Chapter 4. Fragility and violence
This chapter provides an overview of the results in the individual dimensions of fragility – economic, environmental, political, security and social. The overview starts by setting out the different degrees of fragility within each dimension, including a map of fragility in each dimension in the world today, and an analysis of the links between each dimension and the manifestation of different types of violence. The chapter continues with discussion of the overall statistical correlation between violence and different aspects of fragility, and an examination of the co-existence, linkages and contradictions between fragility and resilience.
Direct forms of political and social violence have formed the basis of indicator selection for the security dimension in the OECD fragility framework. While these forms of violence have clear implications for fragility, focusing on them exclusively provides an incomplete picture. According to the World Health Organization (WHO), violence also includes acts of deprivation and neglect.1 Such violence, often termed structural violence, includes any form within a social structure that prevents some of its members from meeting basic needs (Galtung, 1969). Taking this broader view, structural inequality as a form of violence therefore needs to be included in any measure of fragility. In the OECD framework such forms of violence are included in the economic, political and societal dimensions. The following sections explore each of the five fragility dimensions individually, detailing the typology of fragility within each.
Economic dimension
The economic dimension in the OECD fragility framework is determined by risk factors that can either be structural or of a more temporary nature. These include resource rent dependence, the number of vulnerably employed as a proportion of total employment, aid dependence, and the number of youth not employed or in education or training. More traditional macroeconomic variables such as the size of government debt, the GDP growth rate and the rate of unemployment are also important risk factors of economic fragility.
The coping capacity factors that mitigate these economic risks include human capital (levels of education and employment), the ability of the government to regulate policies to support private sector development and the extent of remoteness from world markets. Food security is also important to support the broader economic environment.
The two principal components that summarise these risk and coping capacity factors can be broadly described as long-term drivers of economic growth and labour market imbalances (Table 4.1). They measure two prominent types of economic fragility related to weaknesses in a country’s development capacity and to the lack of economic opportunities for certain groups of population. The combination of these components leads to the highest levels of economic fragility.
Table 4.2 shows the summary characteristics of the economic dimension.
Sub-Saharan Africa has the highest levels of contexts with economic fragility, often high economic dependence, low individual economic opportunity and weak long-term drivers of economic growth (Figure 4.1).
Note: See Annex A for further details of the methodology.
Economic fragility determined by risk factors involving high economic dependence and low individual economic opportunity in combination with poor conditions for economic growth, is correlated with higher levels of violence (Figure 4.2).
Sources: Homicide data from United Nations Office of Drugs and Crime (UNODC, 2013), Global Study on Homicide 2013, with intentional homicide count and rate per 100 000 population, by country/territory (2000-12). Terrorism data from START Global Terrorism Database (2016), and retrieved from www.start.umd.edu/gtd. Battle-related deaths and non-state actor data from Uppsala University (2016), Conflict Data Program (UCDP), UCDP/PRIO Armed Conflict Dataset v.4-2015, 1946-2014 at http://ucdp.uu.se/. Author calculations.
Figure 4.2 shows distinct trends in different groups of economically fragile settings. Homicide rates or social violence are highest in the group of highly economically fragile contexts. These are typified by high levels of resource and aid dependence and constrained economic geography, but some have relatively moderate levels of individual economic opportunity. There is a clear vicious cycle in the extremely economically fragile group; the absence of long-term drivers of economic growth and individual economic opportunity, coupled with high levels of resource and aid dependence drive violence and conflict, which in turn reinforces economic fragility, and so the cycle continues.
The links between economic conditions and violence in any society are well researched and, like fragility, are highly context specific. The 2011 World Development Report demonstrated that drivers of recruitment to criminal organisations are very similar to those for recruitment to rebel groups and terrorist organisations (World Bank, 2011). A survey in six fragile contexts showed recruits for both rebel and criminal organisations are largely young people driven to join for a combination of economic and identity linked reasons such as unemployment, idleness, lack of respect and self-protection. These reasons typically outnumbered reasons of revenge, injustice or belief in the cause (World Bank, 2011).2
Economic conditions, human capital and expectations also interact in important ways. Motive-based push factors for violence are exacerbated in contexts where youth are highly educated and able to critically analyse the gap between how things are and how they feel things should be. A study by Urdal (2012) suggests that an influx of university graduates into a labour market that cannot absorb them can contribute to the radicalisation effect and recruitment of youth into militant organisations in the Middle East and North Africa (Urdal, 2012). Young people also make up the vast majority of migrants to urban areas, leading to overcrowded schools and saturation of the labour market. In the presence of weak social, political and security coping capacities, these factors can become drivers of political violence.
