Executive summary

Violence is one of many factors that can contribute to fragility. However, it is not the only factor, and the presence of violence does not automatically mean that a context is fragile. States of Fragility 2016 places a spotlight on violence, in all its forms, to explore how violence can contribute to fragility, and examine what should be done about it.

Main findings: Violence

First, a disclaimer. Data on conflicts are contested, and the complex and changing nature of social violence makes its forms even more difficult to define and measure. More data on the gender dimensions of violence and conflict are also needed, given the disproportionate impact of violence on women.

Despite this, there is agreement that violence is on the rise. Almost half the world’s people have been affected by some form of political violence over the last 15 years. However, globally, conflict is not the leading cause of violent death. In 2015, more people died violently in countries outside of conflict, including Brazil and India, than in the Syrian Arab Republic (hereafter “Syria”), in absolute numbers. Low- and middle-income countries bear a disproportionately high share of the burden of political and social armed violence, which often impedes development gains.

Violence is multidimensional, complex and evolving. Even in post-conflict contexts, violence can simply change its form as settings, actors and drivers change, challenging the international community’s best intervention efforts. Political transitions, even towards democracy, can provoke violence, with competition over “who sits at the table” and “who gets what”. Armed agents who engage in political violence during wars or periods of domestic instability are highly likely to be involved in organised crime in more peaceful periods, including racketeering, mercenary activity and illicit trafficking.

Violence is increasingly a regional problem. Political armed violence spreads more easily as neighbours’ domestic instability spills over borders, driven by linkages between aggrieved groups and facilitated by global communications, shared ideologies and financial flows. The organised crime-political violence nexus allows political armed groups to finance themselves through proceeds from criminal activities, with illegal resource exploitation, illicit financial flows and the drug trade providing revenue. Violence also drives millions of refugees from their homes, its impacts extending to overburdened neighbouring countries that often are already in distress.

Violence is increasingly driven by domestic political instability. Weak institutions or those with entrenched patronage systems can create vacuums in which elites are able to siphon off public resources with impunity while also perpetuating economic exclusion. Criminal networks and armed groups can also fill these vacuums. Members of excluded groups are more likely to engage in armed violence, particularly if they have recently lost access to power. Conversely, rivalry for inclusion can also be a problem: the distribution of positions, authority and resources among included elites can also drive domestic political instability.

In both political and social forms of violence, civilians are most at risk. Today 30% to 40% of political violence within states is directed against civilians. Weapons of armed violence designed to produce civilian casualties are increasingly available. In 2015, 43 786 people were killed or injured by improvised explosive devices (IEDs); 76% of those were civilians. Among civilians, women, youth and children in particular disproportionately bear both direct and indirect consequences of violence. Research shows that children who have been subjected to violence are more likely to become violent themselves. Globally, an estimated 35% of women have experienced either physical and/or sexual intimate partner violence or non-partner sexual violence. Over half of the 21.3 million refugees in the world today are children under the age of 18.

Rapid and unregulated urbanisation, income and social inequality, concentrated poverty, youth unemployment, policing and justice deficits, and real and perceived insecurity can all contribute to urban violence. Within cities, violence is unevenly distributed, and particularly acute in lower income informal areas. In Bogota, for example, roughly 98% of all homicides occur in less than 2% of street addresses. Several contexts most affected by fragility and conflict will see rapid population growth in their cities before 2030.

Terrorism-related deaths rose by 61% in 2013, with 18 000 people killed in terrorist attacks globally. Most victims were in Afghanistan, Iraq, Nigeria, Pakistan and Syria. Women and girls are particular victims of this form of violence. The rise of violent extremist networks claiming Islamist ideology has led to politically motivated and criminal violence across a vast territory. However, extremist groups come in many forms, as evidenced by a surge in violent hate crimes across Europe and North America.

The global economic impact of violence was a daunting USD 13.6 trillion in 2015, equivalent to 13.3% of global gross domestic product (GDP) or USD 1 876 for every person in the world. Yet development assistance invests only marginally in violence reduction outside of conflict.

Violence is a behaviour reinforced by social norms, and it acts like a contagion. By using the same approach that the World Health Organization (WHO) uses to stop epidemics - interrupt transmission, change behaviour, change norms – policy makers can break the recurrent cycle of violence, stopping its “transmission”.

