This article is the second in a three-part series on terrorism in Pakistan. You can read part one on the areas of origin here.
Are terrorists more likely to emerge from socioeconomically deprived regions of Pakistan?
Part I of this series described some of the major background features of Pakistan. We found that terrorists are better educated than average Pakistani hence unlikely to emerge from impoverished background. However, could the same be said about their areas of origin or are they more likely to emerge from socioeconomically deprived regions of Pakistan? We can answer this question by making use of regression analysis that determines the explanatory power of factors such as poverty, education, migration, religious conservativeness in relation to terrorist supply. Higher supply of terrorist from districts experiencing higher level of poverty can be expected for number of reasons.
First, the opportunities for economic advancement are likely to be fewer in socioeconomically depressed regions thus offering fertile ground for terrorist recruitment. Professor Eli Berman from UC San Diego and Professor Laitin from CISAC Stanford University have shown that the militant sects serve as agency for provision of publics goods that in return earn them recruitment.
Second, Many militant organizations in Pakistan have wings dedicated to providing various kinds of social services in poorer regions such as health and education facilities to attract potential recruits and to cover up their violent activities. Therefore, this channel may also lead to positive correlation between poverty and terrorist supply across Pakistan.
Third, higher incidence of poverty in the region may itself serve as a driving force for radicalization that seeks to subvert the status quo. The notion of rationality in economic theory may have to be extended to take account of this channel of radicalization. Rationality in economic theory considers individuals to be motivated for selfish pursuit of personal gains. However, in societies like Pakistan, organic identities of religion, ethnicity, sect, caste exerts powerful influence on individuals psyche and their motivations. Although we do not have data for determining the economic status of each terrorist, it can be argued that poverty of the locality definitely means higher probability of potential terrorists to be poorer and may motivate individuals to become agents of change through violence.
“In societies like Pakistan, organic identities of religion, ethnicity, sect, caste exerts powerful influence on individuals psyche and their motivations.”
Similarly, we could also expect terrorist supply to be higher from regions characterized by higher level of migration. Migration can lead to social stresses by initiating scramble for resources and hence may sharpen religious and sectarian identities. We also try to figure out the extent to which religious conservativeness is related to terrorist supply. It is expected that high level of religious conservativeness may prove instrumental in making minds susceptible to propaganda of militant outfits because of some commonalities in beliefs particularly on women and religious minorities.
The unit of analysis is the district level which makes third tier of government in Pakistan after national and provincial level. Because of the lack of information on mainstream measures of these variables, relevant proxies have been employed to measure them whose details can be referred to in Table 1
In order to keep the analysis simple to understand for readers from non statistical background, the results are summarized in non technical format in Table 2 that indicate the nature of statistical relationship between terrorist supply on one hand and independent variables on the other hand.