Environmental dimension
The environmental dimension of fragility is determined by external and internal risk factors including vulnerability to natural disasters risk such as earthquakes, floods, droughts, cyclones or tsunamis. Environmental risk is also measured by the quality of air, water and sanitation, as well as by the prevalence of infectious diseases, the number of uprooted people and the vulnerability of household livelihoods. Coping capacities mitigating these risks include strong civil society, rule of law and government effectiveness, and food security.
The principal components that best summarise these risks and coping capacity indicators can be described as household, community and state vulnerability and natural disaster risks. In many contexts, it is the frequent occurrence of natural hazards combined with high household, community and state vulnerability that increase environmental fragility (Table 4.3).
Table 4.4 shows the summary characteristics of the environmental dimension.
Figure 4.3 shows the fragility in the environmental dimension around the world.
Note: See Annex A for further details on the methodology.
The relationship of violence to environmental fragility is complex (Figure 4.4).
Sources: Homicide data from UNODC (2013) with intentional homicide count and rate per 100 000 population, by country/territory (2000-12). Terrorism data from START Global Terrorism Database (2016), www.start.umd.edu/gtd. Battle-related deaths and non-state actor data from Uppsala University (2016) Conflict Data Program (UCDP), UCDP/PRIO Armed Conflict Dataset v.4-2015, 1946-2014 at http://ucdp.uu.se. Author calculations.
Figure 4.4 shows that homicide and battle deaths are largely the same in contexts with moderate and high environmental fragility, and that there is no distinct relationship between differing levels of environmental fragility and interpersonal violence. However, armed conflict and terrorism are more prevalent in moderate and high environmentally fragile contexts.
It is not surprising that violence can manifest where environmental risks such as infectious disease or natural disasters are present, along with low community and state capacity. However, the relationship between environmental risks and fragility related to violence can be complicated. This is seen in the literature on the link between climate change and conflict. One such study, based on meta-analysis of over 60 studies, finds that the magnitude of climate change’s influence on conflict is substantial and statistically significant at many levels of geographical aggregation. According to this study, one standard deviation increase in temperature or extreme rainfall increases the frequency of interpersonal violence by 4% and increases inter-group conflict by 14% (Hsiang, Burke, and Miguel, 2013). But other studies question this research, underlining the mixed and inconclusive results from scientific research on climate change and conflict (Buhaug et al., 2014).
Climate change is an important factor for environmental fragility, however, as it is closely linked with an increased likelihood of natural environmental threats and hazards. Thus climate change should not be seen as a singular driver of conflict but rather a stressor that may lead to heightened risk of violence and conflict in an already fragile setting. The intersection of weak institutions and social fragility with climate change vulnerability is what is referred to as the climate-conflict nexus. More directly, changing and severe weather patterns multiply the threat of conflict in at least two ways. The first is interruption of resource supply leading to greater resource scarcity (Theisen, Gleditsch and Buhaug, 2013). The second is increased natural disaster risk and its potential to trigger population displacement (IPCC, 2012). Contexts with weak institutions, high levels of poverty and agricultural-based economies are particularly vulnerable to these conflict threat multipliers, and are at an increased risk of falling into the climate-conflict nexus.
A 2016 UNOCHA occasional paper on the climate-conflict nexus identified 20 countries which fall into this gap: Afghanistan, Bangladesh, Burundi, Cambodia, Chad, Democratic People’s Republic of Korea, Eritrea, Ethiopia, Haiti, Kenya, Lao People’s Democratic Republic (hereafter Lao PDR), Madagascar, Mauritania, Mozambique, Myanmar, Niger, Pakistan, Somalia, Sudan and Zimbabwe (Bodanac, Hyslop, and Valente, 2016). This paper describes a situation where social unrest, intergroup grievances and gender-based violence can increase if government is unable to provide the resources needed to cope with a changing environment or destruction from extreme weather conditions, or if international climate change adaptation support is insufficient. This, in turn and in combination with political-economic and societal fragilities, may contribute to violent conflict.