The fragility framework: Fragility in the world today

It is now widely recognised that fragility is multidimensional and its challenges are universal. The OECD is therefore committed to a universal, multidimensional fragility framework. Fragility is defined as the combination of exposure to risk and insufficient coping capacity of the state, system and/or communities to manage, absorb or mitigate those risks.

The new OECD fragility framework is built on five dimensions of fragility – economic, environmental, political, societal and security – and measures each of these dimensions through the accumulation and combination of risks and capacity.

Using the fragility framework, 56 contexts are identified as fragile in 2016, with 15 of those classified as extremely fragile. Over 1.6 billion people, or 22% of the global population, currently live in these fragile contexts. Of this group, 27 contexts are low income, 25 are lower middle income and 4 are upper middle income. While the number of people living in extreme poverty will fall globally, the number of extremely poor people living in fragile contexts will increase to 542 million in 2035 from 480 million in 2015.

There are interlinkages between the dimensions of fragility and violence in the world. Homicide rates and social violence rise as contexts become more economically fragile. Armed conflict and terrorism are more prevalent in moderate to highly environmentally fragile contexts. Contexts with high political fragility have high levels of all types of violence, and are often in conflict or have a recent history of conflict. Extremely fragile contexts in the security dimension are among the most violent places in the world. These include Afghanistan, the Central African Republic, Iraq, Pakistan, Somalia, South Sudan, Sudan, Syria and Yemen. Contexts with high societal fragility experience higher battle related deaths and high homicide rates. Higher levels of gender equality are associated with a lower risk of civil conflict within a society.

Effective fragility programming requires effective financing

Total financial flows to fragile contexts including official development assistance (ODA), foreign direct investment (FDI) and remittances increased approximately 206% between 2002 and 2014 in constant terms. Remittances are the largest type of financial flow to fragile contexts followed by ODA and then FDI. Total remittances and FDI grew significantly faster than ODA for fragile contexts over 2002-14. Fragile contexts received the majority of ODA, or almost 64% of all ODA between 2011 and 2014, but the distribution of ODA within these fragile contexts is uneven on both an aggregate and per capita basis. Extremely fragile contexts are more dependent on aid than other fragile situations on average, but the extent of aid dependence varies significantly.

In addition, ODA is often targeted at the symptoms rather than the real drivers and root causes of fragility. Indeed, aid to fragile contexts is often for “firefighting” rather than for long-term structural change.


Fragility is a major issue on the global agenda, and the international community is united – most recently in the Stockholm Declaration – to address it more effectively.

Policy recommendations:

  • recognise that fragility is multidimensional: this will help design better theories of change and programming in at-risk contexts

  • address violence in all its forms: moving from interventions that are focused primarily on conflict and its aftermath to ones that systematically address violence, and its prevention

  • challenge existing, simplistic paradigms about violence: avoid attributing labels of “good” and “bad” to violence, this will help with a better understanding of, and response to, violence

  • invest in prevention: prevention saves lives, resources and money

  • deliver on Stockholm Declaration commitments

  • use domestic policy to promote global peace and security: this can make a real difference to the factors of power, marginalisation and capacity that enable violence around the world.

Programming recommendations:

  • move towards a whole-of-society approach to fragility: better results will come from working with multiple types of actors – individual, community, municipal, provincial and national – and taking a multidimensional, multi-sector approach

  • put people at the centre, recognising that a stable state and strong institutions do not automatically lead to a reduction in violence

  • use the violence lens – power, capacity and marginalisation – to design and deliver programming

  • prioritise reconciliation: to heal the social cleavages that perpetuate and exacerbate violence

  • recognise the critical role of gender in addressing fragility: with tools that bring together gender, violence and fragility issues within one framework

  • be open to experiment, remain flexible and take risks: becoming comfortable with a measure of well-calculated risk, and even programming failure, can have big payoffs

  • learn and build the evidence base: bringing together a broad range of research and policy fields.

Financing recommendations:

  • provide adequate, long-term ODA financing for fragile contexts, and focus funding on the real drivers of fragility: if ODA is to be most useful, it will need to be sufficiently predictable, flexible and long-term

  • develop better financing strategies: the OECD will continue work to promote a better understanding of financial tools and portfolio management in fragile contexts during 2017.