S . C h a n d r a s e k h a r a n d L . B h a t t a c h a r y a 2 0 1 8 EDUCATION OF CHILDREN FROM MIGRANT RURAL HOUSEHOLDS IN INDIA: MOVING TOWARDS A COHERENT POLICY FRAMEWORK This paper was commissioned by the Global Education Monitoring Report as background information to assist in drafting the 2019 GEM Report, Migration, displacement and education: Building bridges, not walls. It has not been edited by the team. The views and opinions expressed in this paper are those of the author(s) and should not be attributed to the Global Education Monitoring Report or to UNESCO. The papers can be cited with the following reference: “Paper commissioned for the 2019 Global Education Monitoring Report, Migration, displacement and education: Building bridges, not walls”. For further information, please contact [email protected]. ED/GEMR/MRT/2018/P1/10 Background paper prepared for the 2019 Global Education Monitoring Report Migration, displacement and education: Building bridges, not walls2 CONTENTS 1. Introduction 5 2. Triangulation of Data 7 a. Out of School Children: Estimates from Census of India 2011 7 b. Age-Specific Attendance Ratio: Estimates from NSSO’s 2014 Survey of Education 8 c. Children in Rural Households with Short-term migrants: Estimates from NSSO’s Surveys 9 d. Children from Households Engaged in Construction Work: Estimates from NSSO’s Survey 11 3. View from Recent Five Year Plans 12 4. Towards a Bottom up Approach 13 5. Conclusion 18 6. References 20 7. Tables & Figures 22 LIST OF TABLES AND FIGURES Table 1: Distribution of Individuals Aged 15-19 years by Education and Work Status 22 Table 2: Age-Specific Attendance Ratio by Household Type 22 Table 3: Age-Specific Attendance Ratio 24 Table 4: Estimates of Number of Rural Households with a Short-term migrant and Children Aged 6-14 Years Living in Rural Households with a Short-Term Migrant 27 Table 5: Status of Current Attendance in Educational Institution among those Aged 15-19 Years in 2011-12 28 Figure 1: Age-specific Attendance Ratio in Rural Odisha 29 Figure 2: Age-specific Attendance Ratio in Rural Maharashtra 293 ABSTRACT1 A large number of children in India are either enrolled but not attending or have dropped out of school. One factor that would explain why children are out of school is migration. Seasonal migration leads to a situation where the children are enrolled as reflected in the school registers but do not attend school. This happens when their parents move in search of work for several months which overlap with the school academic calendar. The periodic absence from school affects their learning outcomes and eventually affects the number of years that they eventually attend school. With the passage of The Right of Children to Free and Compulsory Education Act 2009 (henceforth referred to as RTE), the local authorities are legally obliged to “ensure admission of children of migrant families”. The age-specific attendance ratio for children aged 6-14 years in 2014 is greater than 90 per cent suggesting the effectiveness of RTE. However, using data from Census of India 2011, we provide evidence that educational attainment is lower in high outmigration districts. Field studies also suggest that there are large differences at the sub-national level especially in high outmigration districts. In the absence of reliable estimates of child migration, as a proxy, we provide estimates of children living in households with at least one short-term migrant. Since seasonal migrants are more likely to work in the construction sector, we provide estimates of children living in households with a member who reports working in a construction site. If migration is driven by poverty or distress then it is reasonable to expect that the underlying cause will indeed adversely affect educational attainment of children from these households. From a policy perspective it is truly a challenge to ensure that over 10.7 million children from rural households with seasonal migrants indeed attend and complete elementary education. Acknowledging the complexity and the need for collective action, India is following a bottom up approach. The state governments have recognized the need for empowering the local governments, coordination across departments, collaboration with other states, and engagement with civil society organisations in order to ensure that educational outcomes of children from migrant households are not 1 Written by S Chandrasekhar, Indira Gandhi Institute of Development Research, Mumbai, India (Email: [email protected]) and Leena Bhattacharya, Indira Gandhi Institute of Development Research, Mumbai, India (Email: [email protected])4 affected. The paper sketches the broad contours of the measures taken in this regard. The paper concludes by highlighting the data and resultant knowledge gap and identifying the research that needs to be undertaken to inform policy. 2 Keywords: Right to Education, Short-term Migration, Child Migration, Out of School Children, Learning Outcomes, Policies, Panchayats, India 2 Acknowledgements: We are grateful to Priyadarshani Joshi for comments received on earlier draft of this paper.5 1. INTRODUCTION Leading up to the framing of Constitution of India there was a lively discussion in the Constituent Assembly of India on what would constitute as a fundamental right. For instance, should free and compulsory education for children be enshrined as a fundamental right? At the heart of conversation was the consequence of not being able to deliver on what would be committed to as a fundamental right. M. Ananthasayanam Ayyangar, who would later become the second speaker of the lower house of Indian Parliament, Lok Sabha, asked who would be held responsible if for some reason India failed to introduce free compulsory education within a period of ten years3. As things turned out, free and compulsory education for those 6-14 years of age was enshrined as a directive principle4 of state policy and not as a fundamental right. Over time, there was considerable discomfort in light of the failure to ensure 100 per cent enrolment and eliminating drop outs. Eventually, in 2002, free and compulsory education for children aged 6-14 years became a fundamental right5. By the end of the decade, i.e. in 2009, The Right of Children to Free and Compulsory Education Act (henceforth referred to as RTE Act) was passed by the Indian Parliament. Under the Act the responsibility of the appropriate government and the local authority is to establish a school, while it is the duty of the parents to admit their child to school. While RTE Act covers all children, it recognises the challenge posed by hard-to-reach children including children from migrant families. It specifies that the local authorities need to “ensure