State failure and conflict recurrence

In our ongoing research programme on security and development, we have more recently focused on the analysis of state failure. Most international development agencies, such as DFID or USAID have specific development programmes for fragile states. Other agencies refer to difficult partnerships (The Development Assistance Committee of the OECD) or to low-income countries under stress (World Bank). The main concern is that these countries are failing to provide an adequate environment for poverty reduction. On the other hand, security agencies have more often referred to collapsed or failed states and are mainly concerned with the inability of states to provide security to their own citizens and their potential to cause regional and global destabilization. For the purpose of this discussion the descriptions ‘weak’, ‘failing’, ‘fragile’ and ‘collapsed’ are treated as synonymous.

States can thus ‘fail’ in two distinct senses: they can fail to provide economic development opportunities and they can fail to provide security. The most basic role of the state is to provide physical security to its citizens through maintaining a monopoly of organized violence within the society. Where the government fails to do this and rival organizations of violence emerge, the state descends into civil war. However, in the modern world the demands legitimately placed upon the state extend beyond this basic function of security. Governments in all modern societies play some role as regulators of private economic activity, and as suppliers of public goods such as transport infrastructure, health and education. The quality of regulation and public goods is important for the capacity of citizens to earn a living and thus avoid poverty. Increasingly, as globalization makes economic activity more mobile between countries, the quality of government matters in a relative rather than an absolute sense: governments that are much worse than others are likely to lose economic activities and this will rebound upon their citizens. Hence, a state can fail because its government provides a quality of regulation and public goods that is markedly worse than provided by other governments.

Although it is relatively easy to provide a basic definition of state failure, most of the international development and security agencies do not produce publicly available lists of states they consider to be failing. This is probably due to a number of reasons. They do not want to appear as politically biased and wish to avoid lengthy debates as to why they include country A but not country B. A label of ‘state failure’ may also lead to a change in behaviour of citizens, investors, NGOs and other international agencies, thus possibly making an already bad situation worse. No international agency wants to be responsible for that. A rare example of a published list of failing states is provided by the German Ministry of Development (BMZ 2007). It is a relatively wide definition of failing states and lists 56 countries. It includes poor countries, many at war or having just recently ended civil wars, such as Afghanistan, Liberia and East-Timor. However, this group of failing states also includes countries with poor governance that are neither at civil war nor are they low-income countries. Examples are Belarus, Syria and Venezuela. According to this classification about 1.2 billion people live in failing states. As Figure 1 shows, almost half of all failing states are located in sub-Saharan Africa.

Cost of state failure
In order to gain some idea of the magnitude of the problem it is useful to estimate the cost of state failure. Chauvet, Collier and Hoeffler (2007) estimate the global combined cost of state failure at $270 billion. This is the sum of three distinct components: citizens of the failing state suffer the costs of poverty and civil war, and in addition neighbouring countries suffer from the spillovers from the failing state. No attempt was made to estimate the global costs, so this estimate of $270 billion is likely to be an underestimate. This estimate is the result of three cost components:

Cost of poverty
Good economic policies and governance provide economic opportunities for all citizens and enable people to live a life free of poverty. Chauvet, Collier and Hoeffler (2007) confine their analysis to failing states with low income. They find that during 1990–2001 low-income failing states experienced no economic growth, whereas other developing countries grew at 2.8 per cent per year. One result of this is that a large proportion of the world’s poor, defined as surviving on less than a dollar a day, live in failing states. Although the total population of this country sample makes up only about 7 per cent of the world’s population, 15 per cent of the world’s poor live in these countries. If this current trend continues, nearly 30 per cent of the world’s poor will live in failing states in 2015. The loss in income due to missed economic growth in all these failing states is estimated as a cost of $32 billion per year.

Cost of civil war
Some countries experience state collapse and large-scale internal violence causes large costs to their citizens. Wars kill, but even after the fighting stops the risks to human life and health are considerable. Wars divert precious resources into fighting, during the civil war military expenditure increases and does not return to normal levels after the fighting stops. On average civil wars last for about seven years (Collier, Hoeffler, Söderbom 2004) during which their growth is reduced considerably. Even though post-conflict countries experience a peace dividend in form of increased growth, it takes the country about 21 years to return to the income level pre-war (Collier and Hoeffler, 2004a). The total cost of civil war is calculated at slightly less than $3 billion per year. Compared to the economic cost of state failure this number seems relatively small, but this is due to the fact that civil war is a rare event even in low-income countries.

Costs to neighbours
The negative effects of economic stagnation and civil war spillover to neighbouring countries. Neighbours’ growth rates are significantly reduced by economic and security troubles; Chauvet, Collier and Hoeffler (2007) estimate this growth loss at about 0.6 per cent per annum per neighbour. As the average failing state has 3.5 neighbours, these growth losses add up to lost income of about $237 billion per year.

To summarize, states can fail in three distinct ways: they can hurt citizens in neighbouring countries, they can fail to provide basic security for their own citizens, and they can fail to provide an environment in which poverty reduction is feasible. The relative size of the different cost components is depicted in Figure 2.

