The principle of voluntariness is a cornerstone of refugee return. At times narrowly interpreted to signify consent, voluntariness in fact implies a degree of autonomy in decision making – that is, the ability of refugees to freely choose when, and whether, to return. However, there is only limited understanding of what actually influences refugees’ decision. Primary data collected among 393 refugees and returnees from Iraq, Colombia and Myanmar, complemented by quantitative analysis of refugee returns worldwide between 1995 and 2015, allow for an examination of what drives refugees’ decision to return, or not, to their country of origin.
Existing literature almost unanimously contends that security in countries of origin is a paramount precondition for refugee return, and indeed the role of security is particularly apparent in current discussions of returns to Syria. A quantitative analysis of refugee returns in the period 1995–2015 tentatively suggests that conflict-related deaths in countries of origin may indeed be negatively correlated with the proportion of returns to that country. Reflecting this trend, most of the refugees who wish to return to their countries of origin say they will only do so when there is peace. Yet while security may well enable return, it is not necessarily a motivating factor; among people surveyed, only 16.5% of returning refugees and 19.6% of refugees wishing to return to their countries of origin cited security improvements as the primary reason for return.
Willingness to return does appear to be correlated with refugees’ trust in their country of origin’s security forces. 67.9% of refugees who do not wish to return do not trust the security forces, and a further 20.6% feel actively threatened; in contrast, only 53.9% of the refugees who do wish to return experience mistrust, and none report feeling threatened.
Alongside physical security, a second piece of conventional wisdom dictates that socio-economic conditions in both host countries and countries of origin play an important role in refugees’ decision to return. Results of the study seem to corroborate this. Quantitative analysis reveals that the proportion of returns is negatively correlated with life expectancy and GDP per capita in the host country, suggesting that the likelihood of refugee returns decreases as life expectancy and GDP increase, and vice versa.
However, lived experiences of refugees may differ significantly from national averages of socio-economic well-being, and returns can also take place from host countries with higher standards of living if refugees are marginalised and excluded from their host community, unable to work, and faced with an uncertain future due to lengthy refugee status determination procedures. Among returnees surveyed, poor living conditions in their host country were the most common motivation for return (30.4%).
According to community leaders in Jordan, returns to Iraq are predominantly motivated by lack of income-generating opportunities in Jordan. Meanwhile, Iraqi refugees returning from camps in Syria report particularly difficult hosting conditions. “We lived in a prison, not a camp. They treated us like animals,” reported Dilshad, a returning Iraqi refugee. Staff of non-governmental organisations in Syria made similar observations: “People are returning because of course they want to go home, but also because they are not happy with the services and, perhaps more importantly, because of the fact that they are virtual prisoners in the camps here.”
Attachment and reunification
Conventional wisdom assumes that patterns of return mirror security and socio-economic well-being in host countries and countries of origin. However, evidence suggests that refugees’ decision making may also be influenced by their attachment to their countries of origin even in the face of socio-economic challenges and ongoing insecurity. For example, following the peace agreement in Angola in 2002, rapid spontaneous returns took place. The international community thought these returns premature. According to one Angolan returnee, “UNHCR explained that there would be no food, houses or schools, and they also told us there would be a lot of mines. But even if we don’t have houses, and we don’t have food, and we don’t have schools, we wanted to return to our country because it’s our country.” 
Among the Iraqis, Colombians and Burmese surveyed, missing home was reported as the main driver of return by 28.7% of returnees and 23.2% of the refugees who wished to return. Beyond attachment to country, family reunification is a key driver of return. The most common motivation cited by refugees wishing to return to their country of origin was reunification with family and friends (33.9%).
Implications for policy and practice
Most discussions of refugee returns focus on the importance of restoring security in countries of origin. Often, this is indeed a precondition for return – but security improvements do not automatically result in returns. The international community should not expect perceived milestone achievements towards peace and security to result in large-scale returns but should rather plan for continued support and assistance to refugees abroad.
Myanmar is a case in point. Following the nationwide ceasefire agreement and electoral success of the National League for Democracy in 2015, it was widely assumed that refugees on the Thai–Myanmar border would return to their country of origin. These assumptions resulted in a reduction in support from the international donor community, and therefore a decrease in food rations. In practice, however, the refugees have only limited confidence in the peace process, and the pace of returns has been slow.
Returns prompted by the pressures of unmet basic needs are likely to prove unsustainable. If refugees are returning to their country of origin despite security concerns because they are unable to sustain themselves in their host countries, there is a strong likelihood that these returning refugees will find themselves further displaced. Host States should ensure that refugees have sufficient access to livelihoods and assistance to prevent premature returns from contributing to vicious cycles of displacement. Some host States may believe it is in their interest to encourage rapid return – but if premature returns lead to further displacement, they are not the solution.
Finally, there is a need to better acknowledge the role of intangible factors including attachment to people and place. Understanding the complexity of decision making would improve stakeholders’ ability to plan for return, support refugees and returnees, and safeguard voluntariness.
 All primary data collected by IDMC. Myanmar report forthcoming. Reports on Iraq and Colombia online at:
 Harild N, Christiansen A and Zetter R (2015) Sustainable Refugee Return: Triggers, constraints, and lessons on addressing the development challenges of forced displacement, Global Program on Forced Displacement, p70–74.
 See Hasegawa Y ‘Forced to return? Facilitated return of refugees to Myanmar’ in this issue.