admission of children of migrant families”. The challenge faced by the 3 “In Article 26 it is said that the State should within a period of ten years introduce free compulsory education. Take this as an instance. Let us assume that the State does not do so, then can any court of law enforce it? Against whom? In case a decree is granted by a court of law, who will carry it out? If the Government does not carry it out, can the High Court or the Supreme Court enforce it? Is it open to the Supreme Court to change such a government? With its authority, can it by an officer of the Court, an Amin or a Sheriff, imprison all the Ministers, and bring into existence a new set of ministers?” http://164.100.47.194/Loksabha/Debates/cadebatefiles/C19111948.html 4 Article 45 in The Constitution of India 1949 pertaining to provision for free and compulsory education for children, states: “The State shall endeavour to provide, within a period of ten years from the commencement of this Constitution, for free and compulsory education for all children until they complete the age of fourteen years” 5 A new article 21A was inserted into the Constitution of India and it read as follows: "The State shall provide free and compulsory education to all children of the age of six to fourteen years in such manner as the State may, by law, determine." At the same time Article 45 was substituted, as follows: "Provision for early childhood care and education to children below the age of six years".6 government in ensuring “attendance and completion of elementary education” has proven to be truly daunting in case of children living in rural households with seasonal migrants. If migration is driven by poverty or distress then it is reasonable to expect that the underlying cause, i.e. poverty and lack of livelihoods, will adversely affect educational attainment of children from these households. Seasonal migration also leads to a situation where the children are enrolled as per the school registers but do not attend schools when their parents move in search of work for several months of the year, which also overlap with the school academic calendar. In the official surveys, short-term migrants are defined as those who stay away from home for more than 15 days but less than 6 months. While the six month cut off is standard in surveys to determine usual place of residence, field studies have shown that it is not uncommon for individuals to stay away from villages for even up to 9 months of the year6. The industry of work of short-term migrants also matters. For instance, it is a documented fact that the entire family including the children move and find employment in brick kilns. There is a small literature in the Indian context that has dwelled on the challenge of ensuring uninterrupted schooling of migrant children. Rogaly et al. (2001) study the agricultural labourers who migrate seasonally to West Bengal, a state in the eastern part of India. In the agricultural season, the children are usually out of school as evident from the empty classrooms. Many children work with the parents and generate additional family income. The study points to the inefficiency in schooling system and its failure to adjust to the need of the migrants. Coffey (2013) studied the educational attainment of children in 70 villages in 5 outmigration prone districts of Rajasthan, Madhya Pradesh and Gujarat. She found that children who had migrated the year prior to the survey had lower grade for age than the ones who did not. They were also at a disadvantaged position in terms of ever attending school, or going to school on the day before the survey. Further, she found the gap to be increasing as the child grew older. The central and state governments as well as civil society organisations have recognised evidence from field studies and started taking pro-active steps to ensure that children at risk of migration stay in school. Srivastava and Dasgupta (2010) provide a succinct overview of such initiatives by civil society organisations in Gujarat, Maharashtra and Odisha and highlight the gaps that are needed to be filled. 6 If they stay away from home for a continuous period of 6 months or more these individuals would typically be considered as out-migrants in any household survey.7 2. TRIANGULATION OF DATA In the absence of a comprehensive survey on migration, participation in education and learning outcomes, we triangulate different data sets in order to outline the magnitude of challenge in ensuring that children’s schooling decisions are not affected by the decision of members of households to migrate. We analyse tables from Census of India 2011 to establish that educational outcomes among 15-19 year olds are likely to be worse in high outmigration districts. We then draw upon National Sample Survey Organisation’s (NSSO) Survey of Education conducted in 2014 to provide estimates of age-specific attendance ratio and how this varies across individuals from households which differ by their principal source of income. We provide estimates of number of households in rural India with a short-term migrant and the number of children who lived in such household in 2013. Since a large proportion of short-term migrants are absorbed in construction sector, based on NSSO’s 2011-12 survey (see Government of India, 2014), we provide an estimate of number of children in households where a member reports his/her work place location as a construction site. a. Out of School Children: Estimates from Census of India 2011 For all those aged 5-19 years, Census of India 2011 tables have information on those attending educational institution, and among those attending whether they are also working or not. For those not attending educational institution, we know whether they had ever attended an educational institution or whether they had never attended. For these individuals we also know whether they are working or not. Since the Census tables pertain to the period before RTE Act came into effect, it is not of interest to focus on those aged 6-14 years. Instead, we consider the educational attainment of those aged 15 – 19 years and the sub-national patterns. Estimates from Census of India data show that 48.44 million individuals aged 15 – 19 years were not attending educational institution in 2011 of whom 35.62 million individuals had dropped out and 12.82 million had never enrolled (Table 1). While 8 states, viz. Andhra Pradesh, Gujarat, Karnataka, Maharashtra, Madhya Pradesh, Rajasthan, Uttar Pradesh, and West Bengal, accounted for over two thirds of the individuals who dropped out, three states, viz. Uttar Pradesh, Bihar and Rajasthan accounted for 50 per cent of those who never enrolled. We highlight these patterns since they are likely to