The Impact of Fundamentalist Terrorism on School Enrolment ...

DISCUSSION PAPER SERIES

IZA DP No. 10168

The Impact of Fundamentalist Terrorism on School Enrolment: Evidence from North-Western Pakistan, 2004-09

Sarah Khan Andrew J. Seltzer August 2016

Forschungsinstitut zur Zukunft der Arbeit Institute for the Study of Labor

The Impact of Fundamentalist Terrorism on School Enrolment: Evidence from North-Western Pakistan, 2004-09

Sarah Khan

Georg-August-Universit?t G?ttingen

Andrew J. Seltzer

Royal Holloway, University of London, LSE and IZA

Discussion Paper No. 10168 August 2016

IZA P.O. Box 7240

53072 Bonn Germany

Phone: +49-228-3894-0 Fax: +49-228-3894-180

E-mail: iza@

Any opinions expressed here are those of the author(s) and not those of IZA. Research published in this series may include views on policy, but the institute itself takes no institutional policy positions. The IZA research network is committed to the IZA Guiding Principles of Research Integrity. The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center and a place of communication between science, politics and business. IZA is an independent nonprofit organization supported by Deutsche Post Foundation. The center is associated with the University of Bonn and offers a stimulating research environment through its international network, workshops and conferences, data service, project support, research visits and doctoral program. IZA engages in (i) original and internationally competitive research in all fields of labor economics, (ii) development of policy concepts, and (iii) dissemination of research results and concepts to the interested public. IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper should account for its provisional character. A revised version may be available directly from the author.

IZA Discussion Paper No. 10168 August 2016

ABSTRACT

The Impact of Fundamentalist Terrorism on School Enrolment: Evidence from North-Western Pakistan, 2004-09*

Islamist groups in Afghanistan, Pakistan, and elsewhere have sought to remove females from public life. This paper uses data from Pakistan Social and Living Standards Measurement and the Global Terrorism Database to examine the impact of the Pakistani Taliban's terror campaign in the north-western province of Khyber Pukhtoonkhwa aimed at removing girls from school from the age of 10. Using a difference-in-difference-indifference approach, we show that low levels of exposure to terrorism had little effect on school enrolment. High levels of exposure reduced the enrolment rate for boys by about 5.5 percent and girls by about 10.5 percent. This decline in enrolment, although strongly significant, is far smaller than has commonly been portrayed in the media. Finally, although the Taliban warned students to enrol in madrassas rather than secular schools, we find no evidence that this led to increased madrassa enrolment in the affected regions.

JEL Classification: O15, I25, D74, O53 Keywords: education, terrorism, Pakistan

Corresponding author: Andrew Seltzer Department of Economics Royal Holloway, University of London Egham, Surrey TW20 0EX United Kingdom E-mail: a.seltzer@rhul.ac.uk

* This paper is a revised version of chapter 2 of Sarah Khan's PhD thesis completed at Royal Holloway, University of London. We have benefited from comments and suggestions by Bishnupria Gupta, Juan Pablo Rud, and Diana Weinhold and participants at a seminar at Royal Holloway.

1

I. Introduction

It has been increasingly recognized that conflicts have serious adverse effects on children's educational attainment (UNESCO, 2010) (Justino, 2011). Conflicts can affect schooling through multiple channels. Active conflicts may have direct effects such as the destruction of school buildings or roads needed to reach schools, the loss of teachers to violence or intimidation, the recruitment of children as soldiers, the displacement of students and their families, and the creation of a climate where parents are afraid to let children leave the home. There also exist indirect mechanisms such as reallocation of resources within households; for example, school children needing to find work to replace lost family income. In turn, reduced school attendance may result in a permanent decline in the stock of human capital, both at the individual and more aggregated levels, and to worse health and labour market outcomes. Several studies have shown that these effects can be quite large, depending on the age and enrolment status of children at the time of exposure. Akbulut-Yuksel finds that German children who resided in cities heavily targeted by allied bombings during World War II went on to attain significantly fewer years of schooling and were on average one centimetre shorter than children of same the age in non-targeted areas (Akbulut-Yuksel, 2014). Swee finds that the recruitment of child soldiers during the Bosnian civil war of 1992-95 led to a decrease in the likelihood that students completed secondary education, but had no significant effects on primary education (Swee, 2015). Akresh and de Walque find that school age children exposed to the genocide in Rwanda experienced a drop in completed schooling of almost half a year and were 15 percent less likely to complete 3rd or 4th grade (Akresh and de Walque, 2008).

