Search

Moore (2005) has identified the three central research questions that concern scholars studying refugee flows. First, what circumstances lead people to leave their habitat? Second, why do certain people choose to cross their nation?s borders rather than relocate internally in a different part of the country? Third, what is the basis for refugees to choose to seek asylum in one country over the available alternatives? In this paper, I will focus on the first question. In particular what internal conditions in a country?political and economic?tend to produce high numbers of refugees? I hypothesize that regime type does have a significant impact on the probability that a country will produce refugees. In other words, I test whether political systems that guarantee political participation to its citizens have a lower likelihood of creating refugees than those that do not. I also hypothesize that as the degree of political repression in the country of origin rises, the likelihood of refugee production also increases. In order to code regime type, I have relied on the combined polity score in Gurr?s (2003) Polity IV data set. I relied on the political terror scale developed by Mark Gibney to measure the general of level of repression in a country and threat perception of its citizens from their government. The following independent variables were also tested to explain the likelihood of a country?s level of refugee production?degree of political repression, percentage of violent actions by governmental actors, violence by civilian actors, Gross Domestic Product and ethnic diversity. The World Political and Social indicators data collected by Craig Jenkins has been used to estimate the degree of violence perpetrated by civilian and governmental actors respectively. I have computed the average values for each of these independent variables for the period, 1992-2001. I have used the World Bank 1995 dataset for country wise numbers on GDP per capita and ethnic diversity. I am relying on the binary logistic regression method for this study. I created a dichotomous dependent variable from the refugee outflow data for 1992-2001 by classifying the nations of the world into two categories: high refugee producing countries (number of refugees in a given year > 5000) with an assigned value of 0 and low refugee producing countries with an assigned value of 1 (number of refugees < 5000). In order to test how regime type affects the probability of a country?s level of refugee production, I created a categorical independent variable from the combined polity score from the Polity IV data set. The countries that had a positive polity score were classified as ?democratic? with a value of 0 and those with a negative polity score were assigned a negative polity score were labeled ?non-democratic? with a value of 1. I assigned the value of 1 to the group of non-democratic countries since this was my category of interest. The odds ratio derived from the logistic regression results indicate that non-democratic countries are 1.12 times more likely to be classified as a high refugee producing country compared to democratic countries. Yet, regime type did not attain the level of significance in accounting for a country?s refugee production. The political terror scale value?an indicator of political repression? of a country is the most reliable factor in accounting for the likelihood that it will be a refugee producing country... ..PAT.-Conference Proceeding [ABSTRACT FROM AUTHOR]