persist in the immediate future. We revisit this issue in the next section when we examine, using more recent survey data, age-specific attendance ratio (AAR) among those aged 15-19 years. We seek to understand the patterns in AAR evident in districts which have been classified as high outmigration districts. In its report, the Working Group to Study the Impact of Migration on Housing,8 Infrastructure and Livelihood, Government of India7 identified high outmigration districts (Government of India, 2017b). In order to explore whether there are any differences, we plot the AAR for the age group 7-19 years in rural Odisha8 (Figure 1) and rural Maharashtra9 (Figure 2). It is apparent that the AAR is lower in the outmigration prone districts as compared to those which are not classified as outmigration prone. b. Age-Specific Attendance Ratio: Estimates from NSSO’s 2014 Survey of Education As mentioned earlier, the Census of India tables provide a snapshot of participation in education for the period before RTE became effective. A more recent snapshot is available from NSSO’s Survey of Education conducted between January – June 2014. This survey has information on enrolment, attendance and educational attainment. Unlike the Census of India tables where district level estimates can be generated, the data from NSSO’s survey is representative at the level of the state and for rural and urban regions separately. For details of this survey, see Government of India (2016). We start off with a discussion on the AAR for the age groups 6-14 and 15-19 years (Table 2). The AAR gives the proportion of individuals in an age group who are currently attending educational institutions, irrespective of the grade in which the individual is studying. In both rural and urban India, we find that over 90 per cent of the boys and girls are attending school. There are marginal differences in the AAR across household types, where the type of the household is determined based on its principal source of income. As 7 Among the terms of reference given to the group included the following: to review the trend of various forms of migration in India and its impact on urban areas as well as the implications on rural and small towns, to better understand the reasons for migration and analyze its fall outs, to analyze the migration patterns in various economic groups and its trends and influences, and to suggest policy measures to better tackle migration from the perspective of housing, infrastructure and livelihoods for the migrants. The patterns presented by the Working Group in its report are based on Census of India tables and NSSO data sets. 8 A total of 11 districts from the state of Odisha viz. Balangir, Bargarh, Gajapati, Ganjam, Kalahandi, Kalahandi, Koraput, Malkangiri, Nabarngapur, Sonepur and Raygada were identified as source districts of outmigration districts by the Working Group to Study the Impact of Migration on Housing, Infrastructure and Livelihood, Government of India. http://mohua.gov.in/upload/uploadfiles/files/1566.pdf accessed on 10th April, 2018 9 Smita (2008) identifies seven districts of arid Marathwada region viz., Ahmadnagar, Beed, Dhule, Nandurbar, Nashik, Jalgaon and Jalna as the outmigration districts of Maharashtra.9 compared to the 6-14 year age group, the AAR is markedly lower in case of the 15-19 year old age group10. Unlike in the case of the 6-14 year age group, there is considerable variation in the AAR of 15-19 year age group across household type. Poverty is concentrated among those engaged in casual labour and one observes seasonal migration from these households. Hence it should not be a surprise that the AAR in the 15-19 year group is much lower than the national average in households whose primary source of income is from casual labour. There are marginal differences in the AAR of boys and girls aged 6-14 years across the states of India (Table 3). In Uttar Pradesh, which is also India's most populous state, the AAR of girls in the urban sector is 10 percentage points lower than the national average of 94. When we focus on the 15-19 year age group, we find large differences across the states in the AAR. Among the major states, the AAR is higher than the national average in the southern states, i.e. Andhra Pradesh, Karnataka, Kerala, Tamil Nadu, and Telengana. In contrast, the AAR (for the boys, girls or both) is lower than the national average in the states which account for the bulk of India’s rural poor, i.e. Bihar, Odisha, Rajasthan, Uttar Pradesh, and West Bengal. The survey also has information on the status of current educational attendance and status of current educational enrolment. This allows us to estimate the number of individuals currently enrolled but not attending educational institutions. The following states account for 79 per cent of total children who are enrolled but not attending school: Bihar (9 per cent), Madhya Pradesh (18 per cent), Maharashtra (9 per cent), Odisha (10 per cent), Rajasthan (7 per cent), Uttar Pradesh (14 per cent) and West Bengal (12 per cent). One common aspect among these states is the phenomenon of seasonal migration and it is particularly pronounced in a sub-set of these states. c. Children in Rural Households with Short-term migrants: Estimates from NSSO’s Surveys Estimates of children who are seasonal migrants are unavailable. Nor are estimates available on number of children who are affected by the decision of households to migrate seasonally. Instead, what we have are estimates of number of households with a short-term migrant and number of children aged 6-14 years in households with short-term migrants. These estimates are available from NSSO’s 2013 survey of land and 10 It would be pertinent to note here that the working age population is considered to be 15-59 years.10 livestock holdings which is representative for rural India (Government of India, 2015). A similar estimate for recent years is not available for urban India11. From NSSO’s 2013 survey we have estimates of number of rural households with a short-term migrant and also how they vary by household types. An estimated 10.08 million rural households have at least one short- term migrant, i.e. at least one member of the household stayed away from the village continuously for 15 days or more for employment during the 6 months preceding the survey (Table 4). An estimated 10.7 million children aged 6-14 years live in rural households with a short-term migrant. What we also find is that among those aged 15-19 years living in a household with a short-term migrant 28 per cent of them are either not literate or have not completed their primary education. In contrast, when we consider all individuals aged 15-19 years, less than 18 per cent are either not literate