Nowhere to go? Why do some civil wars generate more refugees than others?

Abstract

Civil wars greatly vary in the number of refugees they generate, ranging from zero to over six millions in a given conflict. Work on this variation has largely focused on “push” factors–deleterious attributes of the home country that lead to refugee flows, such as violence and repression. Yet, few have studied the importance of “pull” factors–attractive features of the potential host countries. Here we show in particular the importance of the expected quality of life in possible destinations. Using data on civil wars from 1951 to 2008, we find that the proximity of democratic and wealthy potential hosts accounts for much of the variation in the number of refugees. Out-of-sample validation methods show that these “pull” factors account for nearly as much predictive power as all the main variables previously identified in the literature combined.