Published on February 16, 2014
An analysis of the social impact of the stipend program for secondary school girls of Khyber Pakhtunkhwa Vaqar Ahmed & Muhammad Zeshan Educational Research for Policy and Practice ISSN 1570-2081 Educ Res Policy Prac DOI 10.1007/s10671-013-9154-4 1 23
Your article is protected by copyright and all rights are held exclusively by Springer Science +Business Media Dordrecht. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”. 1 23
Author's personal copy Educ Res Policy Prac DOI 10.1007/s10671-013-9154-4 ORIGINAL ARTICLE An analysis of the social impact of the stipend program for secondary school girls of Khyber Pakhtunkhwa Vaqar Ahmed · Muhammad Zeshan Received: 18 April 2013 / Accepted: 6 August 2013 © Springer Science+Business Media Dordrecht 2013 Abstract The present study carries out an impact analysis of a conditional cash transfer (CCT) program for secondary-school girls in seven districts of Khyber Pakhtunkhwa province in Pakistan, including Battagram, Bonair, Hangu, Kohistan, Shangla, Tank, and Upper Dir. In 2012 we collected household-level primary data and used a probit model for quantitative analysis. Further, we conducted various focus group discussions and key informant interviews in the target areas. Results show that the chances of female schooling decrease with a rise in family size. The chances of female education increase by 1.8 and by 3.3 % if household heads and their spouses have one additional year of schooling, respectively. Better educational services and rises in family income increase the chances of female ecucation by 11 and 0.3 %, respectively. Finally, socioeconomic awareness, improved economic conditions, and CCTs increase the chances of female education by 5.2, 4.7, and 0.03 %, respectively. Overall, the stipend program (CCTs) shows a pareto improvement. Our results indicate that 35 % of girls will drop out in the absence of a stipend program. The present study recommends that to increase program effectiveness, local-level monitoring and program evaluation may be improved, delays in stipend payments to female students should be reduced, a grievance redressal mechanism for parents and guardians should be introduced, and clear synergies should be developed with other transfer programs. Keywords Education · Poverty · Conditional cash transfers · Gender JEL Classiﬁcation P36 · J16 · J18 1 Introduction The propoor impacts of education have been widely discussed in the literature (Johnson 2013; Ladd 2012). The channels and time period in which education impacts poverty have also been V. Ahmed · M. Zeshan (B ) Sustainable Development Policy Institute, Islamabad, Pakistan e-mail: email@example.com; firstname.lastname@example.org; email@example.com 123
Author's personal copy V. Ahmed, M. Zeshan discussed. Thorpe et al. (2013) claims that female education has a greater impact on their well-being but in a longer time frame. The short-run rigidities in the labor market (particularly in developing countries) do not favor females, making it difﬁcult to overcome poverty in the short run. This claim is supported by Roche (2013), who ﬁnds that educated women reap considerably lower economic rewards for their services. This heterogeneity in returns to education creates difﬁculties for women to rise out of poverty in the short run. However, United Nations Educational, Scientiﬁc and Cultural Organization (UNESCO) believes that education can lead to a reduction in poverty and promote growth in developing countries by gradually reducing anomalies in the labor markets (UNESCO 2012a). Education provides opportunities to deprived marginalized classes and makes them active players in the economic growth process. At the Rio 20 conference held in Brazil in 2012, the United Nations declared that the education gap was the key difference between the developed and developing worlds.1 In response to the ﬁndings of this conference, many developing countries are now set to redesign their national and subnational education policies to address the various forms of poverty and inequality that represent the main hurdles to sustained growth. The key idea here is to use education as an active tool for job-oriented growth and welfare improvements. The Fourth World Conference on Women, held in Beijing in 1995, recognized that female literacy is a key determinant of empowerment for women.2 It enables them to participate in household decision-making processes and makes them active members of household affairs. In this era of rapid development, depriving females of equal opportunities will result in an unsustainable development milieu (see also World Bank 2012). Several past governments in Pakistan prioritized education; however, the sector faces enormous difﬁculties in implementing policy and practice corrections. Following passage of the 18th constitutional amendment, the education policy is now a provincial subject. Provincial governments are now working rigorously toward province-speciﬁc education policies. The setup of society and the prevalant social capital are also important ingredients in the success of any policy. As Pakistani culture is dominated by a patriarchal social structure, a vast body of literature recommends advocacy and outreach programs aimed at promoting literacy. The conservative culture that maintains a status quo in gender roles is also an obstacle in the way of achieving the Millenium Development Goals (MDGs), especially in remote areas of Pakistan where females have little participation in decision making regarding their or their children’s education. Women are supposed to get permission from the household head, who is normally male, before seeking out educational, travel, employment, or business opportunities. Most families traditionally tend to make relatively less investment in the education of girls compared to boys. Figure 1 exhibits the trend over time in female education in Pakistan. Progress is far from satisfactory, particularly in the case of secondary education. It is also important to see the trend in the availability of female teachers in Pakistan (Fig. 2). Due to the conservative norms in parts of Pakistan, parents will only send their female children to school where there are female teachers. Schools without female teachers have seen a decline in female enrollments. Growth in the availability of female teachers, particularly at the primary-school level, has been less than desired. The government’s budgetary allocations and their disbursements over time indicates the low priority attached to education in general. In fact, this raises doubts about the government’s 1 http://www.uncsd2012.org/about.html. 2 United Nations Entity for Gender Equality and the Empowerment of Women (2012), http://www.un.org/ womenwatch/daw/beijing/fwcwn.html. 123