or have not completed their primary education. Before proceeding further, a couple of observations would be in order. First, the classification of households is a reasonable representation of distribution of rural households. Though not strictly comparable, it is similar to the distribution of households by principal source of income as reflected by NSSO’s survey of employment and unemployment 2011-12. Second, in the data we analyse, the category wage/salaried actually comprises of households whose principal income comprises of two types of households: casual labour and regular wage/salary. Based on the distribution evident in NSSO’s survey of employment and unemployment it is reasonable to assume that around 75-80 per cent of these households are engaged in casual labour and the remaining have a regular salaried job. We find that over 88 per cent of households with a short-term migrant are from one of the following types: self-employed in cultivation, self-employed in non-agricultural enterprise or wage/salaried employment (Table 4). The extant literature has established that short-term migrants are observed from households engaged in cultivation but with less than 1 hectare of land and from households working as casual labour12 in either agriculture or non-agriculture of whom a large proportion is landless. The 11 Estimates for urban India can be generated from NSSO’s survey of employment, unemployment and migration 2007- 08. Since this survey is over 10 years old we do not provide any estimates based on this data. 12 There is a specific reason why we dwell on occupation. While the literature has established intergenerational upward mobility in terms of education, there is however stickiness in intergenerational mobility in occupation. Motiram and Singh (2012) compare the outcome of the sons with their fathers using the India Human Development Survey 2004. They find “substantial intergenerational persistence, particularly in the case of low-skilled and low-paying occupations, e.g., almost half the children of agricultural labourers end up becoming agricultural labourers” (p. 56). Evidence from the repeated cross sectional surveys of NSSO too paints a similar picture.11 income of households with less than one hectare of land is less than their consumption expenditure. Hence, individuals from these households migrate seasonally13. Also, the importance of wages in total income of cultivator households varies across land size class. The share of wages in total income declines from over 60 per cent in households with less than 1 hectare of land to below 5 per cent in households with over 10 hectares of land. d. Children from Households Engaged in Construction Work: Estimates from NSSO’s Survey It is the construction sector that has absorbed individuals who are underemployed or unemployed in rural India. The construction sector is also the biggest absorber of short-term migrants (Agrawal and Chandrasekhar 2016). By the turn of the last decade, i.e. by 2011-12 a large number of individuals, 21.84 and 10.01 million in rural and urban areas respectively, reported that the construction site was their place of work. The share of rural and urban construction workers in the labour force in 2011-12 was 22.13 and 7.43 per cent respectively. In 2011-12, a total of 18.7 million rural and 5.1 million urban children aged 6-14 years lived in household with at least one worker in the construction sector14. The evidence from the field studies is that children in households engaged in construction work could be at risk of poor health outcomes and could also be out of school. Households engaged in the construction are more likely to be poor. The average monthly per capita expenditure of rural and urban households in 2011-12 is Rs 1192 and Rs 2311 respectively. In contrast, the average monthly per capita expenditure of rural and urban households whose principal industry is the construction sector is lower at Rs 1010 and Rs 1566 respectively. Since children from households engaged in the construction sector are likely to be from poorer households their schooling outcomes are likely to be 13 In order to deter distress-driven short-term migration, India has a long history of public works and relief works programs that dates back to the pre-independence era. Presently, the Mahatma Gandhi National Rural Employment Guarantee Scheme which offers rural households 100 days employment. Overall there is enough evidence to suggest that these programs have not generated enough days in terms of employment. Hence the jury is out on whether the objectives of the programs were realized. In the same vein, the last word has not been written on the extent to which these programs have reduced distress migration. 14 As a robustness check, we provide an alternate estimate based on the principal industry of work of the household. We find that in 2011-12 a total of 17.6 million and 4.9 million children aged 6-14 years lived in rural and urban households respectively which reported their principal industry of work to be the construction sector.12 affected. As mentioned in the earlier section, with the effective implementation of RTE the proportion of children who never attended school has declined. The question is what happens after the age of 14. We find that the proportion of individuals attending schools in the age group 15-19 years from households whose industry of work is construction is lower than the national average. Boys are more likely to drop out in order to work and supplement household income while girls are more likely to drop out to attend to domestic chores (Table 5). Many workers engaged in the construction sector are seasonal migrants. If the children accompany their parents they are more likely to not attend school and possibly drop out eventually. This aspect is evident in some field studies, particularly in case of families that migrate to work in the brick kilns (Smita 2008). Mosse et al. (2005) describes the condition of the Bhil Adivasis from Western India who migrate seasonally to urban construction sites as a ‘painful and socially disruptive element of their livelihood’. The employers at the construction site look at child labor as a favor to the migrant family, rather than providing the children with facilities of health or education. The discrimination faced by the migrant children during schooling admissions and in classrooms was highlighted in the study. At times the schools refused to admit children in the middle of the year as the children would inflate their drop-out rate once they left. A recent study by Srivastava and Sutradhar (2016) of seasonal migrants who are engaged in the construction sector in the National Capital Region of Delhi found an ambiguous effect of migration