In addition to reducing the overall level of education, conflicts may have important implications for the gender-based education gap. There may exist real or perceived differences by gender in the risk of violence, harassment, or abduction during conflicts; possibly leading to families being less willing to allow girls to attend school. In addition, during a conflict households may reallocate increasingly scarce resources away from daughters and toward sons. Conversely, boys are more likely than girls to be recruited as child soldiers and forced out of school as a result. Previous studies examining genderspecific outcomes have had mixed findings on the relative impact of conflict on boys and girls. Chamarbagwal and Moran examine the effects of the civil war in Guatemala

2

and find that during the peak conflict years (1979-84) Mayan girls in areas worst hit by conflict attained 12 percent fewer years of education than in the pre-war period, while for Mayan boys this gap was 15 percent (Chamarbagwala and Mor?an, 2011). On the other hand, Shemyakina finds that the civil war in Tajikistan had much larger effects on girls than boys (Shemyakina, 2011). Stewart, et. al. study the impact of armed conflict in several African countries and find that in some countries boys' enrolment declines more than girls due to conflict, whereas the reverse is true for other countries (Stewart and Wang, 2001).

An important recent change in the nature of conflict has been the increasing prevalence of Islamic extremist groups with ideological agendas regarding the role of females. These groups follow a strict interpretation of Islamic law that restricts girls from a fairly young age from participating in most activities outside the household, including schooling. In countries including Afghanistan, Nigeria, Pakistan, and Somalia, Islamist movements have engaged in targeted violence at girls and their schools (United Nations, 2016). This sort of targeted violence is likely to amplify the gender-based effects of conflict, and thus conflicts involving these groups are likely to have gender-specific outcomes that are very different from those of other types of conflicts, such as ethnically-motivated civil wars. This paper examines Islamic fundamentalist terrorism and gender gaps in schooling, using the case of the Pakistani province of Khyber Pukhtoonkhwa (henceforth KPK). Beginning in 2007 the Pakistani Taliban engaged in a campaign to remove girls from public life from age 10. Girls under age 10 and all boys were permitted to remain in school, although in some areas they were warned to attend madrassas (religious schools) rather than state schools. The Taliban enforced this edict with violence; bombing schools at night and making threats to both teachers and students. Both boys' and girls' state schools were attacked, although girls schools were disproportionately targeted (Hayat, 2009) (Khan, 2012) (Amin, 2008) (UNESCO, 2010). According to the Global Terrorism Database there were about 80 attacks against schools in 2008 and 2009 (Global Terrorism Database, 2016). Other sources put the numbers much higher (UNESCO, 2010). Individuals speaking out against the edict, such as the teenage blogger and future Nobel laureate Malala Yousafzai, were dealt with harshly.1

1In early 2009, Yousafzai wrote a blog entry exposing how a decree in forbidding girls from attending school was enforced by the Taliban. On 9 October, 2012 Taliban gunmen halted a van transporting Yousafzai and other children home from school, and shot her in the head and neck. When she survived her injuries, a Taliban spokesman promised that they would "finish the job the next time" (Brumfield, 2012).

3

Because the Taliban did not have complete military control of Pakistan, this campaign was limited in scope, focussing primarily on KPK in the northwest of the country, close to the Afghanistan border.2 Even within KPK, there was considerable variation in the pattern of violence. The bulk of the attacks recorded in the Global Terrorism Database occurred in two districts: Peshawar and Swat. Another group of districts experienced at least one attack, but far fewer than either Swat or Peshawar (Bannu, Charsadda, Dir, Hangu, Malakand, Mansehra, Mardan, Nowshera, Swabi, and Tor Ghar). A third group of districts did not experience any attacks (Abbotabad, Batagram, Buner, Chitral, Dera Ismael Khan, Haripur, Karak, Kohat, Kohistan, Lakki Marwat, Shangla, and Tank). The two high exposure districts experienced very different patterns in the timing of attacks. In Swat, the number of attacks was high in 2007 and 2008. However, a successful military operation by the Pakistani Government cleared the Taliban from the area in March 2009. From this point forward, there were no further attacks against schools recorded in the Global Terrorism Database. By contrast, the number of attacks in Peshawar remained high throughout the period of our study.