Author's personal copy An analysis of the social impact of the stipend program Fig. 1 Female gross enrollments (%). Source World Development Indicators Fig. 2 Availability of female teachers (%). Source World Development Indicators claims regarding the seriousness of its commitment toward achieving MDGs. More recently, just 2 % of GDP was allocated for education in the ﬁscal year 2010–2011. This inadequate investment resulted in a literacy rate of only 67 % for males and 42 % for females, whereas the rate is 65 % for males and 28 % for females in Khyber Pakhtunkhwa (the province on which we will focus in this paper).3 Some studies indicate that even these low levels of allocations are not fully disbursed by year end (Government of Pakistan 2012). The literacy rate is low compared to global trends, which is 61 % for young females between the ages of 15 and 25 and 79 % for males in the same age group. UNESCO (2012b) explains how a committed implementation by government of its own policies could raise literacy rates up to 72 and 82 % for males and females, respectively, by the year 2015. However, the challenge is that about half of the female population has never visited a school and 35 % of them live in rural areas. To overcome this situation, in 2000, governments at various levels in the country introduced targeted educational policies. The stipend program for female secondary-school students was also introduced for the same purpose in Khyber Pakhtunkhwa province. 1.1 Khyber Pakhtunkhwa’s context The provincial government of Khyber Pakhtunkhwa initiated a stipend program for female secondary-school students in 2007 that aimed to increase secondary-school enrollments of 3 Source: Pakistan Social and Living Standards Measurement 2010–2011. 123
Author's personal copy V. Ahmed, M. Zeshan Table 1 Literacy rate for 2010–2011 (%)– population 10 years and older Region Urban Rural Total Male Female Total Male Female Total Male Female Total Khyber Pakhtunkhwa 77 50 63 67 29 48 68 33 50 Upper Dir 67 35 52 71 25 50 71 25 50 Shangla 0 0 0 58 16 37 58 16 37 Bonair 0 0 0 53 14 32 53 14 32 Hangu 77 38 56 69 15 41 71 20 44 Tank 81 44 64 54 10 33 57 14 36 Battagram 0 0 0 70 28 49 70 28 49 Kohistan 0 0 0 42 6 26 42 6 26 Source Pakistan social and living standards measurement (PSLM) Survey 2010–2011 Battagram Bonair Hangu Kohistan 14308 Shangla Tank 80 14308 80 2002 2004 2006 2008 2010 2002 2004 2006 2008 2010 Upper Dir 14308 80 2002 2004 2006 2008 2010 Boys’ Enrollments Girls’ Enrollments Fig. 3 Trends in middle-level schooling 2002–2010. Source Bureau of Statistics Khyber Pakhtunkhwa females. The target area of this stipend program comprised seven of the poorest districts, including Battagram, Bonair, Hangu, Kohistan, Shangla, Tank, and Upper Dir. A disaggregated analysis of these districts is provided in Table 1. It is evident that most of these districts have literacy rates below the provincial average, except for Upper Dir and Battagram. Other than the low literacy rate, the gender gap is obvious, especially in rural areas. The enrollment trend in middle-level4 schooling for boys and girls indicates that the gender gap is smaller in those districts where the overall student population is small, for example, the Battagram, Hangu, Kohistan, and Tank districts. On the other hand, the differences are larger in districts where total enrollments are high, such as in the Banair, Upper Dir, and Shangla districts. This explains the fact that gender discrimination is relatively more pronounced in larger districts (populationwise) compared to smaller districts (see Fig. 3 for details). 4 Class 6–8 is middle-level schooling. 123
Author's personal copy An analysis of the social impact of the stipend program Table 2 Genderwise educational statistics of Khyber Pakhtunkhwa province School level No. of schools Male Enrollment rate Female Male Primary (Class 1–5) 14,600 7,583 Participation rate Female Male (%) Budgetary allocation (%) Female (%) 1,610,024 803,204 98 66 Middle (Class 6–8) 1,436 920 142,407 74,714 51 25 51 15 High (Class 9–10) 1,069 310 358,383 128,550 40 17 34 Source PC-1 Document, Planning & Development Department, Government of Khyber Pakhtunkhwa 2011 More speciﬁcally, the gender disparity is low in the Tank, Hangu, and Kohistan districts. This is interesting because these three districts are close to the Afghanistan border and have suffered due to the ongoing war on terror. These primary ﬁndings show that even if parents are interested in educating their female children, there are lacunae in the female schooling system that need to be identiﬁed. The present study aims to precisely identify these factors. This analysis is conducted with the help of quantitative and qualitative techniques discussed in the following sections. It is important here to discuss some programmatic details about this stipend program. Primarily it aims to reduce the gender gap in Khyber Pakhtunkhwa province, which is currently 26 % in secondary-level schooling. To this end, it allocated 30 billion Pakistan rupees (PKR) under the Annual Development Program (ADP) in 2007. This program is part of the Mid-Term Development Framework that aims to reduce gender imbalances to allow the development process to work on a more equitable basis. In 2008, the province had a 57 % literacy rate for males and 20 % for females. The fundamental reason for this low female literacy rate is low income and a conservative social context. In this province, girls are also expected to work with their parents