on schooling of the child. They found evidence of limited access to schooling facilities in the destination area. However, remittances sent by migrants could have a positive effect on educational outcome of children who were left behind at the source area. 3. VIEW FROM RECENT FIVE YEAR PLANS Since 1951, the five year plans have been central to policymaking in India. Recent plans have flagged the issue of educational outcomes of hard to reach children. The eighth plan (1992-97) encouraged volunteer agencies, factories or co-operatives to set up primary schools with flexible schooling days and instructional hours in order to ensure that such children, including the children of seasonal migrants, could attend school full time. The ninth plan (1997-2002) identified the Indian states that were falling behind in terms of providing education to all the children. It expanded the scope of national and state Open Schools in order to bring all children under the net of education and encouraged the role of voluntary organisations to reach this end. Parallely, there were discussions on making education of 6-14 year olds a fundamental right. This put a greater stress on the need for bringing the topic of education to the forefront and an education policy initiative called Sarva Siksha Abhiyan (SSA) was set up in 2001 to universalise elementary education in India.13 In 2002, Article 21A of the Constitution of India was introduced under the 86th Amendment which made Right to Education a fundamental right in India for children under the age of 14. Strategies to reach the ‘hardest to reach’ children were discussed in the tenth plan (2002-07). It involved initiatives under the SSA to identify the children at risk, undertake a need base analysis and design effective strategies. The task of government was to make sure that no child was left behind in terms of receiving quality elementary education. Keeping in mind the challenges posed by different groups of children as well as the need to reach all, the Right to Education Bill was introduced in 2005. Finally in 2009, six decades after the independence, the RTE was passed in India. Owing to the movement of the citizens across the country and the possibility of the children to miss out on schooling, the Act mentions that it is the duty of the local government to ensure admission of the children from the migrant families. Following the framing of RTE, discussion on education of children from migrant families was taken up in the eleventh and twelfth five year plans and within the SSA framework. The eleventh plan (2007 – 12) proposed the plan of walk-in Integrated Child Development Services centres in the railway stations and bus stands, where the migrant children were most likely to arrive. On arrival the children were to be given identity cards and they could use the basic services of health check-up. The identity cards would ensure tracking of children as they moved and could help them in getting placed in nearby schools. The discussion was also reflected in the framing of the twelfth five year plan (2012 – 17). Components of permanent migration as well as seasonal migration were discussed in more details. The plan suggested opening of seasonal schools in the destination area for children of migrant families. It proposed that government schools be turned into ‘Residential School Complexes and seasonal hostels for migrating tribal children’. With the end of the 12th plan, a decision was taken to replace the five year plans with a 15 year vision document. Simultaneously, the emphasis was on cooperative federalism and also improving local governance. 4. TOWARDS A BOTTOM UP APPROACH The idea of involving the local community has a long history in India. It finds a mention in the Government of India’s flagship program SSA. As we discuss later in this section, the RTE Act made the Panchayat central to the whole process. The SSA came into existence with its far reaching vision of ensuring compulsory elementary education in India. While the enrolment rates did increase in the decade of 2000, ensuring14 enrolment and attendance of some groups of children, especially the children from migrant households proved to be a problem15. The SSA framework, among the other guidelines, laid down special provision for the children from the migrant families. First, the high outmigration districts or blocks had to be identified. After that, the children could either be kept back in the source villages in seasonal hostels, where they could attend their schools without any interruption, or schools could be arranged around the worksites16. Teaching volunteers who could move with families would ease the process of educating the children, even if the family migrated to distant places. Under the RTE Act, there is provision for providing special training centres for out of school children and the SSA framework highlights and allocates budget for the same17. Finally, it suggested inter- state collaboration, since the barriers to school and other facilities become much more difficult to provide to the inter-state migrants unless the sending and receiving states collaborate with each other towards providing benefits to the migrants. The options considered to overcome the problem included: making curriculum more child-friendly, having multi-lingual schools with multi-grade textbooks, issuing cards to migrant children and opening of seasonal hostels in the source areas. In 2011, the SSA framework was revised for implementation of the RTE Act. States had the freedom to adapt the strategies outlined in the framework to their own context. In the recent years there have been conversations on covering children from pre- primary to senior secondary within an integrated scheme for school education. In April 2018, a more comprehensive educational policy of Samagra Shiksha Abhiyan covering pre-school to class 12 was prepared. Samagra Shiksha Abhiyan, once again, lays emphasis on community participation and importance of local government to reach the hard-to-reach children. 15 http://planningcommission.nic.in/reports/peoreport/peoevalu/peo_ssa2106.pdf accessed on 10th April, 2018 16 The SSA suggests either the provision of schools at the exact destination, or arranging for transportation facility to and from the school in the vicinity of the worksite. 