This page is not available in English.  Go to English front page.

Global trends in armed conflict

Halvard Buhaug, Scott Gates, Håvard Hegre and Håvard Strand, Centre for the study of civil war, International Peace Research Institute, Oslo (PRIO)

The number of ongoing conflicts has declined since shortly after the end of the Cold War and the severity of armed conflict has generally declied since World War II. This fact sharply contradicts many pessimistic perspectives bolstered by media headlines from Iraq, Afghanistan, and Darfur. Research conducted by the Centre for the Study of Civil War at PRIO, using the most recently updated data collected in collaboration with the Uppsala Conflict Data Program (UCDP) at Uppsala University, indicates a more complex situation, with both reassuring and disturbing trends.

After a period of steady decline in the number of armed conflicts in the world, the downward trend has ended. Data from PRIO and Uppsala University indicates that the number of active conflicts is no longer sinking, but has held steady at 32 for three years in a row. Secondly, we are now in the longest period since World War II without interstate war (those fought between two or more countries). Moreover, we register no new conflicts of any type over the previous two years; this is the first time in the postwar period in which two years have passed with no new conflicts having broken out.

This report serves as a background paper for the Norwegian Ministry of Foreign Affairs’ programme, Refleks,Globalisation and national interests. We focus here on global trends in armed conflict with special emphasis on civil conflict, since it is the most common form of armed conflict.

Thereport is frontloaded with our policy recommendations. Our analysisfollows. Two types of policy recommendations are featured, those that serve to mitigate conflict onset and those that work to end ongoing conflicts. Both types of policies must be addressed in order to sustain the downward trend in armed conflict.

Policy recommendations to decrease the risk of internal conflict onset
  • Promoting economic growth and diversification is the best long term strategy for reducing the risk of conflict. Natural resource based growth requires very good resource revenue management to have positive political effects. Norwegian authorities may contribute to growth and diversification through foreign aid, liberalization of trade in goods exported by lowand middle income countries, and by promoting transparent and redistributive management of resource revenues. Regional development strategies should be emphasized.
  • Promoting greater democracy and improved human rights in low and middle income countries is desirable for a number of reasons but does not necessarily reduce conflict levels. In countries where democratic processes function reasonably well, policies to maintain and develop the democratic institutions will reduce the risk of conflict. In countries where electoral institutions are nonexistent or dysfunctional, however, encouraging constraints on the executive (through mechanisms to ensure budgetary transparency, auditing bodies, and the stimulation of the emergence of an independent press and non-governmental organizations) is more important than promoting free and fair elections. As long as executives can operate with few constraints, the incentives for electoral fraud will be too large.
Policy recommendations to terminate ongoing armed conflict
  • Promote and finance UN and other peacekeeping operations (PKO). In addition, longterm guarantees of rapid intervention resembling the UK guarantee in Sierra Leone should be supported. Such guarantees should be used to press post-conflict governments to reduce military spending which diverts valuable budgetary resources. Military spending in low income countries is in general counterproductive as means of preventing domestic conflict and also leads to regional arms races.
  • Promote postconflict aid. Tieaid to limits on military spending. Aid regularly spills over into military spending (Collier and Hoeffler estimate that 11.4% of development aid leaks into military budgets).[1] Moreover, military spending typically increases substantially during conflicts. Caps on military spending as a condition for development aid is therefore both a necessary and potentially effective tool to rduce risk of conflict recurrence.
  • Promote the role of the Hague, particularly the efforts of the International Court of Justice to criminalize wrongdoing during war. Also promote local post-conflict justice efforts. Work to discourage blanket amnesty forwar crimes.
  • Promote international regimes such as the Kimberley Process to regulate trade in contraband resources used to finance war. While there are holes in the Kimberley Process, and regulatory regimes alone will not stop a conflict, such measures in combination with PKOs and other policies can be effective.
The remainderofthe report is organized as follows:
  • Graphical presentation of trends
  • Armed conflict trends
  • New, recurring, enduring conflicts
  • Geographic trends in armed conflict
  • Number of countries involved in armed conflict
  • Battle death trends
  • Analysis of the factors associated with these trends
  • Factors associated with conflict onset
  • Factors associated with trends


Defining armed conflict

For our analysis we draw on the definition from the UCDP/PRIO Armed Conflict dataset. An armed conflict is a contested incompatibility over government or territory between two organized parties, of which at least one is the government of a state, in which the use of armed force between the parties has resulted in at least 25 battle‐related deaths in a calendar year.