Political dimension
Political fragility in the OECD fragility framework is determined by risk factors such as regime persistence and instability, the presence of state-sponsored violence or political terror, and the levels of corruption. These risks are moderated by coping capacities broadly relating to the quality of political institutions and protections of human rights including, importantly, the legal environment around the protection and rights of women. Combinations of these factors increase instability in political processes, events or decisions and affect the ability to accommodate change and avoid oppression.
The two principal components of political fragility that best summarise these indicators can be broadly described as checks and balances present in political institutions and protection of human rights and political stability. Political fragility thus captures the main drivers of political violence, as well as the mechanisms of accountability and restraint that help prevent the emergence of violence and mitigate its consequences (Table 4.5).
Table 4.6 shows the summary characteristics of the resulting categories of political fragility.
Figure 4.5 shows the resulting political dimensions of fragility.
Note: See Annex A for further details on methodology.
Contexts with centralised state leadership fragility have very low levels of violence, and include the People’s Republic of China (hereafter “China”), Cuba, Lao PDR, the Russian Federation, Saudi Arabia, Thailand and Zimbabwe. The low levels of violence could be statistical distortions; the size of China’s population, for example, has a distorting effect on the rate derived from a per-100 000 population calculation. Despite this, this group’s levels of terrorism, battle deaths and deaths from non-state actors are still low relative to the other categories of political fragility. The presence of strong and centralised political institutions and high levels of state-perpetrated political terror can serve to suppress violence through mechanisms and resources not available to moderately fragile contexts with weaker political institutions. Despite relatively low violence in some forms, these contexts also experience fragility. The centralisation of power can produce systems that are limited in their ability to adapt to evolving situations (Taleb and Treverton, 2015). Further state-sanctioned violence against its citizens is a manifestation of a collapse of state legitimacy, which research has identified as one critical measure of fragility. Furthermore, state repression often forces opposition groups toward other means of expressing dissent including violence (Regan and Norton, 2005).
Figure 4.6 shows that rates of violent death vary across the spectrum of contexts with political fragility. Those with high political fragility have high levels of all types of violence and are in conflict or have a recent history of conflict. In terms of particular types of violence, deaths per capita from terrorism are highest in countries that have some legislative constraints on state power but also high levels of political violence. Such countries include Iraq, Mali, Nigeria and Pakistan. This highlights the strong link between political terror and violence by non-state actors using terrorist tactics. Between 1989 and 2014, almost 90% of all terrorist attacks occurred in countries where violent political terror was widespread (IEP, 2015).
Sources: Homicide data from UNODC (2013) with intentional homicide count and rate per 100 000 population, by country/territory (2000-12). Terrorism data from START Global Terrorism Database (2016), www.start.umd.edu/gtd. Battle-related deaths and non-state actor data from Uppsala University (2016) Conflict Data Program (UCDP), UCDP/PRIO Armed Conflict Dataset v.4-2015, 1946-2014 at http://ucdp.uu.se. Author calculations.
Regimes in fragile contexts can be broadly classed into three types: common interest, redistributive, and weak or failing (Anten, Briscoe and Mezzera, 2012). In the common interest contexts, governments may be willing to increase provision of services but lack the resources to do so. In the redistributive contexts, they may invest in capacity building, but taxes and other resources are collected and redistributed to a select group. Governments in weak or failing contexts lack the ability to build capacity and also do not serve any particular group. As discussed in Chapter 2, conventional indicators of development, such as poverty and inequality, are not on their own predictors of violence. Indeed, many poor contexts are not violent. Large-scale violence, however, does tend to stem from the politicisation of factors such as poverty and inequality (Ncube, Jones and Kayizzi-Mugerwa, 2013; Vallings and Moreno-Torres, 2005). Fragility can manifest in contexts where leaders, elites, and non-state groups and violent actors fall into a cycle of conflict in which violence becomes profitable (Muggah, 2010).
Security dimension
The security dimension aims to capture both the presence of direct violence as well as institutional coping capacities to prevent and mitigate violence. The risk or presence of direct violence is measured by the homicide rate, the level of violent organised crime, the number of deaths from non-state actors, the impact of terrorism, the number of battle deaths from conventional warfare and levels of domestic violence. The coping capacity indicators for security involve the number of police and armed security officers in combination with the presence of the rule of law, and the extent to which the state has control over territory. Also considered is the presence of formal international security alliances, which are associated with lower levels of interstate conflict (Table 4.7).
