume). e case of Afghanistan is
also relevant here. e -led military campaign against the Taliban in seemed
to match a drop in poppy production, suggesting an apparently successful case of
military intervention. Yet the intervention occurred after the Taliban had imposed
a ban, and the military were not tasked with military cracking down on drug
production, tracking, and revenue ows before (Felbab-Brown ). Since
, opium production sharply increased, partly a result of a sharing agreement
11 Nor was major US funding provided for counterdrug activities, contrasting with the USD 750
million dollars allocated in 2003 for the Andean region (see White House 2003 National
Drug Control Update).
Table 3. Conict termination initiatives and resource types (1989-2006)
Implementation Peace after 1 year Peace after 5 years
Resources Share
Sanction
Military Share
Sanction
Military Share
Sanction
Military
Illegal lootables
Narcotics 3/3 0/1 2/2 2/3 — 1/2 1/3 — 0/2
Legal lootables
On-shore oil 2/2 — 3/3 1/2 — 0/3 0/2 — 0/3
Alluvial minerals 6/7 3/3 4/4 4/6 1/3 1/4 4/6 2/2 2/4
Crops — — 0/1 — — — — — —
Timber 5/6 2/2 2/2 2/5 0/2 0/2 4/5 1/2 0/2
Legal non-lootables
Off-shore oil 1/1 0/1 1/1
Deep-shaft minerals — — 1/1 — — 0/1 — — 1/1
Note: gures refer to number of positive cases out of the total number of cases (i.e. initiative
successfully implemented, peace sustained after one year or ve years).
28
LE BILLON
between former warlords now in the Afghan government, with—at the time—the
tacit knowledge (if not agreement) of the government.
In the case of legal lootable resources, sanctions appear to be the most successful
with regard to peace stability, despite a low implementation success rate. Although
sharing appears as a second best option, with a higher rate of implementation
success, all successful cases were only part of broader negotiated peace agree-
ments, rather than prior to a settlement of the conict. Revenue sharing in Papua
New Guinea, for example, was part of a comprehensive agreement signed after
a three-year truce between belligerents. Military interventions to control legal
and lootable resources, while highly successful in terms of implementation, were
very often followed by a resumption of the conict within the next ve years.
In the case of non-lootable resources, military intervention was most successful,
followed by sanctions. is could be explained by the fact that these resources
were controlled by central governments unwilling to respond to sanctions (e.g.
Iraq), or facing politically motivated secessionist groups with which sharing
agreements were not respected or proved unsatisfactory (e.g. Aceh in ,
Chechnya in and Sudan in ).
As discussed in Le Billon and Nicholls , the choice of instruments should
thus not only be dictated by the type of resources, but also by the type of conicts
involved and the mechanisms at play. In turn, the selected instruments could be
articulated with other conict resolution measures so as to sustain or amplify
the positive impacts of instruments on a possible return to peace. Furthermore,
organizations seeking to curtail revenue access for belligerents should also
consider the structure of the industry as well as the capacity and motivations
of intermediaries and authorities along the resource supply chain—as dem-
onstrated in the case of ‘conict diamonds’ and the creation of the Kimberley
Process (Smillie ).
CONCLUSION
is analysis suggests three preliminary ndings of relevance to conict termina-
tion. First, and mostly based on a review of the literature and anecdotal evidence
from the case studies reviewed, the analysis provides qualied support to the
argument that access to resource revenues by belligerents generally prolongs
armed conicts, thereby justifying that conict settlement initiatives should ad-
dress this relationship. is argument is supported by the relatively short delay
within which a number of conicts are settled after resource-focused initiatives
are implemented. Yet renewed hostilities after many of these ‘successful’ interven-
tions indicate that curtailing nancial opportunities is not a panacea. It suggests,
rather, that the importance of resource revenues for the viability and motivation
of rebellion in these conicts may be overemphasized. In this regard, resources
29Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
are rarely the only source of revenues and motivation for belligerents who often
nd ways to adapt their struggle to more dicult economic conditions resulting
from eective resource-focused initiatives (Jean and Run ).