to obtain the basic necessities of life. The stipend program intends to create an incentive for parents to send their children to secondary school because many rural families follow the decades-old tradition of not sending their girls to school beyond the primary level. Table 2 speciﬁes some basic facts of the educational system in Khyber Pakhtunkhwa province. From Table 2 it is obvious that a substantial proportion of girls are not enrolled in secondary schools, which in turn increases the gender gap. The budgetary allocation seems unfair for females despite the government’s claim that it is committed to reducing the gender gap. The incentive of the stipend at the secondary level aims at incentivizing poor families to educate their girls and create a mindset change in the region. In the target districts, girls are the main victims of educational poverty, so this stipend program targets only poor girls. The expected outcome is a reduction in the number of dropouts and improvements in girls’ enrollments at the secondary level, especially in remote areas. In this program, stipend money of PKR 200/month was given to each girl starting in 2006 (between Classes 6 and 10). After 1 year, in 2007, around 0.28 million female students were registered under this program. This project also fulﬁlls the objectives of the National Education policy, Education Sector Reforms, and the Poverty Reduction Strategy of the Government. The conditions imposed on the stipend money brings opportunity costs for households. Supporters of such conditions are of the view that a conditional cash transfer (CCT) allows efﬁcient resource utilization. Hence, the gains from these conditions might outweigh the organizational costs and society might achieve a society would have a better human capital 123
Author's personal copy V. Ahmed, M. Zeshan (Bassett 2008). In a rural setting, in making decisions about their children’s schooling, parents are also interested in the social beneﬁts associated with investment in education. In this case, the imposition of conditions might increase enrollments of females when parents see a stream of higher future income. To obtain the desired results, the disbursement of funds is based on some condition, for example, a female student must attain 80 % school attendance to receive the next installment of funds. The present study also aims to conduct an impact analysis of the conditional cash transfers in Khyber Pakhtunkhwa to date. At the same time, we aim to identify the prospective changes required in this program to enhance its progress. We will examine the various transmission channels through which this CCT in the education sector can impact various forms of poverty. To this end, our approach analyzes the educational and noneducational factors that add to or detract from the intended impact of the transfer program. The rest of the study is organized as follows. In the next section we discuss the contemporaneous literature on this topic, which includes global and national studies. Section 2 elaborates on the research methodology adopted in this study. Section 3 explains the empirical results, and Sect. 4 concludes the study and provides policy recommendations. 1.2 Literature review This section brieﬂy looks at the relationship between female education, poverty, and welfare. Thorpe et al. (2013) has evaluated the impact of female education on the well-being of African American females using a logistic model. It ﬁnds that female education has a greater impact on their well-being but in the longer run. The rigidities in the labor market provide smaller incentives for females in the shorter run. This also implies that females ﬁnd it difﬁcult to overcome poverty in the short run but can avoid poverty in the long run. Similar results are provided by Roche (2013), who ﬁnds self-employed women receive relatively lower economic rewards for their services compared with self-employmed men. This difference in returns to education makes it difﬁcult for women to escape poverty in the short run. Further, Johnson (2013) ﬁnds that graduate-level enrollment has a countercyclical impact on female employment but an acyclical impact on male employment. This result implies that during a recessionary phase, the number of women losing their jobs is higher than the number of males. Hence the labor markets offer a substitution effect in favor of males during recessions. In the following part of this section, we review the existing literature on the role of conditional cash transfer programs and their impact on education and social well-being. We split these studies into global and national studies for ease of creating linkages with this paper. Moreover, in the case of national-level studies, we will also look at the case of Khyber Pakhtunkhwa province and the state of female education. 1.3 Global literature Conditional cash transfers are frequently employed around the world to alleviate education poverty. Along with facilitating education, they also work to change the mindsets of local populations and mold their thinking in favor of positive externalities arising from eduction (see Faith and Vinay 2010). Certain conditions are generally attached to the availability of CCT funds, increasing the efﬁciency of the programs. Skouﬁas and Maro (2008) examine the efﬁciency effects of the Progresa Program in Mexico. This program is designed to increase school enrollments among children so that they can become active participants in the future labor market. These educational grants are provided only to households with children 123