The statistical analysis of these indicators helps to summarise them into two principal components of security fragility, which can be broadly described as rule of law and state control of territory and armed conflict, terrorism, organised crime and interpersonal violence. The first component is a measure of the level of security fragility; the second distinguishes countries according to their predominant type of fragility. By combining these two aspects, security fragility thus captures citizens’ and society’s security vulnerability that emanate from different forms of violence and crime, including political and social violence (Table 4.8).
Figure 4.7 shows that the extremely fragile contexts in the security dimension include Afghanistan, the Central African Republic, Iraq, Pakistan, Somalia, South Sudan, Sudan, the Syrian Arab Republic (hereafter “Syria”) and Yemen, which are amongst the most violent countries in the world. However, rates are almost as high in the next tier of high security fragile group of contexts, which tend to have a mix of weak rule of law, criminal activity and terrorist activity. This group includes countries such as Colombia and Nigeria. The spectrum of types of violence, and the overlaps between them, coincides with the link between conflict and crime, known as the conflict-crime nexus.
Note: See Annex A for further details on methodology.
Figure 4.8 shows that deaths per 100 000 population from all types of violence are notably higher in extreme security fragility contexts; these situations typify what is described as conflict-crime nexus contexts.
Sources: Homicide data from UNODC (2013) with intentional homicide count and rate per 100 000 population, by country/territory (2000-12). Terrorism data from START Global Terrorism Database (2016), www.start.umd.edu/gtd. Battle-related deaths and non-state actor data from Uppsala University (2016) Conflict Data Program (UCDP), UCDP/PRIO Armed Conflict Dataset v.4-2015, 1946-2014 at http://ucdp.uu.se. Author calculations.
Crime and criminal violence can pose as great a threat to government stability as rebel and armed violence (Stepanova, 2010a). Traditionally within the research literature, crime and armed conflict have been differentiated by their motive, with armed violence considered through a political lens and crime associated with a profit motive (Stepanova, 2010b; de Boer, 2015a). However, in recent years there has been growing recognition that this distinction in many cases is artificial; understanding and addressing the conflict-crime nexus has become an important goal for violence prevention and post-conflict peacebuilding.
The overlap of motivations for conflict and crime is an acute source of fragility in post-conflict situations. Profiteers from armed conflict in these situations often transition into crime. Individuals involved in conflict have an especially difficult time reintegrating into normal life once fighting ceases. For generations that grew up during civil war, violence has become normalised, making it difficult to shift away from it (USAID, 2015). Conflict often leads to a breakdown in education, which has carryover effects in post-conflict societies, in that people who lack formal schooling have few opportunities to lead normal lives outside violent or criminal networks even if they want to do so (Özerdem and Podder, 2011).
The presence of crime during violent conflict also has a number of effects on the conflict itself. The presence of illicit trade allows for a steadier flow of funds, stocks and arms that prolongs the violence. It also has an impact on the intensity of violence (de Boer, 2015a). For politically violent groups, access to illicit flows can reduce the need for public support in order to operate, thus allowing rebel organisations to concern themselves less with the consequences of civilian casualties and to engage in more indiscriminate forms of violence (de Boer, 2015a).
The conflict-crime nexus also poses some serious challenges to conflict prevention, resolution and post-conflict peacebuilding. In many cases, the threat posed to fragile contexts from violent crime is outside of the scope of international humanitarian law, limiting the international community’s ability to respond to ensuing crises (de Boer, 2015a). In post-conflict situations, the separation historically made between conflict and crime has meant that criminal factions that may have played critical roles during the conflict are not included in peacebuilding negotiations (de Boer, 2015a). This lack of acknowledgement offers few alternatives to violent criminal organisations in the post-war period.
The amount of resources devoted to internal security varies widely among fragile contexts; the gap is particularly great between extremely fragile contexts and the rest of the world. This has an impact on their effectiveness and capacity in addressing the conflict-crime nexus. Within fragile contexts there is also variance as some authoritarian countries purchase significantly larger amounts of internal security in order to supress dissent. While security expenditures must be considered within a broader context, low spending on security is correlated to weak rule of law and impunity and the crime-conflict nexus present in many of the fragile contexts identified in the OECD framework (Figure 4.9).
