The security situation in Pakistan continues to remain unstable. The country’s natural resources have long been suspected as a catalyst for insurgent and separatist violence. Lootable natural resources, such as emeralds, allow rebels to purchase weapons to enhance their military capacity vis-à-vis the state. The emerald mines in the Swat district in Pakistan have long played a role in financing the Tehrik-e-Taliban’s insurgency operations. The perceived inequalities in the distribution of natural resource wealth is also suspected as a key driver of conflict in Pakistan.
In the province of Balochistan, which contains enormous amounts of copper, gas and gold reserves, Baloch insurgents accuse the central government of funneling the province’s resource revenues to wealthier areas in Pakistan and depriving Balochis of proper financial compensation. While there is a large body of literature that examines how natural resources impact conflict, much less attention has been given to understanding the relationship at the sub-national level, a level of analysis which is critical in providing researchers and policymakers insights into why certain areas within a country experience political violence while other areas remain relatively calm.
Pakistan is a suitable case to empirically test the relationship between natural resources and conflict and to better understand the channels by which resource wealth influence violence due to the country’s long-running insurgencies with violent groups and the accessibility of micro-level data on political violence over time. The findings emanating from this proposed research has implications for both the country’s domestic security and the stability of the region.
This research project poses two related research questions. The first research question is as follows: 1) what is the effect of natural resource wealth on the probability of political violence? Second, do different types of natural resources have varying impacts on the probability of political violence? Lootable resources, including gemstones, are easily extractable and allow rebel groups to purchase weapons to conduct insurgency activities. Non-lootable resources (such as oil and gas) require substantial amounts of fixed capital and are not easily extracted by rebel groups for financial profit. However, rebels can extort mining companies in order to secure funds needed to sustain their insurgencies. Moreover, disagreements with the central government regarding the allocation of resource rents emanating from rebel-held territories is believed to increase the incidence of political violence. Findings from studies that disaggregate natural resources by their types, however, are mixed and should be subject to further theoretical and empirical scrutiny.
This research project focuses on assessing the effects of the following natural resources on political violence in Pakistan which have been suspected in facilitating political conflict: gemstones, gas, oil, coal, and copper. This proposed research will consist of a multi-method approach, consisting of three components: 1) quantitative analysis; 2) geospatial analysis; and 3) fieldwork. The quantitative component will employ a logistic regression model to assess the relationship between natural resources and the probability of political conflict. The unit of analysis is the district-year level. The dependent variable, Political Violence, captures is a dichotomous variable that takes on a value of 1 if there was an act of a political violence (such as riots, political terrorism, bombings) by any non-state group in a district in a given year, and 0 otherwise. Data for the dependent variable will be taken from the BFRS Political Violence dataset in Pakistan which provides information on the number of political violence episodes at the district level from the years 1988-2011. The independent variable of interest, Natural Resources captures the presence of natural resources in each district-year. It is a dichotomous variable that is coded as 1 if there exists gemstones, gas, oil, coal, or copper mines, 0 otherwise. A subsequent model will include separate independent variables for lootable and non-lootable resources to determine if there are any differential effects resulting from each type of resource on conflict dynamics. The variable Lootable Resources captures the presence of gemstone mines in a district-year. This is dummy that equals 1 in all district-years with gemstone mines, 0 otherwise. The variable Non-Lootable Resources captures the presence of gas, oil, coal, or copper mining in a district-year. This is a dummy variable that is coded as 1 if there is oil, copper, coal, or gas mining in a given district-year, 0 otherwise. The model will also control for a vector of variables associated with conflict, including geographical, economic, demographic, political, and institutional variables as well as provide controls for temporal and spatial correlation.