Author's personal copy An analysis of the social impact of the stipend program enrolled in Grades 3–9. The study concludes that these CCTs are effective at increasing the literacy rate and labor force participation over time in Mexico (see also Saavedra and Garcia 2012). These CCT programs are also designed to promote gender equality in the education sector. In this regard, Schurmann (2009) has worked on a female secondary-school stipend project in Bangladesh aiming to increase female enrollments. The program has an overarching objective of increasing female empowerment by generating employment opportunities for them. Using the Social Exclusion Framework, it ﬁnds that harassment, poverty, early-age marriages, and childbirth are responsible for preventing girls from attending school. The results illustrate that the CCT project was able to increase women’s empowerment in various ways via increased female secondary schooling. Furthermore, Barrera-Osorio et al. (2008) asserts that conditional cash transfers should be based on school attendance and part of the money should be paid before admission. Students’ own attitudes should also be considered before granting these scholarships. Along with the provision of educational grants, some transfer programs are designed to invest in the health and nutrition of students, especially female students. Barham (2009) evaluates the performance of a pilot CCT program called Red de Protección Social (RPS), which was implemented in two regions of Nicaragua. RPS was a multipronged approach to promoting the gender equality with balanced human capital. It provides cash payments to females for increasing their participation in education, schooling of children, health services, and nutrition. Results showed that this program had a positive impact on school enrollments. However, its impact was greatly reduced by negative economic shocks such as droughts and dwindling coffee prices. 1.4 Literature on Pakistan The UNESCO (2012a) report proposes that developing countries should spend at least 4 % of their GDP on education. In contrast, Pakistan spends around 2 % of its GDP on education. Furthermore, a lack of schools, inadequate facilities at existing schools, and ghost schools in remote areas are adding to this problem.5 Under the business-as-usual scenario, Pakistan will need 38 years to achieve universal primary education, and it will require approximately 16 years to attain an 86 % literacy rate. There is evidence that education and income poverty reinforce each other. Chaudhry et al. (2010) ﬁnds that education has a direct impact on skill development, which in turn has positive effects on the overall economic growth process. Some indirect effects are also accrued such as awareness about basic rights, health issues such as clean drinking water, and sanitation. Empirical results indicate that education signiﬁcantly alleviates the incidence of absolute poverty. In developing countries, the views of local communities on female education are very important. Most decisions in such countries are based on the perceptions of society regarding education. Shahzad et al. (2011) analyzed the attitude of the community and parents toward female education in Khyber Pakhtunkhwa province. The authors found that people were unaware of the direct and indirect beneﬁts associated with educating female children. Independent Evaluation Group (2011) produced the same results for Punjab province. The short-run impact evaluation shows that CCTs increase female school enrollment by 23 %. Furthermore, they have a signiﬁcant impact on productivity, consumption, and intergenera5 The term ghost school refers to those educational institutions that have a physical infrastructure but no teachers or students. There have been instances where local inﬂuential persons use these state-owned premises for their own interests. 123
Author's personal copy V. Ahmed, M. Zeshan tional human capital accumulation in the long run. In view of the aforementioned factors and given the gaps in the literature speciﬁc to Khyber Pakhtunkhwa province, the objectives of our inquiry in this paper are given as follows. • Review outputs and outcomes associated with the stipend (CCT) allocated to female students attending secondary school in Khyber Pakhtunkhwa province. • Identify changes required in program design and compensatory schemes to enhance progress in schooling for girls. • Evaluate the transmission channels and their capacity to reach the poor. • Evaluate the role played by the program in reducing the gender gap in education indicators. • Assess the monitoring and evaluation system associated with the program. • Review the process of stipend distribution. • Identify the actions needed for better implementation of the program. 2 Methodology This section brieﬂy discusses the quantitative and qualitative tools that are used to assess the impact of stipends on female education. The following chart summarizes the methodology employed in the present study. Our household survey (given the time and physical constraints in this region bordering Afghanistan) was comprised of randomly selected households from three randomly selected districts: Battgram, Hangu, and Shangla. In total, the sample size was 600 households. Further, 25 in-depth key informant interviews (KIIs) were also conducted in each of the speciﬁed districts. We also validated our quantitative output through a qualitative method, for example, focus group discussions (FGDs) were conducted in each district. The household-level questionnaire focused on (a) location variables, (b) employment, income and consumption proﬁle of households, (c) community environment, (d) availability of and access to school, and (e) quality of school infrastructure. This study used a limited dependent variable model for estimation purposes, also known as a probit model. It is a maximum-likelihood estimation technique widely used in the literature (Njong 2010; Geda et al. 2005; Ai and Norton 2003). In this model, female education depends on multiple factors including ﬁnancial well-being, access to education, family size, household bias in favor of male children’s education, awareness, education level of household head, and 123