Second, resource-focused initiatives have dierent levels of successful imple-
mentation and potential association with stable peace. Military interventions
appear to be a deceptive ‘quick-x’: often successfully implemented, these appear
to force the targeted party into a settlement, but fail to be followed by a stable
peace. Military intervention would thus require signicant follow-up to avoid the
recurrence of hostilities. In the light of the Angolan and Sierra Leonean cases, the
deployment of peacekeeping forces with weak mandates following interventions
by external mercenary groups targeting rebel-controlled resource production
areas warrants attention in this regard—in particular stronger mandating of
peacekeepers to militarily intervene in resource control (as in the case of
resolution for the in the Democratic Republic of Congo, see Le Billon
). Revenue sharing seems as successful as military intervention in terms of
implementation and is more rapidly followed by conict settlement, but is also
rarely followed by a stable peace. is nding, however, may reect a timing issue
since agreement on revenue sharing is often concluded as part of a settlement
of a conict. Given the asymmetry between belligerents and the risks of duplic-
ity characterizing many of these sharing agreements, third parties may have a
role in guaranteeing these arrangements. Adequately mandated peacekeeping
forces and an international supervising mechanism for the resource sector can
help provide such guarantees. e Secretary General recently stressed the
importance of supporting mediation, notably for wealth sharing agreements, a
task assigned to the Mediation Support Unit ( ). Sanctions have
a poor overall record in terms of implementation for the period examined, but
major improvements have been noted since the late s in terms of monitoring
and enforcement. Sanctions are furthermore generally lifted only once a conict
is comprehensively settled, possibly contributing to a lasting peace.
ird, the characteristics of the resource sectors targeted seem to aect the
eectiveness of these instruments. is nding does not only argue in favour of
contextualising responses, but also points to some of the dilemmas and limits
of resource-focused instruments. Conicts involving primarily illegal lootable
resources seem best addressed by sharing arrangements; legal lootable resources
by sanctions; and non-lootable resources by military intervention. Responding
to conicts related to narcotics poses a dilemma: sharing arrangements are
rarely an ocial option for governments and even less so for conict responding
countries. As noted earlier, however, a number of governments or government
ocials have nevertheless taken this option to secure a conict settlement, to
support local allies, or to reduce levels of violence (see Snyder, this issue)—not to
30
LE BILLON
mention beneting from narcotics revenues at the individual level or to sustain
the economy. ere is a greater chance that at least some government ocials
will be amenable to such wealth sharing if narcotics are deeply entrenched in
the political economy of the area, support numerous livelihoods (with few al-
ternatives) as well as local revenue reinvestments, politicized and portrayed as
‘emancipatory’ (from poverty or specic elites), and if the state is too ‘weak’ to
impose anti-drugs policies through coercive means or incentives, and the state
is isolated from major donors and sources of capital investment.
e above analysis is admittedly tentative, as it does not address the many
other conditions that aect the settlement of a conict and the likelihood of
war recurrence. In the absence of a multivariate analysis that controls for these
other factors, the ndings of this article should be treated as hypotheses for
further investigation, rather than a demonstration of causal links between con-
ict resolution mechanisms and outcomes. ree further specic studies could
be conducted. e rst is a large- study encompassing all conicts involving
resources since . Such a study is currently being undertaken by Rustad et
al. () using the /Uppsala conict dataset—with preliminary results
suggesting that the type of resource (lootable and non-lootable) does aect sta-
tistically the duration of ‘post-conict’ peace. Wealth sharing appears to be an
eective peacebuilding tool when applied to conicts involving lootable resource.
So far, results have not reproduced ndings from (Le Billon and Nicholls )
and more work is needed to consider the impact of sanctions and the legality of
resources. e second study would consist of a more detailed comparative analysis
of individual case studies, so as to determine more precisely the relative impact
of resource-focused instruments. Whereas sanctions have been the object of
much attention, this has not been the case for military interventions and wealth
sharing initiatives. Such study could be extended to other instruments, such
as certication schemes, judicial measures, and corporate social responsibility
measures. e last study would be a comparative analysis of both international
and domestic conict resolution initiatives across a variety of sectors within
one or several countries. e instruments examined here represent only some
of the initiatives taken to address connections between resources and conicts,
and such a detailed study would broaden scope and depth.
More generally, a more precise analysis would result from a standardization of
the assessment of instrument eectiveness (e.g. standard questionnaires sent to
conict specialists). A more comprehensive approach would require an examina-
tion of the eectiveness of military interventions by domestic groups; a more
detailed dierentiation of the types of economic sharing agreements, including
at dierent scales; as well as an examination of regional and unilateral sanction
regimes. Future analyses could also examine the inuence of the timing and
31Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
complementary of these various initiatives, as well as the inuence of resources
on the capacity and will of external interveners, including the question of com-
mercial interests among interveners (Balch-Lindsay and Enterline ). Finally,
further research could focus on the means by which to establish a credible and
enforceable sharing agreement, and examine how credibility and enforceability
might vary according to the type of resources involved. Learning more about the
context in which conict termination instruments are deployed may improve
their eectiveness and reduce the risk of renewed conicts.