Figure: Armed conflicts

Several points should be noted:

  • Intrastate (civil) conflict is the most common form of armed conflict and this has been the case since World War II.
  • Since 2004there has been no interstate conflict. This is the longest period without interstate conflict since before World War II. Given the relative rareness of interstate conflict, we focus our report on civil conflict (both intrastate and internationalized intrastate conflict).
  • The number of armed conflicts trended upward until 1991-92, when it peaked. Since then the pattern has been a general trend of declining number of conflicts, although the trend has levelled out in recent years. This pattern of decline will serve as a theme in this report.
  • Little change is evident in the number of civil conflicts with external intervention (i.e. internationalized intrastate armed conflict).
  • Interstate conflicts (fought between at least two countries) have been relatively rare.

Table 1 shows in detail the numbers and trends from 2000 onwards associated with Figure 1:

  • Intrastate conflict is clearly the dominant form of armed conflict.
  • Since 2004 there hasbeen no interstate conflict.
  • Internationalized intrastateconflicts (i.e. civil wars with outside intervention) seem to exhibit little trend.

The number of armed conflicts is stable at 32 for four of the last five years; however, by looking at the numbers of new and recurring conflicts per year a more volatile picture emerges.

  • While the annual number of active conflicts is fairly constant in recent years, there is substantial variation in the sample of conflicts from one year to the next. Four conflicts dropped out while another four reappeared after some period of inactivity in 2006; no less than nine of the 32 conflicts in 2005 were inactive during the previous year (and, similarly, nine conflicts active in 2004 were no longer active in 2005).
  • The number of new conflicts, those in which there is no record of previous armed conflict, has been low throughout the 2000s, and there have been no new conflicts after 2004.
  • Recurring conflicts are those cases where conflict re-ignites or escalates after one or more years of very low intensity. We see little trend with regard to these re-activated conflicts.

Figure: Ongoing  conflicts

Figure 2 shows the number of ongoing conflicts along with the number of new conflicts:

  • The number of new conflicts has been fairly constant and generally low.
  • Two periods are associated with increases in the number of new wars; 1958–67, which is associated with decolonialization and the emergence of new states; and 1989–93, which corresponds with the end of the cold war.
  • More generally, in terms of proportion of the number of countries in the world, the frequency of new conflicts has a clear negative trend.
  • The increase in the number of conflicts up to 1992 was not due to an increase in new conflicts, but a steady accumulation of conflicts that either did not end or restarted soon after peace agreements.
  • Correspondingly, the main reason for the decline in the number of conflicts since 1992 is a considerable increase in successful terminations ofconflicts combined with a steady but small decline in the frequency of new conflicts. This is why it is so important to support policies that can serve to prevent the onset of armed conflict.

  • Figure 2 shows a significant recent decline in the number of conflicts fought in Europe and the Americas. European conflict peaked in the early 1990s with the civil conflicts in the Balkans. Armed conflict in the Americas, which affected many countries in Central and South America, reached its zenith in the 1980s.
  • The number of conflicts in the Middle East has remained relatively stable throughout the period, though with a slight increase in the 1980s.
  • Most conflicts are now fought in Asia or Africa and these continents have not contributed much to the decline in the frequency f armed conflict in recent years. We expect this trend toremainunchanged.