Author's personal copy An analysis of the social impact of the stipend program economic conditions of the household. The detailed results of this binary response function are discussed in the next section. 3 Results from survey exercise This section combines our results from household-level survey data, community-level FGDs, and KIIs with government ofﬁcials and community representatives. We examine how the stipend program has impacted the enrollment levels and public awareness about female education. Most of the households believe that stipend money has a favorable impact on female education and that CCTs alleviate their ﬁnancial constraints as well. The districtwise impact of the stipend program reveals that female enrollment has increased in secondary schools in all the surveyed districts. According to the results, 93 % of families have availed themselves of the stipend program, resulting in a 7 % increase in female enrollments over the time period of this program. Just in Hangu and Battagram districts, female enrollment rates increased by 16 and 12 %, respectively. It may be mentioned here that Hangu is a hard-hit district due to the ongoing war on terrorism. After the receipt of stipend money, females devote their relatively greater time to study, which in turn results in a reduced dropout rate from secondary school. Transfer payments have an income effect and make it easier for households to increase their consumption because they have access to stipend money. Because the transfers are in the form of CCTs, though households are free to allocate the cash received (among their competing demands) as they see ﬁt, but with one condition – that school attendance of female children should not drop below 80 %. In this way, we ﬁnd that the stipends reduce poverty, and results indicate that only 49.7 % of the stipend amount is consumed for female education expenses while the rest of it is used to meet household consumption demands. In addition, 58 % of families responded that this stipend money was sufﬁcient, whereas 42 % of families responded that it was not according to their expectations. The beneﬁts associated with stipend money are not uniform across all households. For families earning less than PKR 5,000/month, the stipend amount was the basic source of education for their female children. In our sample this was the case for 82 % of households. For relatively higher income groups, the need for stipend money is relatively less urgent. In the absence of stipend money, the female enrollment rate at the secondary level was found to be low at 65 %, implying that 35 % of female students might drop out in the absence of this stipend. This conﬁrms the importance of such CCTs for female education, particularly for the poorest regions. On the whole, the stipend program shows a pareto improvement in society: someone is better off without making anyone worse off. The stipend program has brought many females to a better position with no adverse effects on society. The data show that around 80 % of households were aware of schools, teaching staff, infrastructure, and the stipend program. About 84 % of households reported that schools provided basic facilities like furniture, chairs, tables, and desks; 94 % of female schools had trained and qualiﬁed teaching staff. Hence the argument of scarcity of resources, lack of trained teachers, and nonavailability of educational facilities is not correct; rather, other factors are responsible for the low enrollment rates. The enrollment rate of girls is 86 % in secondary schools, showing a structural change toward female education. People are realizing the importance of education for females; several community members reported that educated female members can now contribute to the development of subsequent generations. This trend is very encouraging in the Shangla district, where 97 % of females now attend secondary school. However, uneven socioeconomic 123
Author's personal copy V. Ahmed, M. Zeshan conditions make female education less attractive, and in the overall sample, 14 % of girls were unable to attend secondary school because of communal constraints (i.e., related to individual or household harsh living conditions, distance to school, and lack of transportation). The parental economic status matters greatly for female education because a sound family background in terms of education and earnings might facilitate children’s education as well. Results show that 32 % of dropouts were due to severe ﬁnancial constraints as these families were barely able to meet basic household needs. The distance of schools from home and the availability of safe public transportation are important determinants of female education. The former in particular forces parents to send their daughters to informal schools (e.g., madrassah). Around 81 % of the girls in our sample perform household chores, whereas 10 % were working in nonfarm activities. On average, a female child could earn PKR 2,500/month, and this is part of the reason that some respondents opted to drop out of secondary school. 3.1 Results from probit model This section reports the ﬁndings based on our probit model. Household decisions are determined by various factors, including social interaction and economic and cultural aspects, so the present study examines the impact of these factors on girls’ likelihood of attending school. More precisely, these factors include the education level of household heads and their spouse, family size, awareness about the beneﬁts of education, household economic conditions, distance of home from school, and satisfaction with services provided by the school. Our results reveal that all the variables in the model are statistically signiﬁcant except child illness. Family size has a negative impact on female education. A large family size reduces the chances of female education by 0.7 % (Table 3). This phenomenon also explains the impact of resource constraints faced by families, which in turn results in dropout or nonenrolllment. Under such circumstances it is common to see families preferring the education of male over female children (Mogstad and Wiswall 2009). Attitudes toward female education are inﬂuenced by the parents’ education level. Educated parents realize the importance of education for both boys and girls. Our results conﬁrm that one additional year of schooling of the household head increases the chances of female education by 1.8 %. This is also the case with the educational level of the spouse of the household head: one additional year of schooling of the spouse of the household head improves the chances of female education by 3.3 %. Chevalier