References
Auty, Richard. . Natural resources and civil strife: A two-stage process.
Geopolitics (): –.
Balch-Lindsay, Dylan, and Andrew J. Enterline. . Killing time: e world
politics of civil war duration, –. International Studies Quarterly
(): –.
Balch-Lindsay, Dylan, Andrew J. Enterline, and Kyle A. Joyce. . ird-
party intervention and the civil war process. Journal of Peace Research ():
–.
Ballentine, Karen, and Heiko Nitzschke. . Proting from peace: Managing
the resource dimensions of civil war. Boulder: Lynne Rienner.
Bannon, Ian, and Paul Collier . Natural resources and violent conict: Options
and actions. Washington: World Bank.
Basedau, Matthias, and Wolfram Lacher . A Paradox of Plenty? Rent
Distribution and Political Stability in Oil States. German Institute of Global
and Area Studies Working Paper Series .
Binningsbo, Helga M., Indra de Soysa, and Nils Peter Gleditsch. . Green
giant, or straw man? Environmental pressure and civil conict, -.
Population and Environment (): –.
Block, Alan A. . On the inestimable value of the . Crime, Law and Social
Change : –.
Buhaug, Halvard, and Scott Gates. . e geography of civil war. Journal of
Peace Research (): –.
Buhaug, Halvard, Scott Gates, and Päivi Lujala. . Geography, Rebel Capacity,
and the Duration of Civil Conict. Unpublished manuscript.
Buhaug, Halvard, and Jan Ketil Rød. . Local determinants of African civil
wars, –. Political Geography (): –.
Collier, Paul and Anke Hoeer. . On economic causes of civil war. Oxford
Economic Papers : –.
32
LE BILLON
———. . Greed and grievance in civil war. Oxford Economic Papers ():
–.
———. . Resource rents, governance and conict. Journal of Conict
Resolution (): –.
Collier, Paul, Anke Hoeer, and Måns Söderbom. . On the duration of civil
war. Journal of Peace Research (): –.
Cornell, Svante. . Narcotics and armed conict: interaction and implications.
Studies in Conict and Terrorism (): –.
Cortright, David, and George A. López. . e sanctions decade: Assessing
strategies in the s. Boulder: Lynne Rienner.
———. . Sanctions and the search for security: Challenges to action.
Boulder: Lynne Rienner.
Dalby, Simon. . Environmental Security. Minneapolis: Minnesota University
Press.
Dale-Scott, Peter, and Jonathan Marshall. . Cocaine politics: Drugs, armies,
and the in Central America. Berkeley: University of California Press.
Dube, Oeindrila. and Juan F. Vargas. . Commodity price shocks and civil
conict: Evidence from Colombia. Documentos de .
Doyle, Michael W., and Nicholas Sambanis. . International peacebuilding:
A theoretical and quantitative analysis. American Political Science Review
(): –.
Fearon, James D. . Why do some civil wars last so much longer than others?
Journal of Peace Research (): –.
———. . Primary commodities exports and civil war. Journal of Conict
Resolution (): –.
Felbab-Brown, Vanda. . Afghanistan: When counternarcotics undermines
counterterrorism. Washington Quarterly (): –.
Humphreys, Macartan. . Natural resources, conict, and conict resolution:
Uncovering the mechanisms. Journal of Conict Resolution (): –.
Jean, François, and Jean-Christophe. Run, eds. . Économie des guerres
civiles. Paris: Hachette.
Johnson Likoti, Fako. . e military intervention in Lesotho:
peace mission or resource war? International Peacekeeping (): –.
Le Billon, Philippe. . e political ecology of war: Natural resources and
armed conicts. Political Geography (): –.
———. . Fueling war or buying peace: e role of corruption in conicts.
Journal of International Development (): –.
———. . Getting It Done: Instruments of Enforcement. In Natural Resources
and Violent Conict: Options and Actions, ed. Ian Bannon and Paul Collier, –.
Washington: World Bank.
33Colomb ia Internac ional 70, julio a diciembre de 2009: 9-34
NATURAL RESOURCE TYPES AND CONFLICT TERMINATION INITIATIVES
———.
. Fuelling war: Natural resources and armed conflicts. London:
Routledge.
———. . Diamond wars? Conict diamonds and geographies of resource
wars. Annals of the Association of American Geographers (): –.