  • Map 1 shows the location of the centre of all armed conflicts fought in 2006 denoted by red dots.
  • Countries with conflictin 2006 are shaded indark brown. Lightly shaded countries experienced armed conflict sometime between 1989 and 2005.
  • Notethat most conflicts aregeographicallygrouped. The ellipses stretching from Caucasus to the Philippines and the other in the Great Lakes and Horn of Africa demonstrate the tansnational dimensions of civil conflict. Only Colombia and Algeria lie outside a regional nexus of conflict; nonetheless, Colombia used to be part of a conflict zone that stretched from Mexico to Peru. We are not saying that a conflict in the Philippines affects a conflict in Georgia, but that civil conflict spills over borders and destabilizes neighbours. These patterns demonstrate that conflicts cannot be considered in isolation to one another. Regional factors play into domestic conflicts shaping the nature of warfare.


Figure 4 shows two trends: the absolute numbers (left-hand vertical axis) and the share of countries (right-hand vertical axis) involved in armed conflict.

Three distinct peaks are evident:

In terms of share of UN members in conflict, the greatest proportion occurred in the early 1950s with the Korean War, which involved 20 countries;

  • Another peakoccurred in1991. Severalconflictscontributetothistotal,includingthe Gulf War involving Iraq vs. a US-led coalition of 29 in addition to several post-Soviet/Yugoslav intrastate conflicts;
  • Several conflicts in 2004-06 including: Afghanistan (39), Iraq (33) and US and partners vs. al-Qaida (18). Many of the same countries are involved in all three conflicts.
  • In terms of absolute numbers, more countries are getting involved in armed conflict. However, when looking at the proportion of countries involved in war, no obvious trend can be seen, since the number of countries in the world has increased.
  • Conflict involvement also meansverydifferentthingsfordifferentcountries. WhiletheUS andUK are heavily involved in both Iraq and Afghanistan, these figures also include countries such as Tonga, which has contributed only a token number of troops to the Iraqi War.


  • Figure 4 shows the number of battle deaths annually from 1900 to 2005. The figures include civilian deaths, but only deaths which result directly from armed conflict are counted. We do not count indirect effects of warfare such as increased mortality from disease and famine. These numbers also exclude one‐sided violence (genocide, terrorist attacks on civilians) and non‐state conflicts (ethnic/sectarian violence), which excludes the Rwandan genocide, the Nazi Holocaust, the Kampuchea ‘killing fields’, and the Chinese Cultural Revolution.
  • The two World Wars tower over all other conflicts. Even the Korean War, the Vietnam War, and the Chinese Civil War are dwarfed in comparison.
  • Although interstate wars are (and have always been) rare, they have also accounted for the bulk of death (and presumably destruction).
  • Note that while today’s most severe conflicts, in Iraq and Afghanistan, generate massive media attention, their number of battle‐related deaths are quite moderate in comparison to earlier wars.


Figure 6 shows the number of annual battle deaths by region since World War II.

  • The generaltrend in battledeathsisdownward. This is in part due to the decline in the number of conflicts (see Figure 1), but also partly due to at least two other developments: the demise in conventional interstate conflicts and a concentration of intrastate conflicts in non-democratic and very poor countries with limited military capabilities.[2]
  • If we consider the contemporaneous growth in the world’s population, the downward trend is even stronger.
  • Individual conflicts, however, account for most of the peaks. The Korean War and Chinese Civil War produce the first peak. The Vietnam War (1965–75) is seen in the second peak, and the third is mainly made up by the Iran-Iraq War (1980-88), and the Soviet Afghanistan War. The civil war in DRC (1996-2001) can be identified in the last peak.
  • If non-state violence was included in these statistics, we would see an additional peak in 1994 (Rwandan genocide) and, potentially, in the most recent years (Iraq, Afghanistan). However, the overall trend of a decline in conflict-related deaths would conceivably remain unchanged, as earlier conflicts, too, generated substantial indirect casualties.
  • The Korean War is the most severe in terms of annual casualties; however, when summing battle deaths over the duration of the conflict, the Vietnam War is the most deadly.
  • In the mid-1980s, Africa superseded Asia as the continent most severely affected.


