1. How borders shape conflict in North and West Africa

Borderlands in North and West Africa have historically been sites where state control is ephemeral at best due to the practical difficulty of controlling movement in the region. In the last two decades, rebel groups and violent extremist organisations have exploited this weakness to carry out an increasing number of attacks that are essentially staged and launched from a neighbouring country (Radil, Irmischer and Walther, 2021[1]). For example, in early October 2017, a large group of fighters aligned with the Islamic State in the Greater Sahara (ISGS) crossed the Nigerien border and attacked a military patrol outside the village of Tongo Tongo, in the Liptako Gourma, before returning to Mali afterwards. In the ensuing firefight, five Nigerien and four American soldiers were killed, and several others wounded. The Tongo Tongo attack is but one example of a larger dynamic where international borders are routinely crossed by non-state armed groups (Map 1.1). In the first six months of 2021, for example, violent events that caused nearly 60% of fatalities took place less than 100 kilometres from a land border and nearly half of these incidents involved civilians. In response to this increase of conflict in border regions, African states and their international allies have launched several military operations in the hope of crushing transnational insurgents. The more recent of these international efforts is the Takuba Task Force led by French forces between Burkina Faso, Mali and Niger where ISGS and other violent groups are operating.

The growing importance of transnational conflicts in North and West Africa calls for more spatialised approaches that can map how borders shape political violence. While numerous single-case studies throughout the region have shown the salience of borderlands for violent groups, little is known about the overall relationships between political violence and borderlands region-wide. Using spatial analysis, this report examines the increasing prominence of borders and borderlands for state actors and their adversaries in the region (Box 1.1). More specifically, the report maps the changing location of violent events in relation to borders in 21 North and West African countries since the late 1990s (Map 1.2).

Expanding previous efforts by the Sahel and West Africa Club (SWAC) to document the geography and conflict dynamics of the region (OECD/SWAC, 2021[3]; OECD/SWAC, 2020[4]), the report addresses three crucial questions for the future of counter-insurgency operations against violent non-state actors in North and West Africa: Are borderlands more violent than other spaces? Has the intensity of violence increased in border regions over time? Are some borderlands more violent than others? Without answers to such basic questions, taking on more complex issues related to border violence, such as assessing the divergent impacts to civilian populations, considering potential policy responses to such violence, or untangling the root causes of the conflicts that continue to disrupt the region, remains inaccessible at best.

This report examines the presumed link between borders and conflict by analysing whether political violence tends to cluster in borderlands, as recent events in the Liptako Gourma and around Lake Chad seem to suggest, or whether it is rather more spatially dispersed across the region. The wide temporal and spatial perspective adopted in the report facilitates a critical approach to the common assumption that borderlands are inherently more violent than other regions because weak political control at the margins of the state encourages the rise of violent actors facing major military threats in their own country. In contrast to the typical understanding of African borderlands, the report also shows that border regions do not automatically transform into “ungoverned areas” and “safe havens” for rebels and violent extremist organisations.

Using data from the Armed Conflict Location & Event Data (ACLED) Project for 21 states across 23 years shows that nearly 9% of all violent events occurred within 10 kilometres of a border, 15% of such events occurred within 20 kilometres, and 25% occurred within 50 kilometres (Figure 1.1). This relationship also exhibits a classic distance-decay effect. Generally speaking, the further from the border, the fewer violent events are observed. The relationship is punctuated by some exceptions, however, such as the spike in violence observed between 90-99 kilometres due to the existence of large urban centres such as Maiduguri in Nigeria.

The study also confirms that more violent events and fatalities are observed near borders than what could be assumed from their population. In 2018, an estimated 6% of the region’s population lived within 10 kilometres of an international border. Yet, the percentage of violent events within the same distance was more than 1.5 times the percentage of the population (9%). However, once past 30 kilometres, the proportion of population and violent events by distance are quite similar (see Figure 1.1).

