Terrorism Risk and Democratic Preferences in Pakistan Faiz Ur - PowerPoint PPT Presentation

Terrorism Risk and Democratic Preferences in Pakistan Faiz Ur Rehman Paolo Vanin European Doctorate in Law & Economics Faiz & Vanin Terrorism and Democracy 1 / 29

Motivation 9,374 terrorist attacks in Pakistan between 1974-2013 (GTD) 7.5% of the world’s total terrorist attacks have been carried out in Pakistan (GTD) Since 2001, estimated cost of terrorism to Pakistan is $102.5 billion (Pakistan Economic Survey) According to a rough estimate, 50,000 lost their lives to terrorism after 2001 (The Express Tribune) Similarly, 18,760 difgerent types of small scale violent incidents have been occurred between 2001-2013 (ESOC) The goal of a terrorist is to create uncertainty and fear through attacks on individuals and their property (GTD) Faiz & Vanin Terrorism and Democracy 2 / 29

Terrorism Trend in Pakistan, 2000-2012 Faiz & Vanin Terrorism and Democracy 3 / 29

Violence Trend in Pakistan, 2000-2011 Faiz & Vanin Terrorism and Democracy 4 / 29

Spatial Distribution of Terrorism, 2004-2012 Faiz & Vanin Terrorism and Democracy 5 / 29

Spatial Distribution of Violence, 2004-2011 Faiz & Vanin Terrorism and Democracy 6 / 29

Research Question What is the democratic legacy of persistent terrorism shocks? Does terrorism afgect political attitudes and preferences regarding democratic values and institutions? Faiz & Vanin Terrorism and Democracy 7 / 29

Main Results Yes, terrorism risk matters for individuals democratic/political preferences. Faiz & Vanin Terrorism and Democracy 8 / 29

Terrorism, Violence and Individual Preferences: Prior Evidence The economic costs of terrorism and confmict: [Benhabib and Rustichini (1996); Abadie and Gardeazabal (2003); Enders and Sandler (2011); Enders and Olson ( 2012)]. Positive efgects of violence and confmict on social capital: [Colletta and Cullen (2000); Wood (2003); Cramer (2006); Bellows and Miguel (2006, 2009); Blattman (2009); Gilligan et al. (2010); Bateson (2012); Voors et al. (2012)]. Negative efgects of violence and confmict on social capital: [Collier et al. (2003); Collier (2007); Pinchotti and Verwimp (2007); Bratton (2008); Collier and Vicente (2014)]. Faiz & Vanin Terrorism and Democracy 9 / 29

Terrorism, Violence and Individual Preferences: Prior Evidence (cont.) Positive efgects of violence on political participation and institutions: [Cramer (2006); Bellows and Miguel (2006, 2009); Blattman (2009); Bateson (2012)]. Negative efgects of violence on political participation and institutions: [ Bratton (2008); Collier and Vicente (2014)]. Negative efgects of insecurity and organized crime on institutions: [Blanco (2012); Blanco and Ruiz (2013)]. Faiz & Vanin Terrorism and Democracy 10 / 29

Research Design: Survey Preferences : Survey of 6,000 households across difgerent provinces to measure socio-economic, political and religious preferences. Rural-Urban Distribution of the Sample : Rural = 67% & Urban = 33%, based on the latest population census. Gender Distribution of the Sample : Male = 52% & Female = 48%, based on the latest population census. Faiz & Vanin Terrorism and Democracy 11 / 29

Research Design: Terrorism & Democratic Preferences 246.87 5 Property Rights 4.34 0.99 1 5 Panel B: District-Level Terrorist Attacks and Violence Incidents Past Terrorism (2004-2008) 52.00 82.04 1 365 Past Violence (2004-2008) 367.20 0.97 1 1278 Current Terrorism (Jan-May 2009) 10.64 18.53 1 84 Current Violence (Jan-May 2009) 75.47 125.22 1 537 Faiz & Vanin Terrorism and Democracy 1 4.19 Variable 5 Mean Std. Dev. Min Max Panel A: Individual-Level Democratic Values (2009) Democratic Values 23.93 3.85 8 30 Elected Representatives 4.10 1.00 1 Freedom of Expression Independent Judiciary 4.14 0.95 1 5 Freedom of Assembly 4.22 0.92 1 5 Civilian Control over Military 2.79 1.35 1 5 12 / 29

Research Design: Democratic Values (Importance Attributed to Democratic Institutions) (a) Elected Representatives (b) Freedom of Expression (c) Freedom of Assembly (d) Civilian Control over the Military (e) Independent Judiciary (f) Property Rights Faiz & Vanin Terrorism and Democracy 13 / 29