and Lanot (2002) argue that the impact of ﬁnancial constraints on female education is less important than parental educational level. The distance of the school from the home is an important determinant of female education; the chances that a female will receive an education decrease if the school is located far from home. Our results indicate a negative and signiﬁcant impact of an increase in distance of the school from home. The chances of female education decrease by 2.8 % if there is a 1 % increase in the school distance. Furthermore, the quality of education plays a role. Satisfactory educational services have a signiﬁcant impact on female education. The likelihood of female education increases by 11 % if the family is satisﬁed with the educational services provided by the school. Family income is an important determinant of female education in the selected districts because the average family located at a distant place faces severe resource constraints. However, smaller families with higher incomes can invest more in education, but they are few in number. A 1 % rise in household income increases the chances of female education by 0.3 %. This result is consistent with the ﬁndings of Hobcraft and Kiernan (2001). The lack of resources causes poor families to pull their girls out of school and put them in an informal 123
Author's personal copy An analysis of the social impact of the stipend program Table 3 Probit model for females attending school Dependent variable: female attending school = 1, 0 otherwise. Variables Probit model Marginal ﬁxed effects Illness at home −0.11 −0.046 (0.128) (0.051) Spouse education −0.02** −0.007** (0.006) Household size (0.002) HH education 0.08*** 0.033*** (0.023) (0.009) Radio (awareness) 0.018*** (0.002) 0.008* .003* (0.003) HH income 0.05*** (0.005) (0.0015) 0.052* (0.057) Economic conditions of HH 0.13* (0.023) 0.12*** 0.047*** (0.027) (0.010) 0.03 0.011 (0.024) Local economic conditions (0.009) Satisfaction with school −0.07*** −0.028*** (0.020) Distance from high school (0.008) 0.110*** (0.028) 0.07*** 0.0003*** (0.002) Conditional cash transfers 0.28*** (0.072) (0.000) District level controls −0.33*** −0.131*** (0.087) Shangla (0.033) −0.10 −0.038 (0.082) Bonair (0.033) −0.37** −0.14** (0.085) Hangu (0.033) −0.11*** (0.036) Battagram −0.72*** −0.27*** (0.115) Kohistan The probit model uses zero and one as its fundamental units. Therefore, correlations may not be very accurate. This is a limitation of this research study Standard errors are reported in parentheses. ***, **, and * indicate 1, 5, and 10 % levels of signiﬁcance, respectively. HH Household −0.28*** (0.094) Tank (0.038) 0.19*** 0.074*** (0.081) (0.032) −0.81*** – (0.213) – Observations 3,123 – LR Chi2 394.43 Constant 123
Author's personal copy V. Ahmed, M. Zeshan labor situation so that they can earn money to feed the family. The relationship between the attainment of education and income has become more important in the context of increasing income inequalities. Household awareness about the importance of education plays a crucial role in girls’ education. If the family head recognizes the short- and long-term beneﬁts associated with education, then the family will be more interested in the education of girls. The present study employs radio use at home as a proxy of awareness. Results explain that socioeconomic awareness increases the chances of female education by 5.2 %. Economic conditions of the household and of the local area are both essential determinants of female education. Results indicate that improved economic conditions of a household and local area increase the chances of female education by 4.7 and 1.1 %, respectively, consistent with Chevalier (2004). The provincial government in Khyber Pakhtunkhwa province is working under the assumption that CCTs would increase the transitional income of households. In this way, the government might improve the efﬁciency of female education in the province. Our results indicate that this stipend money improves the chances of female education by 0.03 %, which is very low. This result is consistent with actual government data on treatment and control groups (given in the next section), which assert that the impact of CCTs falls after 2 years. Hence, along with the provision of CCTs the government should eliminate other bottlenecks that represent the main hurdles preventing female education in Khyber Pakhtunkhwa province. 3.2 Results from secondary data This section evaluates the impact of stipend money on the treatment group as compared to the control group. The treatment group comprises the cluster of districts where CCTs are distributed in households, whereas the control group comprises the nonprogram areas. The treatment group includes the Battagram, Bonair, Hangu, Kohistan, Shangla, Tank, and Upper Dir districts, whereas the control group includes the Chitral, DI Khan, Karak, Kohat, Lakki, Lower Dir, and Malakand districts. This analysis covers the period 2006–2010. Our examination reveals that from 2006 to 2008, school enrollments in the treatment group increased more rapidly compared to the control group (Fig. 4). However, the relative growth in enrollments did not remain constant for the next 2 years, implying that the impact of CCTs Fig. 4 Enrollment in treatment and control groups (percentage change) 123