The concentration of violence in border regions is not a surprise given the region’s history of political instability. The clarity of the relationship, especially at shorter border distances, suggest that borderlands remain a key spatial locus in states’ struggles to develop and maintain internal sovereignty. The concentration of violence near borders is a reminder that the circulation of money, people, and arms across the region is central to understanding the ebbs and flows of violence from state to state and over time. Yet, the security practices that states develop and deploy to manage affairs in borderlands, which range from border-crossing checkpoints to physical barriers to security patrols to electronic surveillance, are highly spatially variable even along a single border. This means that borderland spaces are neither fully absent of state influence nor likely to be consistently awash in anti-state violence simply because of their distance from centres of state power.

The report examines the role of border regions in the spatial diffusion of violence across North and West Africa since 1997. It shows that borderlands remain central in the shifting of political instability that has affected the region in the last two decades, despite being geographically and politically peripheral. The report also confirms that the location of violence in North and West Africa is highly dynamic over time. For example, most of the major conflict areas of the 1990s are peaceful today and some of the most violent attacks are observed in states that were considered stable a mere 15 years ago.

With this in mind, the study shows that the importance of borders to violence is not an immutable ‘law’ of the region’s political geography. For example, in eras of interrelated conflicts occurring in multiple states, such as the wars in Guinea, Liberia, and Sierra Leone in the late 1990s and early 2000s, borders clearly had a heightened importance for armed groups. The importance of borders for armed groups subsequently declined until 2010 as most of the conflicts of the Gulf of Guinea ended. Since the resumption of major conflicts elsewhere in the region in the early 2010s, borderlands have become increasingly violent, while shifting in space away from the Gulf of Guinea. Both the number of violent events located near borders (Figure 1.2) and their proportion (Figure 5.3) are on the rise, an evolution that is principally due to the degradation of the security situation in West Africa.

Combining quantitative data on the location of violent events with qualitative assessments of the actors involved in these conflicts, the report shows that the political will and military strength of states do shape the development of violent activities in border regions. The report also suggests that border violence is dependent on the ways in which violent non-state actors use borderlands to prepare attacks or mobilise civilian populations. These findings highlight that border conflicts reflect larger political issues, such as the perceived marginalisation of some groups that cannot easily be contained within or ascribed only to the territorial margins of the state. For this reason, such issues can potentially threaten the state’s survival.

The various factors that drive state and non-state actors to go transnational are highly dependent on state and local contexts, which helps to explain why some borderlands are much more violent than others. The two main hotspots of border violence are the Burkina Faso-Mali-Niger tri-borderlands and the Lake Chad basin. In these regions, conflicts are more intense and violent events are more clustered than in the rest of the region (Map 1.3).

This study demonstrates that the political and geographic processes that lead to the concentration of violence in border regions are first anchored in the efforts of states to realise sovereignty within their own borders. Many of the smaller states in the region show a sharp decline in violence as border distance increases due to their size. Others, such as Benin, exhibit a spike at larger distances from borders that reflects the presence of national capitals and other large urban centres. Among the region’s larger states, the relationship between violence and borders is particularly visible in Niger and Cameroon, where insurgencies are focused on border regions. However, the relationship is less clear in Mali and Nigeria because of the multiplication of conflict in several parts of these countries (Figure 1.3).

The significant variations observed across states lead to three important implications. First, the process by which sovereignty ebbs and flows over time is not just a relevant issue for so-called ‘big’ states. Small states can also struggle to project influence across distance as the presence of public authority can wane in borderland regions, even if the absolute distances involved are much smaller than in other states. In Burkina Faso or Liberia, for example, border communities are those least likely to receive visits and resources from agricultural extension services, or benefit from public investments in health or education (OECD/SWAC, 2019[6]; Witinok-Huber et al., 2021[7]).

