Socioeconomic Consequences Of Terrorism Evidence From Pakistan

After the 9/11 incident,Pakistan emerged as the frontline ally in the war against terrorism and undergoes massive damages in social,economic and demographic aspects.Pakistan has been plagued with terrorist incidents during the past couple of decades that makes it the most feasible choice to study the impact of persistent exposure of terrorism on various socioeconomic traits.Pakistan has suffered heavy losses in the form of human lives,both social and physical capital.The foremost objective in this thesis is to study the impact of terrorism on three core aspects i.e.fertility intentions of an individual woman living in a particular household;personal disposable income of an individual;education continuing choice at different educational levels such as primary,middle and secondary education levels.In all three empirical chapters,we employ five data sources.First,the household survey data is obtained from the Pakistan Social and Living Standards Measurement(PSLM).Pakistan Social and Standards Living Measurement(PSLM)is collected by the Pakistan Bureau of Statistics on alternate years.We employ three data waves of PSLM from 2010 to 2015.Second,Labor Force Statistics(LFS),surveyed by the Pakistan Bureau of Statistics(PBS)every year.We exploit data waves from 2001-02 to 2014-15 of Labor Force Statistics(LFS).Both PSLM and LFS cover all the four provinces of Pakistan,except the conflicted areas adjacent to the Pak-Afghan border.Third,the total distance in kilometers from each district in all four provinces to the two exit points in KPK and Balochistan,respectively.We use google maps to find the distance and convert this distance into kilometers by using the National Hurricane Center calculator.Fourth,we employ terrorism-related data from the Global Terrorism Database(GTD).We consider different measures of terrorism such as terrorist incidents,people killed and people injured.The data is obtained from 2000 to 2015.We categorize terrorist incidents into two main classifications.The first one comprises domestic and transnational terrorist incidents.The second classification consists of suicide and non-suicide terrorist incidents.There is a total of 4646 terrorist incidents take place in Pakistan during the period of 2000-15.Finally,we also employ drone attacks from The Bureau of Investigative Journalism(The Bureau of Investigative Journalism,2015).The second chapter of this thesis addresses the question e.g.how the persistent exposure of terrorism influences the decision making of an individual woman to desire for an additional offspring.We employ a pooled cross-sectional household dataset from Pakistan Social and Living Standards Measurement.It covers all the114 districts across Pakistan from 2010 to 2015.We exploit terrorist incidents from 2008-2013.We uncover the impact of terrorist incidents of the second lag on the fertility intentions of a woman.The benchmark findings suggest that terrorist incidents increase the fertility intentions of a married woman that we measure by the individual woman's response to giving birth during each year.We also address the potential endogeneity problem.We employ the interaction of the two variables as our instrumental variable.The two terms are the inverse of the distance to Afghanistan border from each district situated in Pakistan and the log of the number of drone attacks every year,as an instrumental variable.Our benchmark result suggests,a standard deviation increase in the two lags of terrorist incidents explains 0.052 standard deviation increase in fertility intentions.In order to comprehend this slight variation,we put this into another perspective.For every one million married women,a standard deviation increase in terrorist incidents translates approximately 51,650 more births during their lifetime.We reveal very thought-provoking results about the fertility intentions of an individual woman based on our heterogeneous sample.The women with age bracket 31 to 45 years are more sensitive to response terrorist incidents significantly and positively compared to other age groups.The woman possesses either no formal education or lower educational level are also more responsive to terrorist incidents.The women with no availability of pre-natal care are more exposed to positive fertility intentions in reaction to terrorism.The woman whose spouse's personal disposable income lower than average and/or involved in the agricultural sector also reveals significantly in fertility intentions.This finding supports the notion that people with lower income and working in the agricultural industry are more likely to have additional kids to combat future poverty and make sure their smooth potential income in the future.The women reside in intensive terrorist incidents regions and those residing in the districts located within 200 kilometers to the Pak-Afghan border are positively and highly reactive to the two lags of terrorist incidents.We also discover the impact of the terrorist incidents on fertility intentions of a married woman for each year distinctly from 2010 to 2015.The third chapter explores the impact of terrorist incidents on personal disposable income.We exploit pooled cross-sectional data of Labor Force Survey from 2001-02 to 2014-15.We investigate the causal relationship between terrorist incidents that take place during 2000-2015 and the personal disposable income.To quantify and isolate the causal impact,we employ an instrumental variable approach based on the distance to the Afghan border for each district interacted with the number of civilians killed in drone strikes plotted by the US military in