Table 1: Description of variables and hypotheses
|Variable||Proxy||Measurement||Data Source||Rationale||Hypothesized Relationship with Terrorist Supply|
|Diarrhea||Poverty||Number of children aged below 5 suffering from diarrhea as percentage of total children in same age group||Pakistan’s Social and Living Standards Measurement (PSLM)||In line with World Summit on Sustainable Development (WSSD) view that ill health is one of the indicator of poverty||Districts with higher such proportion are expected to have higher supply of terrorists|
|Consultation with private doctor||Poverty||Proportion of prenatal consultation with private doctors as percentage of total consultation in district||PSLM||Pakistan’s health system is segmented into public and private providers of health services. Due to poorly managed public health system, those with relatively higher income consults private doctor that incurs higher monetary charges. Hence higher consultation with private doctor in district is likely to be correlated with higher income level.||Districts with higher prenatal consultation with private doctors are likely to have lower supply of terrorists|
|Access to adequate fuel (Gas and Oil) for cooking||Poverty||Consumption of gas and oil as fuel as percentage of total consumption of fuel||PSLM||This proxy is in line with WSSD view of poverty as one of the basic services that citizens are entitled to.||Districts with higher access to fuel are likely to have lower supply of terrorists|
|Education Score||Education||Index (1-100, lower score indicating lower performance) developed at taking arithmetic mean of enrollment, achievement, attainment and gender parity in education.||Alif Ailaan, a UK funded organization campaigning for higher education in Pakistan
|It measures the actual educational performance of the district||Districts with higher education score are likely to have lower supply of terrorists|
|Rent Housing||Migration||Number of rent housing as percentage of total housing||PSLM||It is obvious that districts with higher proportion of rent housing are likely to have higher migrant population||Districts with higher rent housing (higher migration, hence higher social stresses) are likely to have higher supply of terrorists|
|Gender Parity Index||Religious conservativeness||Arithmetic mean of gender ratios at enrolment and continuation up to class 5||Alif Ailaan||Religious conservative circles in Pakistan disapprove of female education. Even group such as TTP destroyed girls schools as part of their violent campaign in Swat valley. Hence lower gender parity index is likely to be correlated with higher religious conservativeness in locality.||Districts with higher level of religious conservativeness are likely to have higher supply of terrorist|
Table 3: Results
|Variables||Statistical Relationship with Terrorist Supply||Explanation|
|Diarrhea||Positive||Terrorist supply is higher from districts with higher prevalence of Diarrhea|
|Consultation with private doctor||Negative||Terrorist supply is lower from districts having higher proportion of consultation with private doctors|
|Access to adequate fuel (Gas and Oil) for cooking||Negative||Terrorist supply is lower from districts having higher access to fuel for cooking|
|Education Score||Quadratic, terrorist supply increases with education score but after reaching approximately intermediate level starts decreasing||Terrorist supply is higher from districts at intermediate level of education performance|
|Rent Housing||Positive||Terrorist supply is higher from districts having higher rent housing (hence migration)|
|Gender Parity Index||Negative||Terrorist is lower from districts having with higher gender parity (hence religious conservativeness)|
Discussion of results
The results reveal an interesting spatial pattern in terrorist supply in Pakistan. The proxies that has been used to capture poverty seems to have theoretically correct relationship with terrorist supply. It implies that religiously inspired militants in Pakistan are more likely to emerge from relatively poorer districts although they themselves may not belong to impoverished household as indicated by their educational attainment. This is an important finding as it helps filtering the regions who are more likely to be supplier of terrorists. Further, districts with relatively higher level of migration also tend to possess higher level of terrorist supply. Thus we can expect districts with heterogeneous population to offer higher number of terrorists. The case of such metropolitans as Karachi and Lahore readily come to mind who harbors significant migrant population and have also experienced varying levels of militant tensions.
Furthermore, terrorists supply relationship with measure of education follows a non-linear inverted U shape form. It means that the terrorists supply initially increases with higher educational performance but decreases once educational performance exceeds approximately intermediate level. Since terrorists are relatively better off in education then average Pakistanis, it is unlikely that they overwhelmingly emerge from districts that lie at extreme lower end of educational performance. The reason that they are also unlikely to emerge from districts at higher end of educational performance may be due to higher level of development and hence economic opportunities which such educational infrastructure implies. Hence, they are more likely to emerge from districts at average level of educational performance. Finally, the empirical analysis also reveals that terrorists supply is higher from districts characterized by higher level of religious conservativeness as measured by gender parity index. This can be expected because terrorist outfits intend to violently impose many of the beliefs conservative circles in Pakistan also subscribes to such as suppression of religious minorities and women. Thus it can be summarized that terrorists are unlikely to belong to impoverished economic backgrounds but as far as their areas of origin is concerned, they are more likely to emerge from districts characterized by relatively higher level of poverty, migration, religious conservativeness and approximately intermediate level of educational performance.
Part III of this series of article explores background features of terrorists from Pakistan’s majority province also referred to as the “ Punjabi Taliban”. You can read part one here.