Second, because violence may be highly localised in terms of the groups or the grievances involved, conflicts can become entrenched within borderlands. Between Burkina Faso, Mali and Niger, for example, at least two major insurgencies have developed, each fuelled by different extremist organisations that skilfully exploit local grievances left unaddressed by central governments. By implication, a state that struggles from episodes of violence along one border may be at risk along others even if the parties and issues involved seem unconnected. This means that investigations of these local cases should also be placed within the larger context of the variable capacity of states to project influence throughout their territory.

Third, while the political processes that typically result in expressions of borderland violence emerge in localised settings, there are also international and regional politics to consider. States do attempt to manage sovereignty together, whether through bilateral co-operative arrangements or through larger regional partnerships. For example, the Lake Chad Basin Commission’s Multi-National Joint Task Force (MNJTF) and the G5 Sahel’s Joint Force both operate as cross-border counter-terrorism security forces in the region. While there is no evidence that these co-operative politics give rise to the spatial patterns of violence presented in this report per se, military forces tasked with combatting armed groups will often find themselves operating in borderlands and thus potentially may contribute to the insecurity that may already be present in such spaces.

The report also shows that non-state actors tend to respond in a timely and opportunistic manner to state offensives against rebellions and violent extremists. The case of Boko Haram and the Islamic State West Africa Province (ISWAP) in the Lake Chad and of Al-Qaeda affiliated groups in the Central Sahel suggests that pressure exerted by one country on one group results in its relocation to a neighbouring country where military capabilities or political will are weaker. Al Qaeda’s move from Algeria to the Sahara-Sahel is clearly visible on Figure 1.4, which represents the proportion of fatalities involving the Salafist Group for Preaching and Combat (GSPC), Al-Qaida in the Islamic Maghreb (AQIM), and the Group for Supporting Islam and Muslims (JNIM) by country from 1999 to 2021. The figure emphasises the growing importance of Mali and Burkina Faso, where most of the fatalities involving Al Qaeda-affiliated organisations are located. By contrast, Algeria represents a negligible proportion of fatalities involving these groups since 2016.

This report addresses borderland violence with an examination of the geography of violent events in the region from January 1997 to June 2021. The report builds on a database of more than 171 000 violent events that have caused 43 000 fatalities collected from ACLED (ACLED, 2021[2]) (Chapter 3). An original aspect of the report is to combine two different approaches to define borderlands, one that applies a fixed distance from borders and one that calculates their accessibility to the rest of the national territory instead. In this second approach, borderlands are defined as the area that is accessible by road in less than four hours from any border crossing of the region, using local transportation and average speeds determined by topography. This definition based on travel times is well adapted to capture the diversity of borderlands in the region and the ability of state forces and non-state actors to travel across borders (Map 1.4).

The report uses the recent Spatial Conflict Dynamics indicator (SCDi) to map the changing geography of borderland conflict in the region (Chapter 4). Specially designed to capture rapid changes in conflict, the SCDi measures both the intensity and spatial concentration of political violence in 6 540 subnational regions or “cells”, from Dakar to N’Djamena and from Algiers to Lagos (Walther et al., 2021[8]). The indicator identifies four spatial typologies of violence that can be used to determine whether conflicts are intensifying or decreasing, and in which regions. The report shows that political violence is on the rise in West Africa and that borderlands are more prone to high-intensity and clustered violence, particularly in the tri-national region between Burkina Faso, Niger and Mali, and around Lake Chad (Map 1.3). The report then examines the particular role that borders and borderlands play in the diffusion of political violence in North and West Africa (Chapter 5). It first considers the macro space-time trends before taking a disaggregated look at the same patterns across the region.