17 SSA specifies ‘receiving district /State where migrant families are located for some period shall have responsibility for ensuring that education to the children in age appropriate classes continues during the period of migration. It is expected that the Annual Work Plan and Budget of these districts would include activities for education of such children, under Special Training component. The involvement of NGOs in the processes of mapping of migration and planning and implementation of interventions should be actively supported. Funds available under innovation can be used to support activities in an integrated strategy which are not supported under any other Norm of SSA.’15 Following the passage of RTE Act, local authorities were directly made accountable for ensuring that all children aged 6-14 years in India including hard to reach children were in school. The Panchayati Raj Institutions (PRI)18 were made responsible for the following: “identify out-of-school children, largely representing weaker sections and disadvantaged groups, ensuring age appropriate admission through Special Training, motivating parents and creating conditions to free them from domestic/wage earning responsibilities, and monitoring regular attendance of all children and teachers in schools” (Pg 113-14 Government of India 2011). Based on the RTE Act, state governments passed their respective rules19 which had details pertaining to formation of school management committees (SMC) 20 and preparation of school development plans (SDP)21. The SMC could act as a mechanism through which schools could effectively plan for the children of seasonal migrants, including operationalization of seasonal hostels or making other necessary arrangements. Combining the interventions and strategies at the district level, the state government prepares an Annual Work Plan & Budget (AWP&B)22. At national level, the Project Approval Board (PAB) in its annual meeting considers the AWP&B of all the Indian states and Union territories23. An 18 With the passage of the 73rd and 74th amendment of Constitution, India officially had a third and also the lowest tier of government: the Panchayats in rural areas and the urban local bodies in urban areas. The idea was to move towards local self-governance. The Panchayati Raj Institution comprises of Gram Panchayat at the village level, Panchayat Samiti at the block level and Zilla Parishad at the district level. Following the passage of RTE Act, Panchayats were directly made accountable for ensuring that all children aged 6-14 years in rural India including hard to reach children were in school. 19 http://mhrd.gov.in/rte_state_rules accessed on 10th April, 2018 20 The SMC guidelines specify the structure, composition (reservation for women and other groups, chairperson & convener), of SMC and assignment of any special duties to the members. http://mhrd.gov.in/rte_smc accessed on 10th April, 2018 21 Under Clause 21(l) of the RTE Act, the SMC members of the school are responsible for making the SDP, based on which grants are given by the local authority. The SDP is a three year plan for overall improvement of the school, where the members can plan for the additional teacher requirement, infrastructural requirement and financial requirements of the school. 22 The appraisal process of the AWP&B would scrutinize if areas of high incidence of migration have been identified and whether strategies for education of seasonally migrating children have been included in district and State plans. (SSA, 3.8.2.25) 23 http://ssashagun.nic.in/pab1718.html accessed on 10th April, 201816 analysis of the minutes of the meeting for the year 2016-17 shows that a beginning has been made towards ensuring that the children from migrant families have uninterrupted schooling. Recently, states have started the process of source destination mapping of migrants. Gujarat has recognised that the state receives a large number of migrant children from Maharashtra, Rajasthan, Uttar Pradesh and Odisha every year. The state government is keeping track of migrant children, operationalizing seasonal hostels and providing special training for out of school children. Moreover, the state collaborated with civil society organizations and appointed special teams for identification, online entry, enrolment and tracking of migrant children with the help of a migrating monitoring software. The PAB minutes of the meetings in 2016- 17 shows that Gujarat had planned to set up residential seasonal hostels for 9 months covering 34,917 children and non-residential seasonal hostels for 3 months for 22,807 migrant children. Seasonal migration is prevalent in Western Odisha with many households moving to work in the neighbouring state of Andhra Pradesh. Migration is usually family-based and the children accompany their parents to the destination site and work alongside them as a work-unit. The Government of Odisha undertook several active initiatives. In 2012, Odisha signed a Memorandum of Understanding (MoU) with the Government of Andhra Pradesh to encourage collaboration among the two states in order to improve the working and living conditions of the migrants24. Further, there was a need of an institutional mechanism to look deeper into the issue of seasonal migration and as a result, Interstate Coordination Committee was set up in 201225. Additionally, the state recognised the need to achieve convergence among various departments including Department of Women and Child Development, Labour Department, Welfare Department and the Police in order to protect the rights of children at risk, including children moving with their families. The state government has taken into account the insights provided by these organisations. Civil society organisations have been working actively with local communities in high outmigration villages. In 2004-05, as a part of the Learning and Migration Program (LAMP), American India Foundation (AIF) collaborated with partner NGO 24 The MoU was signed as a part of an International Labour Organization Project. 25 There was lack of data on migrants from the block and Gram Panchayat (GP) level in terms of their migration flows and details of the contractors they work with. From October 2012, data collection formats were circulated to the District Labour Officers for collecting migration related information. There were further discussions on developing tracking software to track migrant workers online. Other than tracking and providing benefits to the migrants, various events were held at the GP and block level to raise awareness among the migrants so that they could take better informed decisions.17 Lokadrusti and started operating seasonal hostels in Odisha. The hostels were set up in the school compound where the children could stay behind in the months when their parents stayed away from the village. In 2012- 13, the state Government started providing funds, which are available as part of SSA, for operating the hostels. Efforts have been made to build capacity within school management committee in order to ensure smooth operation of the hostels in the source villages. The caretakers in the seasonal hostels monitor the children, help