|May 17, 2007, Quy Toan Do, Lakshmi Iyer
World Bank Policy Research Working Paper 4228
Since 1945, more than 70 civil wars have occurred around the world, resulting in approximately 20 million deaths and displacing more than 67 million people. Understanding the causes of the onset and continuation of such conflicts is therefore of great importance, and the analysis of the determinants of civil conflict has been the subject of a recent and growing literature in economics. Several cross-country studies, including Collier and Hoeffler (2004), Fearon and Laitin (2003) and Miguel et al. (2004) find that poorer countries face a greater risk of civil war; the results are varied with regard to the effect of social divisions based on ethnicity and religion. A new research paper by Quy-Toan Do and Lakshmi Iyer examines the same issue within one country, Nepal.
Starting in 1996, Nepal has been affected by the “People’s War” begun by Maoist activists, whose main aim was to establish a people’s republic and change the constitution. The conflict spread to all parts of the country, and resulted in a death toll of more than 13,000 people over the ten subsequent years. Two cease-fire agreements, negotiated in 2001 and 2003, had been broken unilaterally by the Maoists. In 2005, the King took over power from the elected representatives, but relinquished power back to them in 2006 following mass protests. The Maoists then began peace talks with the government and have now joined an interim legislature while preparing for upcoming elections to a Constituent Assembly. The election is expected to take place in June 2007.
Studying the spread of conflict within a single country has several advantages over cross-country studies. Variables which differ across countries such as the existing political system, the aims of the rebel movement or the involvement of other countries, are held constant in such a study. The authors are able to construct finer measures of conflict intensity based on the number of casualties and other human rights abuses, rather than just using a dummy variable for whether an area experiences conflict or not. They are also able to separately analyze conflict-related deaths caused by activist forces and by the state, and study the evolution of the conflict intensity over time, while standard cross-country studies usually do not go further than dating the onset and end of civil wars.
The map below depicts the geographical patterns of violent conflict. The conflict started in the districts of Rolpa and Rukum in the mid-Western region, but thereafter spread to almost all districts of Nepal. Areas in the western part of the country (mid-Western region) and the eastern part (Eastern region) have the highest levels of conflict intensity, measured by the number of reported deaths per 1000 population.
The empirical question addressed here is the following: do districts that show high levels of conflict intensity have systematically different economic and social conditions? To answer this question, the authors perform a regression analysis of conflict intensity on a host of potential explanatory variables: geography (elevation, presence of forests), economic development indicators (poverty levels, literacy rates, infrastructure and public services), and social diversity measures (caste polarization, ethno-linguistic fractionalization).
The measures of conflict intensity are based on data provided in the annual Human Rights Yearbooks published by the Informal Sector Service Center, a Nepalese non-governmental organization. Explanatory variables are obtained using data from the 1991 and 2001 population censuses, the Nepal Living Standards Measurement Study Survey 1995-1996, and the Nepal District Profiles based on official data.
The main results of this study suggest that poverty is the single biggest determinant of whether a district will experience conflict. The figure below illustrates this finding. The estimates imply that a 10 percentage point increase in poverty is associated with 24 additional conflict-related deaths in that district. Geographic variables, such as elevation and the presence of forests, which make it easier for insurgents to hide from government forces, also have a significant correlation with conflict intensity. This effect is mainly due to lack of infrastructure: in particular, such areas lack good roads, which prevents government forces from being effective. In contrast, variables measuring social divisions such as caste polarization or linguistic fractionalization do not have a significant association with conflict intensity, suggesting that support for such movements comes from economically deprived areas and not from social divisions.
These basic results do not change if the analysis is repeated using other measures of economic development (literacy rates, infant mortality rates), measures of infrastructure (road density, number of post offices, presence of banks) or different measures of social divisions (caste, ethnic and linguistic diversity). Interestingly, both deaths caused by the Maoists and deaths caused by state forces are significantly higher in poorer areas.
Analyzing the time-line of the conflict, the authors find that the poorest areas have the highest levels of conflict in the early stages. Over time, as the Maoists brought many of these areas under their control, the highest-intensity conflict shifts to places that are somewhat better off. Their results are thus not driven by the fact that poor people are just more likely to die for any level of conflict i.e. that poverty increases the likelihood of being a “collateral damage” victim. It is much more likely that poor people join the rebellion for lack of economic opportunities (the “opportunity cost” view) or that poorer areas have larger grievances against the national government and hence are more likely to support insurgencies (the “grievance” view). While the analysis is unable to distinguish between the two, and both views may be right, the present research reiterates the importance of poverty in explaining violent conflict.
Collier, Paul, and Anke Hoeffler (2004), “Greed and grievance in civil war”, Oxford Economic Papers 56, 563-95.
Fearon, James and David Laitin (2003), “Ethnicity, insurgency, and civil war”, American Political Science Review 97, 75-90.
Miguel, Edward, Shanker Satyanath and Ernest Sergenti (2004), “Economic shocks and civil conflict: an instrumental variable approach”, Journal of Political Economy, 112(4), 725-53.