At the regional level, the report maps violent events and fatalities to determine whether they are more spatially clustered in border regions than elsewhere in North and West Africa. It then examines the temporal evolution of violent events and fatalities over the last two decades to identify several waves of border-related violence. Finally, the report identifies which border regions have become the most violent in recent years and discusses the reasons why conflicts between state and non-state actors tend to affect these regions. The local analysis builds on an examination of several case studies where state forces and violent organisations have conducted numerous operations beyond state boundaries, causing significant numbers of violent events and deaths. These cases are the Mali insurgency and its consequences in the Central Sahel since 2012, the Boko Haram insurgency in the Lake Chad region since 2009, and the conflicts around Chad. The last part of the report discusses the policy implications of addressing violence within border regions in North and West Africa (Chapter 6).

The report focuses on all forms of political violence in the region, including military campaigns, rebellions, terrorism, and communal violence. The originality of the report is to focus on how state forces and non-state actors make use of international borders and borderlands to achieve their political objectives. Such transnational conflicts are defined as armed struggles in which there are cross-border military activities and at least one of the actors involved is a non-state group.

Border regions play a growing and decisive role in the evolution of the security situation in North and West Africa. Conflicts have become more intense and violence is more clustered near borders than in the rest of the region. Borderland violence develops when the ability of the state to project influence is confronted with the ability of violent non-state actors to find refuge or mobilise civilians in peripheral regions. States and their challengers both vie to control borderland regions, which are therefore central to the stability of the state despite often being perceived as marginal. Religious extremist groups, in particular, have been successful at using borderlands to impose an alternative political agenda that builds on a strict interpretation of religious law and attempts to address grievances left unresolved by the state.

Sahelian and Saharan countries have developed different strategies to counter transnational organisations, without managing to establish a new societal model that could simultaneously strengthen national cohesion in a multi-ethnic context and serve as an alternative to religious extremism. The last two decades have shown, on the contrary, that states have continued to rely on two strategies: the delegation of power to political and economic allies in peripheral regions and the clientelist distribution of national resources.

On the one hand, national elites have continued to rely on informal arrangements with politically connected traders at the margins of the state. This approach has resulted in the emergence of a parasitical economy that connects borderlands to the core of the government. These informal networks, which extend from entrepot economies such as Benin, Togo or Gambia to the largest markets of the region, make cross-border co-operation and the harmonisation of economic policies difficult. Some state elites have also invested in numerous ethnic or political militias that often represent minority groups or fight for their own interests, such as the Imghad Tuareg Self-Defense Group and Allies (GATIA) and Dan Na Ambassagou in Mali, the Civilian Joint Task Force (CJTF) in Nigeria or the Koglweogo in Burkina Faso.

On the other hand, states have tried to convince religious extremists to abandon armed struggle. At the end of 2021, after nearly 10 years of civil war, the Malian government, for example, allegedly tried to open talks with Al Qaeda-affiliated leaders, a move vigorously opposed by Mali’s ally France. By doing so, political elites were trying to replicate a strategy that has proven successful in the recent history of the region. This strategy has been so successful that, for some rebel movements, the objective of the conflict is not so much to secede from the state than to claim better access to its resources. In Mali, for example, Tuareg rebellions have often been motivated by future political dividends and rebels have artificially inflated their importance in order to provide steady jobs when they were reintegrated in the Malian army (OECD/SWAC, 2021[3]).

The strategy adopted by national governments to rule borderlands by proxy and negotiate with violent extremists is potentially the most perilous for the future of the region for at least two reasons. First, recent years have amply demonstrated the danger of using ethnic or political militias in place of government forces. These militias, which represent one-third of all actors in conflict in 2020, are too often used as an instrument of political disorder by politicians, religious leaders and community strongmen. Second, there is also no guarantee that religious extremists can be co-opted by political elites in the same way as rebels from previous movements. The new generation of religious extremists that thrive on the inability of states to fully control their own territory is less interested in negotiating a larger share of a government’s revenue than it is in destroying its political order. In contrast to ethno nationalist movements, Jihadist groups argue that modern nation-states are incompatible with religious law and that their borders are irrelevant to the community of believers (Walther, forthcoming[9]). In that sense, they pose an unprecedented existential threat to state elites and their informal arrangements in border regions.