Factors associated with the onset of civil conflict

The decreasing trend in the number of new conflicts can to some extent be explained by trends in factors that have been documened to be systematically related to the onset of civil conflict. Below, we identify the factors that are most robustly associated with intrastate conflict (based mainly on Hegre & Sambanis, 2006)[3], and which feature prominently in our policy recommendations regarding the prevention of new wars.

Demographic factors:
Populous countries have more armed conflicts than small countries, but have fewer conflicts or casualties per capita than small countries.

Countries with a large youth cohort(i.e. a largeproportionofthepopulationinthe15–24year age group)havea highriskofinternalconflict.[4]

Economic Factors:
Low income levels and low rates of economic growth: A disproportionate share of internal conflicts occur in low income and lower middle income countries and in countries with low growth rates. One possible explanation for this relationship: Low income countries have governments with limited capacity to provide public goods and to enforce the monopoly on the use of force

Oil and other natural resources dependence: Countries with a large proportion of primary commodity exports also have a somewhat higher risk of conflict. Political institutions, however, play a strong role in mitigating the destabilizing effects of natural resource dependence. Nevertheless, without strong institutions, dependence on oil and natural resources significantly increases the risk of conflict.

Political Factors:
Inconsistent political institutionsand recentpoliticalinstability:Inconsistent political institutions (i.e. institutions that are neither democratic nor authoritarian) give rise to conflicts as they are relatively non-responsive to popular demands at the same time as they are more inefficient than authoritarian regimes in countering insurgencis. Semi-democratic institutions typically also provide incentives for nationalistic and sectarian politics, and are liable to experience conflict-generating institutional changes. A particularly dangerous constellation of political institutions is a strong unconstrained executive combined with expanded pulic participation. Changes to the design of political institutions (typically partial democratizations) often lead to or are associated with violent conflict.

Regional Factors:
Countries in conflict-prone and non-democratic neighbourhoods have a heightened risk of conflict. Rebel groups often depend on sanctuaries outside the country’s border and enjoy (tacit) support from rivalling neighbouring regimes, which are more easily obtained in an unstable environment and in areas with limited statehood. Conversely, a stable and democratic regional neighbourhood helps economic growth and stimulates non-violent economic activities.


Kinship and Identity Factors:
Ethnic differences are related to violent conflict, but may be less important as causes of large-scale civil war than often argued. However, societies where one ethnic group is dominant but minorities are sizeable are more conflict-prone – particularly when the politically marginalized groups reside in the periphery of the country.

Trends in these factors

Demographic trends:
The increase in the global population (and in the number of independent countries) partly explains the increase in the number of conflicts up to 1992. Yet the probability of dying in battle decreased for the world as whole during this period. The increase in the number of conflicts is counteracted by demographic changes. Globally, the proportion of youth to adults has been decreasing since the late 1970s. This trend, however, has not yet appeared in sub-Saharan Africa.

Economic trends:
The regional variation between regions in income levels and growth rates is considerable; moreover, this variation in growth rates corresponds to changes in the geographic distribution of conflict. Latin America has had moderate economic growth, from a relatively affluent point of departure, and has experienced a considerable decrease in conflict. The rapid growth in East Asia has been associated with a dramatic decrease in conflict intensity from a very high level, although this has not occurred in South Asia. The economic growth in the Middle East has not yet been accompanied with a similarly dramatic reduction in conflict, probably because of the importance of natural resource extraction in these economies. Sub‐Saharan Africa has not had economic growth over the last four decades and has only seen a moderate reduction of conflict levelsafter the end of the Cold war.

Political trends:
The world is moredemocraticthanever. More and more states are democratizing. Democratizationhascomplexeffectsonconflict. Few mature democracies experience serious internalviolence. Nevertheless, the institutional change in itself increases the risk of conflict in the short run, and a change from an authoritarian to a semi-democratic system increases the risk until a transition to a consistent democracy is completed. On the other hand, democratization in one country seems to have a consistent risk-reducing effect on its neighbourhood. Moreover, the strong decline in the frequency of interstate conflict is related to the spread of democracy.