A variety of media outlets have claimed that the campaign had drastic effects on school enrolment, particularly for girls. For example, IRIN News claimed that terrorist attacks lead to the closure of 900 schools and a nearly 60 per cent fall in girls' enrolment between 2007 and 2009 (Hayat, 2009). An Amnesty International report claimed that thousands of children in KPK had been deprived of an education (UK Home Office, 2012). Other reports claimed that teachers remained too intimidated to return to work long after attacks had taken place, even after the military defeat of the Taliban in Swat (UNESCO, 2010). A 2012 article in the Guardian claimed that even three years after the defeat of the Taliban in Swat, girls' return to schools was an ongoing process (Khan, 2012). Although the bulk of media reports claimed that the school bombing campaign had a significant effect on girls' enrolment, some reports claimed that the Taliban's campaign had limited success (Michael and Qasim, 2011). One factor which makes it difficult to assess the veracity of these opposing claims is that the underlying data on school enrolment is not discussed in any report of which we are aware. In many cases the evidence is anecdotal, often based on anonymous interviews with Pakistani Education

2The Taliban had a strong military presence in KPK and in the Federally Administered Tribal Areas. However, the tribal areas are sparsely populated and Taliban actions in these areas were directed primarily against the Pakistani Military.

4

Department officials. We are unaware of any systematic attempt to estimate the overall effects across KPK.

Our study relies on a large random sample of households in KPK to examine the determinants of school enrolment and the impact of terrorism in KPK. We use data covering individual children from the Pakistan Social and Living Standards Measurement (henceforth PSLM), a large, nationally-representative household survey conducted annually since 2004. During the period of our study, the PSLM was carried out all four Pakistani provinces (Punjab, Sindh, Balochistan, and KPK) and the Federal Capital Territory in every year. A sixth administrative unit, the Federally Administered Tribal Areas, was not surveyed in all years. The survey is organised at the level of the household and covers education, health, social capital, marriage and fertility, employment, and economic status. We use the PSLM data from 2004-05, the first year the surveys were conducted, to 2008-09, when military operations removed the Taliban from Swat. We link the data from the PSLM to district-level data on terrorist incidents directed against schools from the Global Terrorism Database (GTD). We use this data to create an exposure to terrorism variable, based on year and district. Although the Taliban campaign continued past 2009, we have ended the study at this point because Pakistan experienced a massive flood which covered one fifth of the country in the summer of 2010. KPK, and in particular Peshawar, experienced some of the worst flooding in the country (OCHA, 2010). Many residents from districts that experienced school bombings were forced to relocate because of the flooding, and it is likely that flooding had a much larger effect on school enrolment than did terrorism.

Our empirical strategy is based on a natural experiment created by variation in exposure to terrorism. We estimate the effects of terrorism on schooling, and in particular the differential effect on girls relative to boys, using a difference-in-difference-in-difference approach. The relevant differences are across sex, year ? grouped in to the pre-attack years (2004-05, 2005-06, and 2006-07) and the post-attack years (2007-08 and 2008-09) ? and extent of terrorism across districts. We run separate regressions for primary school age children (age 5 to 9), who were in principle not affected by the Taliban edict; secondary school age children (10-14), who were affected by the edict; and upper secondary school age children (15-18), who were also affected by the edict but who differed significantly from the their younger counterparts in terms of observable characteristics. In

5

addition, we run separate regressions for religious schools, as the Taliban encouraged both boys and girls to attend madrassas rather than state schools.

Our main result is that the effects of the Taliban's campaign were much more limited than has been suggested by most media accounts. We find no evidence of any decline in enrolment in the low intensity districts. Even in Swat and Peshawar, we find no evidence of a decline in enrolment until 2008-09. The overall decline across all school-age children was on the order of 5.5 percent for boys and 10.5 percent for girls in these districts in 2008-09. As would be expected based on the nature of the Taliban's edict, we find a larger effect for children age 10-18 than those age 5-9. Finally, we find no evidence of students switching from enrolling in state or private schools to enrolling in madrasas. Madrasas accounted for a very small percentage of enrolment in KPK throughout the period.

The outline for the remainder of the paper is as follows. After the introduction, the second section provides a brief history of the conflict in KPK and the Taliban's campaign to remove girls from public life. The third section describes the data used in this paper. The fourth section provides a brief outline of the schooling system in Pakistan and some summary statistics on school enrolment prior to the start of the Taliban's campaign. The fifth section describes our empirical methodology and shows our results. Finally, the sixth section concludes.

II. The Pakistani Taliban and the conflict in Khyber Pukhtoonkhwa

Pakistan has experienced a series of civil conflicts dating back to its founding in 1947, following the partition of Colonial India. The country is currently experiencing sectarian conflict in Balochistan, a protracted low-intensity dispute with India over the region of Kashmir, and an Islamist insurgency in the semi-autonomous Federally Administered Tribal Areas (FATA) and in KPK. It is the insurgency by the Tehrik-i-Taliban Pakistan (or Pakistani Taliban) in KPK, which is the focus of our paper. The Taliban were formed in the Tribal Areas, but the vast majority of their attacks against schools have taken place in KPK. This section briefly outlines the history of the Islamist insurgency and the school bombing campaign.

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download