Author's personal copy An analysis of the social impact of the stipend program starts decreasing after just 2 years. This result indicates some obstacles (other than educationrelated expenses) that discouraged female enrollments in the target districts after 2008. 4 Conclusion In 2006, the government of Khyber Pakhtunkhwa province in Pakistan initiated a CCT in the form of a stipend program for female secondary-school students that aimed to increase secondary-school enrollments among girls. In this program, stipend money in the amount of PKR 200/month was given to each girl starting in 2006. After 1 year, in 2007, around 0.28 million female students were registered under this program. This project shows the commitment of the provincial government to attaining MDGs. The target area of this stipend program comprised seven backward districts, including Battagram, Bonair, Hangu, Kohistan, Shangla, Tank, and Upper Dir. Preliminary analysis indicated that all these districts were below the average provincial literacy rate except Upper Dir and Battagram. Other than the low literacy rate, the gender gap is very obvious in these districts. Our household-level survey exercise reveals that the stipend program has a signiﬁcant impact on female education because it eases household ﬁnancial constraints. It has increased female enrollment in secondary schools, and around 93 % of families have availed themselves of the stipend program, resulting in a 7 % increase in female enrollments. Survey results reveal that 35 % of girls might drop out in the absence of stipend money. Around 80 % of the households were aware of schools, teaching staff, infrastructure, and the stipend program. Approximately 84 % of households reported that the schools provided basic facilities like furniture, chairs, tables, and desks; 94 % of female schools had trained and qualiﬁed teaching staff. The female secondary-school enrollment rate is 86 %, showing a structural change toward female education. However, uneven socioeconomic conditions make female education less attractive, and 14 % of girls were unable to attend secondary school because of constraints related to harsh living conductions, distance from school, and lack of safe transportation. Results showed that 32 % of the dropouts left school due to severe ﬁnancial constraints; they were barely able to meet basic household needs. Females in Khyber Pakhtunkhwa also work in household and nonfarm activities, and 81 % of girls performed household chores, whereas 10 % worked in nonfarm activities (making the decision for education difﬁcult for their households). The results of the probit model reveal that family size has a negative impact on female education. A marginal increase in family size decreases the chances of female education by 0.7 %. One additional year of education of the head of household has a positive and signiﬁcant impact on female education, raising the chances of female education by 1.8 %. This is also the case with the educational level of household head’s spouse, one additional year of schooling for whom improves the chances of female education by 3.3 %. The chances of female education decrease by 2.8 % with a 1 % increase in distance from school. The likelihood of a female attending school increases by 11 % if her family is satisﬁed with the educational services provided by the school. A 1 % rise in family income increases the chances of female education by 0.3 %. The results also show that if the family is aware of the socioeconomic beneﬁts of education, then the chances of female education increase by 5.2 %. The economic conditions of the household and of the local area are essential determinants of female education. Results indicate that improved economic conditions in both households and the local area increase the chances of female education by 4.7 and 1.1 % respectively. Finally, stipend money through CCTs improves the chances of female education by 0.03 %. 123
Author's personal copy V. Ahmed, M. Zeshan This low level of gain associated with stipend disbursement indicates that provision of CCTs must be supplemented by other reforms that would improve community awareness about education, reduce the distance from school, provide safe public transportation to children, and address lacking facilities at female schools. 4.1 Policy recommendations Over the course of this study, we also conducted in-depth interviews with education department ofﬁcials. Therefore, a few of our recommendations focus on process and institutional mechanisms that can further strengthen the efﬁciency of this program. What follow are the main policy recommendations of our study. • For families earning less than PKR 5,000/month, the stipend amount is the basic source of motivation for educating their children. Ensuring a smooth and timely ﬂow of funds can facilitate female education in these households. In the past there have been delays in the receipt of promised stipends, which discouraged parents and in some cases led to discontinuing a child’s education. Such lapses in the government’s own organization need to be addressed. • Cultural constraints make female education difﬁcult in places, and therefore regular awareness campaigns are recommended to reduce the impact of such taboos. Engagement with communities can take the form of local language literature focusing on the beneﬁts of female education, street theaters, and the strengthening of parent–teacher associations. • The distance of schools from home is a major impediment to female education; parents prefer to send their daughters to informal schools (e.g., madrassah in mosques) in cases where formal schools are far away. Accessibility to school should be made easy through the provision of safe public transportation or by constructing new schools in far-ﬂung regions. The latter is only possible if there are teachers willing to serve in the distant areas. • Many females are involved in household chores and other nonfarm activities, which reduces female enrollments in schools. Mandatory female education enforced by law can increase female enrollments in secondary schools. • The impact of ﬁnancial constraints on female education is less important than parental education. Female enrollments can be increased by educating their parents. The aforementioned awareness programs can be customized for parental knowledge. • The stipend money only slightly improves the chances of female education because of the many operational bottlenecks. The impact of stipend money can be increased if government would consider and address the following impediments: (a) Secondary education budget formulation is based upon arbitrary assumptions; a more scientiﬁc approach by the Department of Education of the government of Khyber Pakhtunkhwa would bring greater efﬁciency to the system. (b) An interdepartmental committee is needed to address coordination issues (resulting in delayed processing of stipends), including representatives from the ﬁnance department, planning and development department, education department, Pakistan post ofﬁce, accounting general ofﬁce, and Executive District Ofﬁcers in the government of Khyber Pakhtunkhwa province. (c) A transparent procedure for the monitoring and evaluation of the stipend program is required. The government of Khyber Pakhtunkhwa province is responsible for laying down clear monitoring and evaluation mechanisms for the development budget in the province in accordance with the 18th constitutional amendment in Pakistan. 123