Source: Internal security expenditure data from the International Monetary Fund (IMF). Author calculations.
What is also typical of both the extremely fragile and fragile group of contexts is the presence of weak legal frameworks against gender-based violence. This is important because gender-based violence and gender inequality have been shown to have statistical relationships with security. Gender inequalities are a key manifestation of horizontal inequalities that lead to destabilised societal relations and make societies less resistant to shocks (Baranyi and Powell, 2005). During conflict and/or crisis, the survival of a household can often depend on women’s work in and outside the home.
Societal dimension
The societal dimension in the OECD fragility framework is determined by vulnerability to risks affecting societal cohesion that stem from both vertical (for instance, income inequalities) and horizontal inequalities (inequality among different ethnic, religious, racial or caste groups). Social inequalities related to gender, high urbanisation and large numbers of displaced people in the presence of poor coping capacities also increase societal fragility.
The important variables of societal coping capacity to deal with social and horizontal inequalities are the robustness of civil society, the extent to which citizens have access to justice to address grievances as well the perception of citizen’s access to voice and accountability, meaning the extent to which they can participate in sections of their government, their freedom of expression and the freedom of the media.
The statistical analysis of these indicators generates two principal components of societal fragility, namely access to justice, accountability and horizontal inequality and vertical and gender inequalities. The first component points towards institutional sources of social inequalities, related to the unequal treatment of citizens and discrimination based on ethnicity and other differences between social groups. The second component is determined by inequalities engendered in the economic and/or private spheres, in particular gender-based segregation and income disparities (Table 4.9).
Table 4.10 shows the summary characteristics of the societal dimension.
The results of the societal dimension of fragility are shown in Figure 4.10.
Note: See Annex A for further details of the methodology.
Rates of violence are lowest in the low societal fragility group. Homicide rates are on average five times lower in the low societal fragility group than in the high societal fragility group (Figure 4.11).
Sources: Homicide data from UNODC (2013) with intentional homicide count and rate per 100 000 population, by country/territory (2000-12). Terrorism data from START Global Terrorism Database (2016), www.start.umd.edu/gtd. Battle-related deaths and non-state actor data from Uppsala University (2016), Conflict Data Program (UCDP), UCDP/PRIO Armed Conflict Dataset v.4-2015, 1946-2014 at http://ucdp.uu.se. Author calculations.
Contexts categorised with low societal fragility have lower levels of inequalities (vertical, horizontal and gender), and robust civil society, voice and accountability, and access to justice. Extreme societal fragility sees higher levels of armed conflict fatalities than any other grouping. Many of the contexts that show high fragility in the societal dimension also have fast urbanisation in combination with low levels of accountability and rule of law. This group includes mostly fast-growing sub-Saharan African countries.
Battle-related deaths are higher in contexts in the extreme societal fragility category. These contexts are also typified by high horizontal inequalities and high homicide rates, and high vertical inequalities as measured by income. The correlation between violence and extreme societal fragility is shown in many studies. Horizontal inequality is linked to a number of types of violence (Stewart, 2010; Langer and Stewart, 2013; Brown and Langer, 2010), including communal violence (Mancini, 2005; Fjelde and Østby, 2014); inter-regional inequality and separatist conflict (Bakke and Wibbels, 2006); group mobilisation in civil war (Langer, 2005); the spatial distribution of high-intensity fighting in civil war (Murshed and Gates, 2005); and urban unrest (Østby, 2015; Raleigh, 2015).
Societal factors of gender inequality and gender-based violence (GBV) have been shown to influence overall levels of security. Empirical findings show that levels of domestic gender inequality in political, economic and social spheres are linked to state-level variables concerning security. Research has found that contexts with higher levels of gender equality are less likely to initiate interstate conflict or escalate an interstate dispute (Hudson et al., 2012). Similarly, higher levels of gender equality are associated with a lower risk of civil conflict within a society (Caprioli, 2005). Gender-based violence including intimate partner violence is often a pre-cursor to outbreaks of more endemic conflict. Rape and other forms of GBV are also often weapons of war during conflict, thus perpetuating societal instability. In contexts where GBV and gender inequalities are high, women are also likely to have higher distrust of political institutions and political processes, perpetuating the participation problem (Dryzek, 2014).