them in communicating with their parents and support the children academically after school hours. The Government of Maharashtra used to operate seasonal hostels in high outmigration districts, to discourage child migration. A study conducted by UNICEF in Maharashtra found that the individuals who migrated seasonally did not want their children to stay back in the village in the seasonal hostels26 (UNICEF, 2017). Recognising this, the state government explored alternative community based living arrangements for the children. Parents were encouraged to leave their children with alternate caregivers, while the government provided them with timely meals, stationary and toiletries. Under this arrangement, the parents were less likely to take their children with them when they stayed away from the village (UNICEF, 2017). This led to the Government Resolution in October, 2015 wherein community based arrangement was adopted all across Maharashtra. However, ensuring the smooth processes was not the work of a single stakeholder and community volunteers played a huge role in providing psychosocial support to the caregivers and the left behind children. The village volunteers or Balmitra (friend of children) were selected from the members of the Village Child Protection Committee by the respective Gram Panchayats. They conducted after school programmes for the children and ensured regular communication among the children and the parents. The southern Indian state of Tamil Nadu attracts migrant families from Andhra Pradesh, Assam, Bihar, Chhattisgarh, Karnataka, Odisha, Telangana, West Bengal, and Uttar Pradesh. One challenge posed by inter- state migration is the language barrier faced by the children since the medium of instruction in the destination state would be different from their mother tongue. The government has made concerted efforts at providing multi-lingual books for children. However, finding teachers with good academic qualification and appropriate language skills remained a challenge. Overall, the evidence from the various initiatives suggests 26 It was not surprising as the classrooms were used as residential hostels at night and basic infrastructural facilities could not be provided to the children.18 that the central and state governments have sought to move towards implementing a coherent policy framework for education of children from migrant rural households in India. 5. CONCLUSION Two recent developments suggest the willingness of policy makers to address the issue of migration and its consequences. The Government of India’s Economic Survey 2016-17 recognized the need for formulating appropriate policies to reduce the costs associated with migration by addressing the ‘lack of portability of benefits, legal and other entitlements upon relocation’27. There was an opportunity to make access to the subsidised food grains from the public distribution system portable. Although, an initial draft of National Food Security Act (NFSA) dating back to 2011 did state that the "migrants and their families shall be able to claim their entitlements under this Act, at the place where they currently reside", in its final version NFSA 2013 did not assure portability of benefits. Unlike NFSA, the RTE Act does mention the need to include hard to reach children including migrant children. The other development was the submission of the report of the Working Group to Study the Impact of Migration on Housing, Infrastructure and Livelihood which was constituted by Ministry of Housing and Urban Poverty Alleviation, Government of India. The report called for a strengthening of the ‘legal framework for protection of interests of migrants’. Recognising the lack of timely estimates on migration either from household surveys or Census of India, the Economic Survey 2016-17 argued for the need to use other data sets innovatively in order to arrive at indirect estimates of migration. The Working Group too in its report called for better and timely data. In the foreseeable future, the construction sector will continue to be the one that will absorb the unemployed and under employed. The Economic Survey 2017-18 estimates that the real estate and construction sector will create over 15 million jobs in the next five years. The effect that this will have on the children from these families is difficult to ascertain at this point. Concerted efforts are needed to ensure that children who 27 “Portability of food security benefits, healthcare, and a basic social security framework for the migrant are crucial – potentially through an interstate self-registration process. While there do currently exist multiple schemes that address migrant welfare, they are implemented at the state level, and hence require inter-state coordination of fiscal costs of migration” (Government of India 2017a, p. 277). https://www.indiabudget.gov.in/es2016-17/echap12.pdf accessed on 10th April, 201819 migrate with the parents do not have trouble accessing schools at the destination area. For the ones who are left behind in the source villages, only proactive initiatives recognising ground realities can improve educational outcomes of these children. The SSA is getting adapted by the state government recognising local realities. What is needed is an evaluation of the processes and solutions put in place by the state governments. For instance, how does sector of work, duration and place of migration affect the decision made pertaining to whether the children will stay behind or move with their parents? Further, we do not know enough on how educational attainment differs across children who migrate with the parents, vis. a vis. children who stay back with alternate care givers. In order to understand local priorities, we need to analyse unstructured data, viz. minutes of SMCs, the SDP, minutes of Panchayat meetings and district-level standing committee on education and whether the priorities highlighted in the minutes are reflected in the Panchayat budgets. What is eventually required is a meshing of structured and unstructured information whose analysis will improve deliberations and decision- making at the local level resulting in improved governance.20 6. REFERENCES Agrawal, T., & Chandrasekhar, S. (2016). Labour market outcomes of itinerant workers in rural India. Journal of International Development, 28(8), 1251-1271. Coffey, D. (2013). Children's welfare and short-term migration from rural India. The journal of development studies, 49(8), 1101-1117. Mosse, D., Gupta, S., & Shah, V. (2005). On the margins in the city: adivasi seasonal labour migration in western India. Economic and Political Weekly, 3025-3038. Motiram, S., & Singh, A. (2012). How close does the apple fall to the tree? Some evidence from India on