It seems unlikely that a more inclusive societal model will be put in place in the region to foster national cohesion. For the time being, initiatives to mitigate the impact of border conflicts by states and their foreign partners are more likely to take one or more of the following forms: a securitisation of borders through new troops and technologies, a reinforcement of the transport infrastructure that links border regions to the rest of the country, and an investment in health, education and other public services in border cities that are sorely lacking at the margins of the state. Each of these initiatives has a strong spatial component that should ideally consider the specificities of border regions and the varying local contexts in which border communities interact with state representatives (OECD/SWAC, 2017[10]).

Designing and implementing place-based policies, in addition to national or sectoral policies, is one of the most effective ways to address the political marginalisation of borderlands while promoting their economic centrality within the region. Such place-based policies are part of a larger agenda that seeks to enhance regional integration by promoting agglomeration economies at the local level, national cohesion at the national level, and regional trade at the international level (World Bank, 2009[11]). Policy makers involved in the resolution of conflict in African borderlands should build on this conceptual framework to promote urban density in border cities, reduce the impact of distance between capital cities and peripheries, and facilitate international exchanges within North and West Africa and beyond (OECD/SWAC, 2019[12]).

Given the current regional focus on military operations, future policy initiatives are likely to focus on strengthening the capacity of governments to defend and monitor their borders. Porous borders have long been recognised as a reality in the region and multiple initiatives have been launched in the last two decades to train African armies against transnational groups and develop multinational forces to operate in borderlands, such as the MNJTF in the Lake Chad basin, or the G5 Sahel and Task Force Takuba in the Central Sahel. Beyond these initiatives, further investments in border monitoring resources and technologies would also be needed to help Sahelian and Saharan countries better secure their borderlands. In other words, the process of regional integration promoted by regional bodies such as the Economic Community of West African States (ECOWAS) and international donors should be accompanied by a more effective securitisation of borders. However, international efforts to transfer technologies that can monitor the transnational circulation of goods and people have proved rather disappointing thus far. Imported technologies have been instrumentalised by political elites and contested by border communities when they threatened local livelihoods. For example, the highly sophisticated border post of Kantchari, between Burkina Faso and Niger, that implemented fingerprint scanners and computer registration systems with biometric databases, was burned down only two years after its inauguration and its agents reverted to paper-based control practices (Donko, Doevenspeck and Beisel, 2021[13]).

While the securitisation of borders will likely remain one of the top priorities of African governments and their international allies, it cannot alone lead to a peaceful resolution of the conflicts in the region. Insurgencies emerge when peripheral communities feel marginalised and the state is unable to maintain national cohesion. Improving transport infrastructure is a necessary step towards restoring the legitimacy of the state. Thus far, border regions remain poorly connected to their respective national territories. For example, more than 60 years after the independence of Chad, Mali, and Niger, there are still no paved roads linking Bardaï, Kidal, Timbuktu or Bilma to the rest of the country (OECD/SWAC, 2019[6]). Several decades of underinvestment in borderlands have contributed to the breakdown of national cohesion and reinforced centrifugal tendencies that have been exploited by violent extremists. Investments in transportation infrastructure and related policies should aim to reduce the isolation that currently separates border regions from other regional and national centres, so as to minimise the inconvenience associated with their geographical marginalisation.

Better border security and more connectivity for borderlands would be meaningful efforts toward reducing borderland conflicts. Attention is also specifically needed within border cities as these are key nodes in the regional circulation of goods and people in both North and West Africa. Yet, these urban centres usually lack the public services that would help them develop both as centres of innovation and commercial hubs. This in turn can fuel a sense of alienation and grievance among borderlanders that can be exploited by extremists. For instance, a lack of medical, social and educational services in border cities significantly weakens the ability of the state to present itself as a productive force in the lives of borderlanders. Development policies that aim to promote peace and security in the region should invest more in border cities, especially given that urbanisation is accelerating across Africa. Efforts to improve the lives of borderlanders, targeted specifically to where they increasingly live and work, could undercut the appeal of extremist groups as an alternative to the state.