Correlates of violence
The links between violence and the other dimensions of fragility are shown in Figure 4.12. While correlations are instructive to form a descriptive, rather than causal, point of view, some broad theories of change are reflected in the pairwise relationships between indicators of fragility and particular types of violence. For instance, as discussed in Chapter 2, homicide is linked more with vertical inequalities (measured by the Gini coefficient), while political violence tends to be more linked to horizontal inequalities (measured by the V-Dem dataset capturing whether social groups distinguished by language, ethnicity, religion, race, region or caste enjoy the same level of civil liberties). Broader patterns are also apparent. For example, organised crime, as a form of violence, and gender physical restrictions (measuring domestic violence and the legislative framework to prevent it) are the most correlated to indicators of fragility in other dimensions.
Fragility and resilience
Early research in the resilience field conceptualised fragility and resilience as “opposite ends of a spectrum”. However, more recent work has emphasised that fragility and resilience actually “co-exist” and their relationship is complex and dynamic, with changes in one not necessarily leading to a commensurate change in the other (de Boer, 2015b).
While resilience has had a long etymological history in fields such as ecology, psychology and engineering, it is a relatively nascent field in its application to social systems, cities or states (Muggah, 2014). Normatively, resilience is used as a positive term used to describe a system’s ability to absorb disturbance and still retain its basic function (Rodin, 2014). The positive implication of the term is reflected in the OECD description of resilience as “the ability of households, communities and nations to absorb and recover from shocks, whilst positively adapting and transforming their structures and means for living in the face of long-term stresses, change and uncertainty” (OECD, 2014).
Increasing resilience would seem, by definition, to always decrease fragility. But increasing resilience could potentially in some cases also increase fragility. A common example of this dynamic occurs in fragile contexts where they receive limited essential services from the state. In such circumstances non-state actors may necessarily perform core state functions. While such informal structures increase the coping capacities of the communities involved, the diversity of service provision without integration within a larger state structure can make response and recovery in the aftermath of a negative shock more challenging. More saliently, these situations can potentially introduce opportunistic competition within the state system, with organisations vying for the monopoly on violence. The rise of so-called violence entrepreneurs not only increases the immediate fragility of the community, but also further erodes the authority and legitimacy of the state (Brock et al., 2012).
With respect to violence, the same measure of fragile contexts can be considered positive or negative, depending on the situation. For example, the presence of state forces can build resilience by maintaining law and order, and their absence could be considered an indicator of a government’s ability to provide security to its citizens. However, the impact on violence is dependent on how state force is applied. Indeed, the politicisation of the provision of security may actually increase the likelihood of violence. There are many such cases where it is unclear whether one factor represents a net positive or negative, thus making analysis of fragility and resilience a challenge.
The relationship between fragility and large so-called “black swan” events is also interesting. Black swan events are large, infrequent and unpredictable events; the 11 September 2001 attacks and the wave of Arab revolutions that started in December 2010 are prominent examples. Such events can be taken as a sign of fragility, but a counter-argument is that the ability to withstand large devastating events is evidence of robustness (Taleb and Treverton, 2015). According to this view, Lebanon, for example, experienced 15 years of civil war and sporadic eruptions of violence and conflict for decades since, but has evolved robust systems in the face of competing claims to power. Proponents of this theory describe a system that benefits from volatility as “anti-fragile” (Taleb and Treverton, 2015).
Indicators of fragility and resilience should not be treated as existing in a vacuum; their net effect can be positive or negative depending on combinations of country-contextual factors. The link between fragility and resilience will continue to be a developing field for researchers and practitioners.
References
Anten, L., I. Briscoe and M. Mezzera (2012), “The political economy of state-building in situations of fragility and conflict: From analysis to strategy”, Conflict, January, www.clingendael.nl/sites/default/files/20120100_cru_political_economy.pdf.
Bakke, K.M. and E. Wibbels (2006), “Diversity, disparity, and civil conflict in federal states”, World Politics, Vol. 59/1, pp. 150.
Baranyi, S. and K. Powell (2005), “Fragile states , gender equality and aid effectiveness : A review of donor perspectives”, North-South Institute, August, http://www.gendermatters.eu/resources_documents/UserFiles/File/Resourse/NSI%20gender%20fragile%20states%20donors%20perspectives.pdf.