Figure 2: Cost of State Failure

What can be done about state failure?
The combined total cost of state failure of $270 billion is much larger than the total amount of global development aid (about $80 billion). Even if all donor countries raise aid to the UN target of 0.7 per cent of GNI, this would still only amount to about half of the cost incurred due to state failure. Thus, the total global aid effort is substantially less than what would be contributed by turning around failing states. Failing states, in collaboration with international aid and security agencies, have to find ways to spend aid more effectively and find new solutions to turning around poor governance. Collier (2007) stresses that aid alone will not rescue failing states. Ultimately, reform has to come from within the failing states, but international efforts can assist these internal reforms. One mechanism of ensuring that public money provides services to poor people could be through public expenditure tracking surveys; these surveys state how much of the centrally allocated budget reaches the local service providers. Reinikka and Svensson (2004) pioneered these surveys in Uganda. First, they found that about 87 per cent of the budget for primary education did not reach schools. They then staged a local campaign and informed teachers, parents and pupils how much money their school should receive. Based on this information, local stakeholders demanded to know why there were receiving so little of the allocated funds. When these schools were visited two years later they received almost 90 per cent of the allocated funds. There was still corruption but, due to improved transparency, local stakeholders demanded accountability and hence reduced the opportunity for corruption.

A reform of trade policies would provide new income-generating opportunities for poor countries. An obvious candidate for reform would be the abolition of agricultural subsidies in rich countries. These subsidies for food and non-food items (such as cotton) in the USA and Europe are making it impossible for poorer countries to compete. But poor countries themselves have also generated trade distortions, their high import tariffs have generated industries in which profitability depends on lobbying rather than effective production.

A reform of the international financial system could make corruption more difficult. A large proportion of hot money leaves failing states and is invested in rich countries. A higher degree of transparency would make it more difficult for international banks and funds to invest illegally acquired wealth and a tighter regulation of unaccountable offshore tax havens would reduce the opportunities to send money abroad.

Many failed states are rich in natural resources, for example Angola, Chad and Turkmenistan. In many cases natural resources distort the economic and political system. The presence of natural resources also makes countries more conflict-prone (Collier and Hoeffler 2004b). Typically, these economies remain undiversified and their tax collection efforts are low. This hampers economic development and political accountability of governments to their citizens. Poor governance in countries rich in natural resources results in poverty, corruption and war. A reform of trade in natural resources would not only increase income flows from international extraction companies to governments of poor countries, but would also increase accountability and transparency. Some natural resource initiatives have already started. The Extractive Industries Transparency Initiative supports improved governance in resource-rich countries through the verification and full publication of company payments and government revenues from oil, gas and mining. A further initiative is the Kimberly Process, a certification scheme that imposes extensive requirements on its members to certify shipments of rough diamonds as ‘conflict-free’. This results in reduced financing opportunities for rebel movements and thus reduces the risk of large-scale violent conflict.

Above, state failure was defined in two ways: first the inability of the government to provide public goods to its citizens; and second the inability of the government to provide security. These causes of state failure are closely interlinked. If states have failed due to poor governance and economic policies they are much more likely to experience large-scale violence. Chauvet, Collier and Hoeffler (2007) estimate that the chance of a civil war breaking out in a failed state is about twice as high as in other developing countries. Unfortunately, these countries can then become stuck in a conflict trap: once they have experienced a civil war, countries are even more likely to have another. About 40 per cent of post-conflict countries descend into civil war within a decade (Collier, Hoeffler and Söderbom 2008). Although there is international agreement that civilian security must be guaranteed, this agreement is currently not matched by an international willingness to organise, finance and deploy military force. In 2005 the United Nations committed themselves to protect populations from genocide, war crimes, ethnic cleansing and crimes against humanity. Although this Responsibility to Protect was endorsed unanimously, it has not led to a de facto change of United Nations’ policies. Comparatively speaking, the UN peacekeeping operations budget is very small: in 2007 UN peacekeeping expenditure was about $7 billion, which is only about 0.5 per cent of worldwide military expenditure. So far, there is no international rapid reaction force that could be deployed at short notice, although the European Commission is working towards creating such a force. If we want to prevent failing states becoming stuck in a conflict trap, we need to develop comprehensive assistance strategies, which consist of military assistance (peacekeeping) as well as longer term development aid.


  • BMZ Federal Ministry for Economic Cooperation and Development, Transforming Fragile States – Examples of Practical Experience, Baden-Baden: Nomos, 2007.
  • Chauvet, Lisa, Paul Collier and Anke Hoeffler, ‘The Cost of Failing States and the Limits to Sovereignty’, UN-WIDER Working Paper, 2006.
  • Collier, Paul, The Bottom Billion: Why the Poorest Countries are Failing and What Can Be Done about it, Oxford: Oxford University Press, 2007.
  • Collier, Paul and Anke Hoeffler, ‘Conflicts’, in Lomborg, B. (ed.), Global Crises, Global Solutions, pp. 129–56, 2004a.
  • Collier, Paul and Anke Hoeffler, ‘Greed and Grievance in Civil War’, Oxford Economic Papers, vol. 56, pp. 663–95, 2004b.
  • Collier, Paul, Anke Hoeffler and Måns Söderbom, ’Post-Conflict Risks’, Journal of Peace Research, vol. 45, pp. 461–78, 2008.
  • Collier, Paul, Anke Hoeffler and Måns Söderbom, ’On the duration of Civil War’, Journal of Peace Research, vol. 41, pp. 253–73, 2004.
  • Reinikka, Ritva and Jakob Svensson, ‘Local Capture: Evidence From a Central Government Transfer Program in Uganda’, The Quarterly Journal of Economics, vol. 119, pp.679–705, 2004.