intergenerational occupational mobility. Economic and Political Weekly, 56-65. Rogaly, B., Biswas, J., Coppard, D., Rafique, A., Rana, K., & Sengupta, A. (2001). Seasonal migration, social change and migrants' rights: Lessons from West Bengal. Economic and political weekly, 4547-4559. Government of India (2011) Sarva Shiksha Abhiyan Framework for Implementation. Department of School Education and Literacy. Ministry of Human Resource and Development. Government of India (2014) Employment and Unemployment Situation in India 2011- 12. NSS 68th Round (July 2011 – June 2012), Report No. 554. National Sample Survey Organisation, New Delhi: Ministry of Statistics and Programme Implementation. Government of India (2015) Land and Livestock Holding 2013. NSS 70th Round (January 2013-December 2013), Report No. 572. National Sample Survey Organisation, New Delhi: Ministry of Statistics and Programme Implementation. Government of India (2016) Social Consumption - Education Survey 2014. NSS 71st Round (January 2014- June 2014), Report No. 575. National Sample Survey Organisation, New Delhi: Ministry of Statistics and Programme Implementation. Government of India (2017a) Economic Survey of India. Department of Economic Affairs, Ministry of Finance. Government of India (2017b) Report of the Working Group on Migration, January 2017, Ministry of Housing and Urban Poverty Alleviation, Government of India Smita. (2008) "Distress seasonal migration and its impact on children’s education" Create pathways to access, Research Monograph No. 2. National University of Educational Planning and Administration, New Delhi21 Srivastava Ravi, S., & Dasgupta, R. (2010). Small Steps Lead the Way. Srivastava, R., & Sutradhar, R. (2016). Labour migration to the construction sector in india and its impact on rural poverty. Indian Journal of Human Development, 10(1), 27-48. UNICEF (2017) Interstate Consultation on Migration and Children: Practice to Policy organized by Government of Maharashtra, UNICEF-Maharashtra, Indira Gandhi Institute of Development Research (IGIDR), Mumbai and Harvard T Chan School of Public Health, Harvard University, USA.22 7. TABLES & FIGURES Table 1: Distribution of Individuals Aged 15-19 years by Education and Work Status Marginal Worker Main Worker Not working Currently Attending 3,099,280 1,861,493 67,120,992 Ever Attended and Currently not Attending 6,682,121 12,224,607 16,714,852 Never Attended 2,731,669 3,617,210 6,474,229 Source: Authors’ Calculation from Census of India Tables 2011 Table 2: Age-Specific Attendance Ratio by Household Type Male Female Age 6 -14 Age 15 - 19 Age 6 -14 Age 15 - 19 Household Types Rural Self-employment in agriculture 95 66 91 61 Self-employment in non-agriculture 94 65 91 60 Regular wage/Salary earning 94 70 95 72 Casual labour in agriculture 91 48 90 43 Casual labour in non-agriculture 88 48 88 4823 Others 93 80 89 67 All Households 93 61 91 58 Urban Self Employed 94 73 94 72 Regular wage/Salary earning 95 76 96 78 Casual Labour 89 44 87 50 Others 98 88 97 83 All Households 94 70 94 71 Source: Authors’ Calculation from NSSO’s Survey of Education24 Table 3: Age-Specific Attendance Ratio State Rural Male Rural Female Urban Male Urban Female 6 -14 15 - 19 6 -14 15 - 19 6 -14 15 - 19 6 -14 15 - 19 1 Jammu & Kashmir 96 81 96 70 94 74 97 75 2 Himachal Pradesh 96 84 99 86 100 84 100 74 3 Punjab 96 62 97 66 98 81 96 78 4 Chandigarh 96 77 71 69 99 88 94 89 5 Uttaranchal 99 76 97 63 96 82 96 78 6 Haryana 96 75 98 69 93 74 88 72 7 Delhi 85 91 91 45 95 67 96 76 8 Rajasthan 92 72 86 48 93 74 88 63 9 Uttar Pradesh 89 58 85 58 87 60 84 65 10 Bihar 90 60 88 56 93 72 93 72 11 Sikkim 97 90 98 95 100 75 100 82 12 Arunachal Pradesh 93 85 91 87 98 86 88 76 13 Nagaland 92 94 98 84 95 89 98 85 14 Manipur 98 87 91 72 98 94 98 88 15 Mizoram 96 66 93 76 100 88 100 83 16 Tripura 99 76 99 55 98 72 99 7725 17 Meghalaya 95 79 95 58 100 79 82 89 18 Assam 96 64 97 60 96 71 99 69 19 West Bengal 92 49 95 62 93 67 97 68 20 Jharkhand 94 68 92 60 96 57 94 70 21 Odisha 95 46 92 41 96 56 93 55 22 Chhattisgarh 95 56 93 75 97 61 96 73 23 Madhya Pradesh 92 55 88 50 94 68 95 70 24 Gujarat 92 55 89 42 93 72 91 74 25 Daman & Diu 90 45 99 46 72 68 95 97 26 D & N Haveli 98 30 91 24 100 59 100 64 27 Maharashtra 97 64 96 59 94 73 96 78 28 Andhra Pradesh 95 69 94 50 97 78 97 71 29 Karnataka 96 62 93 53 97 67 97 66 30 Goa 100 74 100 89 100 78 100 69 31 Lakshadweep 100 58 100 89 100 78 100 54 32 Kerala 99 87 100 87 100 90 100 88 33 Tamil Nadu 99 67 99 75 100 66 100 80 34 Puducherry 100 100 100 99 98 68 100 76 35 A & N Islands 99 58 94 59 100 93 100 10026 36 Telengana 97 69 96 66 98 84 98 60 All India 93 61 91 58 94 70 94 71 Source: Authors’ Calculation from NSSO’s Survey of Education 201427 Table 4: Estimates of Number of Rural Households with a Short-term migrant and Children Aged 6-14 Years Living in Rural Households with a Short-Term Migrant Household Classification Number of Households with a Short-Term Migrant Estimates of Number of Children in Households with Short-Term Migrant Self-employment in cultivation 3,904,647 (38.7) 4,318,552 (40.3) Self-employment in livestock farming 84,292 (1) 122,053 (1) Self-employment in other agricultural activities 481,011(4.8) 318,820 (3) Self-employment in non- agricultural enterprise 1,214,885 (12) 1,203,379 (11.3) Wage/salaried employment 3,782,610 (37.5) 3,955,086 (36.9) Others (pensioners, remittance recipients, etc.) 612,685 (6) 803,453 (7.5) Total 10,080,130 (100) 10,721,343 (100) Source: Authors’ Calculation from NSSO’s Survey of Land and Livestock Holdings 201328 Table 5: Status of Current Attendance in Educational Institution among those Aged 15-19 Years in 2011- 12 Households whose Industry of Work is Construction All Households Male Female Male Female Rural Urban Rural Urban Rural Urban Rural Urban Never Attended 11.68 7.48 12.76 9.45 5.62 3.74 8.99 5.26 Dropout to Supplement Household Income 22.36 28.4 3.1 8.54 17.98 15.93 4.65 4.17 Dropout to Attend Domestic Chores 1.06 0.89 22.43 14.83 1.69 0.48 17.12 10.05 Dropout Due to Other Reasons 11.61 9.85 16.02 16 9.08 7.32 14.15 10.34 Attending 53.29 53.38 45.69 51.18 65.64 72.53 55.09 70.19 Total 100 100 100 100 100 100 100 100 Source: Authors’ Calculation from NSSO 2011-12 Survey of Employment and UnemploymentFigure 1: Age-specific Attendance Ratio in Rural Odisha Source: Calculations based on Census of India, 2011 Tables Figure 2: Age-specific Attendance Ratio in Rural Maharashtra Source: Calculations based on Census of India, 2011 Tables 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 7 8 9 10 11 12 13 14 15 16 17 18 19 Outmigration Districts Other Districts 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 7 8 9 10 11 12 13 14 15 16 17 18 19 Outmigration Districts Other Districts