Bøås, M., Å. Tiltnes, and H.Flatø (2010), “Comparing the cases”, Background paper for the World Development Report 2011, World Bank.
Bodanac, N., D. Hyslop and R. Valente (2016), “Understanding the climate-conflict nexus from a humanitarian perspective: A new quantitative approach”, OCHA Policy and Studies Series, May, https://docs.unocha.org/sites/dms/Documents/OCHA_OP17_climatechange_online.pdf.
Brock, L. et al. (2012), Fragile States, Polity Press, Cambridge.
Brown, G.K. and A. Langer (2010), “Horizontal inequality and conflict: A critical review and research agenda”, Conflict, Security & Development, Vol. 10/1, pp. 27-55.
Buhaug, H. et al. (2014), “One effect to rule them all? A comment on climate and conflict”, Climatic Change, Vol. 10.
Caprioli, M. (2005), “Primed for violence: The role of gender inequality in predicting internal conflict”, International Studies Quarterly, https://doi.org/10.1111/j.0020-8833.2005.00340.x.
de Boer, J. (2015a), “The crime-conflict ‘nexus’: State of the evidence”, July, http://i.unu.edu/media/cpr.unu.edu/attachment/1665/OC_05-The-Crime-Conflict-Nexus.pdf.
de Boer, J. (2015b), “Resilience and the fragile city”, Stability: International Journal of Security and Development, Vol. 4/1, pp. 1–7, https://doi.org/10.5334/sta.fk.
Dryzek, J.S. (2014), “Political inclusion and the dynamics of democratization”, American Political Science Review, Vol. 90/3, pp. 475–487, https://doi.org/10.2307/2082603.
Fjelde, H. and G. Østby (2014), “Socioeconomic inequality and communal conflict: A disaggregated analysis of sub-Saharan Africa, 1990–2008”, International Interactions, Vol. 40/5, pp. 737-762.
Galtung, J. (1969), “Violence, peace and peace research”, Journal of Peace Research, Vol. 6/3, pp. 167–191.
Hsiang, S.M., M. Burke and E. Miguel (2013), “Quantifying the influence of climate on human conflict”, Science (New York, N.Y.), Vol. 341/6151, https://doi.org/10.1126/science.1235367.
Hudson, V. et al. (2012), Sex and World Peace, Columbia University Press.
IEP (2015), Global Terrorism Index 2015: Measuring and Understanding the Impact of Terrorism, http://economicsandpeace.org/wp-content/uploads/2015/11/Global-Terrorism-Index-2015.pdf.
IPCC (2012), Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation, https://www.ipcc.ch/pdf/special-reports/srex/SREX_Full_Report.pdf.
Langer, A. (2005), “Horizontal inequalities and violent group mobilisation in Côte d’Ivoire”, Oxford Development Studies, Vol. 33/1, pp. 25-45.
Langer, A. and F. Stewart (2013), “Horizontal inequalities and violent conflict: Conceptual and empirical linkages”, Working Paper, No. 14, Centre for Research on Peace and Development, Leuven, Belgium, https://soc.kuleuven.be/web/files/12/80/wp14.pdf.
Mancini, L. (2005), “Horizontal inequality and communal violence: Evidence from Indonesian districts”, CRISE Working Paper, No. 22, Centre for Research on Inequality, Human Security and Ethnicity, University of Oxford, Oxford, UK, https://assets.publishing.service.gov.uk/media/57a08c6ded915d3cfd0013aa/wp22.pdf.
Muggah, R. (2014), “Deconstructing the fragile city: Exploring insecurity, violence and resilience”, Environment and Urbanization, Vol. 26(2011), pp. 1–14, https://doi.org/10.1177/0956247814533627.
Muggah, R. (2010), “Stabilising fragile states and the humanitarian space”, in M. Berdal and A. Wennmann (eds.), Ending Wars, Consolidating Peace: Economic Perspectives, The International Institute for Strategic Studies London.
Murshed, S.M. and S. Gates (2005), “Spatial horizontal inequality and the Maoist insurgency in Nepal”, United Nations University WIDER, Research Paper, No. 2003/43, http://isites.harvard.edu/fs/docs/icb.topic700752.files/Murshed_Gates_Spatial_Inequality_Maoist_Nepal_RDE_9_2005.pdf.