Pakistan.The institution behind multiplying by the number of civilians is that more terrorist incidents take place in response to civilians' death We apply years and district fixed effects,demographic controls,and economic controls,employ instrumental variable strategy to overcome omitted variable bias and reverse causality,and we isolate the causal influence of terrorism on personal disposable income.Our baseline findings suggest that a one-unit increase in the past year terrorist incidents(with an average of 5.42),causes 4.14 percent decrease in the personal disposable income i.e.for every million rupees income there will be a reduction of approximately 41,400 rupees.In simple words,if there would no terrorist attack takes place,an individual can earn almost 4 rupees more for every100 rupees income.Based on heterogeneous results,we also separate the influence of various terrorism types on personal disposable income.The transnational and suicide terrorist incidents have no significant relationship with personal disposable income.The individual with no formal and lower educational level(primary school or lower)belong to a particular household and living at a district level,also has a negative influence on his personal disposable income.The sole proprietors are less secure in the current business environment in Pakistan.Their income is more receptive to terrorism.The continuous lockdown takes place after persistent terrorist incidents.The individuals work in private organizations have more job risk compared to the individuals working in government organizations.Their income is also negatively influenced by terrorist incidents.The personal disposable income of individuals involved in non-elementary occupations is also adversely affected by terrorism.These non-elementary occupations include cleaners,helpers,labor in construction and agriculture,and street related service workers.Mostly terrorist incidents take place in urban areas.The firms located in urban areas have a negative impact on terrorism.Moreover,only the domestic and non-suicide terrorist incidents are significantly negatively affecting the personal disposable income of individuals.The fourth chapter examines the influence of terrorist incidents on educational continuation decision making at different educational levels such as primary,middle and secondary education levels.After employing five waves of Pakistan Social and Living Standards Measurements(PSLM)surveyed on alternate years during 2006-2015.We exploit terrorist incidents from 2005 to 2014.The reason for this time span is that we disclose the impact of lag year terrorist incidents on the education continuation decision.The baseline findings suggest that terrorist incidents significantly reduce the possibility of education continuation of individuals at different education levels.The terrorist incidents may be endogenous because of omitted variable bias as well as reverse causality.Thus,we resort to the instrumental variable strategy to identify the causal effect.Following Ostwald(2015),we employ lag of average total terrorist incidents in neighboring districts as an instrumental variable.There is no possibility of migration between the provinces.There are almost 114 districts in Pakistan out of which approximately more than 100 districts absorb terrorist incidents.Our benchmark findings in chapter four suggest that terrorism has an adverse impact on all levels of education i.e.obtaining primary school & above education,middle school & above education and high school & above education.However,the primary school educational level is adversely affected the most by terrorism.The IV results suggest that a standard deviation increase in terrorist incidents during the past year causes a reduction in the possibility of education continuation by 2.65 percentage points in the overall sample.In simple context,a standard deviation increase in the lag 1-year terrorist incidents,for every one million people,26,501 fewer individuals choose to continue their education in the overall sample of primary school & above education level.The terrorist incidents have a less adverse impact on education continuation decisions by individuals belong to middle school & above and high school & above educational level.We divide the overall sample into different categories.Based on these subsample analyses,we find that the influence of terrorism has strong heterogeneity.First,we reveal the terrorism impact on the education continuation subject to the classification of terrorism.The findings suggest that domestic and non-suicide terrorist incidents are significant and negatively related to education continuation decision making.However,transnational and suicide incidents are too weak or place no significant effect on decision making as well.The second sample classification is based on the income level of the areas within the urban region of each district.There are four categories of such income classification i.e.no-income,low-income,middle-income and high-income areas.Terrorist incidents place much more adverse effect on the education continuation decisions in the‘no-income' classification area2.In addition,terrorist incidents have a similar and significant adverse impact on education investment decisions in both low-income and middle & high-income classification regions but lower in magnitude compared to the no-income classification.The terrorist incidents put more hostile influence on education investment decisions of females than males.The terrorist plagued regions create many hurdles for the females to continue their education.More females than males stop their education after persistent exposure to terrorist incidents.