Key message 2 – The economic effects of civil war often last well beyond the conflict period and can spill over to other countries

It is useful to understand the different channels through which violence impacts the economy to better prioritise what can be done before, during, and after a conflict to minimise economic losses. In addition to the immediate, direct effects of insecurity and armed violence on economic activity, there are indirect, longer-term impacts. This includes impacts that spill over to other countries – such as through trade and refugee flows. Other longer-term effects on human health and human capital can have a knock on impact on the economy.

GDP, employment, and trade

There is a large body of research that finds civil wars have large and significant aggregate impacts on macroeconomic indicators such as GDP and trade (e.g. Collier, 1999; Abadie and Gardeazabal, 2003; Martin et al., 2008). More evidence has emerged recently on the channels of this immediate disruption.

Recent microstudies have shown, for example, that the insecurity spread by violence can create large increases in labour and transport costs. For instance, a study on the 2007 Kenyan presidential election found that electoral violence drove labour costs up by 70% (Ksoll, Macchiavello, and Morjaria, 2009). Another study found that insecurity caused by Somali pirates in the Gulf of Aden and the Indian Ocean led to an increase in shipping costs of about 10% (Besley, Fetzer and Mueller, 2012).

Refugees and displacement

In addition to affecting trade, violence can drive large scale human displacement, which in turn can have destabilising regional and global impacts. In fact, the number of people running away from civil war violence is far larger than the number of fatalities globally.

During the average civil war, 600,000 people leave their country. About 80% of these refugees return within a year after the conflict ends, but 10% still have not yet returned even a decade after peace (Mueller et al., 2016, based on UNHCR statistics). Countries that neighbour states in civil conflict host about 11,000 refugees on average and considerably more in some cases: Pakistan hosts more than 1.5 million Afghan refugees, for example. According to UNHCR, over 50 million people are now either refugees or internally displaced, with an estimated 93 billion USD in associated costs.2 Refugee streams at this scale can destabilise whole countries or regions.

Health, education, and human capital

Avoiding a humanitarian crisis can be considered a long-term investment in human capital. Exposure to conflict during childhood or adolescence in particular can have large, persistent effects on health, education, and labour productivity outcomes for a generation. A study on the effects of Burundi’s civil war on health found that an extra month of exposure to the conflict reduced child height significantly (Akresh, Bundervoet and Verwimp, 2009). Similarly, a study on the long term impacts of the 1967–70 civil war in Biafra, Nigeria, which killed 1–3 million people, estimates that exposure to violence led to height reductions – an indication of poor health – in both children and adolescents (Akresh, Bhalotra, Lene and Osili, 2012). These effects are estimated to have reduced the income of those most affected by 1.5–3% per year of conflict exposure (Mueller, 2013, calculations based on Case and Paxson, 2008).

Like health, education also suffers during conflict. A study of abducted child soldiers in Uganda found they completed nearly a year less schooling than their peers on average (Annan and Blattman, 2010). The loss was not made up for later and led to a significant drop in earnings. Reduced education outcomes may translate into eventual impacts on wages similar to health deterioration. A study on political violence during the 1980s and 1990s in Peru found exposure to violence before school-age leads to about half a year less schooling per year of exposure (Leon, 2012). This implies a decrease in wages by 1.08–1.52%.3

In the aftermath of conflict, there is an important role for aid and redevelopment strategies to focus on children and adolescents to prevent human capital losses that can last a whole generation.


  • 2 According to the Internal Displacement Monitoring Centre (IDMC).
  • 3 Author calculations based on combining Leon’s (2012) estimates of the effects of political violence on schooling with Duflo’s (2001) estimates on the wage returns to education.