Factors that explain the increase in conflict terminations – the changing role of the international community

End of the Cold War:
During the Cold War the United States and Soviet Union financed a number of armed groups and governments as a key aspect of figting so-called proxy wars (examples include Nicaragua, El Salvador, Angola, and Mozambique). The end of the Cold War meant that armed conflicts no longer have as ready a source of external source of financing.

New political climate for the UN Security Council:
With the end of the Cold War came increased political latitude for the UN, particularly the Security Council. The new climate has allowed an expansion in the number, scope, and mandate of peace-keeping operations by the UN and other international bodies.

International justice:
Since the Cold War, the role of the Hague has been expanding. The International Court of Justice (ICJ) has played a particularly important role in the post-conflict prosecution of war crimes.

Associated with a change in the role of the ICJ has been a trend toward the criminalization of war, whereby individuals are held personally responsible for wrongdoing in the context of conflict. Earlier, states were held accountable within international law. The threat of international legal consequences may deter potential belligerents; but of ourse it could also encourage a hard line among those who are already engaged in armed conflict. It also presumes that the perpetrators will be prosecuted.

While amnesty and impunity are still common, post‐conflict judicial mechanisms are much likely to be employed in the post Cold War era. South African style truth ommissions and restitution to victims in particular are more common.


Transnational dimensions ofconflict:
Reliance on contraband financing and extortion requires tacit compliance by neighbouring countries, or their inability to enforce borders.

Thiscomplianceisavailableinashrinking numberofstates. International regimes, suchastheKimberley Process, have sustained this trend.

Politicalandeconomicchangesaffectentire regions. Civil war doesnotoccurinisolation butaffects entire neighbourhoods. Peace-buildingstrategiesthereforeneed tobe regionally focused.

Future research

The conclusions of this report are mostly based on research for which there is proper documentation. In several areas, however, the basis for guiding policies is relatively weak. Researchon thefollowingissuesshouldbestimulated

Our argument regarding the trend toward criminalization of war is largely based on observations from individual cases. A more systematic treatment of the normative and empirical basis for this conclusion would help ascertain whether the argument holds outside these cases.

It is often argued that an increasing proportion of inter-group violence occurs without the explicit participation of states. Data for such violence, however, are much sparser than corresponding data for state-based violence. To be able to identify and explain trends it is necessary to invest heavily in data collection.

A topic we do not address relates to the effect of climate change.

Climate change may lead to intense resource competition and large-scale migration which again may result in inter-group conflict. However, neither of these links is systematically documented.

Policy recommendations should be accompanied by assessments of both costs and benefits of suggested conflict-reducing interventions. Such assessments are rare. The papers in the Copenhagen Consensus[5] attempt to assess and compare solutions to global challenges, conflicts being one of them, to help policy-makers choose between alternative strategies. For the challenge of conflict prevention, however, the research community only has weak models for conflict prediction. Research to improve our ability to predict has a great potential to improve the quality of policy recommendations.

[1] Collier, Paul & Anke Hoeffler, 2007, Unintended Consequences: Does Aid Increase Military Spending? Oxford Bulletin of Economics and Statistics 69(1): 1–27; see p. 12.

[2] Lacina, Bethany Ann; Nils Petter Gleditsch & Bruce M. Russett, 2006. ‘The Declining Risk of Death in Battle’, International Studies Quarterly 50(3): 673–680.

[3] Hegre, Håvard & Nicholas Sambanis, 2006. ‘Sensitivity Analysis of Empirical Results on Civil War Onset’, Journal of Conflict Resolution 50(4): 508–536.

[4] Urdal, Henrik, 2006. ‘A Clash of Generations? Youth Bulges and Political Violence’, International Studies Quarterly 50(3): 607–630.

[5] Bjørn Lomborg (ed.) Global Crises: Global Solutions, Cambridge University Press

Last updated: 04.12.2007
Refleks - utenrikspolitisk ordskifte