Ncube, M., B. Jones and S. Kayizzi-Mugerwa (2013), “Drivers and dynamics of fragility in Africa”, Africa Economic Brief, Vol. 4/5, pp. 1–16, http://www.afdb.org/fileadmin/uploads/afdb/Documents/Publications/Economic_Brief_-_Drivers_and_Dynamics_of_Fragility_in_Africa.pdf.
OECD (2014), “Guidelines for resilience systems analysis: How to analyse risk and build a roadmap to resilience”, OECD, Paris, www.oecd.org/dac/Resilience Systems Analysis FINAL.pdf.
Østby, G. (2015), “Rural–urban migration, inequality and urban social disorder: Evidence from African and Asian cities”, Conflict Management and Peace Science, Vol. 1/25.
Özerdem, A. and S. Podder (2011), “Disarming youth combatants: Mitigating youth radicalization and violent extremism”, Journal of Strategic Security, Vol. 4/4, pp. 63–80, https://doi.org/10.5038/1944-0472.4.4.3.
Raleigh, C. (2015), “Urban violence patterns across African states”, International Studies Review, Vol. 17/1.
Regan, P.M. and D. Norton (2005), “Greed, grievance, and mobilization in civil wars”, Journal of Conflict Resolution, Vol. 49/3, pp. 319–336, https://doi.org/10.1177/0022002704273441.
Rodin, J. (2014), “The resilience dividend: Being strong in a world where things go wrong”, Public Affairs, https://ssir.org/articles/entry/the_resilience_dividend.
START (2016), Global Terrorism (database), https://www.start.umd.edu/gtd.
Stepanova, E. (2010a), SIPRI Yearbook 2010: Armaments, Disarmament and International Security.
Stepanova, E. (2010b), “Armed conflict, crime and criminal violence”, in Stockholm International Peace Research Institute (ed.), SIPRI Yearbook 2010, pp. 37–60.
Stewart, F. (2010), “Horizontal inequalities as a cause of conflict: A review of crise findings”, World Development Report 2011, background paper.
Taleb, N.N. and G.F. Treverton (2015), “The calm before the storm”, Foreign Affairs, February, Vol. 1–7, www.relooney.com/NS4053/00_NS4053_134.pdf.
Theisen, O. M., N.P. Gleditsch and H. Buhaug (2013), “Is climate change a driver of armed conflict”, Climatic Change, Vol. 117, pp. 6-13.
UNODC (2013), Global Study on Homicide 2013: Trends, Contexts, Data, United Nations Office for Drugs and Crime, www.unodc.org/documents/gsh/pdfs/2014_GLOBAL_HOMICIDE_BOOK_web.pdf.
Uppsala University (2016), Conflict Data Program (UCDP), UCDP/PRIO Armed Conflict Dataset, v.4-2015, 1946-2014, http://ucdp.uu.se.
Urdal, H. (2012), A Clash of Generations? Youth Bulges and Political Violence.
USAID (2015), Organized Crime, Conflict and Fragility, September, http://pdf.usaid.gov/pdf_docs/PA00KPM5.pdf.
Vallings, C. and M. Moreno-Torres (2005), “Drivers of fragility: What makes states fragile?”, Poverty Reduction in Difficult Environments, Vol. 7, http://ageconsearch.umn.edu/bitstream/12824/1/pr050007.pdf.
WHO (2002), World Report on Violence and Health, World Health Organization, Geneva, http://apps.who.int/iris/bitstream/10665/42495/1/9241545615_eng.pdf.
World Bank (2011), World Development Report 2011, World Bank, Washington, DC, https://doi.org/10.1596/978-0-8213-8439-8.
Notes
← 1. Violence is defined by the World Health Organization as “the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment or deprivation” (WHO, 2002).
← 2. Aggregates of the most common responses to two questions: “what is the main reason why young people join rebel groups?” and “what is the main reason why young people join gangs?”, in surveys conducted in Colombia, Côte d’Ivoire, Democratic Republic of the Congo, Mali, Sierra Leone, and the West Bank and Gaza Strip. For details on the survey methodology, see Bøås, Tiltnes and Flatø (2010).