———. . Natural resources, peacekeepers and contracts: improving resource
control and management in ‘post-conict’ situations. Working Paper, Liu
Institute for Global Issues, .
Le Billon, Philippe, and Eric Nicholls . Ending ‘resource wars’: Revenue
sharing, economic sanction or military Intervention? International Peacekeeping
(): –.
Le Pichon, ibault. . Drug Tracking as a Security reat in West Africa.
Vienna: United Nations Oce on Drugs and Crime.
Licklider, Roy. . e consequences of negotiated settlements in civil wars,
–. American Political Science Review (): –.
Lujala, Päivi. . Classication of natural resources. Edinburgh: Joint
Session of Workshops.
———. . Deadly combat over natural resources: Gems, petroleum, drugs,
and the severity of armed civil conict. Journal of Conict Resolution ():
–.
Lujala, Päivi, Nils Peter Gleditsch, and Elisabeth Gilmore. . A diamond
curse?: Civil war and a lootable resource. Journal of Conict Resolution ():
–.
Lujala, Päivi, Jan Ketil Rød, and Nadja ieme. . Fighting over oil: Introducing
a new dataset. Conict Management and Peace Science (): –.
Mack, Andrew, and Asif Khan. . e ecacy of sanctions. Security
Dialogue (): –.
McCormack, Robert J. . Caribbean sovereignty and the war on drugs:
Historical factors and current perspectives. Caribbean Journal of Criminology
and Social Psychology (/): –.
McCoy, Aalfred.W. . e Politics of Heroin: Complicity in the Global Drug
Trade. Chicago: Lawrence Hill & Co.
Mehlum, Halvo, Karl Moene, and Ragnar Torvik. . Institutions and the
resource curse. Economic Journal : –.
Nitzschke, Heiko, and Kaysie Studdard. . e legacies of war economies:
challenges and options for peacemaking and peacebuilding. International
Peacekeeping (): –.
Regan, Patrick. M. . ird party interventions and the duration of intrastate
conicts. Journal of Conict Resolution (): –.
Regan, Patrick. M., and Aysegul. Aydin. . Diplomacy and other forms of
interventions in civil wars. Journal of Conict Resolution (): –.
34
LE BILLON
Regan, Patrick M., Richard Frank, and Aysegul Aydin. . Diplomatic interventions
and civil wars: A new dataset. Journal of Peace Research (): –.
Ross, Michael. . Oil, drugs, and diamonds: How do natural resources vary
in their impact on civil war? In e political economy of armed conict: Beyond
greed and grievance, ed, Karen Ballentine, and Jake Sherman, –. Boulder:
Lynne Rienner.
———. . How do natural resources inuence civil war? Evidence from thirteen
cases. International Organization (): –.
———. . A closer look at oil, diamonds, and civil war. Annual Review of
Political Science : –.
Rustad, Siri, Helga Binningsbo, and Philippe. Le Billon. . Do resource-
related peacebuilding initiatives build peace? Paper presented at the annual
conference of the International Studies Association, February , in New York,
United States.
Rustad, Siri, Jan Ketil Rød, Wenche Larsen and Nils Petrer Gleditsch. .
Foliage and ghting: Forest resources and the onset, duration, and location
of civil war. Political Geography (): –.
Sherman, Jake. . Burma: Lessons from the cease-res. In Beyond greed and
grievance: e political economy of armed conict, ed, Karen Ballentine and Jake
Sherman, –. Boulder: Lynne Rienner.
Smillie, Ian. . What lessons from the Kimberley process certication scheme?
In Proting from peace: Managing the resource dimensions of civil war, ed. Karen
Ballentine and Heiko Nitzschke, –. Boulder: Lynne Rienner.
Snyder, Richard. . Does lootable wealth breed disorder? A political economy
of extraction framework. Comparative Political Studies (): –.
de Soysa, Indra. . Ecoviolence: Shrinking pie, or honey pot? Global Environmental
Politics (): –.
Stedman, Stephen John. . Implementing peace agreements in civil wars:
Lessons and recommendations for policymakers. Policy Paper Series on
Peace Implementation.
eisen, Ole Magnus. . Blood and soil? Resource scarcity and internal armed
conict revisited. Journal of Peace Research (): –.
SG. . Report of the Secretary-General on enhancing mediation and its
support activities. //, April .
Weinstein, Jeremy. . Resources and the information problem in rebel
recruitment. Journal of Conict Resolution (): –.
———. . Inside rebellion: e politics of insurgent violence. Cambridge:
Cambridge University Press.
• • •