Author's personal copy An analysis of the social impact of the stipend program Acknowledgments The authors are economists at the Sustainable Development Policy Institute. We would like to acknowledge the ﬁnancial support of UNDP and the government of Khyber Pakhtunkhwa and Pakistan and the technical advice of two internal referees. References Ai, C., & Norton, E. C. (2003). Interaction terms in logit and probit models. Economics Letters, 80, 123–129. Barham, T. (2009). A healthier start: The effect of conditional cash transfers on neonatal and infant mortality in rural Mexico. Journal of Development Economics, 94(1), 74–85. Barrera-Osorio, F., Bertrand, M., Linden, L. L., & Perez-Calle, F. (2008). Conditional cash transfers in education design features, peer and sibling effects evidence from a randomized experiment in Colombia. NBER Working Paper Series, No. 13890. Bassett, L. (2008). Can conditional cash transfer programs play a greater role in reducing child undernutrition?. Social Protection Discussion Papers, No. 46687. Chaudhry, I. S., Malik, S., Hassan, A. U., & Faridi, M. Z. (2010). Does education alleviate poverty?. International Research Journal of Finance and Economics: Empirical Evidence from Pakistan. 52. Chevalier, A., & Lanot, G. (2002). The relative effect of family characteristics and ﬁnancial situation on educational achievement. Education Economics, 10(2), 165–81. Chevalier, A. (2004). Parental education and child’s education: A natural experiment. London School of Economics and IZA Bonn, No. 1153. Faith, G., & Vinay, C. (2010). Conditional cash transfers: Myths and ﬁndings, No. 01. Geda, A., Jong, N., Kimenyi, M. S. & Mwabu, G. (2005). Determinants of poverty in Kenya: A household level analysis. University of Connecticut working paper series. Government of Pakistan (GoP) (2012). Economic Survey of Pakistan. Ministry of Finance. Government of Pakistan (GoP) (2013). Review of public sector Development Program in Pakistan. Islamabad: Planning Commission. Hobcraft, J., & Kiernan, K. (2001). Childhood poverty, early motherhood and adult social exclusion. The British Journal of Sociology, 52(3), 495–517. Independent Evaluation Group. (2011). Do conditional cash transfers lead to medium term impacts?: Evidence from a female school Stipend Programme in Punjab. Washington DC: World Bank. Johnson, M. T. (2013). The impact of business cycle ﬂuctuations on graduate school enrollment. Economics of Education Review, 34, 122–134. Ladd, H. F. (2012). Education and poverty: Confronting the evidence. Journal of Policy Analysis and Management, 31(2), 203–227. Mogstad, M., & Wiswall, M. (2009). How linear models can mask non-linear causal relationships an application to family size and children’s education’. Research Department, Discussion Papers No: Statistics Norway. 586. Njong, A. M. (2010). The effects of educational attainment on poverty reduction in Cameroon. Journal of Education Administration and Policy Studies, 2(1), 001–008. Roche, K. (2013). Reconciling gender differences in the returns to education in self-employment: Does occupation matter? Journal of Socio-Economics Volume, 44, 112–119. Saavedra, J., & Garcia, S. (2012). Impacts of conditional cash transfer programs on educational outcomes in developing countries. RAND Population Research Center. Schurmann, A. T. (2009). Review of the Bangladesh female secondary school stipend project using a social exclusion framework. Journal of Health, Population, and Nutrition, 27(4), 505–517. Shahzad, S., Ali, R., Qadeer, M. Z. H., & Khan, M. S. (2011). Community attitude towards female education. International Journal of Academic Research, 3(1), 970–973. Skouﬁas, E., & Maro, V. D. (2008). Conditional cash transfers, adult work incentives, and poverty. Journal of Development Studies, 44(7), 935–960. Thorpe, R. J, Jr, Szanton, S. L., Bell, C. N., & Whitﬁeld, K. E. (2013). Education, income and disability in African Americans. Ethnicity and Disease, 23, 12–17. UNESCO. (2012a). Youth and skills: Putting education to work. EFA Global Monitoring Report. UNESCO. (2012b). Seminar on girls right to education. Pakistan: Islamabad. United Nations Education, Scientiﬁcal and Culturla Organization. (2012). Tuation analysis of the education sector. Pakistan. World Bank. (2012). On norms and agency: Conversations about gender equality with women and men in 20 countries. Working Paper, 74191. 123
... of Khyber Pakhtunkhwa ... the stipend program for secondary school girls of ... analysis of the social impact of the stipend program ...
... stipend program for secondary school girls ... analysis of the social impact of the stipend ... for secondary school girls of Khyber Pakhtunkhwa ...
... stipend program for secondary school girls ... analysis of the social impact of the stipend program for secondary school girls of Khyber Pakhtunkhwa
... poverty and social impact analysis of the project titled: Stipend Program for Secondary School Girls ... secondary school girls of Khyber Pakhtunkhwa.
... of Stipend Program for Secondary School Girls of ... Social Impact Analysis of Stipend Program for Secondary School Girls of Khyber Pakhtunkhwa ...
Poverty & Social Impact Analysis ... Secondary School Girl students of Khyber Pakhtunkhwa ... Analysis (PSIA) of Provision of Stipend to Secondary ...
Muhammad Zeshan, Sustainable Development Policy ... Poverty and Social Impact Analysis of Stipend Program for Secondary School Girls of Khyber Pakhtunkhwa ...
... Social Impact Assessment Study and ... of a 'Stipend Programme' initiated by government of Khyber Pakhtunkhwa to increase girls enrolment at secondary ...