Research Design: Defjnition of Terrorism According to the Global Terrorism Database (GTD), a terrorist incident should have the following features. The incident must be intentional. The incident must entail some level of violence or threat of violence. There must be evidence of an intention to coerce and intimidate to convey message to a larger audience (or audiences) than the immediate victims. The perpetrators of the incidents must be sub-national actors. The act must be aimed at attaining a political, economic, religious or social goal. Faiz & Vanin Terrorism and Democracy 14 / 29

Research Design: Matching District Level Attacks with Individual Level Preferences Aggregated (2004-2008) and disaggregated (year by year) district level terrorist attacks are mapped with the individuals democratic attitudes. We assume that every individual has the same probability to expose to a terrorist attack in a given district. Faiz & Vanin Terrorism and Democracy 15 / 29

Control Variables at Individual and District levels Demographic Controls: Gender, age, age square, marital status, education level, population density and urbanization. Economic Controls/Opinions: Nominal income, assets, deprivation index, land reforms, inequality and economic expectations. Language/Culture Controls: Five main languages speak across Pakistan. Religious Controls/Opinions: Sect type, number of prayers per day, recitation of the Holy book, frequency of recitation per week and opinion about the role of the Sharia law. Faiz & Vanin Terrorism and Democracy 16 / 29

Methodological Challenges irrespective of people socio-economic-political preferences. Terrorism and Democracy Faiz & Vanin we used an IV identifjcation strategy. IV Identifjcation Strategy: If the above problems still persist, presence of terrorism risk. show the shift in preferences from one institution to another in the Alternative Test: Preferences on military institution is used to main objective of terrorism is to create an environment of fear Confounding Factor(s): This problem is minimized through the individual level social, economic and political preferences. The Reverse Causality: District level attacks are exogenous to disaggregated samples ( gender, urbanization and education level). spurious regression, we repeat the base line regression for difgerent Heterogeneous Efgects: To further minimize the problem of controls. inclusion of several demographic, economic, cultural and religious 17 / 29

Empirical Strategy 2004-2008 period ( past terrorism ); Terrorism and Democracy Faiz & Vanin democracy socio-economic determinants of individual preferences for Our baseline estimating Probit equation is: 18 / 29 (1) ′ X i + δ ′ Z d + φ p + ε i log DV i = β log PT d + γ DV i represents the democratic values of individual i (who resides in district d , belonging to province p ) in 2009 ( democratic values ) PT d is the district-level total number of terrorist attacks in the X i is a set of individual control variables Z d is a set of district-level controls φ p are province dummies; and ε i is an error term Controls X i and Z d comprise demographic, religious and

Baseline Evidence: Terrorism and Democratic Values Y Y N N N Economic Controls Y Y Y Y N N Demographic Controls Y Y Y Language Controls Y N Terrorism and Democracy Faiz & Vanin Y N N N N N Religious Controls Y Y N N N Y Y (1) -0.028*** [0.003] [0.001] -0.030*** -0.031*** -0.029*** -0.027*** -0.021*** [0.003] Past Terrorism (log) (6) (5) (4) (3) (2) [0.003] [0.003] N 0.036 Province Dummies 0.297 0.295 0.303 0.182 0.181 R-squared [0.003] 3,947 3,983 4,293 5,626 5,626 5,626 Observations 19 / 29

Terrorism and Distinct Democratic Values Y Y Y Y Y Economic Controls Y Y Y Y Y Y Demographic Controls Y Y Y Language Controls Y Y Terrorism and Democracy Faiz & Vanin Y Y Y Y Y Y Religious Controls Y Y Y Y Y Y Y E.Rep. -0.026*** [0.006] [0.007] -0.015** -0.021*** -0.067*** -0.026*** -0.045*** [0.014] Past Terrorism (log) P.Rights I.Judic. M.Control F.Assemb. F.Exp. [0.005] [0.007] Y 0.12 Province Dummies 0.23 0.15 0.23 0.20 0.21 R-squared [0.006] 4,082 4,060 4,029 4,061 4,067 4,060 Observations 20 / 29

Past and Current Exposure to Terrorism and Violence Y Y Y Demographic Controls Y Y Y Y Economic Controls Y Y Y Language Controls Y Y Y Y Y Religious Controls Y Y Y Y Faiz & Vanin Terrorism and Democracy Y Province Dummies (1) [0.004] (2) (3) (4) Past Terrorism (log) -0.030*** [0.003] Past Violence (log) -0.020*** [0.003] Current Terrorism (log) -0.007* Current Violence (log) 0.295 -0.024*** [0.003] Observations 3,947 3,947 3,947 3,947 R-squared 0.297 0.283 0.290 21 / 29