Out of the Frying Pan into the Fire? Migration from Fragile States to Fragile States

Anke Hoeffler


Fragile states contributed 18 million migrants and 8 million refugees in 2000. More than 20% of these migrants and more than half of the refugees settle in other fragile states. Thus, migration is likely to be both a consequence and a possible cause of conflict and fragility. This paper asks why people from fragile states would want to move to another fragile state. Is it simply a question of jumping out of the frying pan into the fire – that migrants from fragile states have no other options than to settle in another fragile state?

The study analyses a new set of global data on the sources and destinations of migrants, revealing that economic factors, such as the pull of higher incomes in destination countries, are important. The paper concludes by discussing how migration from fragile states in search of higher incomes and greater wellbeing is an important development strategy that should be supported.

Key findings:

  • People leave countries with low incomes and are attracted to countries with higher incomes. However, the pull of higher incomes is greater than the push of low incomes. If origin country incomes increase by 1%, the stock of migrants decreases by 0.2%, but if destination country incomes increase by 1%, the stock of migrants increases by 0.3%. Migration also depends on comparable incomes in the destination countries.
  • Relative income is captured as the ratio between comparable countries of destination and the income of the chosen destination. For example the comparable income for migration to a neighbouring country is the income of other neighbouring countries. This shows that migrants prefer to emigrate to countries with comparably higher incomes. On the other hand, economic opportunity (measured as the growth rates in the country of origin and destination) does not appear to be significant.
  • Political factors are less strong. Autocracy does not appear to push people out; this is perhaps because severe autocracies manage to control their borders more tightly and make emigration more difficult. Thus, even if people want to leave autocracies they are less able to do so. There is a slight pull factor from democratic regimes. If there is a one point improvement in the polity score, immigration increases by 4%.
  • Extreme political events – like international and civil wars in the previous decade – do result in more migrants. International wars increase migration by about 350%. However, international wars are very rare events; more common events are civil wars. A civil war increases migration by about 10%.
  • Geography is important, too. Neighbouring countries receive more migrants; the further away the destination country the less migration we observe. However, there are a number of interaction effects. For example, the obstacle of distance is less severe if the incomes in the destination as well as origin country are higher. The latter suggests that migration is an investment. Incomes in origin countries have to be relatively high to make the costly choice of emigration to a far away country. The results suggest that migrants from poor countries are more likely to go to neighbouring countries, whereas migrants from middle income countries are more likely to go to a richer country, even if it is a long distance away.
  • Former colonial relationships do not in themselves determine migration, but migrants from relatively well-off former colonies are more likely to emigrate to the former colonial power.
  • The most important determinant of migration appears to be the stock of existing migrants, or diaspora. The larger the existing diaspora the greater the subsequent migration flow. The results here suggest that a 1% increase in the diaspora increases migration by 0.8%.


Hoeffler, A. (2013). Out of the Frying Pan into the Fire? Migration from Fragile States to Fragile States. OECD Development Cooperation Working Paper 9. Paris: OECD Publishing.