A final important point to note is that as violence has surged in some border regions, borderland populations have also been the ones that have primarily borne the consequences. No matter the mix of policy initiatives that may be undertaken across the region to aim to reduce borderland violence, protecting the lives and livelihoods of civilians, rather than only killing insurgents, should be the main concern of African states and their allies (OECD/SWAC, 2020[4]). West African women, in particular, are facing unprecedented levels of targeted political violence, with worrying implications for their participation in the agricultural sphere, pastoral livelihoods and political involvement (Kishi, forthcoming[14]). Violence against women is increasing in border regions, where the control of civilian populations has become one of the key issues at stake between governments and non-state armed groups (Walther, 2020[15]). In the absence of a comprehensive database that would report the gender of both victims and perpetrators of political violence, the amplitude of the violence against women in general and in borderlands in particular remains understudied. Without specific attention to the ongoing and deepening insecurity of borderland populations, especially within the Burkina Faso, Niger and Mali tri-border area and around Lake Chad, none of the efforts described above are likely to be achievable. Protecting the lives and well-being of civilians in the immediate term must become a priority of any longer-term security and development strategies in the region.


[2] ACLED (2021), Armed Conflict Location & Event Data Project, https://acleddata.com.

[5] Dobson, J. et al. (2000), “LandScan: A global population database for estimating populations at risk”, Photogrammetric Engineering and Remote Sensing, Vol. 66/7, pp. 849–857.

[13] Donko, K., M. Doevenspeck and U. Beisel (2021), “Migration control, the local economy and violence in the Burkina Faso and Niger borderland”, Journal of Borderlands Studies, https://doi.org/10.1080/08865655.2021.1997629.

[14] Kishi, R. (forthcoming), “Political violence targeting women in West Africa”, West African Papers, OECD Publishing, Paris.

[3] OECD/SWAC (2021), Conflict Networks in North and West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/896e3eca-en.

[4] OECD/SWAC (2020), The Geography of Conflict in North and West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/02181039-en.

[6] OECD/SWAC (2019), “Businesses and Health in Border Cities”, West African Papers, No. 22, OECD Publishing, Paris, https://doi.org/10.1787/6721f08d-en.

[12] OECD/SWAC (2019), “Regional Integration in Border Cities”, West African Papers, No. 20, OECD Publishing, Paris, https://doi.org/10.1787/f41ef7ab-en.

[10] OECD/SWAC (2017), Cross-border Co-operation and Policy Networks in West Africa, West African Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264265875-en.

[1] Radil, S., I. Irmischer and O. Walther (2021), “Contextualizing the relationship between borderlands and political violence: A dynamic space-time analysis in North and West Africa”, Journal of Borderlands Studies, https://doi.org/10.1080/08865655.2021.1968926.

[15] Walther, O. (2020), “Femmes et conflits en Afrique de l’Ouest”, Notes ouest-africaines, No. 28, OECD Publishing, Paris, https://doi.org/10.1787/d5004dd3-fr.

[9] Walther, O. (forthcoming), “Security and trade in African borderlands – An introduction”, Journal of Borderlands Studies.

[8] Walther, O. et al. (2021), “Introducing the Spatial Conflict Dynamics indicator of political violence”, Terrorism and Political Violence, https://doi.org/10.1080/09546553.2021.1957846.

[7] Witinok-Huber, R. et al. (2021), “Gender, place, and agricultural extension: A mixed-methods approach to understand farmer needs in Liberia”, The Journal of Agricultural Education and Extension, Vol. 27/4, pp. 553-572, https://doi.org/10.1080/1389224X.2021.1880453.

[11] World Bank (2009), World Development Report 2009: Reshaping Economic Geography, World Bank, Washington, D.C., https://openknowledge.worldbank.org/handle/10986/5991.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2022

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.